rr280 - real time evaluation of health and safety ...hse health & safety executive real time...

HSEHealth & Safety

Executive

Real time evaluation of health and safetymanagement in the National Health Service

Prepared by the Occupational Health and Safety AdvisoryService (OHSAS) and the University of Aberdeen

for the Health and Safety Executive 2004

RESEARCH REPORT 280

HSEHealth & Safety

Executive

Real time evaluation of health and safetymanagement in the National Health Service

Karen J M Niven PhD MSc FFOH FIOSH RSPOccupational Health and Safety Advisory Service

(OHSAS) and University of AberdeenNavy HouseStuart Road

Rosyth Industry ParkRosyth

Fife KY11 2BJ

The design of an impact evaluation of a health and safety management system in the National HealthService (NHS) was informed by a review of relevant literature, which identified a design comprising sixmain aspects. These were a longitudinal design; inclusion of comparison groups; an intervention that wasof interest to the NHS; a participative style; multiple measurement methods and multiple indicators ofeffectiveness.

Field study data was generated using a prospective longitudinal before-and-after design with a multiplebaseline. Seven NHS Trusts participated; two of which were used as comparison groups. The interventionwas a safety management workbook, introduced only to the test group. Evaluation of the workbook impacton safety performance involved two identical phases, approximately 12 months apart. Each phasecomprised of a staff opinion questionnaire survey, based on previously validated work; and a new HSEmethodology involving analysis of accident data to derive costs, which could be linked to management rootcauses.

The most frequently encountered system failure was that of risk assessment, with planning, implementing,measuring and reviewing as the main root causes. Estimated extrapolated costs were between 0.06% and1.44% of the running costs of the NHS. Responses from the questionnaires showed significant differencesbetween the Trusts and a significant improvement in staff opinion in some safety climate dimensions.

This report and the work it describes were funded by the Health and Safety Executive (HSE) Withadditional funding/support from: Scottish Executive Health Department, Directorate of HumanResources,University of Aberdeen, Tayside Primary Care NHS Trust and Fife Primary Care NHS Trust. Itscontents, including any opinions and/or conclusions expressed, are those of the authors alone and do notnecessarily reflect HSE policy.

HSE BOOKS

ii

© Crown copyright 2004

First published 2004

ISBN 0 7176 2911 2

All rights reserved. No part of this publication may bereproduced, stored in a retrieval system, or transmitted inany form or by any means (electronic, mechanical,photocopying, recording or otherwise) without the priorwritten permission of the copyright owner.

Applications for reproduction should be made in writing to:Licensing Division, Her Majesty's Stationery Office, St Clements House, 2-16 Colegate, Norwich NR3 1BQ or by e-mail to [email protected]

iii

ACKNOWLEDGEMENTS

The research described in this report has a single author. This is potentially misleading and does little justice to the individuals and organisations without whose participation, commitment and enthusiasm this work would have been impossible. My grateful thanks and unbounded admiration for their tenacity are extended to them.

The funding bodies The Health and Safety Executive The Scottish Executive, Department of Health Tayside Primary Care NHS Trust Fife Primary Care NHS Trust

The research team Elaine Ferguson – Research Assistant Louise Webster – Research Administrative Assistant

The staff and long-suffering principal contacts within the participating Trusts Dumfries and Galloway Acute and Maternity Hospitals NHS Trust Fife Primary Care NHS Trust Highland Primary Care NHS Trust Lomond and Argyll Primary Care NHS Trust Lothian University Hospitals NHS Trust Tayside Primary Care NHS Trust Tees and North East Yorkshire NHS Trust

Other sources of advice and encouragement John Cairns – University of Aberdeen, Health Economics Research Unit John Cherrie – University of Aberdeen, Department of Environmental and Occupational Medicine Liz Archibald – Cormack Consulting Amanda Ridings - Originate Alistair Cheyne - Loughborough University, Centre for Hazard and Risk Management Andrew Turner – Focused Marketing

v

CONTENTS

ACKNOWLEDGEMENTS ................................................................................................iii

EXECUTIVE SUMMARY .............................................................................................. ix

INTRODUCTION ............................................................................................................xiii

1 LITERATURE REVIEW .........................................................................................1

1.1 Literature review methodology ……………………………………….......11.2 Findings ....................................................................................................2

1.2.1 Literature review objective 1 ...........................................................2 1.2.2 Literature review objective 2 ...........................................................4 1.2.3 Literature review objective 3 ...........................................................5

2 METHODS ...............................................................................................................7

2.1 Practical factors relevant to the research design ......................................72.1.1 Variability between NHS Trusts.......................................................7

2.1.2 Recruitment of research participants ...............................................7 2.1.3 Relationship between researchers and Trusts....................................82.2 Evaluation methodology design (Research objective 1)............................9

2.2.1 Objective measurement component ..................................................9 2.2.2 Subjective measurement component............................................... 11

2.3 Intervention design (Research objective 2) ............................................. 14 2.3.1 Method of introduction ................................................................ 15

2.4 Research design to measure change in health and safety performance (Research objective 3) ........................................................ 16

2.4.1 Method of monitoring workbook use.............................................. 17 2.4.2 Study composition ......................................................................... 18

2.5 Data analysis strategy .............................................................................. 21 2.5.1 Incident data ................................................................................. 21 2.5.2 Questionnaire data ........................................................................ 22 2.5.3 Workbook monitoring data ............................................................ 23

3 RESULTS ............................................................................................................... 24

3.1 Trust descriptive/observational results.................................................... 243.2 General incident findings......................................................................... 26

3.2.1 “Special” category........................................................................ 31 3.3 Risk control system failures and root causes of incidents....................... 37

3.3.1 Risk control system failures (HSE level 2)...................................... 38 3.3.2 Management root causes (HSE level 1).......................................... 41

3.4 Cost of incidents....................................................................................... 44 3.4.1 Staff/patient incidents .................................................................... 46

3.4.2 Special category ............................................................................ 48

vi

3.4.3 Management root causes of incidents ............................................ 49 3.4.4 Categories, factors and types of incidents ...................................... 53

3.5 Questionnaire data................................................................................... 59 3.5.1 Response rates .............................................................................. 60 3.5.2 Questionnaire confirmatory factor analysis ................................... 62 3.5.3 Dimension score responses............................................................ 63

3.5.4 Analysis of responses by Trust ....................................................... 64 3.5.5 Analysis of responses by job category ............................................ 65 3.5.6 Analysis of responses by individual members of staff ..................... 66

3.5.7 Questionnaire respondents’ additional comments .......................... 67 3.6 Results from workbook usage monitoring (Project phase 3).................. 67

3.6.1 Telephone survey........................................................................... 67 3.6.2 Visual inspections.......................................................................... 71

3.7 Triangulation of results ........................................................................... 72 3.8 Summary of findings................................................................................ 76

3.8.1 General findings............................................................................ 76 3.8.2 Special category ............................................................................ 77 3.8.3 Identification of risk control system failures and root causes of incidents.................................................................................... 77

3.8.4 Costs of incidents .......................................................................... 77 3.8.5 Staff opinion survey ....................................................................... 78 3.8.6 Monitoring and audit of the use of the management system workbook ...................................................................................... 79

3.8.7 Triangulation of results ................................................................. 79

4 DISCUSSION ......................................................................................................... 80

4.1 Health and safety management performance evaluation methodology.. 80 4.1.1 Performance evaluation based on the use of reported incidents ..... 80

4.1.2 Use of the HSE root cause analysis methodology........................... 82 4.1.3 Investigator bias and its implications for the training and development of safety advisers ........................................................................... 85

4.1.4 Costs and economic aspects .......................................................... 86 4.1.5 Health and safety culture............................................................... 89

4.1.6 Triangulation of health and safety measurement methods .............. 90 4.2 Health and safety management intervention .......................................... 91

4.2.1 The degree of use of the workbook................................................. 92 4.2.2 The time interval between the benchmarking phases ...................... 92

4.3 Health and safety management performance change measurement methodology ............................................................................................. 94

4.3.1 Intervention objectives................................................................... 94 4.3.2 Conceptual basis ........................................................................... 94 4.3.3 Research format and design........................................................... 94

4.3.4 Threats to external validity ............................................................ 95 4.3.5 Threats to internal validity ............................................................ 95

4.3.6 Outcome measurement .................................................................. 97 4.3.7 Other limitations ........................................................................... 97

4.4 Assessment of the degree to which the research aim and objectives have been achieved ........................................................................................... 97

vii

4.5 Summary of conclusions and recommendations ..................................... 98 4.5.1 Health and safety management performance evaluation methodology.................................................................................. 98 4.5.2 Health and safety management intervention................................... 99 4.5.3 Health and safety management performance change measurement

methodology................................................................................ 100

APPENDICES

APPENDIX 1 HSE ROOT CAUSE ANALYSIS METHODOLOGY ......................... 101 APPENDIX 2 COST INCLUSIONS AND EXCLUSIONS.......................................... 116 APPENDIX 3 COVERING LETTERS, QUESTIONNAIRE AND

QUALITY CONTROL SYSTEM FOR INPUTTING QUESTIONNAIRE DATA.................................................................... 118 APPENDIX 4 MANAGEMENT SYSTEM WORKBOOK.......................................... 125 APPENDIX 5 PROTOCOL FOR BRIEFING OF WORKBOOK HOLDERS ............................................................................................. 158 APPENDIX 6 PROTOCOL FOR FOLLOW-UP OF WORKBOOK HOLDERS ............................................................................................. 163 APPENDIX 7 INCIDENT DESCRIPTIVE DATA ...................................................... 166 APPENDIX 8 SPECIAL CATEGORY INCIDENT DESCRIPTIVE DATA ............. 186 APPENDIX 9 INDIVIDUAL TRUSTS BREAKDOWN OF TELEPHONE INTERVIEW RESPONSES ................................................................. 208

REFERENCES ................................................................................................... 213

BIBLIOGRAPHY ................................................................................................... 218

ix

EXECUTIVE SUMMARY

A high priority for the NHS is ensuring patient safety and minimising staff absence due to incidents and work-related illness. The research aim was to evaluate whether it was possible to reliably measure change in health and safety performance when a formal health and safety intervention was introduced to the UK healthcare sector. The aim was supported by three research objectives, all of which were informed by a review of relevant literature.

The first objective was to design a health and safety performance measurement methodology. This included a new method, developed by the HSE, for assessing costs, and health and safety management system root causes of reported incidents that met a predetermined set of inclusion criteria. This quantitative data was supported by a qualitative set via a staff opinion questionnaire survey, based on previously validated work, administered to a stratified 10% sample of staff from participating Trusts.

The second objective was to evaluate the use of a health and safety management system workbook, developed by the author for the research. Use of this was intended to improve health and safety management performance. It was issued to selected staff from participating Trusts, who were asked to use it for approximately a year. Workbook users were monitored during this period by telephone interviews and visual inspection of workbooks. Some Trusts already had a workbook-based health and safety management system in place. They were used as a comparison group and did not receive the workbook or the follow-up monitoring.

The third objective was to design and implement a field study to measure change in health and safety performance. Seven NHS Trusts participated in this research; two being the comparison group. The method developed to measure health and safety performance, was used to establish a baseline of health and safety performance in all Trusts (phase 1), before the workbook was introduced to the test group. At the end of the research, health and safety performance was again measured (phase 2). The results, including the monitoring data on level of use of the workbook, were examined to establish whether change in health and safety performance could be detected. Field study data was therefore generated using a design which included: use of a before-and-after (longitudinal) design; use of comparison groups; use of an intervention that was of interest to participating NHS Trusts; use of a participative style which involved the participating Trusts; use of multiple measurement methods and multiple indicators of effectiveness.

The results from the performance measurement methods (to support the first objective) showed that the risk control system of risk assessment was the major source of management root causes in both phases. These root causes were associated with planning, implementing, measuring and reviewing of the management system. Therefore improvements in the planning and implementation of risk management should be a priority for action within the NHS. Where management root causes could be assigned, incidents involving staff were nearly three times more frequent and represented approximately four times the cost than for patients. The total cost assigned to accidental injury was an order of magnitude higher than any other category, while staff absence and replacement labour represented approximately 75% of the total incident factor costs. Also person to person assault, slips and trips, and patient lifting/handling represented 75% of the incident type costs.

x

For the questionnaire, data pooling of data from individual Trusts was avoided because their results were statistically significantly different from each other (P<0.05). There were significant correlations (P<0.05) between the rank order of some staff opinion questionnaire dimensions, captured incidents and workbook use. Higher reported use of the workbook was significantly associated with both higher opinion on the questionnaire dimension of working environment and lower captured incident rate. There was no correlation found between reported incident rate and the other performance measurement methods.

Use of the management workbook was low. For example, use of workbooks decreased significantly over time (P<0.01), with approximately one third of workbook holders having dropped out by the end of the monitoring period and overall perceived workbook usefulness decreased significantly over time (P<0.001). Nevertheless all Trusts reported a significant increase in the use of particular workbook sections (P<0.001) but low implementation was confirmed by a visual inspection of a 10% sample of workbooks.

There was little evidence of change in health and safety performance between the two benchmarking phases. The exception to this was in the responses from the questionnaires, which showed significant differences between all Trusts (P<0.05) and a significant improvement (P<0.05) in staff opinion in some safety climate dimensions for two Trusts. A significantly lower (P<0.05) opinion on health and safety was noted in the medical and dental staff in one Trust. This effect had disappeared by the second project phase and was thought to be linked to the annual intake of NHS junior doctors, which coincided with the first data collection phase in the Trust. The mean number of reported incidents rose by 24% in the second phase, which was marginally significant (P=0.06). There was no significant difference between reported incidents the test and comparison Trusts (P=0.49 for phase 1 and P=0.58 for phase 2).

In the discussion, the wide variation in incident reporting rates was thought to be more likely to be related to reporting culture rather than health and safety performance. However, the smaller group of captured incidents was thought to offer a potentially useful alternative performance indicator and raised questions about the cost effectiveness of collecting data on every reported incident.

A potential limiting factor was inter-investigator bias in benchmarking root causes of incidents, so far as learning lessons from incidents was concerned. It was concluded that there exists a need for improvements in consistency and in incident investigation skills for safety practitioners and others with health and safety management duties as well as a need for a common standard for incident reporting and costing. It was concluded that incident reporting in the NHS has largely been used as a performance indicator and that the process of learning from adverse events has been largely underdeveloped.

Management root causes could not be identified for approximately half of the total incidents reported (defined as a “special” category of incident). Most (67%) were not thought to have been preventable. The potential for many of this type of incident to occur in the NHS should be investigated and debated further to establish whether this phenomenon is largely confined to healthcare and how to interpret their existence in the context of health and safety management performance.

A number of issues were not captured by either the HSE incident analysis method or the Trust incident reporting systems. For future use these should be refined to improve their ability to

xi

effectively capture and manage the risks from cases of occupational ill health; hospital acquired infections and other patient-related incidents; and the potential for impact by civil litigation.

The costing strategy used in the method was conservative. Estimated extrapolated national costs were between 0.06% and 1.44% of the running costs of the NHS, which were low compared to previously published data. The potential future use of such a conservative costing strategy is discussed. So far as the evaluation of the workbook was concerned, it failed to produce a significant effect on health and safety performance. It was not possible to conclude that it produced an effect that was too weak to be detected because, had it been more fully used, there might have been a detectable effect. Therefore, with further refinement it could be a useful tool within a health and safety management system and it was recommended that, following refinement, it be made freely available. It was apparent that a simple approach to health and safety management, which did not ensure the workbook was used and implemented, would not produce significant change in performance. It was recommended that the issue of how to improve implementation and use should be addressed, possibly via a multi-factorial strategy.

In the performance change measurement field trial it was possible to conclude that, despite the fact that minimal change in performance was detected, the measurement methodology had the ability to detect change in health and safety performance. For example the questionnaire could detect significant change in staff opinion and therefore represented a potentially useful measure of culture change. The absence of significant change in questionnaire responses from five of the seven Trusts may represent evidence that the underlying health and safety culture in those Trusts was relatively stable. The key message of the research was that there exist potentially effective methods to evaluate health and safety performance but that to make a lasting effect on change in performance represents a greater challenge, possibly requiring a less simple, multi-factorial strategy, which takes account of both systems and behavioural approach. In addition to the areas identified above, the main area for further work was to seek further cause and effect evidence for the major finding of the research, that is the significant correlation between the use of the workbook and lower captured incident rates and higher staff opinion for some questionnaire dimensions.

xiii

INTRODUCTION

The United Kingdom (UK) National Health Service (NHS) employs more than 1 million people at a cost of around £17.7 billion for salaries and £1.2 billion for agency staff to cover for vacancies and staff absences(1). There are chronic staff shortages caused by vacancies (approximately 3.5% across the whole sector) and staff absences (in 2001/2002 this was running at an average of 4.9% across all NHS Trusts)(1). Staff incidents and work-related illness are known to be major factors in these staff absences(1).

So far as patients are concerned, the Department of Health (DoH) has concluded that in the past the healthcare sector has underestimated the scale of unintended harm or injury to patients and that the issue of patient safety has become a high priority for improvement(2).

There is therefore a need to examine more closely factors that might lead to improvements in these areas and how these improvements might be measured and evaluated. This introductory Chapter therefore comprises of three parts:

Part 1: Explores the background to health and safety management in the context of the strategic policy approaches of both the European Union (EU) and UK. The UK tactical approach to health and safety management and how this applies to the healthcare sector is then described; Part 2: Considers measures currently used to describe current UK health and safety performance, explores how this performance might be improved and how change might be evaluated; Part 3: Sets out the research aim and objectives.

BACKGROUND TO POLICIES ON HEALTH AND SAFETY MANAGEMENT

EU and UK Current health and safety policy in the UK has been largely influenced since 1974 by the Health and Safety at Work etc. Act (HASAWA)(3). Since the late 1980’s the EU “Framework Directive” (89/391/EEC) brought about the enabling of the UK Management of Health and Safety at Work Regulations 1992(4) and 1999(5). These regulations made explicit the implied requirement in HASAWA to manage health and safety risks. It required risks to be assessed and controlled using an approach that was integrated with other business management activities. UK policy on the management of health and safety has therefore tended to embrace the use of workplace risk assessment as an instrument through which the technical aspects of risk are evaluated(6).

The healthcare sector The DoH and the Scottish Executive (SE) are together responsible for overseeing occupational health and safety strategy in the NHS in England, Wales and Scotland. They have a multi-factorial strategy, which involves targets aimed at patients, employees, the general public and employers(7, 8). So far as employees are concerned, these targets center on reductions in incidents and sickness absence.

xiv

In the NHS an “incident” had a broad meaning covering a number of different types of event. The most widely used definition of an incident in the NHS at the time of the research was1:

�� Accidents: defined as any incident, no matter how small, which did or could have adversely affected any person, not caused deliberately (e.g. some acts of violence or fire);

�� Violence/abuse/harassment: defined as any incident involving verbal abuse, unsociable behaviour, racial or sexual harassment or physical assault, whether or not injury was a result;

�� Clinical incidents: defined as any incident related to patient treatment or care which did, or could have, resulted in an adverse outcome (e.g. treatment error, medical equipment failure);

�� Ill health: defined as any case of known or suspected work or environment related ill health (e.g. infection, dermatitis, asthma);

�� Fire and security: defined as any incident, no matter how small, involving fire or fire warning systems (including false alarms);

�� Vehicle incidents: defined as any incident involving a vehicle, excluding vandalism or theft (regarded as a security incident);

�� Complaints: defined as any adverse formal or written communication from a person or organisation which required a response;

�� Other incidents: defined as near misses (an incident which had the potential for serious consequences); accidental property damage or loss; environmental incidents (e.g. accidental discharge to drains or the atmosphere); or food safety incidents.

In 1996 the National Audit Office (NAO) published the results of an audit of thirty acute Trusts in England(9). They found that health and safety management systems were poorly developed and characterised by a generalised lack of standardisation in the recording, costing and definitions of accidents.

Since then the healthcare sector has been encouraged to improve its health and safety management. For example, the Health and Safety Commission (HSC) Strategic Plan 2001-2004 for the UK was published in October 2001(10). As one of the eight priority programmes within the strategic plan, the health service has received, during 2001/02, targeted Health and Safety Executive (HSE) inspections on the management of: violence; manual handling; and slips/trips(11). These inspections included arrangements for both staff and patients. Enforcement action in England and Wales increased by 225% (from 8 to 26 improvement notices) and was thought by HSC to be a causal factor in a slow but steady decline in the rate of incidents reported to HSE(11). Conversely the NAO reported that in 2001/02 the overall number of recorded incidents increased by 24%(1). Reasons for these differences are unclear but may represent an improving culture of incident reporting in the NHS with fewer incidents meeting the HSE reporting criteria(12). Nevertheless these conflicting data should be interpreted with caution until reasons can be established with more certainty.

Two DoH publications, “An organisation with a memory”(13), and “Building a safer NHS for patients”(2) set out to develop a system in England and Wales for adverse clinical incident reporting and establishing a system for reducing the risk of unintended harm to patients. This was intended to establish a mechanism for learning from adverse events. The establishment of the National Patient Safety Association (NPSA) in July 2001 activated the means for implementation of a third 1 Based on that incorporated in the “SAFECODE-Plus” risk and quality management suite, widely used within the NHS (http//www.safecode.co.uk). Throughout this report the above definition of an “incident” will be used unless otherwise stated.

xv

document “Doing less harm”(14). This document set out key implementation requirements for healthcare providers, such as incident investigation and root cause analysis as well as implementation and monitoring of improvement strategies.

So far as the staff of the NHS are concerned, a set of standards for the effective management of occupational health and safety services has been published for both England and Wales(15) and Scotland(8). This has been combined with financial incentives via the insurance schemes covering the NHS in England and Wales (Clinical Negligence Scheme for Trusts(16) (CNST)) and Scotland (Clinical Negligence and Other Risks Indemnity Scheme(17) (CNORIS)). These schemes include basic standards for health and safety management, broadly based on HSE guidance(18).

The above initiatives have therefore laid a foundation for the development of management systems for occupational health and safety risks for staff and patients in the NHS.

HEALTH AND SAFETY PERFORMANCE IN THE UK

The health and safety system in the UK is well established, integrated across industry sectors, based on tripartism and social involvement(19). A major tactic of the HSE to achieve progress with its market forces policy has been to strongly advocate that positive health and safety management represents good business sense.

Their “Good health is good business” (GHGB) campaign ran for five years between 1996 and 2001 with the aim of increasing awareness of occupational health and safety in the workplace and improving employers’ competence in managing health risks in the workplace. The campaign was found to have been associated with significant improvements in employer attitude and the quality of their approaches to risk management(20). The evaluation had an acknowledged selection bias, in that organisations with a positive predisposition towards health and safety were more likely to sign up for the campaign. Indeed some organisations were not convinced. For example Monnery(21) questioned the concept that incident management was always good for business and suggested that other factors, such as the moral and legal aspects of health and safety and potential loss of reputation were better motivational factors for line managers.

However, proposals for reform of the arrangements for employers’ liability insurance(22), may help to redress the balance in the distribution of the costs of health and safety failures. This proposal is one of ten strategic aims of the “Revitalising health and safety” strategy (RHS), which contains Government and HSC targets for improving health and safety performance over the ten years to 2010(22). The targets seek to:

�� Reduce the number of working days lost per 100,000 workers from work-related injuries and ill-health by 30% by 2010;

�� Reduce the incidence rate of cases of work-related ill-health by 20% by 2010; �� Reduce the incidence rate of fatalities and major injuries by 10% by 2010.

This approach has been supplemented by the long-term occupational health strategy for England, Scotland and Wales “Securing health together” (SH2) (23). This seeks to establish evidence-based best practice in five key areas (compliance, continuous improvement, knowledge, skills and support).

xvi

An assessment of the likelihood of success of reforms to employers liability insurance(24), based on the experiences of other countries, has concluded that employers could be motivated to improve occupational health and safety via insurance premium incentives. The financial penalty for non-compliance however would require to be high, estimated at approximately 3% of payroll costs(24).

STRATEGIES FOR IMPROVING HEALTH AND SAFETY PERFORMANCE

Within the UK there is an abundance of readily available guidance on the practicalities of achieving robust health and safety management systems(25). The cornerstone is “Successful health and safety management” (HSG65), published by HSE in 1991(25) with the second edition in 1997(18). It has come to be a “best seller” for HSE, with its lay language and readily remembered mnemonic POPIMAR to signpost the seven main steps of a closed-loop management system (policy, organising, planning, implementing, measuring, audit & review).

HSE have also issued guidance on behavioural based safety(26), recommended as a method that can affect the number of incidents and thereby positively influence health and safety management performance.

A British Standard (BS8800:1996)(27) on occupational health and safety management systems has been developed. Based on HSG65, it currently has the status of voluntary guidance and cannot therefore be used for certification purposes. However, the subsequent standard (OHSAS: 18001 detailing the specification(28) and OHSAS: 18002 giving guidelines for implementation(29)) has the potential to form the basis of an assessment and certification scheme for a UK health and safety management system. BS8800 and OHSAS 18001/18002 are fundamentally the same in terms of structure and requirements. Presently, neither is recognised as either an UK or International standard despite the fact that all major certification bodies are certifying to OHSAS 18001 using OHSAS 18002 to provide generic guidance. Whether a formal certification scheme will result or not remains unclear, given the current UK national policy to allow market forces to bring about improvement in workplace health and safety(30).

The Scottish Executive Department of Health (SEDoH) were interested in the improvement of health and safety management within the healthcare sector and funded the author to develop a health and safety management tool to assist with this process.

This workbook was designed by the author to assist line managers with their health and safety management responsibilities. Its' use is described in Chapter 2.

SEDoH were also interested in whether use of this tool could help to improve health and safety management performance in the healthcare sector and commissioned the author to investigate this. This required a method for the measurement and evaluation of change that might occur as a result.

STRATEGIES FOR EVALUATING CHANGE IN HEALTH AND SAFETY PERFORMANCE

The need for more research to help inform policy and strategy on improving workplace health and safety has been consistently identified (e.g.(23, 31-33)).

In the UK much effort has been directed at evaluating costs of accidents and ill health both to employers and the economy as a whole (9, 34-38). The purpose of this effort has generally been to

xvii

encourage performance improvement, (i.e. used as an incentive rather than measurement tool). Cost estimates have ranged from between 1% and 5% of annual running costs for a hospital, incurred by that hospital(34, 36), to between 1% and 2% of UK total Gross Domestic Product (GDP) for the economy(37, 38). The range of estimates was due to differences in approaches and difficulties in measuring cost, but nevertheless suggested that the costs of accidents and ill health are significant to both employers and the economy as a whole. Auditing has also been extensively used in health and safety performance evaluations(39-41). Its main uses have been to:

�� Identify strengths and weaknesses of occupational health and safety management systems; �� Measure occupational health and safety performance; �� Establish the extent of legal compliance; �� Identify and define areas for improvement; �� Compare organisational performance with an established standard.

Gay and New(42) described the view of HSE on the use of auditing as a means of formally assessing health and safety management systems. The authors regarded auditing as an essential regular part of a management system and identified that it had further potential to be linked to costs and causes of accidents. The basis for this was the underlying paradigm that where audit identified weaknesses in health and safety management systems, there were more accidents and higher costs. Conversely, where a component of a management system was managed effectively there could be expected to be fewer accidents and lower costs. Therefore, if causes of accidents were tracked back to their root cause in the health and safety management system, these should correlate with health and safety management audit findings.

HSE subsequently commissioned the development of a new health and safety management performance measurement tool, based around the potential linkages between audit, costing and root cause analysis of incidents(34, 43).

To do this HSE identified a three-level model to define health and safety management (described below):

�� Workplace precautions (“Level 3 arrangements”) to protect workers health and safety (e.g. ensuring appropriate machinery guarding or the use of safety helmets on construction sites). These were regarded as part of;

�� Risk control systems (“Level 2 arrangements”) (e.g. systems for management of hazardous substances or fire precautions. These were, in turn, part of;

�� Management arrangements (“Level 1 arrangements”) for health and safety (e.g. policy or planning)(18).

In addition, HSE commissioned a literature review(43) of incident investigative tools which could be used for root cause analysis (defined as the level 1 management system causes of incidents). The resulting report concluded that none of the techniques available at the time adequately addressed every stage in the investigation process and that those models that attempted to capture all the relevant points did so at the expense of comprehensibility. The tool with most potential was identified as Management Oversight Risk Tree (MORT). It took the form of a generic fault tree analysis of an incident, which required the identification of factors that must have been present for it

xviii

to occur. However, the full fault tree was very large and it was judged to be impractical for all but high hazard industries, such as those in the nuclear and petrochemical sectors.

As a result of this review, HSE then commissioned the development of a new root cause analysis method for industry in general and linked it to a costing methodology, which was a simplified version from that originally published in 1993(35) and 1997(36). Finally, a pilot study in a hospital setting was conducted by HSE and the author(34). The purpose of this was to evaluate the ability of the method to be used as an investigative tool, to consistently identify costs and rates of incidents that could be mapped onto safety management failures that were consistent with the elements of the HSE model(18). The pilot study also included a health and safety management systems audit, conducted by HSE inspectors. The results of the audit and the root cause analysis were broadly found to compare well. For example, where the HSE audit identified aspects of the health and safety management system to have less than legal compliance (such as in the planning and implementing aspects of the health and safety management system), the root cause analyses of incidents tended to identify similar weaknesses (by virtue of more incidents and higher costs). Conversely, areas identified in the audit as being well managed (such as policy and communication) had fewer incidents and lower costs.

It was concluded that, in practice, an audit component was an unnecessary component because the pilot study had provided evidence that the outcome of the root cause analysis was similar to that obtained by audit. Additionally, the approach was thought to offer the potential for greater sensitivity and objectivity than auditing alone, as well as sharing with audit the potential advantages of use in benchmarking, baseline or before-and-after measurements, thus offering the potential for an effective novel method for measuring change in performance.

The HSE method had not been published because of the need for further validation to establish more objective field data on the performance of the method. HSE commissioned the author to commence the validation process by including the method in the research described in this report.

RESEARCH AIM AND OBJECTIVES

This Chapter has highlighted questions on methods for defining, evaluating and improving health and safety performance.

The aim of this research is therefore to evaluate whether it is possible to reliably measure change in health and safety performance when a formal health and safety intervention is introduced to the healthcare sector. This aim will be achieved by setting the following three research objectives:

Objective 1: The design of a methodology for evaluation of health and safety performance. This design will include the HSE methodology described above but will also be supported by an in-depth critical literature review to assess whether the HSE methodology would benefit from additional components; Objective 2: The evaluation of the strengths and weaknesses of the use of the health and safety management workbook-based intervention tool for the improvement of health and safety performance. This will also be supported by the literature review; Objective 3: The design of a research methodology to measure change in health and safety performance. This will also be supported by the literature review and will also include a field study within the healthcare sector to test the methodology.

1

1 LITERATURE REVIEW

The definition of a safety intervention that was identified as most closely suited to the research aim and objectives was:

“An attempt to change how things are done in order to improve safety”(44)

The main objective of the literature review was to identify suitable measures for evaluating health and safety performance; to evaluate the design of interventions to improve health and safety management performance; and to identify a suitable design for a field study to measure change in health and safety performance.

This Chapter first summarises the review methodology, then describes the findings. The last section forms conclusions about the implications for research designs most suited to intervention studies in the field of occupational health and safety management.

1.1 LITERATURE REVIEW METHODOLOGY

As observed by others(45-50) an issue for researchers in occupational health and safety management is the breadth and depth of the literature, which is spread across several disciplines. Therefore, the search was carried out across professional and scientific disciplines, and included several different data sources and literature collections (listed below).

The review was conducted during parts of 2002 and 2003. Relevant studies were retrieved, using a systematic approach. Searches were restricted to articles in English, and conducted using combinations of the following search words:

Accident, assessment, change, evaluation, impact, incident, intervention, management, measurement, meta-analysis, prevention, review, root-cause and safety.

The following sources of information were used:

�� Biomedical - BioMedNet; Cambridge Scientific Abstracts (CSA); Cochrane Library; Cumulative Index to Nursing and Allied Health (CINAHL) Database; Internet Database Service; MEDLINE; NeLH; OVID.

�� General - Emerald Fulltext; Emerald Management Reviews; Ingenta Services; NESLI (National Electronic Site Licence Initiative); ZETOC –BL.

�� Government - Canada – EU Cooperation on Workplace Safety & Health (www.eu-ccohs.org/); Department of Health (http://www.doh.gov.uk); European Agency for Safety & Health at Work (http://europe.osha.eu.int/); Health & Safety Executive (http://www.hse.gov.uk/); International Labour Organisation (ILO) (http://www.ilo.org/); National Institute for Occupational Safety & Health (NIOSH) (http://www.cdc.gov/niosh/homepage.html); Scottish Executive Publications online.

�� Health & safety - Directory of Occupational Health and Environmental Hygiene Links (http://www.agius.com/hew/links); Embase (covering occupational health, environmental health and ergonomics databases); International Occupational Safety and Health Information Centre (CIS); The Institution of Occupational Safety & Health (http://www.iosh.co.uk).

2

�� Life sciences - Edina CAB Abstracts; ISI Web of Science Citation databases. �� Science & engineering - EDINA Compendex (indexes to engineering articles and conference

proceedings); European Network for Process Safety (SAFETYNET)(http://www.safetynet.de/); Science Direct.

�� Social science - Applied Social Sciences Index and Abstracts (ASSIA); BIDS International Bibliography of the Social Sciences (IBSS); PsychINFO.

Additional information was obtained by checking author names and key publications via citation indices.

1.2 FINDINGS

A noticeable general feature was that, despite regular overlapping of the publication years of the articles examined, few of the published reviews evaluated the same studies. For example Livingston et al(43) did not cite the same work on accident causation theories published earlier by a study group commissioned by the Advisory Committee on the Safety of Nuclear Installations (ACSNI)(48) despite having clear objectives to review occupational health and safety-related intervention studies. Reasons for this are unclear, but one possible cause could have been the multidisciplinary nature of the subject area and the breadth and depth of the literature.

1.2.1 Literature review objective 1 (To identify suitable measures for evaluating health and safety performance)

Four main approaches to health and safety performance evaluation were identified. They were: (a) Methods that could be used in an objective manner, such as those based on audit or incident

data; (b) Subjective methods such as safety culture/climate evaluations; (c) Economic evaluations; (d) Methods that used more than one type of method.

(a) Performance evaluation methods that could be used in an objective manner. It was intended to use the method developed by HSE, which was based on the analysis of incident data. There was a large literature base identified on the use of incident data to evaluate health and safety performance. There was also a large literature on methods that were based on the use of audit.

So far as limitations to using incident investigation as a performance indicator was concerned, a major potential source of bias was the degree of intra- or inter- investigator consistency (47) with an extensive literature base from the 1980s and 1990s, supporting the theory that both experts and novices exhibited systematic biases in their health and safety assessments, with strong evidence that experts may have become over-confident. It was clear that evaluation methods that used techniques of incident investigation or workplace inspection should attempt to control for sources of inter- and intra-investigator bias.

Another significant limitation of the use of incident data as a performance measurement method was the potential for differences in incident reporting conventions between and within organisations. This could be further confounded by the possibility that, as staff became more aware of health and safety, they might tend to report more incidents(51). This effect could have a

3

profound confounding effect on performance evaluation. The main method to control for this risk would be to use more than one evaluation method, preferably linked via triangulation. None of the identified methods attempted to link analysis of incidents to the health and safety management system of an organisation, a finding also noted by Hale et al(52). Within the NHS this was thought to be a negative feature since there were strong drivers to ensure robust health and safety management systems were developed(1, 8, 15, 53, 54). An evaluation method that was linked to the health and safety management system would also be more likely to be understood by staff working in the NHS. Therefore the HSE method was the only method, which could be used for this purpose.

The use of audit has been well established since the 1970s as a means to evaluate health and safety management performance(51, 55). Using audit to evaluate health and safety management performance was identified as a potentially powerful method. This was also acknowledged by HSE who ensured that the pilot work for their method was validated against an HSE inspection audit(34). This study found good agreement with the findings of the HSE root cause analysis method and the audit. This meant that resources could be devoted to the root cause investigation of incidents. Had this finding not been present it would have been necessary to have considered the inclusion of an audit component as part of the evaluation.

(b) Subjective performance evaluation methods. Guldenmund(56) and Hale and Hovden(57) have reviewed the theory and research base on the nature of safety culture and concluded that the scientific approach was immature given that there was no single accepted model or definition of safety climate or safety culture.

Given there was no prerequisite by the sponsoring bodies for the research project for a subjective performance evaluation method, the literature search therefore sought methods that were:

�� Fully developed and ready for use. This was because the research objectives were not intended to include the development and validation of a qualitative method;

�� Applicable to a range of occupational groups across a diverse industry such as the NHS; �� Applicable to general health and safety management rather than a specific hazard or group of

hazards.

The method that offered most potential was that of Cox and Cheyne(58) who published a nine-dimension methodology for assessing general health and safety management culture in all staff groups in the offshore industry (the Loughborough Safety Climate Assessment Toolkit). The method was based on a systems approach to organisational culture, which had been refined using focus groups, factor analysis, and field trials. The questionnaire, along with a “toolkit” for use in data analysis were freely available via the Internet(59). This was assessed as potentially adaptable for use as an evaluation method in the field of health and safety performance evaluation in the NHS. Although developed for the offshore industry its question set was sufficiently generic to offer potential for use in the NHS, subject to confirmatory factor analysis, such as that conducted by Brown and Holmes(60) and Dedobbeleer and Beland(61).

(c) Economic Evaluations. The author has reviewed economic evaluations in the field of health and safety in healthcare(62). It was concluded that including a costing aspect in the research design, with other performance evaluation indicators was regarded as potentially useful. Justification for this would be to

4

investigate whether cost could be linked to other performance evaluation methods. This was a potential use of costing data that has not been widely explored, except by HSE and the author(34).

(d) The use of more than one method. Many authors have advocated that more than one method should be used to measure health and safety performance (e.g. (56, 58, 63-73)). Guldenmund(56) identified that few studies have attempted to establish correlation (“triangulation”) between safety performance measures (such as accident/incident data or safety audits) and safety culture or climate assessments, and called for these relationships to be explored in more detail.

HSE linked their root cause analysis approach with previously published guidance on health and safety management(74) and the costs of incidents(35). Their new methodology therefore linked incident investigation, root cause analysis and costing(34). This method, however, did not include any aspect of measurement of safety climate. If this were included the potential for triangulation of data would be greatly strengthened.

1.2.2 Literature review objective 2 (To evaluate the design of an intervention to improve health and safety management performance)

To fulfill the project research aim required a formal health and safety intervention to be introduced into the NHS. For this to be realistically achievable across the whole sector it was regarded as needing to possess the following features:

�� Acceptable to NHS Trusts, otherwise willingness to participate in the research would be unlikely to be secured;

�� Able to be implemented within current resources (i.e. cost-effective). This was because cost and spending targets and priorities tended to be set at national level and Trusts had comparatively restricted freedom to act outwith these targets;

�� Generic, so as to be applicable across the wide variety of types of services and staff groups within the NHS;

�� In keeping with other health and safety drivers within the NHS (such as from HSC(10, 11), DoH(2, 75) and SE(8)).

The intervention workbook2 used for the research was developed to ensure that it met the above criteria. In practice this meant taking a general systems approach, which had a long history in the NHS.

Nevertheless, it was acknowledged that there were limitations to a systems-based approach of this type. These risks were:

�� The intervention might be overly generic to make an impact on health and safety performance at local departmental level. Although an iterative approach may have been desirable for the research in this report it was regarded as insufficiently generic for application within the NHS as a whole. The risk was addressed by ensuring that examples, relevant to the NHS were included in the workbook and local conventions and terminology were used.

2 The workbook is described in Chapter 3.

5

�� The intervention might not be sufficiently simple to be used intuitively by line managers as a relatively “stand alone” resource. It was therefore regarded as important to use a combined approach, using a written guidance/workbook approach supplemented by education and training as a cost-effective delivery mechanism;

�� The intervention might not produce an effect strong enough to be detected. To address this issue the workbook was aimed at line managers who had acknowledged legal responsibilities for health and safety management. Additionally it was written in a non-technical language and Trusts were encouraged to ensure that managers appreciate the need to use it regularly;

�� The effectiveness of the method might be reduced because it did not make use of health and safety management programmes based on psychological or behavioural models. Its relatively under-developed nature meant that this approach was, at the time, insufficiently generic for use as the sole approach in a general intervention.

Therefore, on balance, a systems-based approach(57) based on currently accepted guidance on health and safety management(18) was most likely to harmonise with other NHS drivers and therefore represent a potentially effective tool for the NHS

1.2.3 Literature review objective 3 (To identify a suitable design for a research study to measure change in health and safety performance)

The purpose of a research design has been defined as ensuring that there are measures in place to organise research activity, including the collection of data, in ways that are most likely to achieve the research aims(68). Therefore the overarching strategy for the literature review was to search for research designs that had the potential to do this.

Meta-analysis was known to be a powerful method for assessing the validity of research by statistically combining results of comparable studies(76-78). Given their potential for identifying strong research designs, meta-analyses and systematic reviews in occupational health and safety evaluation and intervention studies were specifically sought. Only four meta-analyses of performance change evaluation (45, 79-81) and four systematic reviews of intervention studies (46, 50, 73, 82) were identified. This was thought to be a low incidence of these types of publication, particularly considering the wide-ranging multidisciplinary professional and scientific basis of the search strategy.

A possible explanation was that the primary research base on which the reviews were based was insufficiently robust to support this type of analysis. For example, Shannon et al(73) were unable to make quantitative comparisons between ten studies. This was supported by Rivara and Thompson(83) who attempted to conduct a meta-analysis to assess the effectiveness of different strategies to prevent falls from heights in the construction industry. Their review methodology was based on Cochrane collaboration guidance(76, 83) and identified only three studies suitable for review. However, the methodological quality was of such poor quality (no controls or appropriate multivariate analyses to control for potential confounding factors) that no attempt was made to undertake the meta-analysis and few conclusions about the effectiveness of the intervention strategies were made.

Oliver et al (80) conducted a meta-analysis of twenty-one papers, which met strict pre-determined inclusion criteria for hospital fall prevention programmes. The authors concluded that there was a tendency for the most rigorously controlled trials to produce the smallest effects. For example, Lingard and Rowlinson conducted a behaviour based approach to safety management in the

6

construction industry in Hong Kong(84). In common with Harper et al(85), the authors used a “multiple baseline across groups” design but also included a reversal aspect, cited as being particularly suited to measurement of safety behaviour(44). The results were mixed in terms of detected improvements in performance.

Conversely, it was noticeable that the studies most likely to quote impressive effects on health and safety performance following interventions were usually of the non-experimental type (e.g. (48, 86)). Non-experimental before-and-after designs were not regarded as suitable for evaluating change in health and safety performance because of methodological limitations of the research design resulting in uncontrolled sources of bias.

The main finding of the review has been that there exist in the literature base severe limitations in scientific rigor relating to impact evaluations of health and safety interventions. A possible reason for these limitations was thought to be that the subjective benefits of health and safety intervention programmes have been so impressive and the likelihood of impressive effects so great that there have been few demands for more reliable data(62). Other factors that may have also contributed to the lack of a robust literature base:

�� A general lack of scientific training for safety practitioners(87, 88); �� A general lack of funding for evaluation research(45, 89); �� Ethical issues relating to research designs that might withhold risk control measures from

control groups(90); �� The pragmatic nature of health and safety(91); �� Rule-driven basis of prevention strategies(92, 93).

Hillage et al(94) suggested a comprehensive strategy for improving the evidence base, including ensuring evaluation was included in the planning of interventions; developing a stronger evidence base by conducting more systematic reviews and meta-analyses; and adopting minimum standards. Taking into account the findings of the literature review, it has been concluded that a suitable research design should posses the following features:

�� A base of currently accepted UK national guidance on safety management; �� An economic evaluation component; �� A linkage between the causes of incidents and accidents to the health and safety management

system; �� An intervention in the form of written guidance or work book format, supplemented by an

educational component; �� A quasi-experimental or experimental design with baseline measurements; �� A comparison group; �� Random sampling where possible; �� Validated methods; �� A control strategy for potential inter and intra-observer bias; �� More than one performance evaluation method including both objective and subjective

methods; �� An attempt to correlate the results of the performance evaluation methods (triangulation); �� Sample sizes of sufficient size to allow statistical inferences to be made, where possible; �� Validity checks to ensure that pooling of data from different sites was appropriate before

proceeding with analysis.

7

2 METHODS

This Chapter describes the research methodology and identifies how each aspect combined and contributed to the research design. The Chapter has five parts:

Part 2.1: Practical factors relevant to the research design; Part 2.2: The evaluation methodology design (research objective 1); Part 2.3: The intervention design (research objective 2); Part 2.4: The research design for measuring change in health and safety performance

(research objective 3); Part 2.5: The data analysis strategy.

These will be considered in turn.

2.1 PRACTICAL FACTORS RELEVANT TO THE RESEARCH DESIGN

2.1.1 Variability between NHS Trusts

Within the NHS there was a high degree of variability between individual operational units (i.e. NHS Trusts), as a result of population, demographic, regional, and cultural factors. There was also local autonomy within broad, national performance targets and so individual units within the various Health Authorities or Boards tended to have different priorities and approaches to the development of their arrangements for the management of health and safety.

It was recognised that these potential differences should be taken into account, particularly during the analysis of results by testing the appropriateness of combining results from different trusts. This was done by multiple comparison testing (see Section 3.5 on data analysis).

2.1.2 Recruitment of research participants

Random selection of Trusts was not practicable because participation in the fieldwork required their agreement and for them to be willing to be involved. This involvement included:

�� Commitment to all phases of the project over its 30 month duration; �� Co-operation with the requirements of the methodology; �� Allocation of resources (e.g. those with operational responsibilities for health and safety

working within the participating Trusts were expected to be fully involved, especially in day to day contact with the research team).

Therefore a convenience sample approach was adopted (after Harper et al(85)). A call for volunteers was made via the NHS in Scotland Directors of Human Resources (HR) forum. This forum was chosen because the project had the formal support of the HR Directorate within the SEDoH. Additionally, the majority of NHS Trusts in Scotland had assigned executive responsibility for health and safety management to their HR Directors, which was a positive feature of the recruitment strategy. In England the HSE had contacts with NHS Trusts who had previously expressed willingness to participate in project work.

8

Using this approach it was intended to recruit between six and nine Trusts representing a mixture of urban and rural locations and acute and primary care sectors. This number was selected for pragmatic reasons in that it was estimated at the beginning that the project team could devote sufficient time to support and guide the Trusts with this number of participants. Also, this number was thought likely to generate sufficient data for useful inferences to be made. At the end of this process seven NHS Trusts participated in the research.

Variability was expected in the degree of development of health and safety management systems between the Trusts. In the event it was possible to identify two distinct groups from the volunteers.

One group comprised two Trusts with an existing formal health and safety management system, typified by the presence of a workbook in the possession of line managers. This group was particularly interested in health and safety performance. For these Trusts, the performance measurement exercise was the main interest, so feedback on findings was provided, via regular meetings and interim reports. They were assigned to the test group.

The other group of five Trusts did not have a formal, systematic health and safety management system based on a workbook approach. The Trusts comprising this group were interested in introducing one to their organisation. The project offered the opportunity to implement a system and to evaluate its impact. They were assigned to the control group.

The participating trust were therefore as follows:

Trust A: A test primary care Trust (PCT); Trust B: A test PCT; Trust C: A test PCT; Trust D: A test PCT; Trust E: A test acute Trust; Trust F: A control acute Trust; Trust G: A control PCT.

2.1.3 Relationship between researchers and Trusts

Recognising that non-random selection of the participating Trusts would limit the ability to make statistically significant inferences about the findings, the field study relied on the commitment and participation of the Trusts who agreed to take part.

Ultimately the health and safety management system was introduced for the benefit of the participating Trusts, and it was regarded as important that they had ownership from the beginning. This was because the researchers would eventually withdraw leaving the Trusts able to take forward the day-to-day management of the system. It was therefore in the best interests of all concerned to be involved from the beginning.

For data collection on the scale proposed it was vital that positive working relationships were built with the participating organisations to ensure co-operation. This was achieved by the following combination of measures:

�� Regular, informal dialogue with the main Trust contacts to ensure a positive relationship between the participants and the researchers;

9

�� Ensuring regular verbal and formal written feedback at each milestone of the project; �� Adopting an approach that was as flexible as possible, while recognising the need to minimise

sources of bias within the research design. For example, some Trusts wished more assistance with data processing than others, while some requested that some of the investigation of incidents were undertaken by members of the research team while others were content to undertake this activity themselves, under supervision (these differences were not thought likely to have introduced bias but were documented);

�� Regular project meetings. The project team met at least weekly to discuss day-to-day issues, while formal meetings with the Trusts were usually held at least quarterly.

2.2 EVALUATION METHODOLOGY DESIGN (RESEARCH OBJECTIVE 1)

The research design ensured that data was collected on both objective and subjective aspects of health and safety performance.

The method comprised tracking reported incidents and investigating them to establish the objective components. It had been developed on the paradigm that weaknesses in the health and management system underpinned the occurrence of incidents. Linking causes of incidents to the underpinning health and safety management system of an organisation would therefore provide a measure of its’ performance. This was a novel approach, replacing the need for health and safety management systems audits to be conducted in parallel with incident investigation. The pilot work conducted by the author(34) and described in Chapter 1 provided evidence that this was a valid strategy.

There were three objective components to the research design. These were:

1. Identification of incident management system failures; 2. Root causes of incidents; AND 3. Costs of incidents.

There were also two subjective parts. These were:

4. A staff opinion survey of a stratified random sample of NHS staff; 5. Measurement of staff views on the usefulness of the intervention used as the vehicle to

introduce the health and safety management system to the participating Trusts.

Each of these aspects will be described in turn.

2.2.1 Objective measurement component

The HSE method identified in Chapters 1 and 2 was adopted as the measurement method for this aspect of the evaluation. The method will be described below but the detailed methodology, including flow chart, incident checklist and the recording forms, with an example to illustrate their use, is included in Appendix 1.

Only those incidents that met criteria that had been pre-defined by HSE were included. These were:

�� Whether the incident caused actual harm or had the potential to cause harm; �� Whether the incident was within the power of The Trust to control;

10

�� Whether the cost of the incident exceeded £5 or 15 minutes of costed staff time.

The process that was undertaken by the investigators is summarised below:

�� Initial follow up of the incident with the person who reported it; �� Establishment of whether the incident met the above inclusion criteria. If not the incident was

excluded from the study); �� For each incident included in the study, an investigation pack was allocated, noting the

date/time of incident on all paperwork; �� Detailed information was obtained about the incident. The form used to record this data

contained prompts, which encouraged the investigator to consider three major groupings of control measures (i.e. physical, behavioural and organisational controls) (data capture form Appendix 1 Form 1A);

�� Failed risk systems were then identified (e.g. control of contractors, management of violence and aggression) (also recorded on Appendix 1: Form 1A);

�� The main contributions to costs were then grouped against whether they applied to staff, objects, materials, or equipment (recorded on Appendix 1: Form 1B). A detailed specification for participants as to which staff and non-staff costs to include was available (set out in Appendix 2);

�� An Events and causal factors chart was drawn up to assist with the identification of any further areas for follow-up;

�� Details were recorded on an Evidence table (Appendix 1: Form 2A); �� Management system root causes were then identified (e.g. policy, organising, planning etc.)

(Appendix 1: Form 3) and added to the Evidence table (Appendix 1: Form 2B); �� Costs arising from the circumstances and outcome of each incident were further described by

assigning them to relevant headings within each of three different groups. These groups were:

1. Incident category: accidental injury; accidental property damage; fatality; fire; ill health; near miss; other; physical violence; theft; vandalism; verbal abuse/threatening behaviour.

2. Incident factor: absence; cleaning up; damage/repair/staff treatment; hiring/purchasing; initial response to accident; lost production/wasted time; patient treatment; replacement labour; transport.

3. Incident type: contact with electricity; contact with equipment/machinery; cut with sharp material/object; exposure to fire; exposure to harmful substance; fall from height; hot or cold contact; manual lifting/handling; needlestick/sharps injury; other; patient lifting/handling; person to person assault; slip/trip/fall on same level/stairs; struck against something; struck by an object;

�� Verification was undertaken that no costs or other contributory factors were outstanding before completing the case;

It can be seen from the above that the approach was iterative and that the evidence was recorded in such a manner to render it capable of being audited.

All incidents that were reported via each Trusts incident reporting system were assessed against the above criteria. Those incidents that met one or more criteria were included in the study (i.e. those that were “captured”).

11

The Trusts used reporting schemes that were similar to each other, based on the IR1 form(9, 95), which was produced specifically for use in the NHS, and in this respect could be regarded as comparable.

Nevertheless, these systems were undoubtedly prone to local reporting bias. In addition, the introduction of a health and safety management system could have resulted in an increase in reporting of incidents, due to an increased awareness of the need to report. These possibilities were borne in mind during the analysis and interpretation of the results.

It was also recognised that there was a need to control potential investigator bias during the incident investigations and root cause analyses. The ability to demonstrate consistency of investigation and root cause analysis was recognised as important to the credibility of the research design. A combination of measures was therefore implemented to ensure as high a degree of intra- and inter-observer consistency as possible.

These measures included initial training for investigators, which was undertaken by staff from the HSE, at the start of the first data collection period. Refresher training supervised by HSE, at the start of the second phase, a year later was also undertaken. In addition, all investigation and follow-up decisions in all seven participating Trusts were under the control of the author. In cases where there was ambiguity, a decision was reached following discussion between the author, researcher, investigator and relevant experts (e.g. ergonomists or specialist clinicians such as psychiatrists) as judged appropriate by the research team.

2.2.2 Subjective measurement component

The method selected was the safety climate questionnaire and toolkit developed by Cox and Cheyne(59). This model was chosen because it was based on a systems approach to organisational culture and had been piloted, as a potential benchmarking tool, under the auspices of a joint industry and HSE partnership.

The questionnaire was regarded as validated and reliable in use because of extensive analysis of pilot data, conducted in the offshore industry(58). This included factor analysis, internal-scale consistency and alternate forms reliability tests.

The questionnaire consisted of 43 questions aimed at eliciting individual views on nine dimensions of safety climate, grouped into four categories. These were:

Category (a): Organisational context covering four dimensions: 1. Management commitment 2. Communication 3. Priority of safety 4. Safety rules & procedures

Category (b): Social environment covering two dimensions: 5. Supportive environment 6. Involvement

Category (c): Individual appreciation covering two dimensions: 7. Personal priorities and need for safety

12

8. Personal appreciation of risk

Category (d): Work environment covering a single dimension: 9. Physical work environment

Each question was answered by ticking a box to indicate the response. The whole questionnaire took between 10 and 15 minutes to complete. A free-text area was included for respondents to note any additional information if they wished.

Each item was scored by assigning a value to the question response as follows (the higher the score the more positive the opinion):

1 - strongly disagree 2 - disagree 3 - neither agree nor disagree 4 - agree 5 - strongly agree

Some questions had been negatively worded. The guidance provided with the questionnaire required these to be coded by subtracting the item score from 6 (e.g. a score of 2 on a negatively worded item would be reversed to a score of 4). Scores were then averaged for each question, across each Trust. Since the dimensions in the questionnaire had different numbers of questions they were standardised before being compared, by converting the scores to a 1 to 10 scale, by dividing the actual score by the total possible score and then multiplying by 10.

The resulting standardised dimension scores then had the following meaning:

2 - strongly disagree 4 - disagree 6 - neither agree nor disagree 8 - agree 10 - strongly agree

The lower the score below 6 (representative of the mid-point on the scale) the more negative the opinion(58).

The questionnaire, and detailed guidance for use, had been published for use by managers and safety professionals within the offshore oil extraction industry(59). This detailed guidance was followed closely.

Questionnaire confirmatory factor analysis The questionnaire had been developed for the offshore industry(58). To ensure that the scales in the original questionnaire were appropriate for use in a healthcare setting, confirmatory factor analysis (CFA) was undertaken. The data was subjected to CFA using version 6 of the EQS (structural equation) program(96). Internal-scale reliability (or consistency) was also examined to investigate the degree to which the various questions measured different aspects of the same concept. Finally the data was examined to establish the degree of fit for each Trust.

13

Questionnaire sampling strategy Staff lists were obtained from the Trusts and random 10% sample, stratified by job title/grade was selected. The lists obtained from the Trusts were ordered alphabetically according to job title/grade so that individual staff groups were segregated. Every tenth person on this list was then sampled, to produce the stratified random sample.

Questionnaires were sent to named individuals so that they could be followed up. This was done so that change in opinion, if it were present, could be detected with minimal threats to the validity of the results from confounding history effects. A potential source of bias in the results by using this approach was that individuals might have been concerned about issues of confidentiality. This was not raised as an issue during a pilot exercise (described below). Nevertheless the potential was recognised and the wording of the covering letters was chosen to reassure staff that their responses would be treated in strict confidence. The letters are reproduced in Appendix 3.

Questionnaires were distributed at the start of a twelve week window during which incident data was also investigated, to ensure that the opinions expressed in the survey were matched in time to the incident data. Those individuals who were classified as non-responders at the end of the 12-week period were not followed up any further.

To ensure an optimum response rate, an opt-out system was included. This allowed individuals to inform the project team that they were unlikely to be available for a second questionnaire round. Those who had not responded within 4 weeks were issued with a reminder. Those formally opting out and those who did not respond to one reminder, within the 12 week window, were replaced with an additional individual, randomly selected from the relevant area in the stratification.

Questionnaire pilot and quality control Although its designers had extensively validated the questionnaire, it had never before been used in the healthcare sector. The wording of two of the questions was altered to reflect different terminology used in the NHS.

These changes were not thought to be likely to have an impact on the final analysis. Nevertheless the use of the modified NHS-version was piloted to ensure that the wording was relevant to staff from the healthcare sector. The pilot was carried out in two departments within a Trust not participating in the main study, resulting in 22 completed questionnaires (100% response from those issued).

An additional question was included for the pilot only to elicit any concerns regarding the changed wording in the questionnaire. This question was worded:

“I found the questions in this survey easy to understand and apply to my situation”.

The responses to this question were:

Strongly agree 1 respondent Agree 7 respondents Neither agree nor disagree 10 respondents Disagree 3 respondents Strongly disagree 1 respondent

14

Since most (82%) respondents were either positive or neutral in their responses. This was judged to be reasonable and it was therefore decided to proceed with the questionnaire in its modified form, but without the additional question on ease of use.

A quality control system for questionnaire data entry was developed, using a C-Chart, a form of statistical process control(97, 98). This required that each time a batch (a batch was done in one sitting, so there could be more than one batch in a day) of questionnaires was entered onto the database, five forms were randomly selected, from across the whole batch, and checked for errors. If an error was found the whole batch was rechecked. Once all the data had been entered, the points were added and divided by the total number of data entries to calculate the final percentage error rate.

2.3 INTERVENTION DESIGN (RESEARCH OBJECTIVE 2)

The intervention was a health and safety management system, introduced, via a record keeping based system in a workbook format. It was designed to be a manager’s tool rather than a reference resource. This was because background information and guidance was already available to managers via their professional associations and HSE publications that were freely available to the general public and employees of the Trusts3. Also, users in the pilot had expressed a desire for a pragmatic document, which gave step-by-step guidance on required action. The author therefore designed a new workbook for the research described in this report. It is reproduced in Appendix 6-4. Its purpose was intended to provide a common standard with which the Trusts could implement a formal health and safety management system, based on HSG65 (74, 99).

It was written in a style that followed the general guidance of the Plain English Campaign4. The design of the workbook followed the same structure and terminology of HSG65, which was familiar to managers. Thus the document comprised five main Sections, which are described below:

�� Policy. Including space for a general health and safety policy as well as other policies on specific risks (e.g. policy on the management of violence and aggression), whereby a clear direction for the organisation to follow is described;

�� Organising. Including management structure and arrangements for delivering the policies. In this section are lists of health and safety roles and responsibilities, as well as data on consultation with staff, records of meetings, and training needs analyses;

�� Planning and implementing. Including the process for recognising, assessing and controlling workplace risks. Action planning is encouraged and various examples of risk assessments are given (e.g. manual handling and display screen equipment assessments);

�� Measuring performance. Including pro-active methods (e.g. workplace inspections) and reactive methods (e.g. incident data analysis) to monitor against policy standards. A checklist for workplace inspections is given as well as ideas for other activities, such as safety tours;

�� Audit and review. Including arrangements for a systematic review of performance based on data from monitoring and audits. A managers’ self-audit is included and the reader is guided to undertake this activity at the start of the process and to use the findings to form a yearly plan, with quarterly reporting milestones.

3 e.g. http://www.hse.gov.uk/sources/ or http://www.tuc.org.uk/h_and_s/index.cfm 4 http://www.plainenglish.co.uk/

15

An economic tool, based on the technique of option appraisal had also been developed by the author in response to a stated need for managers to be able to use economic assessment of their control measures as they undertook risk assessment(100). This method was also included in the workbook, although managers were encouraged to seek specialist help from either the author or their local safety adviser if they identified a complex assessment that might benefit from the inclusion of an economic option appraisal.

Early drafts of the workbook were piloted by initially being shown to colleagues and to employees within a NHS Trust that was not part of the fieldwork. After further modification the author formed a small working group, comprising representatives from management, staff and the health and safety department of the Trust. This group discussed and agreed how the workbook should look and how it should be introduced and used. The consensus was that it was most suited to line managers with responsibility for more than one tier of staff and that training in use was required. The third draft was used by a group of 12 managers, within the same Trust, who were asked to use the workbook for approximately three months and to offer suggestions for final refinement. The author interpreted the feedback translated this into modifications of the workbook and produced the version that was used for the research.

In its final form the workbook represented a useful, user-friendly health and safety management tool, which had proved to be popular with users. It was therefore judged to be suitable for use in the research project.

2.3.1 Method of introduction

Although use of the workbook was planned as a standardised system, a degree of flexibility with the different Trusts was necessary. This was because once the workbooks were issued it was important for the users to have ownership of the system, since at the end of the research the workbooks were intended to remain in situ.

This generic workbook was shown to the participating Trust contacts who were asked to specify minor amendments so that it could be tailored to the terminology, management arrangements, recording conventions and culture of each Trust. Examples of typical changes included:

�� Inclusion of an organisational diagram of the management structure of the Trust; �� Inclusion of contact details for relevant individuals and departments (e.g. the safety Trust safety

adviser or infection control department); �� Inclusion of risk assessment forms used by the Trust (these tended to be different in each Trust); �� Use of font, point size and layout for individual Trust documents.

The changes were not regarded as likely to introduce significant differences between the Trusts since the modified workbooks contained the same sections and guidance. The research team incorporated the detailed changes into the workbook and arranged for sufficient copies of the tailored version to be made available. Each set of workbooks therefore had the corporate identity of the individual Trusts. The decision as to who should be designated as a workbook holder, and therefore how many workbooks were necessary, was also agreed with the Trusts. As described above, the workbook was designed for middle managers, such as a charge nurse with responsibility for several wards. The numbers issued to each Trust are shown in Table 1. From this it can be seen that there were large proportional differences between Trusts.

16

Table 1 Numbers of workbooks issued per Trust

Test Trusts Number of employees Number of workbooks issued Trust A 824 36 Trust B 3,378 23 Trust C 1,922 29 Trust D 5,636 40 Trust E 1,853 40 TOTAL 13,613 177

These differences proved to be necessary because each Trust differed in their management structure (for example, some were much less devolved than others). The number issued to each Trust was based on a balance of factors to ensure all areas within the Trusts were covered by a workbook-holder. If each person within a Trust who was responsible for managing staff had a workbook there would have been multiple repetitions of risk assessments. Conversely, if only the most senior managers were issued with a personal copy, insufficient detail of local arrangements would result. Therefore because the grade and job description of workbook holders was broadly similar between Trusts these differences were not regarded as significant to the performance of the workbook.

Issuing of workbooks to named recipients was done via a two-hour briefing, during which time the background to the project was explained and each section of the workbook was explored in detail. A detailed framework for the briefings was prepared and the author always carried out the training (the training protocol is reproduced in Appendix 5). A representative from HSE sat in on a number of the briefings and approved the delivery against the protocol to confirm consistent delivery.

The briefings took place over a 22 week period, between 10th November 2000 and 17th April 2001. This staggered introduction represented a multiple baseline across groups design(85) (44).

2.4 RESEARCH DESIGN TO MEASURE CHANGE IN HEALTH AND SAFETY PERFORMANCE (RESEARCH OBJECTIVE 3)

The aim of the research was to evaluate whether it was possible to reliably measure change in health and safety performance in a NHS setting. A study design was therefore needed to control, as far as possible, for inter-trust differences, which also allowed for reliable performance measurement both before and after an intervention.

The method adopted was a prospective empirical field epidemiology study using a longitudinal cohort observational study design of the before-and-after type, with a multiple baseline. The literature review had identified this as a suitable design that had been validated for use in health and safety impact evaluation research(44).

The Trusts were assigned to one of two cohorts: a test group; and a comparison (control) group5 (64). Those without a formal system were assigned to the test group, which received the

5 Rakel et al defined the difference between a control group and a comparison group as whether the evaluation programme used a randomised or non-randomised sampling strategy. The method adopted uses non-random sampling of the participating Trusts, therefore the control group is, strictly speaking, a comparison group.

17

intervention, while those with an existing workbook-based system already in use were assigned to the control group, which did not receive the intervention. This was a valid strategy because any effect that might be noted in the results from the test group and not the control group could have been as a result of the intervention. Given that the research aim was to assess the effect of the introduction of a formal health and safety management system the Trusts with an existing system did not require another. Another factor was that the Trusts assigned to the control group were not disadvantaged by not receiving an intervention that could have benefited their health and safety performance.

2.4.1 Method of monitoring workbook use

The workbooks were issued via a closed-loop quality management system, so that monitoring of usage could be carried out. Therefore each workbook was given a unique identification code, which specified both the Trust and the identity of the holder. Additionally, each page recorded the section of the workbook, the title of the section, the issue number and date and cumulative page number. This allowed for the possibility of changes being made and tracked. The Trusts senior management and the author both held copies of these codes.

A protocol for workbooks monitoring was developed for the test Trusts only. All workbook holders were contacted by telephone for feedback regarding their use of the workbook in accordance with a telephone interview protocol (Appendix 6). Each person was contacted on three separate occasions and asked the same questions. This occurred within three designated windows, each lasting for four months (between 1st April 2001 and 31st March 2002). Approximately 10% of those individuals who were recorded as using their workbooks were selected at random for a visual examination of their workbooks.

Any confounding effect produced by the presence of the researchers could have been controlled for in the research design if the control trusts had also been included in the monitoring scheme. However, in practice the two control trusts had had their safety management systems in place for several years and personal contact by the researchers were thought to be liable to produce a different reaction from the test Trusts. This could have introduced complex methodological confounding of the results. This was because the workbook holders in the test Trusts all had been advised, as part of their briefing, when and how they would be contacted. There was therefore little surprise when this contact occurred. The workbook holders in the control trusts had not received a briefing, because their system was already in place and training had previously been given.

Therefore, although regular contact was made with the workbook holders to gauge usage of the system via the monitoring protocol, other forms of contact were deliberately avoided, to minimise confounding effects from the research team. The protocol itself was designed to keep contact to the least necessary to gauge usage (for example, making contact three times in a year rather than four, making contact via the telephone and only visually inspecting 10% of workbooks). This effect is discussed further in Chapter 4.

The Trusts were intended to be the principal catalysts for development and integration on a day-to-day basis of the Trust management system. They had been encouraged to adopt the workbook

However, parts of the strategy used random sampling so, for simplicity, this group is referred to as the control group throughout this report.

18

system and to manage its use as they would any other management initiative. The project team took no part in these activities, other than to be aware of what had been implemented.

2.4.2 Study composition

Over 23,500 employees were encompassed by the seven Trusts. They comprised five test Trusts and two control Trusts. There was a mixture of acute and primary care Trusts in both the test and control groups, described in Table 2:

Table 2 Description of Participating Trusts

Test Trusts Control Trusts Type of Trust No. of employees Trust A Primary Care 824 Trust B Primary Care 3,378 Trust C Primary Care 1,922 Trust D Primary Care 5,636 Trust E Acute 1,853 Total number of participants in test group 13,613 Trust G Primary Care 3,978 Trust F Acute 5,944 Total number of participants in control group 9,922

To ensure that the distribution of the questionnaire was even across staff groups, the sample was stratified by selecting every 10th name from a list of staff, which had been split into eleven occupational groups, which were common to the payroll departments of all the participating Trusts. These were (alphabetical order):

�� Administration and clerical (A&C) �� Domestic and catering �� Estates (e.g. joiners, plumbers, engineers, maintenance workers) �� Medical and dental �� Nurse manager �� Nursing and midwifery (qualified) �� Nursing and midwifery (unqualified) �� Other (e.g. Chaplain) �� Professional, technical and Professions Allied to Medicine (PAMs) (unqualified) �� PAMs (qualified) �� Senior manager

The study design then followed the sequence in Figure 1:

19

TEST group CONTROL group

Benchmarking 1

Intervention “Workbook”

Benchmarking 2

Benchmarking 1

No intervention

Benchmarking 2

Compare benchmarking 2 with benchmarking 1

Compare benchmarking 2 with benchmarking 1

Compare TEST/CONTROL

Job Category & Trust

Figure 1 Relationship between test and control groups

Each measurement method had its own indicator(s) of effectiveness. For the objective methods the measures were: change in number of reported incidents; pattern of root causes of incidents; change in pattern of system failures as causal factors within incidents; and change in costs of incidents. The subjective method indicator of effectiveness was change in dimension score for safety climate opinion. The field study plan therefore involved five consecutive phases over 30 months (summarised in Table 3.

20

Table 3 Summary of project phases

Phase 1 (Section 2.2)

Recruitment of participating Trusts Training Publicity campaign

Phase 2 (Sections 2.3 & 2.4)

1st Benchmarking Phase lasting 12 weeks, consisting of: -Incident screening -Incident costing -Incident investigation -Management root cause analysis -Issue/collation of questionnaires to 10% of staff


Introduction of workbook system to five participating “test” Trusts: -Introduced via 2-hour briefing sessions -Used over a 12 month period -Monitored via telephone interviews of workbook holders -10% visually inspected to assess completion of documentation Re-training and publicity reminders also undertaken

Phase 4 (Sections 2.3.& 2.4)

2nd Benchmarking Phase lasting 12 weeks, consisting of: -Incident screening -Incident costing -Incident investigation -Management root cause analysis -Issue/collation of questionnaires to 10% of staff (same individuals as for 1st benchmarking phase)


Data analysis

Phases 1 and 5 involved recruitment of Trusts and data analysis respectively. The test Trusts were encouraged to work on their developing health and safety management systems during the middle phase, Phase 3.

Phases 2 and 4 were identical measurement benchmarking phases where health and safety performance is evaluated. The literature review and guidance from HSE project supervisors helped to inform the final decision on the features and optimum design of the measurement method. It was designed to have an objective component (based on costs and management root cause analysis of reported incidents) and a subjective component (based on staff opinion of the safety climate within their workplace).

All investigators attended training in the implementation of the methodology. Prior to the first data collection benchmarking phase (phase 2), HSE personnel familiar with the methodology, provided the training for all Trusts. This training lasted a full day. Prior to the second phase (phase 4) the author, under the supervision of HSE, carried out the re-training.

A detailed week-by-week project plan was drawn up and used throughout to assess progress. The timescale is shown in Figure 2.

21

Figure 2 Project timescale

2.5 DATA ANALYSIS STRATEGY

All data was entered onto a spreadsheet computer software package (Microsoft Excel) and analysed using this or the Statistical Package for the Social Sciences (SPSS) (version 10)(101).

Much of the data was of the ordinal type. The preferred analysis strategy with this data was to use parametric tests, where possible. Researchers in the behavioural sciences have differed as to whether the more powerful parametric significance tests are appropriate for use with ordinal data(102). The social sciences have been divided between those whose position is that the use of parametric tests on ordinal data is incorrect on both theoretical and practical grounds(103). At the other extreme are those who advocate that this approach is usually acceptable(104). The approach adopted in this work took a more neutral approach, advocated by a majority of authors(102, 105-108), by accepting the use of parametric testing where the data appeared to approach a normal distribution.

Therefore as a general principle, descriptive summary statistics were explored, by examining distributions of the data, their means and medians, prior to estimations of confidence intervals or hyporeport testing. This was done, to test the data for normality before making a decision as to whether to use parametric testing methods in preference to non-parametric methods.

The 5% significance level was used unless otherwise indicated in the results.

2.5.1 Incident data

A computerised database was provided by the HSE for the initial analysis. The outputs of this package were:

�� Summary management system failures and root causes of incidents; �� Frequency of occurrence; �� Costs of incidents; �� Magnitude of cost associated with system failures and root causes.

22

Subsequently incident data was analysed by profiling and other descriptive methods, followed by statistical analysis. The following non-parametric tests were used:

�� Wilcoxon rank sum W test (also known as the Mann-Whitney U test) was chosen because it is suitable for small sample sizes and where normality of the data cannot be assumed. This test was used to compare the following general types of pairs of ordinal variables:

�� Staff numbers and rates between the two project benchmarking phases; �� Numbers of reported incidents between phases; �� Numbers of incidents captured by the project inclusion criteria, across both Phases; �� Incident root causes data and risk control system failures. �� Spearman’s rank-order correlation was used for examination of the rank order of Trusts, by way

of their incident reporting and capture rates as well as costs of incidents and ranks of frequency of incident root causes;

�� Kruskal-Wallis one-way analysis of variance (ANOVA) was used to examine relationships within the ordinal root cause data.

Parametric tests used were:

�� Pearson’s product moment correlation matrices to investigate the relationship between the ratio-scaled variables of incident reporting rates and Trust size. This test was also used to attempt to triangulate results from all data;

�� Independent and paired sample T-testing was used to compare incident data between test and control Trusts and to examine the proportions of incidents reported against those captured by the project inclusion criteria.

2.5.2 Questionnaire data

Evidence for normality of the questionnaire response data was obtained by visual inspection of their distributions. Kruskal-Wallis one-way ANOVA was used to examine the mean dimension scores for each Trust. One-sample Kolmogorov-Smirnov testing was carried out to test the data for normality and to provide supportive evidence that the data could be treated as normal, by using parametric methods. Analysis of Variance (ANOVA) on dimension score responses between Trusts was undertaken. Comparisons within each data collection phase were made using multiple comparison tests. Differences in responses between different job categories were explored using Scheffe multiple comparison testing. The Scheffe test was used since it is suitable for use with unequal sample sizes. The test is of particular value to identify where differences lie following an ANOVA, which indicates differences in the means under testing6 (107).

Changes in responses from individuals between the two data collection phases were analysed using paired sample T-tests.

�� Power: Based on the pilot data, with a standard deviation of 0.83 units in dimension score measurements, a sample size of 1200 would have 99% power in detecting a 0.1 shift in dimension score (say, from 6.2 to 6.3). This premise was dependent on the data from the

6 http://www.richland.cc.il.us/james/lecture/m113/post_anova.html

23

individual Trusts being pooled to create a single sample. A sample size of 200 (if the data was not pooled) would have 96% power in detecting a 0.2 shift in dimension score.

2.5.3 Workbook monitoring data

This data was nominal and was firstly analysed using descriptive methods. Associations between telephone interview responses were subsequently analysed using Cramer’s V.

24

3 RESULTS

As described in the introduction research design comprised both objective and subjective aspects. The three objective components were:

1. Identification of risk control system failures leading to the establishment of; 2. Root causes of incidents; 3. Costs of incidents.

The two subjective methods were:

4. Staff opinion survey; 5. Monitoring and audit of the use of the management system workbook, used as the vehicle to introduce the health and safety management system to five of the seven participating Trusts.

Following a general description of each of the participating Trusts, the results from each aspect of the research design will be described in turn.

3.1 TRUST DESCRIPTIVES/OBSERVATIONAL RESULTS

The seven participating Trusts were described, in general terms, in Chapter 2. Further background and more detailed descriptive information are included below.

Trust A 824 employees Trust A provided community and hospital based services over 10,000 square miles, in the north of Scotland. It had a budget of approximately £118 million, 18 hospitals and nine Local Health Care Co-operatives (LHCCs). The project was restricted to two main areas, representing some 60% of the total staff complement. These were Acute Mental Health and Learning Disabilities and a LHCC.

The Trust operated the IRIS incident reporting system within the “Safecode” suite of risk management software programmes7. Incident reports were produced quarterly. Local managers reported incidents captured by the requirements of the RIDDOR Regulations(109) to HSE

Trust B 3,378 employees The Trust provided a range of specialist services to a population of 800,000 in the north-east of England. It employed approximately 3,500 staff, with a budget of approximately £90 million for patient care. Its specialist services included mental health and learning disabilities services; community mental health services.

The entire Trust was included in the research project. Incidents (both clinical and non-clinical) were reported using a single page form. All incidents were given a risk rating by the risk management department with trend analysis produced quarterly. All RIDDOR reporting was undertaken centrally.

7 http://www.show.nhs.uk/sehd/mels/1999_18.doc

25

Trust C 1,922 employees The Trust provided primary health and mental health care services to communities in the west of Scotland. The Trust was configured into two LHCC’s and a Mental Health Directorate:

Locality Managers/Department Heads were responsible for incident reporting. Copies of all incidents were forwarded to the Occupational Health and Safety Department where a central database was maintained. Analysis was provided quarterly to localities and General Managers. The same form was used for both clinical and non-clinical incidents. Regular reports were provided to the Trust Management Team, including an annual report. RIDDOR reporting was devolved to locality managers/Risk Co-ordinators.

Trust D 5,636 employees The Trust provided primary care, community and hospital based services and supporting services to a population of approximately 400,000 in the east of Scotland. A number of the Trust's services were delivered in partnership with other statutory and voluntary organisations and the Trust had close links with and provided teaching facilities for two local Universities. The Trust employed approximately 5,500 staff and had a budget of around £234 million made up of £125 million for hospital and community health service budget and £109 million for primary care services.

Incident reporting was largely centralised although incident investigation and RIDDOR reporting was delegated via the line management function.

Trust E 1,853 employees The Trust came into being on 1st April 1994. It was the main provider of district general care and maternity services to a population of approximately 147,000 covering an area of approximately 2,500 square miles in south-west Scotland. In 1999/2000 the Trust reported that it employed 1841 staff (1417 whole time equivalents) with a projected income of approximately £50 million and a capital programme of approximately £0.8 million.

Incident reports were followed up by the in-house occupational health service.

Trust F (Control) 5,944 employees The Trust provided a comprehensive range of acute adult and paediatric care to a large city in central Scotland. 12,500 staff provided a wide range of specialist services for approximately 142,000 adults and 19,000 children each year.

The area of the Trust participating in the project was a single large teaching hospital site as it had an established health and safety management system already in place. All incidents were reported on a single initial report form regardless of type or consequence. A Trust-wide database allowed the analysis of trends. RIDDOR reporting was centralised in the Health and Safety office.

26

Trust G (Control) 3,978 employees This was a Primary Care Trust responsible for primary services, community and inpatient mental health services, child health and a range of other NHS services throughout a semi rural area in the east of Scotland. The Trust was moving through Area Redevelopment Teams to reshape mental health services towards a more community-based approach.

Incident and RIDDOR reporting was delegated to four localities.

3.2 GENERAL INCIDENT FINDINGS

The data collection periods were at different times to ensure that there was a multiple baseline(44, 85) but all periods were the same length (12 weeks or 84 days). The Trusts were allocated to a time period that was most convenient to them. The periods were as follows:

Phase 1: 1st July 2000 – 22nd September 2000 Trusts C, E 1st August 2000 – 23rd October 2000 Trusts A, B, F 1st September 2000 – 23rd November 2000 Trusts D, G

Phase 2: 1st January 2002 – 25th March 2002 Trusts A, B, F, G 1st February 2002 – 25th April 2002 Trusts C, D, E

Summary descriptive data is shown in Table 4. The first two columns show the staff head-count and the whole-time equivalent (WTE) staff numbers for both phases of the project. Not all of the Trusts were able to provide staff head count data for the second phase but all were able to provide WTE data. Therefore only WTE data was used later in the results to calculate rates.

The third column shows the WTE data, standardised for the length of the data collection period (84 days), with the total incidents reported in the fourth column. The average reported incident rate (expressed in per person working years) is in the fifth column.

The last two columns show the number of incidents included in the study (referred to as “captured” by the selection criteria) and the percentage of incidents captured.

27

Tabl

e 4

Gen

eral

inci

dent

resu

lts fr

om b

oth

phas

es

Trus

t St

aff h

ead

coun

t W

hole

-tim

e st

aff

equi

vale

nt

(WTE

)

Pers

on w

orki

ng

days

(WTE

*84

days

)

Tota

l in

cide

nts r

epor

ted

Mea

n re

porte

d in

cide

nt ra

te

(per

per

son

wor

king

ye

ar8 )

Tota

l in

cide

nts

incl

uded

in st

udy

(“ca

ptur

ed”)

Perc

enta

ge in

cide

nts

capt

ured

Trus

t A 1

st p

hase

65

4 54

,936

46

51.

91

40

8.6%

Tr

ust A

2nd

pha

se

824

662

55,6

08

488

(+5%

) 1.

98

47

9.6%

Tr

ust B

1st p

hase

2,

977

250,

068

2,03

4 1.

82

20

1.0%

Tr

ust B

2nd

pha

se

3,37

8 2,

969

249,

396

2,92

7(+4

4%)

2.63

16

0.

6%

Trus

t C 1

st p

hase

1,

334

112,

056

583

1.17

33

5.

7%

Trus

t C 2

nd p

hase

1,

922

1,40

6 11

8,10

4 80

0 (+

37%

) 1.

53

42

5.3%

Tr

ust D

1st p

hase

4,

011

336,

924

1,99

2 1.

33

59

3.0%

Tr

ust D

2nd

pha

se

5,63

6 3,

927

329,

868

2,09

3(+5

%)

1.42

12

5 6.

0%

Trus

t E 1

st p

hase

1,

430

120,

120

290

0.54

31

10

.7%

Tr

ust E

2nd

pha

se

1,85

8 1,

5699

131,

796

220

(-24

%)

0.38

20

9.

1%

Trus

t F 1

st p

hase

4,

121

346,

164

547

0.36

32

5.

9%

Trus

t F 2

nd p

hase

5,

944

Con

trol

4,15

0 34

8,60

0 57

7 (+

5%)

0.38

30

5.

2%

Trus

t G 1

st p

hase

2,

787

234,

108

635

0.61

30

4.

7%

Trus

t G 2

nd p

hase

4,

290

Con

trol

2,92

2 24

5,44

8 10

02 (+

58%

) 0.

92

26

2.6%

To

tal 1

st p

hase

17

,314

1,

454,

376

6,54

3 1.

01

245

3.7%

To

tal 2

nd p

hase

23

,852

17

,605

1,

478,

820

8,10

7 (+

24%

) 1.

24

306

3.8%

8 Ass

umes

225

wor

king

day

s per

yea

r (45

wee

ks x

5 d

ays p

er w

eek)

9 T

his i

ncre

ase

was

due

to ta

king

ove

r hot

el se

rvic

es st

aff

29

The numbers of staff covered by the research ranged from 824 (representing 654 WTE at the first baseline) within Trust A, to 5,944 (4,150 WTE) within Trust F (control). The difference in WTE between the two phases did not vary by more than 5%, with the exception of Trust E which rose by 10% as a consequence of taking over hotel services staff from another unit during the study. Nevertheless the differences between the mean WTE numbers between the two phases were not significant (Wilcoxon rank sum W test P=0.20).

The total number of incidents reported was 6,543 in Phase 1 and 8,107 in Phase 2. The number of incidents reported varied between Trusts, with an approximate order of magnitude between the lowest (255 for Trust E (mean of both phases)), and the highest (2,481 for Trust B (mean of both phases)). The mean number of reported incidents rose by 24% in the second phase. However, in Trust E fewer incidents were reported in the second phase (mean fell by 24%). The overall reduction was marginally significant (Wilcoxon rank sum W test P=0.06). However, given the small sample size it would be prudent to assume a lack of significance.

The number of incidents reported from the control Trusts was not found to be different from the test Trusts (Independent samples t-test P=0.49 for phase 1 and P=0.58 for phase 2, although because of the small sample size the confidence intervals were wide (between –1728 and +2760)).

The total number of incidents that met all three incident inclusion criteria, (i.e. the total number of incidents captured) was 245 for phase 1 and 306 in phase 2, an increase that was not significant at the 0.05 level (Wilcoxon rank sum W test P=0.61).

The captured incidents represented an average of 3.7% of the total number reported (range 0.6% to 10.7%). This is a very small proportion, brought about by large number of reported incidents, which did not meet the inclusion criteria largely because of their trivial nature. The mean number of incidents captured per Trust was 40, with more captured in phase 2 (35 in phase 1 against 46 in phase 2), again an increase that was not significant (Wilcoxon rank sum W test P=0.73).

Although the total numbers of reported incidents were broadly within the same order of magnitude for each Trust across both data collection phases, intra-Trust variation between the two phases was noticeable (between +5% for Trust A and Trust D and +58% for Trust G (control). Therefore the relationship between the reported incidents were standardised by person working years. However, the resultant differences between rates within each Trust across the two phases (column 5 in Table 4) were not significant between the two phases (Wilcoxon rank sum W test P=0.61).

The mean incident-reporting rate increased from 1.01 incidents reported per person per year to 1.24 incidents reported per person per year. This was marginally significant (Wilcoxon rank sum W test P=0.09). Due to the small sample size it is regarded as prudent to regard the result as non-significant. The rank order of reporting rates (expressed as per person working year) is shown in Table 5.

30

Table 5 Rank order for incident reporting rates between both phases

Phase 1 Average reported incident rate (per person working year)

Phase 2 Average reported incident rate (per person working year)

Trust F (Control) 0.36 Trust F (Control) 0.38 Trust E 0.54 Trust E 0.38 Trust G (Control) 0.61 Trust G (Control) 0.92 Trust C 1.17 Trust D 1.42 Trust D 1.33 Trust C 1.53 Trust B 1.82 Trust A 1.98 Trust A 1.91 Trust B 2.63 MEAN 1.01 MEAN 1.24

None of the Trusts changed their rank by more than one place between the two phases of data collection. Three Trusts maintained their rank between phases while the ranking interchanged between the remaining four Trusts (Trusts C and D, and A and B). The acute Trusts (Trust F and Trust E) and the control Trusts (Trusts F and G) had, between them, the lowest rates in both data collection phases. The four test PCT Trusts had the highest reporting rates. The Spearman rank order correlation coefficient between the rates for the two phases was 0.919 (P<0.01), confirming a significant relationship.

This suggests that the rates of reported incidents were independent of the size of the Trust. This was confirmed when the Pearson correlation between WTE and incident reporting rate (per person per year) were examined and found to be unassociated (P=0.48)).

Intuitively this seems likely, given the different client groups within the two types of Trust and the likelihood of recording more incidents related to the management of patients with mental health problems and learning difficulties in the PCTs. However, the variation between the five PCTs was approximately four-fold (between 0.61 and 2.63 incidents per person per year). This points to the variation in reporting rates being more related to the reporting culture and conventions within the individual Trusts, rather than differences between the type of Trust.

Despite the large variation in numbers of incidents reported, the numbers actually captured by the study criteria were relatively low (between 0.6% and 10.7% of the total number of incidents reported), because of the large numbers of incidents that were reported but not captured. These captured incidents can be regarded as the most serious or the reported incidents. The ratio of serious to minor (i.e. captured versus non-captured) reported incidents is sometimes regarded as a measure of safety culture – the lower the proportion of serious incidents the better the (reporting) culture (e.g. (66)). The low percentage of captured incidents adds weight to this hyporeport.

The rank order of the rates of captured incidents, across the two data collection phases, is shown in Table 6.

31

Table 6 Rank order for incident capture rates between both phases

Phase 1 Average captured incident rate (per person working year)

Phase 2 Average captured incident rate (per person working year)

Trust B 0.018 Trust B 0.014 Trust F (control) 0.021 Trust F (control) 0.020 Trust G (control) 0.029 Trust G (control) 0.025 Trust D 0.039 Trust E 0.037 Trust E 0.058 Trust D 0.083 Trust C 0.066 Trust C 0.084 Trust A 0.164 Trust A 0.193 MEAN 0.056 MEAN 0.065

The difference between phases of the mean rates of number of incidents captured per person per year was not significant (Wilcoxon rank sum W test P=0.61). The ranking of individual Trusts was maintained across the two data collection phases with the exception of Trusts D and E. The lowest capture rate was Trust B and the highest was Trust A.

Although Trust B had the lowest capture rate it also had one of the highest reported incident rates (see Table 4). When this relationship was explored further with all of the Trusts there was no correlation between rates of reported and captured incidents (P=0.54 for phase 1 and P=0.70 for phase 2). This indicates that the rates of captured reported incidents were not linked to rates for all reported incidents.

Unlike rates of reported incidents, there was an association between the percentage of captured incidents and the size of Trust. This was significant in the first phase (P=0.01) but only marginally significant in phase two (P=0.09). Given the small sample size it is prudent to interpret this finding with caution but it may be that the more serious incidents are linked to some other source than reporting culture, possibly safety performance. This possibility is explored further in Chapter 4.

The test and control Trusts were not significantly different from each other in either phase for either reported or captured incidents (P=0.50 for both phases).

3.2.1 “Special” category

At the start of the field work it was anticipated that most incidents that were identified would have readily identifiable management root causes associated with them(43). This was because of general acceptance of the paradigm that accidents are largely a consequence of deficiencies in the health and safety management system(99).

However investigation of approximately half of the incidents captured did not reveal readily identifiable risk control system failures or management root causes. These incidents fell into a “grey” area between obvious system failures and root causes and meeting the incident inclusion criteria. For example:

�� By the time a junior doctor reported a needlestick incident and it was subsequently identified for further investigation, the doctor had moved to another placement elsewhere in the UK and new contact details were unavailable;

�� In the care of elderly, frail patients, individual clinicians routinely had to make professional judgments as to when it was appropriate to encourage mobilisation so that independence,

32

and ultimately quality of life, could improve. This benefit had to be balanced against the risk that, in so doing, the patients’ risk of falling increased;

�� Some mental health patients were being encouraged to take a more active part in society by being given periods of parole, during which time their freedom was increased. The purpose of this care approach was to build trust with the aim of the patient being able to re-integrate into society. This aim was in the context of a subjective balance with the risk that the patient might abscond or self-harm.

It was judged inappropriate to simply exclude these cases from the study since, on balance they were judged to fall within the inclusion criteria already described. They were therefore assigned to a third “special” category of incident types10.

After the completion of the first data collection exercise special category incidents were identified as belonging to one or more of four different types11. These are listed below but also discussed further in Chapter 4:

�� Health and safety systems were in place within the bounds of what was considered (by the research team) to have been reasonably practicable. An example of this type of incident was where a patient fell, who had been assessed as independently mobile, with no history of falls and already under general observation. It was judged that the same assessment would have been made in similar circumstances;

�� Insufficient information emerged from investigations to come to reliable conclusions. An example of this type of incident was the fall of a patient where the fall was unwitnessed and the patient was unable to offer any explanation for what happened;

�� With hindsight the incident may have been preventable but, based on the facts that were known at the time, the risk was unlikely to have been foreseen. An example of this type of incident was where a gust of wind smashed an open window. The window was normally opened for ventilation and had never before caused problems;

�� The key to assessing how well the risk was managed was associated with individual practitioners’ clinical judgement. An example of this type of incident was where a patient on a soft-food diet was being re-introduced to textured foods. The patient choked and had to be taken to the casualty department for treatment.

For some incidents of this type it was subsequently possible to identify management root causes. Where an incident was identified as having the potential to be included within the special category its various merits were discussed in a case-conference by the project team and with the involvement of various specialists, independent of the project, where necessary. Extensive measures were applied to ensure that every identified possibility was examined in detail before a consensus was reached as to whether the incident was to be assigned to the special category group and, if so, to which type.

It can therefore be assumed that the assignment to the categories was internally consistent. However, a potential limitation was that the need for a special category was only identified during the first data collection phase, in response to the increasing numbers of incidents that had

10 These incidents are listed and described in Appendix 8

11 The distinction between the types was not exact and there were occasions where more than one special category type was assigned to a single incident. For example, if a patient who was not assessed as being at risk from falling, and had never fallen before (risk possibly preventable but not foreseeable) fell but the exact circumstances were not witnessed and the patient was not able to offer any explanation as to what happened (insufficient information to come to a reliable conclusion), the incident would be assigned under both types.

33

defied efforts to identify system failures and root causes. It is therefore possible that the consistency was less reliable with incidents that occurred at the start of phase 1. These incidents were reviewed again during the second phase, to minimise this effect, but this could only achieve partial assurance, given that so much time had elapsed since they had occurred.

Table 7 shows the total numbers of incidents reported; those that were captured by the inclusion criteria; and those that were assigned to the special category. The table also shows the rates (per person per year) for incidents for which root causes could be identified (column 4) and those assigned to the special category (column 6).

35

Tabl

e 7

Inci

dent

s sp

lit b

y m

anag

emen

t roo

t cau

se a

nd s

peci

al c

ateg

ory

from

bot

h ph

ases

Tr

ust

Tota

l in

cide

nts

repo

rted

Tota

l in

cide

nts

capt

ured

Num

ber

of

inci

dent

s (r

oot

caus

es a

ssig

ned)

Aver

age

inci

dent

rat

e (r

oot

caus

es

assig

ned)

(p

er

pers

on w

orki

ng y

ear)

Num

ber

of in

cide

nts

(roo

t ca

uses

not

ass

igne

d) (

i.e.

spec

ial c

ateg

ory)

Aver

age

inci

dent

rat

e (r

oot

caus

es

not a

ssig

ned)

(i.e

. spe

cial

cat

egor

y)

(per

per

son

work

ing

year

) Tr

ust A

1st

pha

se

465

40

20

0.08

20

0.

08

Trus

t A 2

nd p

hase

48

8

47

27

0.11

20

0.

08

Trus

t B 1

st p

hase

2,

034

20

15

0.01

5

0.01

Tr

ust B

2nd

pha

se

2,92

7 16

11

0.

01

5 0.

01

Trus

t C 1

st p

hase

58

3 33

16

0.

03

17

0.03

Tr

ust C

2nd

pha

se

800

42

13

0.02

29

0.

06

Trus

t D 1

st p

hase

1,

992

59

28

0.02

31

0.

02

Trus

t D 2

nd p

hase

2,

093

125

34

0.02

91

0.

06

Trus

t E 1

st p

hase

29

0 31

22

0.

04

9 0.

02

Trus

t E 2

nd p

hase

22

0 20

16

0.

03

4 0.

01

Trus

t F 1

st p

hase

54

7 32

16

0.

01

16

0.01

Tr

ust F

2nd

pha

se

577

30

20

0.01

10

0.

01

Trus

t G 1

st p

hase

63

5 30

10

0.

01

20

0.02

Tr

ust G

2nd

pha

se

1002

26

10

0.

01

16

0.01

To

tal 1

st p

hase

6,

543

245

127

0.03

11

8 0.

03

Tota

l 2nd

pha

se

8,10

7 30

6 13

1 0.

03

175

0.03

36

The number of incidents captured was particularly influenced by an increase in those from the special category in Trust C (increased from 17 to 29) and Trust D (increased from 31 to 91). Both increases were for incidents involving violence and aggression. These resulted in an increase in the total number of special category incidents of 48% between the two phases. However, the differences between the Trusts were not significant (Wilcoxon rank sum W test P=0.69).

Some of the cases included repeated incidents involving a single patient. It could be argued that, once an incident had occurred its re-occurrence could be predicted and therefore should have a root cause. This applied to some of the incidents and, in these cases, the repeated incidents were root caused. However, some were still classed as special category. These were where, for example, the decision to maintain a care plan was a clinical judgment such as a decision to continue to give parole to a patient with a history of absconding, so that they could continue to attempt rehabilitation.

The total number of incidents that were found to have management root causes associated with them (i.e. those minus the special category incidents) were 127 in phase 1 and 131 in phase 2. This represented an increase of 3% between the two phases, which was not found to be significant (Wilcoxon rank sum W test P=0.87)12.

There were also more special category incidents between phase 2 and phase 1 (175 special category in phase 2 against 118 in phase 1) but again this increase was not significant (Wilcoxon rank sum W test P=0.69). This is further discussed in Section 4.4.3.

The percentage of special category incidents did not differ significantly between Trusts (Spearman’s rank order correlation P=0.7). The percentage of special category incidents that were assigned to each of the four categories identified above is shown in Figure 3. It can be seen that most incidents were assigned to the category where the project team judged that avoidance of the incident was out with what could be thought of as reasonable (50% of number of incidents). One quarter of the incidents was based on a clinical judgment that was under the responsibility of the clinician in charge of the patient (26%). The risk was not foreseeable in 17% of the special category incidents and the remaining group was characterised by insufficient information on which to base a judgment (7%).

The percentage of special category incidents that were assigned to each of the four categories identified in 3.2.1 is shown in Figure 3.

12 These incidents are listed (for both phases) and described in Appendix 7

37

Figure 3 Comparison between number and cost of special category incidents

The traditional assumption has been that most, if not all, incidents can be prevented by the implementation of robust risk management policies and procedures(18, 25, 110). The special category incidents are in potential conflict with this paradigm and may represent a consequence of the culture and inherent risks within the NHS whereby a percentage of this type of incident may always exist. It may also be an outcome of the wide definition of “incidents” used in the NHS, which was described in the introductory chapter. The issue is discussed further in Chapter 4.

3.3 RISK CONTROL SYSTEM FAILURES AND ROOT CAUSES OF INCIDENTS

The HSE root cause analysis methodology used in the research was based on their three-level model, described in Chapter 1. This provided a mechanism for identifying management root causes (level 1 failures) from existing workplace precautions (level 3 failures) using identified risk control system failures (level 2) as a means of linking the two(99). Therefore the identification of risk control system failures for each incident was an important pre-requisite to identifying the underlying management root causes.

50

7

17

25

51

16 16 16

0

10

20

30

40

50

60

Outwith reasonablepracticability

Insufficientinformation

Risk was notforeseeable

Risk managementdecision based onclinical judgement

Numbers of incidents (%)Cost of incidents (%)

38

The process identified in Chapter 2 for assessing incidents against the inclusion criteria and then identifying sufficient details about each incident to allow risk control systems and root causes to be identified consistently proved to be challenging for both the project team and the Trust representatives. The process undoubtedly raised confidence in the research results but it was time consuming to undertake.

In practice the investigators in the individual Trusts tended to be resistant to attempts by the research team to modify their normal approach to investigation. Most issues centered on evidence for conclusions reached regarding root causes of incidents. These normally involved requests from the research team for more information on which to base judgments. These requests were generally liable to be interpreted, particularly by safety practitioners with several years’ experience, as potentially calling their competency into question. Incident investigation seemed to have taken on an almost artistic form with the Trust investigators generally taking the view that they could usually establish a root cause with the most meager of basic information, making assumptions (“gut instinct”) based on and justified by their detailed knowledge of the organisation. This issue will be further discussed in Chapter 4.

Nevertheless, the measures taken, to ensure consistency of investigation between Trusts (described in Chapter 3), were thought to have been effective and that the results can therefore be regarded as robust.

The risk control system failures identified will be described in Section 3.3.1. The management root causes will be described in Section 3.3.2.

3.3.1 Risk control system failures (HSE Level 2(99))

A range of 17 different risk control system failures was identified in phase 1. The number of times a root cause was found in each system is shown in Table 8. The risk control systems ranged (alphabetically) from control of contractors to work organisation. The level 1 management root causes, based on HSE guidance(18), ranged from policy to reviewing. Failures within the risk control system for risk assessment predominated with 48% of the identified root causes assigned to this system. With the exception of the risk control systems for procedures (10%) and training (11%) all of the other systems represented less than 10% of the total percentage of root causes identified.

39

Table 8 number of management root causes associated with risk control system failures in seven NHS trusts (Phase 1)

Root cause Risk control system Policy Implem-

enting Control Plan-

ning Compet-ence

Commun-ication

Coop-eration

Meas-uring

Review-ing

Control of contractors

0 1 1 0 0 1 0 2 2

Disposal 0 6 2 2 0 2 2 5 5 Drug administration

0 1 0 1 0 0 0 0 0

Emergency procedures

0 7 1 4 4 6 4 7 6

Handling sharps 1 0 0 0 0 0 0 0 0 Inspections 0 2 0 2 0 0 0 2 1 Maintenance 1 3 0 2 0 0 0 1 1 Management of violence and aggression

0 9 3 5 3 2 2 7 6

Procedures 4 15 1 4 2 5 2 11 11 Procurement 0 2 0 0 0 0 0 0 0 Responsibilities 0 7 0 0 0 0 0 0 0 Risk assessment 0 60 16 36 9 17 21 58 62 Security 6 0 0 0 0 0 0 0 0 Supervision 1 4 0 1 0 0 0 3 3 Training 1 12 3 10 5 6 6 11 8 Work environment

9 0 0 0 0 0 0 0 0

Work organisation

0 5 2 2 0 0 0 1 1

TOTAL 23 134 29 69 23 39 37 113 109

The most noticeable feature in Table 8 was the extent to which the risk control system of risk assessment dominated the others. For example, the number of times that management root causes were assigned in any of the other systems did not exceed 15, whereas for the system of risk assessment, planning was assigned over 30 times, measuring over 50 times and implementing and reviewing in excess of 60 times each. This indicates that the main management root causes of incidents in the first phase were related to these four root causes in the risk control system of risk assessment.

In the second data collection phase 18 different systems were identified, one more than in the first phase. The pattern is shown in Table 9. Fifteen of the systems were shared between the two phases. Two were unique to the first phase (supervision, and maintenance) while three were only identified in the second phase (control of infection, stress management, and hazard reporting).

40

Table 9 number of management root causes associated with risk control system failures in seven NHS trusts (Phase 2)

Root cause Risk control system Policy Implem-

enting Control Plan-

ning Compet-ence

Commun-ication

Coop-eration

Measur-ing

Review-ing

Control of contractors

0 3 1 3 1 1 1 2 2

Control of infection

0 2 0 1 0 1 0 0 2

Disposal 0 4 0 3 0 2 0 4 4 Drug administration

0 1 0 1 0 0 0 1 1

Emergency procedures

2 7 1 6 1 0 0 6 7

Handling sharps

0 17 7 12 7 9 0 9 10

Hazardreporting

0 1 0 0 0 1 0 1 1

Inspections 0 1 1 1 0 0 0 1 1 Management of violence and aggression

0 21 9 20 18 7 13 22 21

Procedures 8 7 0 3 1 2 1 4 3 Procurement 0 1 0 0 0 1 0 1 1 Responsibilities 0 10 0 0 0 0 0 0 0 Risk assessment

0 51 10 31 12 14 2 39 47

Security 2 1 0 1 0 0 0 1 1 Stress management

2 0 0 0 0 0 0 0 0

Training 1 9 6 8 1 4 1 6 6 Work environment

1 8 5 8 5 0 5 0 0

Work organisation

1 0 0 0 0 0 0 0 0

TOTAL 17 144 40 98 46 42 23 97 107

The total numbers of root causes assigned to risk control systems was not significantly different between phase 1 and phase 2 (Wilcoxon Signed Ranks Test P=0.515). The system of risk assessment was again the dominant source of management root causes, as noted in phase 1 (Table 8), although the percentage had significantly reduced from 48% in phase 1 to 34% in phase 2 (P=0.02). Four other risk control systems had marginally significant reductions in numbers of assigned root causes. These were:

Disposal (P=0.07); Procedures (P=0.05); Training (P=0.06); Work organisation (P=0.07).

However, root causes assigned to the system of management of violence and aggression rose significantly from 6% in the first phase to 21% in the second (P=0.01). The system of handling sharps also increased significantly from a negligible percentage (0.2%) in the first phase to 12%

41

in the second (P=0.02). The system of control of contractors also rose but this increase was only marginally significant (P=0.07).

There were undoubtedly more incidents involving violence and aggression reported in the second phase, as compared to the first. It is unclear whether this was a real increase in incidence of this type of incident or as a consequence of better reporting possibly due to heightened awareness.

In the first phase the only management root cause identified for the system of handling sharps was policy. In the second phase seven additional root causes were evident in the system for handling sharps. These were in the areas of planning, implementing, control competence, communication, measuring and reviewing. The NHS has been engaged in major campaigns to raise awareness regarding the risk of needlestick injuries or potential occupational exposure to blood-borne viruses. It is therefore possible that this has improved reporting.

However, it is also possible that the effect could be caused by a systematic error brought about by increased familiarisation by the research team with the root causation methodology in the second phase when compared with the first. Also, in the root cause methodology, if an incident was identified as not having a policy there was no further assessment against the other criteria. This was because HSE had reasoned that the absence of a policy was indicative of the absence of any other aspects of a management system. This approach was not as strict as might be assumed since, during investigations, the presence of an informal local arrangement that was known to staff was taken to be a policy for the purposes of project data capture. These points are discussed further in Chapter 4.

3.3.2 Management root causes (HSE Level 1)

The number of times a management root cause(99) was identified within the incidents captured for all seven Trusts is shown, for both Phases, in Figure 4. Given there was no significant difference between the test and control data for risk control system failures and management root causes, the data was combined. This was done so that the overall visual impression of differences between the individual management root causes could be revealed.

42

Figure 4 Management root causes in seven NHS Trusts

The range for phase 1 was between 23 times for policy and 134 times for implementing. The four main root causes most frequently identified were implementing (134 times), planning (69 times), measuring (113 times), and reviewing (109 times).

The range of citations for phase 2 was between 10 times for policy and 116 times for implementing. In Phase 2 the number of times management root causes were cited was, on average, approximately 14% less than in the first phase. Control, planning and competence increased by 31%, 13% and 57% respectively. The same four root causes as in phase 1 were the most frequently identified (implementing (116 times), planning (78 times), measuring (81 times), and reviewing (92 times)).

The differences in the numbers of cited management root causes between the two phases were not significant (Wilcoxon rank sum W test P=0.92).

The rank order was broadly similar between the two phases. Table 10 shows the ranks split into two arbitrary groups. These are the major group (cited more than 50 times) and the minor root causes group (cited less than 50 times).

0

20

40

60

80

100

120

140

160

Polic

y

Impl

emen

ting

Con

trol

Plan

ning

Com

pete

nce

Com

mun

icat

ion

Co-

oper

atio

n

Mea

surin

g

Rev

iew

ing

Num

ber o

f man

agem

ent

root

cau

ses

iden

tifie

d

Phase 1 Phase 2

43

Table 10 Rank order of frequency of citation of management root causes

Phase 1 Phase 2 Implementing (134 times) Implementing (116 times) Measuring (113 times) Reviewing (92 times)Reviewing (109 times) Measuring (81 times)

Major management root causes

Planning (69 times) Planning (78 times)Communication (39 times) Control (38 times) Co-operation (37 times) Competence (36 times)Control (29 times) Communication (27 times)Competence (23 times) Co-operation (19 times)

Minor management root causes

Policy (23 times) Policy (10 times)

Implementing, planning, and policy held the same rank between phases (1st, 4th and 9th). Reviewing and measuring shared 2nd and 3rd ranking between phases, and communication, co-operation, control and competency shared the rankings between 5th and 8th. The Spearman order correlation coefficient was 0.695 (P<0.05), indicating a significant correlation between the ranks of the root causes between the two phases.

This ranking was broadly similar to the findings of the pilot work(34), where the main root causes were found within the areas of planning and implementing. Measuring and reviewing were not identified as significant root causes within the pilot, although control and competence were highlighted. These two root causes were among the least frequently cited root causes in the main project. A possible explanation for this could be related to history effects between the pilot work, which was undertaken in 1998, and the main study conducted in 2000 and 2001.

It may also be that the root causes identified in the pilot site were characteristic of that hospital and that each Trust might have a distinctive root cause “profile” related to the relative strengths and weaknesses of its health and safety management system. Evidence for this hypo-report was inconclusive. For example, Trusts A, D, E, F, and G were significantly associated between the first and second phases (P<0.05). However, Trusts B and C were not. The pattern in the first phase for Trust C was not significantly correlated with any other Trust except that of Trust G in the second phase (Spearman’s rank order correlation P=0.03). Conversely Trusts E and G were significantly associated with all of the other Trusts (with the exception of Trust C) (P<0.05). The remaining four Trusts had varying degrees of associations with the others.

Another interpretation might be that more risk assessments were carried out but the main cause of incidents was that these had not been implemented sufficiently robustly. This is consistent with subjective observations within the Trusts whereby, in most cases, measures had been put in place to enhance aspects of the health and safety management system during the period following the first data collection phase. However, unless these were adopted and implemented by line managers and staff, incidents with this root cause were predominant.

This results from the risk control systems failure and root cause analyses was potentially useful information for the participating Trusts, in that it gave an indication of where practical effort might be best directed for maximum impact in reducing incident rates. In other words, if effort was directed at the planning, implementing, measuring and reviewing aspects of the system of risk assessment, then some incidents might be prevented or their severity reduced.

This information was communicated to all participating Trusts at feedback sessions conducted at the end of the second data collection phase.

44

3.4 COSTS OF INCIDENTS

As described in Chapter 2 and Appendices 1 and 2, costs of incidents were collected for both phases.

Those incidents that had management root causes associated with them were assigned costs in both phases, while the costs of the special category incidents were only included in phase 2. This was because, as described in 3.2.1, the need for a special category was only identified during the first data collection phase. Costs were not collected for these incidents during the initial investigation and once the special category groupings were established it was judged that too much time had elapsed to obtain reliable information on which to base cost estimations.

With the exception of some staff opportunity costs, all costings were carried out to the same protocol as in the pilot work(34). This exception was to include costs for staff taken away from their normal duties by the effects of the incident. This was because the bulk of costs associated with incidents were incurred by staff absence and replacement labour. The breakdown across each Trust is shown in Table 11. The costs for each phase incurred by each Trust, is shown along with the costs incurred as a result of the incidents that were allocated to the special category.

Table 11 Costs incurred by Trusts in both phases

13 Note that the costs were at 2000 prices for both data collection phases 14 Figures in (brackets) represent the percentage increase/decrease in cost between phase 1 and phase 2

13 Trust A Trust B Trust C Trust D Trust E Trust F (Control)

Trust G (Control)

Total

Phase 1 127incidents

£3,494 20incidents

£22,285 15incidents

£1,960 16incidents

£3,867 28incidents

£8,470 22incidents

£4,198 16incidents

£6,822 10incidents

£51,096

Phase 2 131incidents

£14,779 (+323%)14 27incidents

£11,272 (-49%) 11incidents

£3,592 (+83%) 13incidents

£7,711 (+99%) 34incidents

£1,207 (-86%) 16incidents

£12,734 (+203%) 20incidents

£1,914 (-72%) 10incidents

£53,209 (+4%)

Phase 2 special category 159incidents

£15,986

21incidents

£256

5incidents

£4,563

13incidents

£13,948

91incidents

£3,852

4incidents

£6,602

9incidents

£10,387

16incidents

£55,593

Phase 1 rate (cost per person working day (WTE*84)

£0.06 £0.09 £0.02 £0.01 £0.07 £0.01 £0.03 £0.04 (mean)

Phase 2 rate (cost per person working day (WTE*84)

£0.27 £0.05 £0.03 £0.02 £0.01 £0.04 £0.01 £0.04 (mean)

46

In the first data collection phase, the total cost of incidents that met the inclusion criteria for all seven Trusts was £51,096 over twelve weeks (127 incidents). The costs incurred by each Trust varied between £1,960 (Trust C) and £22,285 (Trust B). In the second phase, the total cost was £53,209, an increase of 4% from the first phase (131 incidents)15. The range was between £1,914 (Trust G) and £12,734 (Trust F). These were both control Trusts.

Costs varied considerably both between Trusts and within phases. Between the first and second phase costs increased in four Trusts and decreased in three. The largest change occurred in Trust A where an increase of +323% in total costs occurred between the two phases. Trust F (control) increased by just over 200% (+203%). Costs fell for three Trusts (-49% for Trust B, –72% for Trust G (control) and –86% for Trust E). However, the overall total average cost only changed by +4% between phases.

This may represent the inherent variability in costs of incidents since the differences between the test and control Trusts was not significant in either phase (Mann-Whitney U Test: P= 0.86 (phase 1); P=1.0 (phase 2)). Between the first and second phase the costs across all seven Trusts were not found to be significantly different (Wilcoxon rank sum W Test P=0.87).

So far as rates were concerned (cost per person working day (i.e. WTE*84)), these ranged between £0.01 per person working day and £0.27 per person working day. These rates were not significantly different between phases (Wilcoxon rank sum W Test P=0.20)

3.4.1 Staff/patient incidents

Incidents involving staff were nearly three times more frequent and represented approximately four times the cost than those for patients. The distribution of incidents incurred by staff, patients and others (e.g. members of the public or property damage) is shown in Figure 5. It can be seen that the two phases were broadly similar and that most incidents involved staff (70% in phase 1 and 69% in phase 2). Incidents involving patients accounted for 20% and 29% for Phases 1 and 2 respectively. In the “other” category there were 10% in phase 1 and 2% in phase 2.

15 Costs were assigned within both phases at 2000 prices

47

Figure 5 Number of incidents classified as staff, patient or other (not including special category incidents)

Figure 6 Cost of incidents classified as staff, patient or other (not including special category incidents)

26

89

12

38

90

3

0

10

20

30

40

50

60

70

80

90

100

No. of patient incidents No. of staff incidents O ther

No.

of i

ncid

ents

Phase 1 Phase 2

£9,534

£39,826

£1,730

£9,640

£43,508

£60£0

£5,000

£10,000

£15,000

£20,000

£25,000

£30,000

£35,000

£40,000

£45,000

£50,000

No. of patient incidents No. of staff incidents Other

Cos

t

Phase 1 Phase 2

48

When cost was taken into account (Figure 6) this distribution was maintained with incidents involving staff incurring the highest cost (78% in phase 1 and 81% in phase 2). Incidents involving patients accounted for 19% and 18% of the costs incurred in phases 1 and 2 respectively. In the “other” category the costs represented 3% in phase 1 and 1% in phase 2.

3.4.2 Special category

So far as special category incidents were concerned, costs were only obtained for the second data collection phase, as explained earlier. There were 159 special category incidents in the second phase accounting for a total loss of £55,593. This is a significant finding in that the total costs for these incidents was more than either of the two data collection phases for the incidents where root causes could be assigned.

With the exception of Trust B the highest costs were incurred by the primary care trusts. In addition, most special category costs were attributed to incidents involving patients, which is contrary to the findings for those incidents for which root cases could be established, as shown in Figure 7.

Figure 7 Cost of incidents classified as staff, patient or other (phase 2 and special category incidents)

Figure 7 shows that the cost of patient incidents was approximately three and a half times (3.64) that of the captured incidents in phase 2. Special category staff incidents were approximately two and a half times less than patient incidents from phase 2. When the phase 2 and special category costs were combined, costs for incidents involving staff (£59,896) were still in excess of that for patients (£44,736).

£9,640

£43,508

£60

£35,096

£4,146

£16,388

£0

£5,000

£10,000

£15,000

£20,000

£25,000

£30,000

£35,000

£40,000

£45,000

£50,000

Cost of patient Incidents Cost of Staff Incidents Cost of Other

49

Therefore costs for staff were consistently higher than those incurred for patients. Possible reasons for this may be that the management of health and safety risks for patients is more controlled than that for staff. This is discussed further in Chapter 4.

3.4.3 Management root causes of incidents

Each incident usually had multiple management root causes. In the HSE methodology the software had the facility to allow more than one management root cause to be assigned per incident. The software then divided the total cost for the incident by the allocated number of management root causes to obtain a cost per management root cause for that incident. The method did not include an evaluation of the relative importance of each management root cause for each incident so the costs were assigned equally. This may have reduced the sensitivity of the method but ensured reproducibility between incidents and phases.

The total cost assigned to any particular management root cause was then obtained from all of the incidents. The numbers of incidents and their costs allocated to identified management root causes (i.e. not including the special category incidents) for both phases are shown in Table 12.

The pattern and distribution of costs was broadly similar to the root causes associated with failed systems (Table 12), in that the main areas of weakness were related to the categories of planning, implementing, measuring, and reviewing. For example, in Trust E costs decreased for every management root cause in the second Phase while in Trust A only communication cost decreased. The pattern of costs for each Trust was similar for both data collection Phases.

50

Tabl

e 12

Num

ber o

f inc

iden

ts a

nd c

osts

per

man

agem

ent r

oot c

ause

16 P

hase

2 d

ata

in (b

rack

ets)

Trus

t A

Trus

t B

Trus

t C

Trus

t D

Trus

t E

Trus

t F

(Con

trol)

Trus

t G

Cont

rol)

Ove

rall

Tota

ls Fo

r 7

Trus

ts N

umbe

r of i

ncid

ents

20

(2

7)16

15 (1

1)16 (1

3)28 (3

4)22 (1

6)16 (2

0)10 (1

0)12

7(1

31)

Ave

rage

cos

t per

man

agem

ent r

oot c

ause

Po

licy

£7

(£56

2)

£31

(£23

9)

£1,3

20

(£0)

£1

58

(£5)

£3

2 (£

3)

£190

(£

5,48

7)

£263

(£

132)

£2

,001

(£

6,42

8)

Impl

emen

ting

£792

(£

4,69

7)

£5,7

26

(£3,

963)

£3

01

(£1,

239)

£9

61

(£3,

070)

£1

,420

(£

260)

£8

03

(£3,

387)

£2

,334

(£

323)

£1

2,33

7 (£

16,9

39)

Con

trol

£179

(£

321)

£2

8 (£

389)

£2

6 (£

779)

£1

38

(£44

) £1

,255

(£

17)

£477

(£

0)

£0

(£11

) £2

,103

(£

1,56

1)

Plan

ning

£3

49

(£3,

652)

£2

,785

(£

385)

£2

5 (£

939)

£3

01

(£1,

103)

£2

37

(£23

0)

£519

(£

1,17

8)

£785

(£

303)

£5

,001

(£

7,79

1)

Com

pete

nce

£0

(£1,

398)

£2

8 (£

10)

£0

(£52

) £1

51

(£42

) £1

70

(£12

) £1

82

(£11

0)

£245

(£

270)

£7

76

(£1,

895)

C

omm

unic

atio

n £1

86

(£36

) £0

(£

7)

£6

(£0)

£2

21

(£56

2)

£1,3

18

(£21

) £4

94

(£1,

235)

£2

(£

270)

£2

,227

(£

2,13

0)

Co-

oper

atio

n £3

57

(£1,

094)

£3

,565

(£

0)

£0

(£2)

£1

71

(£44

9)

£225

(£

193)

£3

0 (£

885)

£2

(£

0)

£4,3

50

(£2,

622)

M

easu

ring

£812

(£

1,49

0)

£5,3

42

(£3,

332)

£0

(£

26)

£883

(£

1,17

3)

£2,4

78

(£23

7)

£794

(£

279)

£2

,327

(£

303)

£1

2,63

6 (£

6,84

1)

Rev

iew

ing

£812

(£

1,52

7)

£4,7

80

(£2,

947)

£2

82

(£3,

592)

£8

83

(£1,

262)

£1

,335

(£

235)

£7

09

(£17

4)

£864

(£

303)

£9

,665

(£

7,00

2)

TRU

ST T

OTA

L £3

,494

(£

14,7

79)

£22,

285

(£11

,272

) £1

,960

(£

3,59

2)

£3,8

67

(£7,

711)

£8

,470

(£

1,20

7)

£4,1

98

(£12

,734

) £6

,822

(£

1,91

4)

£51,

096

(£53

,209

)

52

Total cost increased in four Trusts (Trusts A, C, D and F (control)) and decreased in the other three (Trusts B, E and G (control)). The total costs incurred increased by approximately 4% between the two phases.

Statistical testing revealed that the management root cause costs assigned to the test and control trusts, were not significantly different from one another in either phase (Mann-Whitney U-Test phase 1 P=0.57, phase 2 P=0.38). The costs for management root causes in each Trust were not significantly different between phases (Wilcoxon rank sum W Test P=0.87). Spearman’s rank-order correlation indicated a correlation coefficient between both phases of 0.334 (P<0.01) indicating a significant association (i.e. no significant change between the two phases).

The cumulative distribution of individual incident costs (i.e. from highest to lowest cost, not cumulative over time) across all seven Trusts, for both phases, is shown in Figure 8. The majority of incidents were relatively low cost (for example, in Trust A during the first data collection phase, a small group of 6 incidents accounted for approximately 85% of the total cost incurred).

Figure 8 Cumulative % accident costs

This is in keeping with the HSE concept of accident ratio pyramids, whereby there was an inverse relationship between the severity of the incident outcome and the number of incidents that exhibited that outcome(36). Also, the universal Pareto 80:20 law appears to broadly apply,

0

20

40

60

80

100

0 20 40 60 80 100

Cumulative % numbers of incidents

Cum

ulat

ive

% c

ost

Trust A PCT Trust B Trust C PCTTrust D PCT Trust E Trust F UH (Control)Trust G PCT (Control) Trust A PCT (2) Trust B (2)Trust C PCT (2) Trust D PCT (2) Trust E (2)Trust F UH (2) (Control) Trust G (2) PCT (Control)

53

whereby approximately 80% of the costs were attributed to approximately 20% (actually 10% - 40%) of the observed incidents(111).

This is potentially useful to NHS Trusts if they could focus on prioritising attention on the 20% highest cost incidents to influence safety performance. To further evaluate whether it is possible to predict for the Trusts which incidents on which to concentrate, further examination of the categories, factors and types of incidents is included in the following section (Section 3.4.4.)

3.4.4 Categories, factors and types of incidents

As described in Chapter 2, the costs arising from the circumstances and outcome of each incident was further described by assigning them to relevant headings within each of three different groups of incident category, factor and type. Appendix 7 contains a short description of each incident and their cost, category, and type.

Incident category Categories of incidents are shown in Figure 9.

Figure 9 Total cost per incident category (both phases)

The total costs incurred as a result of accidental injury were an order of magnitude higher than any other category in both phases. Costs from physical violence were nearly four times that of the first phase.

£0

£5 ,000

£10 ,000

£15 ,000

£20 ,000

£25 ,000

£30 ,000

£35 ,000

£40 ,000

£45 ,000

£50 ,000

Acci

dent

al in

jury

Phys

ical

vio

lenc

e

Fata

lity

Thef

t

Nea

r mis

s

Acci

dent

al p

rope

rty d

amag

e

Vand

alis

m

Oth

er

Verb

al a

buse

/thre

aten

ing

beha

viou

r

Fire

Ill h

ealth

A c cid e n t ca teg o ry

Cos

t

P ha se 1 P ha se 2

54

Table 13 shows the average cost per incident category. It can be seen that most incidents were assigned to the category of accidental injury (65% in phase 1 and 56% in phase 2). None of the other categories exceeded 5% of the total cost, with the exception of physical violence (10% in phase 1 and 15% in phase 2) and near miss in the first phase (12%).

Table 13 Average costs per incident category (all Trusts)

No costs were incurred for the category of ill health in phase 1. This was likely to be due to the relatively long timescales involved, but may also be related to relatively poor reporting mechanisms for ill health, as compared to other incidents, and therefore less likely to be picked up by the research design. Three cases of ill health were identified in the second phase. These were:

�� A member of staff absent from work due to prolonged work related stress; �� An enrolled nurse contracted an elbow infection from a splash of urine; �� Pressure sores were identified in an elderly patient.

This general lack of incidents from the category of ill health is important to consider when interpreting the costing data, since it is likely that this represents an underestimate of the true cost to the NHS of inadequate health and safety management. For example, it seems inconceivable that there could have only been one case of pressure sores in the whole sample from both data collection phases. It seems more likely that the study design or the in-house Trust reporting system or culture did not capture them. This is discussed further in Chapter 4.

Incident factor Total cost by incident factor for both phases is shown in Table 14.

Incident category (phase 1 total cost)

No. of incidents

Average cost/category of incident (phase 1)

Incident category (phase 2 total cost)

No. of incidents

Average cost/category of incident (phase 2)

Accidental injury (£44,529)

83 £537 Accidental injury (£32,013)

74 £433

Physical violence (£2,409)

13 £185 Physical violence (£11,075)

20 £554

Fatality (£2,091) 1 £2,091 Fatality (£1,718) 3 £573 Theft (£1,171) 2 £586 Theft (£0) 0 0 Near miss (£211) 15 £14 Near miss (£426) 5 £85 Accidental property damage (£308)

3 £103 Accidental property damage (£0)

0 0

Vandalism (£148) 4 £37 Vandalism (£281) 1 £281 Other (£131) 5 £26 Other (£2,122) 16 £133 Threatening behaviour (£72)

1 £72 Threatening behaviour (£160)

6 £27

Fire (£0) 0 0 Fire (£16) 1 £16 Ill health (£0) 0 0 Ill health (£5,397) 3 £1,799

55

Table 14 Total cost per incident factor

Incident factor Cost (phase 1)

Incident factor Cost (phase 2)

Absence £21,350 Absence £25,798 Replacement labour £18,122 Replacement labour £18,130 Patient treatment £9,184 Patient treatment £7,846 Damage repair/staff treatment

£2,344 Damage repair/staff treatment

£739

Initial response to accident £73 Hiring/purchasing £400 Transport £67 Transport £120 Hiring/purchasing £20 Lost production/ wasted time £119 Cleaning up £4 Initial response to accident £58 Lost production/ wasted time

£0 Cleaning up £0

Staff absence incurred the greatest total cost in both phases, followed closely by replacement labour. These two Factors accounted for approximately 75% of the total costs. Patient and staff treatment accounted for approximately 20% of costs with the remaining 5% spread between the remaining factors.

The costs per incident factor were not significantly different when the test Trusts were compared with the control Trusts (Kruskal-Wallis one-way analysis of variance (P=0.85 (Phase 1); P=0.77 (Phase 2)). For all Trusts the differences between the two phases were also not significant (Wilcoxon rank sum W test P=0.33). Spearman’s rank order correlation coefficient was 0.76 (P<0.01) indicating a significant correlation between the factors for all Trusts across both phases (i.e. no significant change between the two phases).

Incident type The total cost by incident type is shown in Table 15. This shows some variation between the two phases, although the same three types incurred most of the cost in both phases. These were person to person assault, slip trip or fall on the same level/stairs and patient lifting/handling. These three types accounted for 71% of the total cost in the first phase and 78% in the second.

For all Trusts the differences in total cost in incident types between the two phases were not significant (Wilcoxon rank sum W test P=0.69). Spearman’s rank order correlation coefficient was 0.76 (P<0.01) indicating a significant correlation between the types for all Trusts across both phases (i.e. no significant change between the two phases).

56

Table 15 Total cost per incident type (all Trusts)

Incident type Phase 1 Incident type Phase 2 Patient lifting/handling £22,565 Slip, trip or fall on same level/stairs £20,231 Slip, trip or fall on same level/stairs £10,874 Patient lifting/handling £12,223 Person to person assault £2,974 Person to person assault £9,273 Struck by an object £2,909 Other £6,369 Manual lifting/handling £2,583 Manual lifting/handling £1,848 Other £2,500 Fall from height £1,231 Fall from height £2,091 Struck by an object £481 Contact with equipment/machinery £1,824 Needlestick/sharps injury £431 Struck against something £1,797 Contact with electricity £400 Cut with sharp material/object £676 Exposure to fire £297 Needlestick/sharps injury £268 Hot or cold contact £243 Hot or cold contact £28 Cut with sharp material/object £101 Exposure to harmful substance £7 Struck against something £53 Exposure to fire £0 Contact with equipment/machinery £24 Contact with electricity £0 Exposure to harmful substance £4 TRUST TOTAL £51,096 TRUST TOTAL £53,209

Three groups of incident types were apparent:

�� Those that tended to exceed £10,000 total cost (slip trip or fall on the same level/stairs and patient lifting/handling);

�� Those that tended to exceed £5,000 total cost (other and person to person assault); �� Those that tended to fall below £5,000 total cost (the remaining types).

This means that the first two groups of type of incidents were the most costly for the participating Trusts.

The other group (64 incidents (32 in each phase)) was those that did not fit within an incident type. Examples of these were:

�� A fire alarm was activated by burnt toast. A patient who was unsupervised was making the toast, despite a Trust policy that all toast making should be supervised17;

�� A puppy in a patient’s home bit a community psychiatric nurse; �� A visitor, driving a high-topped van, struck an overhead archway while driving under an

archway in the hospital.

The classification in the HSE methodology did not include groupings for incident types involving animals or transport. It is not known why this was done, given that these two types are included in the Reporting of Injuries, Diseases, and Dangerous Occurrences Regulations 1995 (RIDDOR)(12). These may be useful additional Types to include in any future revisions of the methodology.

The average incident type cost is shown in Table 16. 17 This incident was not assigned to the exposure to fire type because a fire had not occurred and there was evidence that toaster was not Trust property

57

Tabl

e 16

Ave

rage

cos

t per

inci

dent

type

(all

Trus

ts)

Inci

dent

type

(pha

se 1

tota

l cos

t) N

o. o

f in

cide

nts

Aver

age

cost/

type

of i

ncid

ent

(pha

se 1

) In

cide

nt ty

pe (p

hase

2 to

tal c

ost)

No.

of

inci

dent

s Av

erag

e co

st/ty

pe o

f inc

iden

t (p

hase

2)

Fall

from

hei

ght (

£2,0

91)

1 £2

,091

Sl

ip/tr

ip/fa

ll on

sam

e le

vel/s

tairs

(£

20,2

31)

14

£1,4

45

Patie

nt li

fting

/han

dlin

g (£

22,5

65)

15

£1,5

04

Patie

nt li

fting

/han

dlin

g (£

12,2

23)

14

£873

St

ruck

by

an o

bjec

t (£2

,909

) 3

£970

Pe

rson

to p

erso

n as

saul

t (£9

,273

) 15

£6

18

Slip

/trip

/fall

on sa

me

leve

l/sta

irs

(£10

,874

) 15

£7

25

Man

ual l

iftin

g/ha

ndlin

g (£

1,84

8)

4 £4

62

Man

ual l

iftin

g/ha

ndlin

g (£

2,58

3)

8 £3

23

Con

tact

with

ele

ctric

ity (£

400)

1

£400

C

onta

ct w

ith e

quip

men

t/mac

hine

ry

(£1,

824)

6

£304

Fa

ll fr

om h

eigh

t (£1

,231

) 6

£205

Pers

on to

per

son

assa

ult (

£2,9

74)

13

£229

O

ther

(£6,

369)

32

£1

99

Stru

ck a

gain

st so

met

hing

(£1,

797)

9

£200

St

ruck

by

an o

bjec

t (£4

81)

3 £1

60

Cut

with

shar

p m

ater

ial/o

bjec

t (£6

76)

4 £1

69

Expo

sure

to fi

re (£

297)

2

£149

O

ther

(£2,

500)

32

£7

8 H

ot o

r col

d co

ntac

t (£2

43)

2 £1

22

Nee

dles

tick/

shar

ps in

jury

(£26

8)

16

£17

Cut

with

shar

p m

ater

ial/o

bjec

t (£

101)

4

£25

Hot

or c

old

cont

act (

£28)

2

£14

Nee

dles

tick/

shar

ps in

jury

(£43

1)

26

£17

Expo

sure

to h

arm

ful s

ubst

ance

(£7)

1

£7

Stru

ck a

gain

st so

met

hing

(£53

) 4

£13

Expo

sure

to fi

re (£

0)

0 £0

C

onta

ct w

ith e

quip

men

t/mac

hine

ry

(£24

) 3

£8

Con

tact

with

ele

ctric

ity (£

0)

0 £0

Ex

posu

re to

har

mfu

l sub

stan

ce (£

4)

1 £4

58

As with total costs, the incident types average cost could be grouped into three:

�� Those that tended to exceed £1,000 average cost (fall from height (phase 1 only), patient lifting/handling (phase 1 only), and slip, trip or fall on the same level/stairs (phase 2 only));

�� Those that tended to exceed £500 average cost (struck by an object (phase 1 only), slip, trip or fall on the same level/stairs (phase 1 only), patient lifting/handling (phase 2 only) and person to person assault (phase 2 only));

�� Those that tended to fall below £500 average cost (the remaining types).

As with total cost, the same three types of incident had the highest average cost (i.e. person to person assault, slip trip or fall on the same level/stairs and patient lifting/handling).

However, struck by an object and fall from height also fall within the high cost group. this was due to a small number of high cost incidents in phase 1.

When this finding was examined further, the differences in average cost in incident types between the two phases were not significant (Wilcoxon rank sum W test P=0.68). However, Spearman’s rank order correlation coefficient was 0.45 (P<0.1) indicating a marginal rejection of the null hypo-report of no correlation (i.e. no significant difference between the two phases.

Table 17 shows the range of costs for each incident type.

Table 17 Range of costs per incident type (all Trusts)

As with total and average cost, incident type cost ranges could be grouped into three:

�� Those where the cost range tended to exceed £1,000. Approximately half of the incident Types (47%) fell within this grouping (patient lifting/handling, slip, trip or fall on the same level/stairs, struck by an object, manual lifting/handling, struck against something (phase 1 only), person to person assault (phase 2 only) and other);

18 NA = not applicable as there was either none or only a single incident in this type.

Incident type Cost range(phase 1)

Cost minimum/ maximum (phase 1)

Cost range (phase 2)

Cost minimum/ maximum (phase 2)

Patient lifting/handling £9,847 £2 -£9,849 £2,682 £6 -£2,688 Slip/trip/fall on same level/stairs £5,706 £7 -£5,712 £7,022 £57 -£7,078 Struck by an object £2,244 £66 -£2,310 £7,260 £2 -£7,262 Struck against something £1,334 £5 -£1,339 £35 £3 -£37 Manual lifting/handling £1,204 £5 -£1,209 £1,525 £14 -£1,540 Other £1,064 £4 -£1,069 £3,383 £2 -£3,385 Contact with equipment/machinery £704 £6 -£710 £17 £2 -£19 Person to person assault £667 £22 -£689 £7,260 £2 -£7,263 Cut with sharp material/object £227 £10 -£237 £64 £4 -£68 Needlestick/sharps injury £27 £4 -£31 £96 £2 -£98 Hot or cold contact £26 £14 -£40 £26 £109 -£134 Fall from height NA18 NA £650 £6 -£650 Exposure to fire NA NA £265 £16 -£280 Contact with electricity NA NA NA NA Exposure to harmful substance NA NA NA NA

59

�� Those that tended to exceed £100 cost range, which represented approximately one third (27%) of the types (contact with equipment/machinery (phase 1 only), cut with sharp material/object (phase 1 only), fall from height (phase 2 only) and exposure to fire (phase 2 only));

�� Those that tended to fall below £100 average cost, which represented approximately one tenth (13%) of the incident types (needlestick/sharps, and hot or cold contact).

Two of the types were not included (contact with electricity and exposure to harmful substance). This is because a range could not be established for either phase because either zero or one incident occurred in either type.

For all Trusts the differences in cost range for incident types between the two phases was not significant (Wilcoxon rank sum W test P=0.35). Spearman’s rank order correlation coefficient was 0.59 (P<0.05) indicating a significant correlation with the range of types in both phases (i.e. no significant change between the two phases).

It should be noted that the incident types in the first grouping (where the range tended to exceed £1,000) all had minimum costs that were less than £10 with the exception of one type (struck by an object), which was less than £100. This represents weak evidence that many of the incident types could incur high costs. However, there is much stronger evidence that some types consistently tended to incur higher costs. Using the cost range data as the measurement criteria these were (alphabetical list):

�� Manual lifting/handling; �� Other; �� Patient lifting/handling; �� Person to person assault; �� Slip/trip/fall on same level/stairs; �� Struck against something; �� Struck by an object.

This indicates a potential cost-effective prioritisation strategy for Trusts, based on placing effort to control risks from incident types that incur highest cost.

Whether they should be prioritised in this way is, according to Hale(112), linked to whether they had the same pattern of underlying causes. As has been shown earlier (section 4.3.3.), the majority of incidents had the risk control system of risk assessment and, specifically, the management root causes of planning and implementing. This is evidence that the underlying causes were the same and that the above strategy could be potentially effective.

3.5 QUESTIONNAIRE DATA

As described in Chapter 2, the safety climate questionnaire used was that developed by Cox and Cheyne(58). Each questionnaire consisted of 43 questions aimed at eliciting individual views on nine dimensions, grouped into four categories. These were:

Category (a): Organisational context covering four dimensions: 1. Management commitment 2. Communication 3. Priority of safety 4. Safety rules & procedures

60

Category (b): Social environment covering two dimensions: 5. Supportive environment 6. Involvement

Category (c): Individual appreciation covering two dimensions: 7. Personal priorities and need for safety 8. Personal appreciation of risk

Category (d): Work environment covering a single dimension: 9. Physical work environment

Within this section of the Chapter the response rates and job category distribution will be described followed by general analysis of the dimension score results and the responses grouped by Trust, job category, and individual, across both data collection phases. The section ends with analysis of questionnaire respondent’s additional comments.

3.5.1 Response rates

The numbers of questionnaires issued, received and derived response rates are shown in Table 18. The data for phase 2 are shown in brackets. For example in Trust A, 82 questionnaires were issued at the start of phase 1. By the end of the second phase 43 valid questionnaires had been returned. The overall project response rate was therefore 52%. Since the questionnaire sample was 10% of Trust staff this represented 5.2% of the total Trust staff (final column).

Table 18 Response rates to the questionnaires issued (1st & 2nd phases19)

Trust Number of questionnaires issued

Number of completed questionnaires received

%Response rate

Change in response rate between phases

Overall project response rate (%)

% of Trust staff

Trust A 82 (56) 56 (43) 68% (77%)

+ 13% 52% 5.2%

Trust B 338 (171) 171 (114) 51% (67%)

+31% 34% 3.4%

Trust C 194 (81) 81 (74) 42% (91%)

+117% 38% 3.8%

Trust D 561 (337) 337 (274) 60% (81%)

+35% 49% 4.9%

Trust E 185 (135) 135 (94) 74% (70%)

-5% 51% 5.1%

Trust F (Control)

592 (220) 220 (145) 37% (66%)

+78% 25% 2.5%

Trust G (Control)

397 (207) 207 (167) 52% (81%)

+56% 42% 4.2%

TOTAL 2349 (1207) 1207 (911) MEAN 55%

(76%) +46% 42% 4.2%

SD 13% (9%) +41% 10% 1.0%

Therefore, it can be seen that in the first data collection phase 2439 questionnaires were issued with 1207 returned satisfactorily completed (55% response rate). In the second phase 1207 questionnaires were issued to the same individuals as before. This resulted in 911 valid 19 2nd benchmarking data in (brackets)

61

questionnaires being returned (76% response rate). This group of respondents represents the individuals who completed two valid questionnaires resulting in an overall project response rate of 42%, representing 4.1% of the total headcount available to the project across the seven Trusts.

The highest response rate in the first data collection phase was 74% (Trust E) and the lowest 37% (Trust F (control)). The rates in the second phase varied between 91% (Trust C) and 66% (Trust F). The differences across Trusts were not significantly different (Kolmogorov-Smirnov Test P=0.94).

The response rates were all higher than in the first phase (between +13% for Trust A and +117% for Trust C), with the exception of Trust E, which fell by 5%. This was not anticipated in that the proportion of those who responded during the first phase represented the entire pool for the second. Reasons why Trust E differed from the other Trusts are unclear.

As described in Chapter 3, efforts were made to optimise the response rate throughout the 12 week sampling phases by sending a reminder letter. However, it was regarded as important that the questionnaires were completed during the same time period as the set of reported incidents. Therefore individuals who had not returned a valid questionnaire after the end of these time periods were not followed up any further.

This data represent an overall project response rate for all seven Trusts of 42% (between 24% (Trust F (control)) and 52% (Trust A)). This is an average of 4.1% of all staff employed during the data collection phases. This is regarded as a good response rate, given the challenge of ensuring that the same individual completed and returned two questionnaires with a year between the events.

Since the responses from individuals in the second phase was dependent on those returning completed questionnaires during the first phase it was necessary to check that each occupational group was adequately represented in both phases. The occupational group data was compared across the two phases and the results represented in Figure 10. This confirms that, despite fewer questionnaires being received in the second phase, a broad balance of respondents across occupational groups was maintained between the two phases and that no group was under-represented.

62

Figure 10 Questionnaire job category distribution

3.5.2 Questionnaire confirmatory factor analysis

The questionnaire had been developed for the offshore industry(58). To ensure that the scales in the original questionnaire were appropriate for use in a healthcare setting, confirmatory factor analysis (CFA) was undertaken.

The data was subjected to CFA using version 6 of the EQS (structural equation) program(96). The original nine-factor model(58) had a relatively good fit for the data (CFI=0.85)20. The CFA for the healthcare data revealed a relatively good fit for the data with no new factors revealed (CFI=0.84).

Internal-scale reliability (or consistency) was examined to investigate the degree to which the various questions measured different aspects of the same concept. The accepted level for this statistic (Cronbach’s alpha) is approximately 0.7(58). All the alphas reported were in the range 0.62 to 0.89 (original range 0.53 to 0.84), which was regarded as acceptable reliability.

It was therefore concluded that the original scales were appropriate for use in a healthcare setting, without modification. However, the instrument could not be regarded as able to distinguish whether a lower opinion score was indicative of a more critical attitude as a result of climate improvement.

20 Cox and Cheyne identified that a value of approximately CFI=0.9 was accepted as indicating good model fit.

0 1 2 3 4 5 6 7 8 9 10

Estates

Senior manager

Professional, Technical & PAMs (qualified)

Medical & Dental

Other

Domestic & Catering

Nursing & Midwifery (qualified)

Professional, Technical & PAMs (unqualified)

Nursing & Midwifery (unqualified)

A&C

Nurse manager

Unknown

Staf

f gro

up

Questionnaires received as a % of total trust staff

Phase 1 Phase 2

63

3.5.3 Dimension score responses

The dimension scores were calculated from the responses for each of the nine dimensions(59). The resulting standardised dimension scores then had the following meaning (with a score below six (scale mid-point) indicating increasingly low opinion(58)):

2 - strongly disagree 4 – disagree 6 – neither agree nor disagree 8 – agree 10 – strongly agree

Summary dimension scores obtained for both data collection phases are shown in Table 19.

Table 19 Questionnaire mean response dimension scores (1st & 2nd phases)21)22

Dimension�

Trust �

Manage-ment commit-ment

Commun-ication

Priority of safety

Safetyrules

Supp-ortive environ-ment

Involve-ment

Pers-onal prio-rities

Apprec-iation ofrisk

Workenviron-ment

MeanTrust score

Trust A 6.4 (6.6)

6.5(6.5)

6.3(6.7)

6.2(6.0)

7.0(7.0)

6.4(6.7)

7.8(7.6)

6.7(6.7)

5.9(6.0)

6.6(6.7)

Trust B 7.2 (7.0)

6.8(6.8)

7.0(7.0)

7.0(7.0)

7.3(7.4)

6.9(7.1)

7.9(8.1)

7.0(7.0)

6.3(6.2)

7.0(7.1)

Trust C 6.6 (7.0)

6.6(6.7)

6.8(7.1)

6.3(6.5)

7.2(7.4)

6.8(6.9)

8.0(8.1)

6.9(6.9)

5.7(5.8)

6.8(6.9)

Trust D 7.2 (7.2)

6.8(6.9)

7.1(7.2)

6.8(6.9)

7.3(7.5)

6.9(7.0)

7.9(8.1)

7.1(7.2)

6.1(6.2)

7.0(7.1)

Trust E 6.8 (6.7)

6.4(6.4)

6.8(6.7)

6.4(6.3)

7.1(7.1)

6.5(6.8)

7.6(7.8)

7.1(7.1)

6.0(5.8)

6.7(6.8)

Trust F (Control)

6.7(6.7)

6.5(6.4)

6.5(6.5)

6.5(6.4)

7.0(7.2)

6.5(6.9)

7.6(7.7)

6.8(6.9)

5.7(5.6)

6.6(6.7)

Trust G (Control)

7.0(6.9)

6.7(6.6)

7.0(6.9)

6.9(6.8)

7.3(7.3)

6.9(6.9)

7.8(7.9)

7.2(7.1)

6.1(6.2)

7.0(7.0)

Mean 6.9 (6.9)

6.6(6.6)

6.9(6.9)

6.7(6.7)

7.2(7.3)

6.7(6.9)

7.8(7.9)

7.0(7.1)

6.0(6.0)

6.9(6.9)

Median 7.1 (7.1)

6.8(6.8)

7.0(7.0)

6.7(6.7)

7.3(7.3)

6.7(7.3)

8.0(8.0)

7.0(7.5)

6.0(6.0)

6.9(7.0)

SD 1.4 (1.4)

1.3(1.3)

1.5(1.5)

1.4(1.5)

0.9(0.9)

1.6(1.6)

1.0(1.1)

1.3(1.2)

1.4(1.4)

0.9(0.9)

Gross change for most scores was either an increase (53%) or no change (24%). The scores for the remaining 23% decreased. There was also a visual impression that three of the nine dimension scores (supportive environment, involvement and personal priorities) were consistently higher between the two phases, across all the Trusts. The mean Trust score was also higher for all Trusts with the exception of Trust G, which stayed the same.

Statistical testing was used to examine further these trends. The null hypo-report used for interpretation of the results was that an increase in dimension score represented an improvement

21 2nd benchmarking data in (brackets) 22 The mean dimension scores are out of a possible score of 10 and, the higher the score, the more positive the opinion. A score of 6.0 equals a neutral opinion so any scores above this level can be regarded as marginally positive.

64

in opinion of the aspect of safety culture. However, a decrease in score could also represent greater awareness of safety culture whereby individuals had become more critical. This could represent an early stage in improving safety culture and is discussed further in Chapter 4.

3.5.4 Analysis of responses by Trust

The dimension scores represented ordinal data. Non-parametric hypo-report testing was therefore used initially to compare responses across all seven Trusts during both data collection phases. Kruskal-Wallis One-way Analysis of Variance for independent samples was chosen as a validated, reliable test for data that are amenable to ranking(103, 106, 113). The results are shown in Table 20. All were highly significant (P<0.01) indicating that the null hypo-report (that the responses were the same from each Trust) should be rejected and the responses treated as if they were from seven distinct sources.

Table 20 Kruskal-Wallis test for mean dimension scores across both phases

Dimension (grouped by Trust) Chi-square Degrees of freedom Significance Mean response per person 75.010 13 0.000 Management commitment 68.748 13 0.000 Communication 43.138 13 0.000 Priority of safety 56.341 13 0.000 Safety rules 57.315 13 0.000 Supportive environment 49.581 13 0.000 Involvement 31.886 13 0.002 Personal priorities 42.518 13 0.000 Personal appreciation of risk 31.677 13 0.003 Work environment 50.331 13 0.000

As discussed in Chapter 2, where possible, the use of parametric testing was the preferred approach for this data provided it was not characterised by gross inequality of intervals(108, 113, 114). To do this the distribution of the data was examined to check whether it approximated to normality. The distribution of the data was examined and found to be close to normality. The means and medians of the questionnaire responses were found to overlap within one standard deviation of the mean in every dimension for both data collection phases for each Trust (P<0.01). Since the mean and median are the same in a normal distribution this was strong evidence of approximate normality within the data. It was decided that this evidence, combined with the near-normal distribution shape, was sufficiently persuasive to proceed with ANOVA.

When ANOVA was carried out the mean responses to the various dimensions for each Trust were significantly different from each other for both phases of the project (P<0.001). This confirmed the result from the Kruskal-Wallis test that the null hypo-report (that the questionnaire responses were not different between individual Trusts) should be rejected. This finding held true when mean dimension scores were analysed within each Trust (Tables 21 and 22). This showed that, for the first data collection phase, all dimension scores differed between Trusts (P<0.05). For the second data collection phase all scores differed between Trusts (P<0.01) with the exception of the dimensions of involvement and appreciation of risk.Subsequent analysis of pooled questionnaire data was therefore avoided with responses from individual Trusts analysed individually.

65

Table 21 One-way ANOVA for mean dimension scores (1st data collection phase)

Sum of squares df Mean square F Sig. Management commitment Between Trusts 78.78 6 13.13 7.276 .000 Communication Between Trusts 31.86 6 5.311 3.300 .003 Priority of safety Between Trusts 63.47 6 10.57 4.859 .000 Safety rules Between Trusts 61.40 6 10.23 5.297 .000 Supportive environment Between Trusts 23.10 6 3.851 4.385 .000 Involvement Between Trusts 40.20 6 6.700 2.717 .013 Personal priorities Between Trusts 21.19 6 3.533 3.235 .004 Appreciation of risks Between Trusts 33.98 6 5.665 3.529 .002 Work environment Between Trusts 52.32 6 8.721 4.787 .000 Mean response/person Between Trusts 35.66 6 5.944 7.339 .000

Table 22 One-way ANOVA for mean dimension scores (2nd data collection phase)

Sum of squares

df Mean square F Sig.

Management commitment Between Trusts 40.965 6 6.827 3.388 .003 Communication Between Trusts 33.515 6 5.586 3.180 .004 Priority of safety Between Trusts 54.079 6 9.013 4.141 .000 Safety Between Trusts 71.267 6 11.878 5.343 .000 Supportive environment Between Trusts 16.185 6 2.697 2.993 .007 Involvement Between Trusts 9.535 6 1.589 .647 .693 Personal priorities Between Trusts 25.697 6 4.283 3.909 .001 Appreciation of risks Between Trusts 16.383 6 2.730 1.764 .103 Work environment Between Trusts 44.606 6 7.434 3.905 .001 Mean response/person Between Trusts 27.937 6 4.656 5.571 .000

3.5.5 Analysis of responses by job category

The responses for each Trust were categorised according to the twelve job categories identified in Chapter 2. Multiple comparison testing was carried out, using the Scheffe test. The null hypothesis was that there was no difference in mean response between staff groups within each Trust.

Trust F (a control Trust) was the only Trust to have significant differences between staff groups. Table 23 shows these differences. The mean questionnaire responses for the A&C staff differed significantly from the nursing and midwifery (unqualified) group. The medical and dental group differed significantly from the nursing and midwifery group (both qualified and unqualified), the professional, technical and PAMs (both qualified and unqualified) group, and the senior manager group. This finding was mostly confined to the first data collection phase (with the exception of the nursing and midwifery (unqualified) group, and the professional, technical and PAMs (unqualified) group. Possible explanations for this finding will be discussed further in Chapter 4.

66

Table 23 Trust F Multiple Comparisons by Job Category (Showing Only Significant Associations) †††

Job Category23 �� Medical & Dental A&C

Nursing and Midwifery (qualified) <0.01 (phase 1) Nursing and Midwifery (unqualified) <0.001 (phase 1) <0.05 (phase 2) <0.05 (phase 1) Professional, Technical & PAMs (qualified) <0.01 (Phase 1) Professional, Technical & PAMs (unqualified) <0.05 (both phases) Senior Manager <0.05 (phase 1)

3.5.6 Analysis of responses by individual members of staff

The research design meant that two questionnaires were available for each individual, with an approximate year in between each questionnaire. Since it was already known that the responses from within each Trust were different individual the data was analysed to investigate whether any change could be detected in individual responses between the two phases by grouping the responses, across all job categories, by individual Trust (Table 24).

Table 24 Paired samples T-tests (individual responses grouped by Trust)

Dimension Asymp. Sig. (2-tailed) Trust

ATrust B

Trust C Trust D Trust E

Trust F

Trust G

Management commitment +0.453 -0.200 +0.127 +0.714 -0.786 +0.805 -0.226 Communication +0.869 +0.705 +0.341 +0.529 +0.774 -0.153 -0.246 Priority of safety +0.160 +0.486 +0.255 +0.025* -0.178 -0.787 -0.668 Safety rules & procedures -0.403 -0.425 +0.036* +0.005* -0.807 -0.090 -0.578 Supportive environment +0.617 +0.728 +0.117 +0.021* +0.982 +0.236 -0.557 Involvement +0.405 +0.608 +0.718 +0.218 +0.065 +0.060 -0.416 Personal priorities/ need for safety

-0.368 +0.720 +0.618 +0.004* +0.067 -0.361 +0.145

Personal appreciation of risk +0.688 +0.776 +0.634 +0.011* +0.690 +0.262 -0.584 Physical working environment +0.566 -0.586 +0.216 +0.135 -0.227 -0.117 +0.788 Mean response per person +0.499 -0.970 +0.079 +0.002* +0.839 -0.747 -0.387

* = Differences significant at the 0.05 level

The only significant change in opinion was found in Trust D, with positive significant changes in 6 of the 9 dimensions. Trust D was also the only Trust to show a significant overall positive change in the mean response per person (P<0.05). Trust C also showed significant improvements in staff opinion in the dimension of safety rules & procedures.

The results show that the questionnaire methodology could detect significant change in staff opinion and was therefore a good measure of culture change. The absence of significant change in five of the seven Trusts may be evidence that the underlying health and safety culture within these Trusts is relatively stable.

23 A distinction was made between unqualified and qualified staff in the categories of Nursing & Midwifery and Professional, Technical and PAMs. This was because the staff databases enabled this distinction to be made.

67

3.5.7 Questionnaire respondents additional comments The pattern of additional comments was similar between Trusts. Approximately two-thirds of those who returned questionnaires did not make additional comments (68% for phase 1 and 69% for phase 2).

Of the remaining approximate third that did comment, approximately one tenth (9%) were either positive or neutral for both phases. Approximately one tenth of comments related to specific issues within the employing Trust (9% for phase 1 and 11% for phase 2). The remaining approximate tenth focused on health and safety management issues within the Trust (11% for phase 1 and 9% for phase 2). Less than 4% of respondents reported that they found the questionnaire difficult to interpret (3% for phase 1 and 2% for phase 2).

3.6 RESULTS FROM WORKBOOK USAGE MONITORING (PROJECT PHASE 3)

As described in Chapter 2, the workbooks were issued to each Trust via a series of briefings to a pre-defined schedule. The number issued to each Trust was agreed in partnership with the Trusts. This took account of the Trust line management structure so that workbooks were issued to managers with responsibilities for a number of staff (e.g. nurses with responsibility for several wards rather than a single ward were specifically encouraged to take part). The final selection of individuals to be designated as workbook holders was however the responsibility of the Trusts.

The participating Trusts were encouraged to introduce measures to ensure the workbooks were used but, other than this, the research team took no further part in ensuring their usage was maximised. The research team took on a monitoring role to estimate and compare the levels of workbook usage in each of the five Trusts that had been issued with workbooks (the two control Trusts were not issued with workbooks and were not included in the monitoring).

3.6.1 Telephone survey

Attempts to contact each workbook holder by telephone were made during three time-periods (the method used is detailed in Appendix 6). Contact was made with the workbook holder and answers to nine questions were collated.

The time periods and percentages of successful contacts made are summarised below:

�� Period 1: o 1st April 2001 - 31st July 2001 o 86% successful contacts (153 workbook holders out of 177)

�� Period 2: o 1st August 2001 – 30th Nov. 2001 o 84% successful contacts (149 workbook holders out of 177)

�� Period 3: o 1st Dec. 2001 – 31st March 2002 o 64% successful contacts (114 workbook holders out of 177)

There was a drop from over 80% successful contacts in the first two periods to 64% in the third period. This was as a result of difficulties in making contact with the workbook holders. There were two main reasons for this, approximately equal in occurrence:

�� The workbook holder had been moved to another area or job and no one had contacted the research team or resumed ownership of the workbook;

68

�� Contact could not be made with the workbook holder despite repeated attempts (a typical reason was that the individual was too busy to participate or otherwise declined).

Therefore approximately one third of workbook holders had dropped out by the end of the monitoring period. The percentage responses for all questions asked during the three periods for all five test Trusts are shown in Figure 11.

Figure 11 Workbook holder telephone interviews feedback from three monitoring periods

Significant associations were tested using Cramer’s V, which is a suitable test for investigating associations between nominal variables. The results showed that the percentage of workbook holders did not change significantly (P<0.1) although use of the workbook reduced significantly

Note: Error bars show standard error

of the mean

0

10

20

30

40

50

60

70

80

90

100

Are

you

still

the

hold

er o

f the

wor

kboo

k?

Hav

e yo

u us

ed th

ew

orkb

ook?

Hav

e yo

u fo

und

the

wor

kboo

k he

lpfu

l?

Hav

e yo

u us

ed th

em

anag

ers'

aud

it?

Hav

e yo

u us

ed th

e w

orkb

ook

gene

rally

?

Hav

e yo

u fo

cuss

ed in

on

parti

cula

r sec

tions

?

Hav

e yo

u im

plem

ente

d an

yco

ntro

l mea

sure

s as

a re

sult

of a

risk

ass

essm

ent?

Hav

e yo

u us

ed th

e op

tion

appr

aisa

l?

Hav

e yo

u an

y su

gges

tions

for i

mpr

ovem

ents

?

Question posed

% o

f w

ork

bo

ok h

old

that

rep

lied

yes

1 Apr - 31 Jul 2001 1 Aug - 30 Nov 2001 1st December 2001 to 31st March 2002

69

over the course of the 12 months of use (P<0.01). This is regarded as low implementation of the workbook.

If the respondent indicated that they were no longer the workbook holder or had not used the workbook the interview was terminated. If the workbook had been used the next three questions were asked. These were intended to gain information on whether the workbook had been used in a general sense or whether particular sections had been used. The percentage that had used the workbook in a general sense had reduced significantly (P<0.05) although the percentage that focused on particular sections had nearly doubled between the first monitoring period and the last (P<0.001). Over the three monitoring periods the use of the manager’s audit did not change significantly (P<0.1).

The percentage that found the workbook helpful decreased by almost half in the third monitoring window (P<0.001).

The percentage who had implemented control measures as a result of risk assessment increased significantly in the middle window (P<0.01) but fell back to just below that of the first window by the time of the third window, resulting in no overall significant change (P<0.1).

The percentage of workbook holders that used the option appraisal was less than 5% across all three windows and did not vary significantly (P<1.0). Verbal feedback from users indicated that the method might benefit from development to make it easier to use. There may also be implications for redesigning training in use of the method. However, the option appraisal was always intended for use in circumstances where the solution to a risk control problem was not clear, and where there were financial implications. The circumstances where the option appraisal could be useful were therefore likely to be limited. This means that the low usage could also reflect the low number of circumstances where it was needed, rather than an avoidance of use.

There was no significant change in the percentage of users who made suggestions for improvement (P<0.1)

The results of the significance testing for individual Trusts, across the time intervals, (Cramer’s V) are shown in Table 25.

Table 25 Workbook holder telephone interview feedback associations

Cramer’s V Question Trust A Trust B Trust C Trust D Trust E Are you still the holder of the workbook? -0.346 -0.002* -0.001* -0.700 +0.096 Have you used the workbook? +0.040* -0.777 -0.083 -0.001* -0.001* Have you used the workbook generally? +0.377 -0.753 -0.024* -<0.001* -0.069 Have you focused on particular sections? +<0.001* +<0.001* +<0.001* +0.645 +0.027* Have you used the manager’s audit? -0.367 -0.856 -0.003* +0.228 +0.700 Have you found the workbook helpful or not?

+0.066 -0.001* -0.003* -<0.001* -<0.001*

Have you implemented any control measures as a result of risk assessment?

+<0.001* -0.635 -0.094 -0.016* -0.020*

Have you used the option appraisal? +0.081 +0.268 ZERO -0.004* -0.004* Have you any suggestions for improvement?

-0.661 +0.002* +0.083 -0.788 -0.471

* = Differences significant at the 0.05 level

70

The workbook holder telephone interview results for individual Trusts are shown in Appendix 9. General conclusions can be drawn from visual inspection of these five Figures. These are:

�� Numbers of holders of the workbooks tended to decrease (significantly in Trusts B and C); �� Usage of workbooks tended to decrease (significantly in Trusts D and E), although in Trust

A usage increased significantly; �� Perceived workbook usefulness decreased significantly in four Trusts (Trusts B, C, D and

E); �� Usage of the manager’s audit did not change significantly except in Trust C, where its use

decreased; �� General use of the workbook tended to decrease (significantly in Trusts C and D); �� All Trusts reported increased use of particular sections (significantly in four Trusts (Trusts

A, B, C and E)); �� Implementation of control measures as a result of risk assessment tended to decrease

(significantly in Trusts D and E). In Trust A implementation increase significantly; �� Use of the option appraisal decreased significantly in two Trusts (Trusts D and E); �� Suggestions for improvement tended to decrease but increased significantly in Trust B.

Although it can be seen from the above descriptive data that individual Trusts varied in their responses, statistical testing was undertaken to establish the degree of association between the Trusts in terms of the percentages of positive responses to the telephone interview questions. No statistically significant differences were found between any of the Trusts (k-sample chi-square between P=0.80 and P=0.99). Table 26 shows the results of non-parametric correlation testing (Spearman’s Rank Order Correlation) for the percentages recorded during the third interview period (the other two periods produced similar results). The null hypothesis was that there was no relationship between Trusts. The table shows significant observed correlation between all Trusts, indicating that the null hypothesis should be rejected thus confirming highly significant correlation.

Table 26 Between trust non-parametric correlations of workbook holder telephone interview feedback (third telephone interview period)

Spearman’s rank order correlation Trust A Trust B Trust C Trust D Trust E Correlation coefficient .975(**) .904(**) .819(**) .775(*) Sig. (2-tailed) <0.001 <0.001 <0.01 <0.05

Trust A

N 9 9 9 9 Correlation coefficient .975(**) .936(**) .836(**) .865(**) Sig. (2-tailed) <0.001 <0.001 0.005 <0.005

Trust B

N 9 9 9 9 Correlation coefficient .904(**) .936(**) .856(**) .846(**) Sig. (2-tailed) <0.001 <0.001 <.005 <0.005

Trust C

N 9 9 9 9 Correlation coefficient .819(**) .836(**) .856(**) .785(*) Sig. (2-tailed) <0.01 <0.005 <0.005 <0.05

Trust D

N 9 9 9 9Correlation coefficient .775(*) .865(**) .846(**) .785(*) Sig. (2-tailed) <0.05 <0.005 <0.005 <0.05

Trust E

N 9 9 9 9 ** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

71

If the percentage of positive answers to the telephone interview questions were assumed to represent an indicator of workbook implementation, it can be concluded that the degree of implementation was similar for each Trust. The implications of this finding in terms of the incident reporting and climate measurement changes are discussed in Chapter 4.

3.6.2 Visual inspections

The visual inspection of a 10% random sample of workbooks in each Trust was carried out during a 12 week period between 27th September 2001 and 10th December 2001, towards the end of the second telephone survey period and the beginning of the third. This means that the workbooks had been in place at least six months. Summary results are shown in Figure 12.

Figure 12 Use of documentation within a 10% random sample of workbook holders

Approximately 80% of holders who took part in the visual inspection had used their workbook, with very few having had pages removed and, encouragingly, approximately 75% had added items, such as policies or written risk assessments. The workbook sections used most were

0 10 20 30 40 50 60 70 80 90

Is the workbook holder accurately recorded?

Are cooperation measures entered or a marker included?

Is there evidence of quarterly reports having been completed?

Is there evidence of safety inspections having been carried outor a marker included?

Are national/trust policies listed with locations or a markerincluded?

Are items from departmental safety meetings documented or amarker included?

Are there training records in the workbook or a marker included?

Are there operational procedures listed or a marker included?

Are there objectives entered following completion of the audit?

Was the model business case used?

Insp

ectio

n qu

estio

n

Percentage of workbooks inspected exhibiting some element of documentation or a marker

72

policies, roles and responsibilities, measures to ensure co-operation, evidence of safety inspections, and quarterly reports.

However, the finding of low implementation following the telephone interviews was generally confirmed by the visual inspection in that, of the 19 questions asked within the inspection, there were only 5 questions where more than 50% of workbooks inspected exhibited some element of implementation.

Workbook sections used least were operational procedures, and incident reporting (possibly because all Trusts had a separate system). The model business case was not used in any of the workbooks inspected. Proactive management activities were among the least used aspects of the workbook. For example, less than one third of workbooks inspected had completed the managers’ audit (27%) with even fewer of this group using the outcome of the audit to set objectives (11%). Evidence of action or forward planning was found in only 37% of workbooks.

The impact of this low implementation on health and safety management performance is discussed further in Chapter 4.

3.7 TRIANGULATION OF RESULTS

As discussed in Chapter 1, Guldenmund recommended attempting to correlate safety performance measures with climate assessments(56). Although the data from the incident analysis and the workbook usage monitoring were not directly comparable with the climate survey results because their units and sampling strategy were different, broad categorisation using the ranks of the Trusts within each category was undertaken.

Tables 27 to 29 show the results of correlation testing (Spearman’s rank order correlation) for both phases. Tables 27 and 28 include:

�� Trust rank for the nine safety climate dimensions plus mean dimension score per person (from rank 1, representing the lowest mean dimension score to 5, representing the highest);

�� Trust rank for both the total reported incidents and those captured by the inclusion criteria and for which root causes could be assigned (i.e. minus the special category incidents) (from rank 1, representing the highest incident rate to 5, representing the lowest).

Table 29 also includes:

�� Trust rank of workbook usage (this was taken from the mean percentage of respondents to the telephone interviews that reported they had used the workbook (data in Appendix 8) (from rank 1, representing the lowest mean percentage reported use to 5, representing the highest).

73

Tabl

e 27

Ran

k or

der c

orre

latio

n m

atrix

bet

wee

n fiv

e te

st tr

usts

rank

ing

for h

ealth

and

saf

ety

clim

ate

mea

sure

s, a

nd re

porte

d an

d ca

ptur

ed

inci

dent

s (p

hase

1) (

Spea

rman

’s ra

nk o

rder

cor

rela

tion)

(* =

Sig

nific

ant c

orre

latio

n (2

-taile

d)).

DIM

ENSI

ON

Ca

ptur

ed

inci

dent

s Cl

imat

e m

ean

Com

mun

-ic

atio

n In

volv

e-m

ent

Man

age-

men

t co

mm

i-tm

ent

Pers

onal

ap

prec

iatio

n of

risk

Pers

onal

pr

iorit

ies

Prio

rity

of safe

ty

Repo

rted

inci

dent

sSa

fety

ru

les/

pr

oced

ures

Supp

ortiv

e en

viro

nmen

tW

orki

ng

envi

ronm

ent

Cap

ture

d in

cide

nts

<0

.01*

0.

10

<0.0

1*

0.10

0.

62

0.28

<0

.05*

1.

00

<0.0

5*

<0.0

1*

0.19

Clim

ate

mea

n <0

.01*

0.10

<0

.01*

0.

10

0.62

0.

28

<0.0

5*

1.00

<0

.05*

<0

.01*

0.

19

Com

mun

icat

ion

0.10

0.

10

0.

10

0.19

1.

00

0.28

<0

.05*

0.

62

0.28

0.

10

0.39

In

volv

emen

t <0

.01*

<0

.01*

0.

10

0.

10

0.62

0.

28

<0.0

5*

1.00

<0

.05*

<0

.01*

0.

19

Man

agem

ent

com

mitm

ent

0.10

0.

10

0.19

0.

10

0.

19

0.87

<0

.05*

0.

75

<0.0

5*

0.10

0.

10

Pers

onal

ap

prec

iatio

n of

ris

k

0.62

0.

62

1.00

0.

62

0.19

0.50

0.

50

0.19

0.

28

0.62

0.

39

Pers

onal

pr

iorit

ies

0.28

0.

28

0.28

0.

28

0.87

0.

50

0.

39

0.87

0.

75

0.28

0.

87

Prio

rity

of

safe

ty

<0.0

5*

<0.0

5*

<0.0

5*

<0.0

5*

<0.0

5*

0.50

0.

39

0.

87

0.10

<0

.05*

0.

28

Rep

orte

d in

cide

nts

1.00

1.

00

0.62

1.

00

0.75

0.

19

0.87

0.

87

0.

87

1.00

0.

75

Safe

ty ru

les &

pr

oced

ures

<0

.05*

<0

.05*

0.

28

<0.0

5*

<0.0

5*

0.28

0.

75

0.10

0.

87

<0

.05

<0.0

5*

Supp

ortiv

e en

viro

nmen

t <0

.01*

<0

.01*

0.

10

<0.0

1*

0.10

0.

62

0.28

<0

.05*

1.

00

<0.0

5*

0.

19

Wor

king

en

viro

nmen

t 0.

19

0.19

0.

39

0.19

0.

10

0.39

0.

87

0.28

0.

75

<0.0

5*

0.19

74

Tabl

e 28

Ran

k or

der c

orre

latio

n m

atrix

bet

wee

n se

ven

trust

s ra

nkin

g fo

r hea

lth a

nd s

afet

y cl

imat

e m

easu

res,

and

repo

rted

and

capt

ured

in

cide

nts

(pha

se 1

) (Sp

earm

an’s

rank

ord

er c

orre

latio

n) (*

= S

igni

fican

t cor

rela

tion

(2-ta

iled)

).

DIM

ENSI

ON

Ca

ptur

ed

inci

dent

s Cl

imat

e m

ean

Com

mun

-ic

atio

n In

volv

e-m

ent

Man

agem

ent

com

mitm

ent

Pers

onal

ap

prec

iatio

n of

risk

Pers

onal

pr

iorit

ies

Prio

rity

of safe

ty

Repo

rted

inci

dent

sSa

fety

ru

les &

pr

oced

ures

Supp

ortiv

e en

viro

nmen

tW

orki

ng

envi

ronm

ent

Cap

ture

d in

cide

nts

0.

19

0.36

0.

08

0.19

0.

52

0.63

0.

27

0.41

<0

.05*

0.

41

0.57

Clim

ate

mea

n 0.

19

<0

.05*

<0

.01*

<0

.05*

0.

18

0.12

<0

.01*

0.

76

<0.0

5*

<0.0

1*

<0.0

5*

Com

mun

icat

ion

0.36

<0

.05*

<0.0

5*

0.09

0.

59

0.05

<0

.05*

0.

34

0.16

<0

.05*

0.

07

Invo

lvem

ent

0.08

<0

.01*

<0

.05*

<0.0

5*

0.34

0.

09

<0.0

1*

0.82

<0

.05*

<0

.01*

0.

07

Man

agem

ent

com

mitm

ent

0.19

<0

.05*

0.

09

<0.0

5*

0.

09

0.64

<0

.01

0.70

<0

.05*

<0

.05*

<0

.05*

Pers

onal

ap

prec

iatio

n of

ris

k

0.52

0.

18

0.59

0.

34

0.09

0.94

0.

15

0.38

0.

18

0.21

0.

18

Pers

onal

pr

iorit

ies

0.63

0.

12

0.05

0.

09

0.64

0.

94

0.

15

0.18

0.

59

0.09

0.

48

Prio

rity

of

safe

ty

0.27

<0

.01*

<0

.05*

<0

.01*

<0

.01*

0.

15

0.15

0.88

0.

05

<0.0

5*

<0.0

5*

Rep

orte

d in

cide

nts

0.41

0.

76

0.34

0.

82

0.70

0.

38

0.18

0.

88

0.

70

0.43

0.

48

Safe

ty ru

les &

pr

oced

ures

<0

.05*

<0

.05*

0.

15

<0.0

5*

<0.0

5*

0.18

0.

59

0.05

0.

70

0.

07

0.07

Supp

ortiv

e en

viro

nmen

t 0.

41

<0.0

1*

<0.0

5*

<0.0

1*

<0.0

5*

0.21

0.

09

<0.0

1*

0.43

0.

07

<0

.01*

Wor

king

en

viro

nmen

t 0.

57

<0.0

5*

0.07

0.

07

<0.0

5*

0.18

0.

48

<0.0

5*

0.48

0.

07

<0.0

1*

75

Tabl

e 29

Ran

k or

der c

orre

latio

n m

atrix

bet

wee

n fiv

e te

st tr

usts

rank

ing

for h

ealth

and

saf

ety

clim

ate

mea

sure

s, re

porte

d an

d ca

ptur

ed

inci

dent

s an

d w

orkb

ook

use

(pha

se 2

) (Sp

earm

an’s

Ran

k O

rder

Cor

rela

tion)

(* =

Sig

nific

ant c

orre

latio

n (2

-taile

d)).

DIM

ENSI

ON

Capt

ured

in

cide

nts

Clim

ate

mea

n Co

mm

un-

icat

ion

Invo

lve-

men

t M

anag

e-m

ent

com

mi-

tmen

t

Pers

onal

ap

prec

-ia

tion

of

risk

Pers

onal

pr

iorit

ies

Prio

rity

of safe

ty

Repo

rted

inci

dent

sSa

fety

ru

les &

pr

oced

ures

Supp

ortiv

e en

viro

nmen

tW

orkb

ook

use

Wor

king

en

viro

nmen

t

Cap

ture

d in

cide

nts

<0

.05*

0.

17

<0.0

1*

0.09

0.

55

0.09

0.

32

0.55

<0

.01*

0.

22

0.61

0.

43

Clim

ate

mea

n <0

.05*

0.10

<0

.01*

. <0

.05*

0.

39

<0.0

5*

0.28

0.

62

<0.0

1*

.019

0.

39

0.28

C

omm

uni-

catio

n 0.

17

0.10

0.10

<0

.05*

0.

50

<0.0

5*

<0.0

5*

0.62

0.

10

0.10

0.

19

<0.0

5*

Invo

lvem

ent

<0.0

5*

<0.0

1*

0.10

<0.0

5*

0.39

<0

.05

0.28

0.

62

<0.0

1*

0.19

0.

39

0.28

M

anag

emen

t co

mm

itmen

t 0.

09

<0.0

5*

<0.0

5*

<0.0

5*

0.

19

<0.0

1*

0.10

1.

00

<0.0

5*

<0.0

5*

0.28

0.

19

Pers

onal

ap

prec

iatio

n of

risk

0.55

0.

39

0.50

0.

39

0.19

0.19

0.

62

0.28

0.

39

0.28

0.

39

0.62

Pers

onal

pr

iorit

ies

0.09

<0

.05*

<0

.05*

<0

.05*

<0

.01*

0.

19

0.

10

1.00

<0

.05*

<0

.05*

0.

28

0.19

Prio

rity

of

safe

ty

0.32

0.

28

<0.0

5*

0.28

0.

10

0.62

0.

10

0.

87

0.28

<0

.05*

0.

50

0.19

Rep

orte

d in

cide

nts

0.55

0.

62

0.62

0.

62

1.00

0.

28

1.00

0.

87

0.

62

0.75

0.

75

0.50

Safe

ty ru

les &

pr

oced

ures

<0

.05*

<0

.01*

. 0.

10

<0.0

1*

<0.0

5*

0.39

<0

.05*

0.

28

0.62

0.19

0.

39

0.28

Supp

ortiv

e en

viro

nmen

t 0.

22

0.19

0.

10

0.19

<0

.05*

0.

28

<0.0

5*

<0.0

5*

0.75

0.

19

0.

62

0.39

Wor

kboo

k us

e0.

61

0.39

0.

19

0.39

0.

28

0.39

0.

28

0.50

0.

75

0.39

0.

62

<0

.05*

Wor

king

en

viro

nmen

t 0.

43

0.28

<0

.05*

0.

28

0.19

0.

62

0.19

0.

19

0.50

0.

28

0.39

<0

.05*

76

The results indicated a number of significant correlations. In the first phase (Table 27) the Trust rank of captured incidents was significantly associated with the Trust ranks of four of the nine questionnaire dimensions (involvement, priority of safety, safety rules & procedures and supportive environment) and the overall climate questionnaire mean score.

Significant associations with overall climate questionnaire mean score were present in both phases. In addition, there was a lack of correlation between climate dimensions and reported incident rates. This indicates that the Trust rank of incidents for which root causes could be assigned was potentially a better indicator of health and safety performance than total incident reporting rate.

In other words, higher overall mean questionnaire dimension score was significantly correlated with lower incident capture rate.

When the control Trusts were included in the first phase rankings the findings were less clear with the only significant correlation between captured incidents and questionnaire response rank was with the dimension of safety rules & procedures (Table 28). Trust rank of reported incident rate was again not significantly correlated with any other group.

For the five test Trusts in the second phase (Table 29) Trust rank of percentage reported workbook use was also included. The Trust rank for use of the workbook was significantly correlated with Trust rank for the climate dimension of working environment. As well as the overall climate questionnaire mean score the questionnaire dimensions of involvement and safety rules & procedures were significantly correlated with Trust rank for captured incident rate.

Therefore higher reported use of the workbook was significantly correlated with both higher overall mean questionnaire dimension score for working environment and lower captured incident rate.

This data should be interpreted with caution, given the small sample size (n=5) and the absence of data on workbook usage in the control Trusts. However, implications of these significant correlations are discussed in Chapter 4.

3.8 SUMMARY OF RESULTS

3.8.1 General findings

�� Between phase 1 and phase 2 the mean number of incidents reported increased by 24% although the deference was only marginally significant at the 0.05 level (P=0.06);

�� Between phase 1 and phase 2 there was highly correlated rankings of reported and captured incident rates for each Trust (P<0.01);

�� Reporting rates were not associated with the size of the Trust (P=0.48); �� Capture rates were significantly associated with the size of the Trust in phase one (P=0.01)

but only marginally associated in phase 2 (P=0.09); �� The numbers of incidents captured represented an average of 6% of the total that were

reported; �� Between phase 1 and phase 2 there was an increase in the number of incidents captured,

although it was not significant (P=0.41); �� Between phase 1 and phase 2 there was no correlation between Trust reporting and capture

rates (P=0.54);

77

�� The mean reporting and capture rates for the control Trusts were not significantly different (P=0.50 for both phases) from the test Trusts.

3.8.2 Special category

�� A challenge exists for those conducting incident investigation within the NHS to identify clearly cases where management root causes cannot reasonably be established;

�� There was an increase in special category incidents between phase 1 and 2, although not significant (P=0.69);

�� Most special category incidents were either not foreseen or outwith reasonable practicability (67% of special category incidents);

�� The remaining group involved clinical judgements that attempted to balance potentially conflicting health and safety and patient care factors (25%). There was a small proportion where there was insufficient information on which to base a decision (7%);

�� The traditional paradigm that most, if not all, incidents can be prevented by implementation of robust risk management arrangements has been challenged by the special category incidents. In the NHS it is likely that a percentage of incidents will always fall into the special category.

3.8.3 Identification of risk control system failures and root causes of incidents

So far as results that were relevant to research objective 1 (The design of a methodology for evaluation of health and safety performance) were concerned:

�� The risk control system of risk assessment was the major source of management root causes in both phases;

�� The four major management system root causes were those of planning, implementing, measuring and reviewing;

�� There were no significant differences between the test and control trusts in either phase of the project.

So far as results that were relevant to research objective 3 (The design of a methodology to measure change in health and safety performance) were concerned:

�� The numbers of root causes assigned to risk control systems was not significantly different between phases (P=0.51);

�� The number of times risk assessment was identified as a risk control system failure decreased significantly between phases (p=0.02), while violence and aggression (p=0.01) and handling sharps (p=0.02) significantly increased. This could indicate better reporting of incidents involving violence and aggression and sharps;

�� Effort on the Implementation of control measures as a result of risk assessment was thought likely to produce most impact on health and safety performance;

�� The differences in the numbers of cited management root causes between the two phases were not significant (P=0.92);

�� There was significant correlation between the ranks of frequency of citation of root causes between the two phases (P<0.05).

3.8.4 Costs of incidents


78

�� The total cost for special category incidents (£55,593) was more than either of the two data collection phases (Phase 1: £51,096; Phase 2: £53,209) for the incidents where root causes could be assigned;

�� Most special category costs were attributed to accidents involving patients; �� Where root causes could be assigned, incidents involving staff were nearly three times more

frequent and represented approximately four times the cost than those for patients; �� The Pareto 80:20 rule broadly applied to the incident costing data in that approximately

80% of costs were attributed to approximately 20% of the observed incidents; �� The total cost assigned to the incident category of accidental injury was an order of

magnitude higher than any other category in both phases. Costs from physical violence increased by nearly four times between phases;

�� The category of ill health may underestimate the true cost to the NHS; �� The incident factors staff absence and replacement labour costs were approximately 75% of

the total costs; �� For incident types, person to person assault, slip trip or fall on the same level/stairs and

patient lifting/handling were approximately 75% of the total costs.


�� There was no significant difference between the two phases for total costs per incident factor or type (P=0.33 (incident factor), P=0.69 (incident type);

�� Total incident factor costs for the test Trusts were not significantly different from the control Trusts (P=0.85 (phase 1), P=0.77 (phase 2));

�� The average costs for incident types were not significant between phases (P=0.68). However, the rank order of average cost of incident type between the two phases was marginally correlated (P<0.1);

�� The difference in cost range for incident types was not significant between phases (P=0.35).

3.8.5 Staff opinion survey


�� There was a 42% response rate from individuals who completed two valid questionnaires; �� The distribution of questionnaire response data was approximately normal allowing

parametric statistical testing to be applied; �� Pooling of data from individual Trusts was avoided because their results were statistically

significant from each other (P<0.05); �� A significant difference (P<0.05) in the opinion of medical and dental staff was found in

one Trust in phase 1.


�� Significant change in opinion was found in two Trusts (P<0.05). In one Trust a single dimension showed improvement while in the other six of the dimensions none showed improvement;

�� The questionnaire methodology could detect significant change in staff opinion and therefore represents a potentially useful measure of culture change;

79

�� The absence of significant change in five of the seven Trusts may be evidence that the underlying health and safety culture within these Trusts is relatively stable.

3.8.6 Monitoring and audit of the use of the management system workbook

There was generally low implementation of the workbook across all Trusts. Evidence for this conclusion included:

�� Over time the percentages of holders of workbooks tended to decrease, although not significantly (P<0.1);

�� Within Trusts use of workbooks decreased significantly over time (P<0.01) with approximately one third having dropped out by the end of the monitoring period;

�� General workbook use decreased significantly (P<0.05); �� All Trusts reported a significant increase in the use of particular sections (P<0.001); �� Use of the manager’s audit did not change significantly (P<0.1); �� Overall perceived workbook usefulness decreased significantly over time (P<0.001); �� Implementation of control measures as a result of risk assessment did not change overall

(P<0.1); �� Use of the option appraisal and suggestions for improvement did not vary significantly

(P<0.1). �� There were no statistically significant differences between individual Trusts in terms of

their responses to telephone interview questions (P<1.0). This was confirmed by the presence of significant rank order correlations between Trusts (P<0.05);

�� Visual inspection of a 10% sample of workbooks also confirmed low implementation of the workbooks.

3.8.7 Triangulation of results

�� Test Trust rank order of captured incident rates (i.e. incidents for which health and safety management system root causes could be established) was significantly correlated with climate questionnaire mean score in both phases (P<0.05);

�� Test Trust rank order of captured incident rates was also significantly correlated with the questionnaire dimensions of involvement, priority of safety, safety rules & procedures and supportive environment in the first phase and involvement and safety rules & procedures in the second phase (P<0.05);

�� When the control Trusts were included in the first phase rankings only significant correlation between captured incidents and questionnaire response rank was with the dimension of safety rules & procedures (P<0.05);

�� Trust rank of reported incident rate was not significantly correlated with any other group; �� The Trust rank for use of the workbook was significantly correlated with Trust rank for

captured incidents and the climate dimension of working environment (P<0.05).

80

4 DISCUSSION

There were two main findings in the research. These were, firstly, that there were significant correlations between the performance evaluation components. The second major finding was that there was little change in performance between the health and safety evaluation phases.

The discussion, conclusions and recommendations arising from these findings comprises of four parts:

�� Part 1: Consists of a discussion and evaluation of the individual components of the health and safety management performance evaluation method and how they might be refined to optimise their potential to reveal correlations;

�� Part 2: Comprises a discussion and evaluation of the strengths and weaknesses of the health and safety management workbook so as to form conclusions as to its impact on the lack of change in health and safety performance;

�� Part 3: Also comprises a discussion and evaluation of the strengths and weaknesses of the health and safety management performance change measurement methodology as to its impact on the lack of change in health and safety performance;

�� Part 4: Concludes with an assessment of the degree to which the research aim and objectives have been achieved.

4.1 HEALTH AND SAFETY MANAGEMENT PERFORMANCE EVALUATION METHODOLOGY

There are six sections to this part of the discussion. These are:

�� Performance evaluation based on the use of reported incidents; �� The use of the HSE root cause analysis methodology; �� Investigator bias and its implications for the training and development of safety advisers; �� Costs and economic implications; �� The Health and safety climate assessment; �� Triangulation of health and safety measurement methods.

Each will be considered in turn.

4.1.1 Performance evaluation based on the use of reported incidents

The Health services sector is currently a priority area for the HSC. One reason for this is because it is a major employer. The NHS employs approximately 1.1 million people, with more than 0.5 million in the private healthcare sector. Although it is not regarded as a high hazard industry, high rates of incidents are routinely reported to HSE(109).

This presents an interesting paradox because the activity of reporting incidents for the purpose of deriving health and safety performance data has been extensively encouraged in the NHS, in order to facilitate comparisons of performance between and within Trusts (1, 9).

In addition, Government and HSCs national targets(22) for improving health and safety performance over the ten years to 2010 are ambitious and partly reliant on incident reporting

81

data24. The Health Services Advisory Committee (HSAC) of the Health and Safety Commission (HSC) set a ten-year target:

“By 2004, health and safety performance across the health care sector will present a model for other sectors to emulate.”(54)

This aim was supported by a number of objectives, including one that set targets for the reduction of the incidence and costs of work-related incidents and ill health. In NHS Scotland incident reporting data (defined in the introductory Chapter) have been routinely collected since 200125 and are set against the context of a national strategy for health and safety performance targets(115).

It has been planned to use these minimum dataset indicators to monitor ongoing performance in occupational health and safety within the NHS(116). The numbers and rates of reported incidents were to be two of the indicators used.

Therefore it can be concluded that there exists in the NHS a strong culture of reporting incidents, no matter how trivial. The “if in doubt – report it” message, has been strongly advocated and implemented at local level by senior management and safety departments alike. The outcome of this strategy could be observed in the research results whereby some Trusts were found to deal with as many as 250 reported incidents per week.

However, in Chapter 3 it was shown that the number of reported incidents (between 255 and 2,481) and the reported incident rate (between 0.36 and 2.63 per person working years) differed between Trusts by an approximate order of magnitude. This is regarded as wide variability.

The cause of this variability could be related to differences in reporting rates between the Trusts but it could also represent an indicator of health and safety performance in that, Trust staff with more awareness of health and safety might report more incidents, a possibility also highlighted by the NAO(1). This is likely given that the majority of reported incidents were found to have little or no potential for harm. However, the total incident reporting rates did not significantly correlate with any of the other performance measures in the triangulation conducted in Chapter 3. It is suggested that his makes the variation more likely to be related to reporting culture, rather than health and safety performance, a limitation of using incident rates that has already been identified (e.g.(44)).

The captured incident rates also differed by an approximate order of magnitude (between 0.02 and 0.19 per person working years) and were a much smaller percentage of the total incidents reported (between 0.6% and 10.7% of the total numbers of incidents reported), however they were found to correlate significantly with workbook use and one of the questionnaire dimensions. Therefore this might be a more useful performance indicator for future use.

Nevertheless there remain a number of unresolved limitations with the method as it currently stands. These are: �� As discussed above, the main purpose of collecting incident data in the NHS has been to

benchmark health and safety performance between Trusts. However, there is frequent

24 The targets were to reduce the number of working days lost from work-related injuries and ill-health by 30%, to reduce the incidence rate of cases of work-related ill-health by 20% and to reduce the incidence rate of fatalities and major injuries by 10% (achieving half the improvement by 2004) 25 at the time of writing the first data collection for the year ending 31 March 2001 was still ongoing had had not yet been published. It is therefore not yet possible to compare the research data with this routine data.

82

reference to learning lessons from incident data in health and safety publications referred to by safety practitioners working in the NHS (e.g. (1, 9, 18, 25, 35, 36, 117)). It is questionable as to whether there is sufficient clarity between Trusts as to the distinction between these two purposes. There is work to do to ensure that policy makers, occupational health and safety professionals, line managers and local staff understand why they are collecting and reporting incident data;

�� Although the Trusts participating in the research tended to use a similar approach to incident reporting, there is an undoubted variation in conventions for reporting clinical versus non-clinical incidents and also accidents and ill-health between Trusts. This is because the reporting mechanisms tend to be different for these types of incident. For example, clinical incidents (such as infections, clinical negligence etc.) tend to be reported via the clinical network (such as director of medicine or nursing) where as non-clinical incidents tend to be reported through the health and safety department. Cases of ill health tend to be reported or recorded by the HR or occupational health departments, where as accidents tend to be reported via the health and safety department. Communication and collaboration between these various departments, who tend to use their own reporting systems, has been highlighted as lacking by the NAO(1). The recent moves towards more multidisciplinary occupational health and safety services within the NHS may offer potential for improvements in this area(8, 23).

Therefore it is concluded and recommended that national strategy based on the use of incident rates to benchmark health and safety performance should use a tighter definition that the current one (described in the introductory Chapter), possibly similar to that used in the research (i.e. the presence of actual or potential for damage; that the incident was under the control of the Trust; and incurring more than £5 cost or 15 minutes time). In any case it is recommended that efforts be made within the NHS to work towards a common standard for incident reporting, supported by guidelines, which clarify the purpose for which the incident reports are used, in order to better control the risk of reporting inconsistencies.

4.1.2 Use of the HSE root cause analysis methodology

The previous section (4.1.1.) explored the use of reported incident data as performance indicators. The other use of reported incident data is to learn lessons so as to control health and safety risks by identifying underlying causes (e.g. (18, 118)).

In this regard, it has been suggested that there are six key features for a suitable incident analysis method (118). These are that the method should:

�� Distinguish between events and underlying causes; �� Allow for the grouping and pooling of accident data; �� Identify remedial action to prevent future accidents; �� Offer reliability between different investigators; �� Be pragmatic and capable of being used by non-specialists; �� Allow conclusions to be reached and prioritised action plans to be produced.

The HSE methodology used in the research26 met all of these criteria. The circumstances of the incident could be linked to risk control system failures and thereby to the root causes within the health and safety management system. (99). This information could then be used to identify remedial action, in other words, to learn lessons. Therefore, the purpose of the root cause analysis aspect of the HSE methodology was different from that of the incident frequency data.

26 Described in Chapter 3 and reproduced in Appendix 1.

83

The methodology provided information to allow conclusions to be formed that, not only was risk assessment the most significant underlying system failure, but that it was associated with the root causes of planning and implementing, and that they were the most significant management system root causes. This has implications for action planning for the employer. This finding was communicated to the participating Trusts at the end of the second phase of the research project. It is therefore not known if any action they took had an affect on health and safety performance.

In some respects the finding of a lack of risk assessment at the root of most of the incidents is unsurprising. For example, in the personal experience of the author, many safety practitioners working in the NHS would subjectively concur, as this is what they find on a day-to-day basis. On the other hand, the requirement for risk management systems has been very much part of the rhetoric of the HSE since the early 1990’s (when Crown immunity was lifted from the NHS). HSE has issued a plethora of advice and guidance on the systematic management of health and safety with risk assessment at its core. It is therefore disappointing that, despite the considerable efforts of nearly 15 years, the message still has not apparently translated into effective control of risk.

Therefore, it is recommended that priority should be given to ensuring that the system of risk assessment was operating effectively and, in particular, the planning and implementation of the system, it is possible that improvement in health and safety performance might result. This approach would certainly direct effort where it was most needed in the health and safety management system.

Another strength of the method was that its use revealed the existence of the group of incidents, which were thought to be unpreventable, those grouped under the title “special category”. Others have noted this phenomenon. For example, Neale et al (119) estimated that approximately half of a group of adverse incidents that occurred in two hospitals were not preventable while Oliver et al(80) also noticed that some health and safety risks in a healthcare setting were difficult to manage. For example, for frail, elderly patients recovering from acute illness, a delicate balance of risk factors is needed because the risk of falling can be challenging to prevent without adversely affecting rehabilitation, a finding also present in the current research, where the special category incidents represented approximately half of those incidents that met the research inclusion criteria and also about half the total cost.

Therefore, although unexpected at the start of the research, they must be regarded as an important group of incidents. It would undoubtedly be possible to argue that some of the incidents could have been categorised differently, but the overall picture remains that, within healthcare, there exist incidents, which do not fit within the adopted health and safety management model.

Although the special category comprised incidents that occurred to both staff and patients, the legal obligations are different. So far as staff is concerned, legislation in the UK centers, within the Health and Safety at Work etc. Act 1974, on the concept of risk control “so far as is reasonably practicable” (SFARP) which is, in effect, a requirement to undertake a cost-benefit analysis whereby risk is weighed against the sacrifice (in money, time or trouble) involved in averting the risk. The decision as to whether risk was controlled in any particular set of circumstances is generally left to the legal system to decide, with its large body of legal precedent on which to base its judgment.

So far as incidents involving staff and patients are concerned the legal test is one of a common law “duty of care” against which claims of negligence or clinical negligence (broadly defined as

84

“careless conduct which injures another”) can be submitted, via the civil courts, again drawing on the body of legal precedent to establish on the “balance of probability”, whether the duty of care was breached(120). So far as patients are concerned the definition of reasonable practicability was one applied by the research team, as a convenient label for this group of incidents. However, it must be remembered that this does not constitute a legal definition.

It is concluded that the special category results have an impact on the definition of reasonable practicability within the NHS as an industry. It is therefore recommended that this issue must be debated further.

Certainly the concept that nearly all incidents have underlying management system root causes(74) and that they are, therefore, preventable seems to be challenged by this group of incidents. It may be that healthcare is a “special case” in that its client group is, by definition, vulnerable and that some incidents, particularly those occurring to patients, may be inevitable, taking account of the needs of the patient and available resources.

However, if this is the case for some incidents, it should not be used as an excuse to avoid the implementation of robust risk management procedures. Nor should the apparent special category of an incident be used as justification for curtailing detailed investigation of serious incidents.

This is not judged to be likely, so far as patient care is concerned. This is because, in many ways clinical management in the NHS is more developed than health and safety management. For example, there exist organisations such as the National Patient Safety Association (NPSA)27 and the National Institute for Clinical Excellence (NICE)28, set up as a Special Health Authority for England and Wales on 1 April 1999 as part of the NHS to provide patients, health professionals and the public with authoritative, robust and reliable guidance on current “best practice” on individual health technologies (including medicines, medical devices, diagnostic techniques, and procedures) and the clinical management of specific conditions. Managers can and should continue to apply these robust management procedures to all aspects of their work, managing risk in the broadest sense.

The author has classified these incidents as a special category. This may be misleading in that the category was not intended to be definitive but rather a convenient label under which to group the incidents. A more appropriate description for the group may be necessary. They all share a common link in that they are incidents occurring as an outcome of the healthcare industry. It is therefore recommended that, in future, investigations of incidents should bear in mind the possibility of a classification of “intrinsic incidents”.

It is therefore concluded that the HSE root cause analysis method was a powerful tool, which could identify consistently health and safety management system weaknesses and, importantly, identify whether there exist incidents in the data set, which cannot be prevented within the bounds of accepted best practice.

Limitations of the method were that it was useful as a research tool but if it were used for more routine applications, further development is recommended. In particular its performance against a range of industries should be established and how it might be applied to workplace ill health. The method would also benefit from conversion to a software tool, possibly based on an expert system, to ensure it is intuitive and easy to use. This may assist with some of the more laborious tasks associated with processing investigation data. 27 http://www.npsa.nhs.uk/ 28 http://www.nice.org.uk/cat.asp?c=20

85

4.1.3 Investigator bias and its implications for the training and development of safety advisers

Wagenaar has suggested that approaches to incident analysis should be reliable in that various independent analysts should reach the same conclusions(118). The literature review (Chapter 1) identified evidence that the current UK standard of incident investigation was poor in that wide variations in approach, and therefore unreliability, were commonplace.

This was confirmed during the incident investigations and root cause analyses described in Chapters 2 and 3, which were typified by a challenge to maintain consistency against the pre-set standard. The consensus approach that was adopted and described in Chapters 2 and 3 has also been discussed by Culvenor and others, who found that consistency of judgments of teams were better than the average of the individuals that comprised the team(121, 122). This adds weight to the conclusion that the results of the incident investigations were, as a result of the control measures, internally consistent.

The main limitation with adopting this consensus approach was not with the research project where the thorough investigation was resources and planned for. The challenge lies in achieving incident investigation consistency within and between safety professionals working on a day-to-day basis, usually with limited time and resources to devote to thorough data gathering prior to incident analysis. This is particularly relevant in the NHS where, the results in Chapter 3 have shown, some Trusts were processing up to 250 incident reports each week. This begs the question as to whether the purpose of collecting the incident data was sufficiently clear within the healthcare organisations.

For example, HSE have suggested that collecting information on injuries and ill health should not present major problems for most organisations but that learning lessons from these incidents can prove more challenging(18). The author has found this to be the case both before and during the research, where the collection of incident data and the preparation of trend analyses and reports seemed the paramount purpose of some health and safety departments. As described in Chapter 2, thorough investigation of incidents seemed to be a secondary task. An illustration of this was when HSE issued a consultative document(123) on proposed changes to the Reporting of Diseases and Dangerous Occurrences Regulations (RIDDOR) 1995(109). Informal discussions with the health and safety practitioners participating in the research project indicated that their perception of the impact of a proposed new duty to investigate all RIDDOR incidents was that it would not be an onerous task. This was thought to be because the task was already routinely undertaken.

Therefore it can be concluded that the purpose of incident reporting in the NHS has been largely as a performance indicator (already discussed above in Section 4.1.1.) and that the process of learning from adverse events has been comparatively underdeveloped and largely based on analysis of frequency, something Hale has cautioned against(112). Hale has also suggested that there may be differences between the causes and underlying conditions between minor and major accidents(112). The use of minor incidents and near misses as a performance measure was thought by Hale to be erroneous and particularly inappropriate if extrapolated to apply to performance in major hazard control. In the experience of the author it was common for underlying causes of minor incidents and near misses to be assumed to be the same as more serious incidents. This illustrates further the underdeveloped culture of learning from adverse events in health and safety in healthcare.

In the search for reasons to explain this underdeveloped situation it is possible to look at training for safety professionals, which has concentrated on techniques of incident investigation

86

and root cause analysis for industries with the highest hazards29. For “softer” industries, such as healthcare, without engineering data on which to base failure probability calculations, the professional safety adviser has to rely on a combination of subjective judgment based on their professional background, experience and interviews or discussions with those involved.

However, this position may be starting to change. For example, training for safety advisers has begun to shift towards producing agents of change rather than the more traditional adviser/officer role. For example, Swuste and Arnoldy have reported on how they have developed their postgraduate masters’ course in health and safety at the University of Delft to meet this need(88). This approach is in response to a perceived lack of progress with management of health and safety using traditional methods.

4.1.4 Costs and economic aspects

If it is assumed that the two twelve week data collection phases were typical, the costs represent an extrapolated equivalent annual total cost for the seven Trusts of £221,416 for phase 1 and £230,572 for phase 2. Taking the speculation to the entire NHS in Scotland this would equate to a cost for captured incidents of approximately £1.3 million per annum (£1,284,996 for phase 1 and £1,338,133 for phase 2). This assumes 23,852 staff in the seven participating Trusts and 138,426 total staff employed in the NHS in Scotland, as of September 2001(124). For the NHS in England the equivalent figure (based on a total staff compliment of just under 1 million (948,275)) is approximately £9 million per annum30 (£8.8 million in Phase 1 and £9.2 million in Phase 2).

However, a cost range may be more appropriate given the wide cost variation observed between the Trusts. If the extrapolated annualised costs for the NHS were based on the lowest rate observed (£2.25/person/annum (based on £0.01 per person working day * 225) the overall losses were of the order of £300,000 (£311,458) in Scotland and approximately £2million (£2,133,619) for England. If the highest observed rate were used on which to base the estimate (£60.75/person/annum), the potential losses increase to just under £8.5million (£8,409,379) in NHS Scotland and just over £57.5million (£57,607,706) in England.

This is a wide range of estimated costs. If the 2000/2001 running costs for the Scottish NHS are taken into account (running cost estimated at approximately £2.5 billion31) incidents accounted for between 0.01% and 0.31%. The equivalent annual estimate for the NHS in England and Wales (which receives proportionately less funding) is between 0.06% and 1.44% (running cost estimated at approximately £40billion (£39,883million)).

These costs are relatively low and represent approximately 50% of those identified in the pilot work reported in Chapter 1 and approximately 20% of that estimated by HSE in 1992(35). They estimated, using a total loss approach, that direct financial and opportunity costs amounted to 5% of a hospital’s running costs (incurred by accidents alone). HSE’s estimation of the costs to UK employers of work injuries and non-injury accidents for health and social work in 1995/96 was approximately £240million(125), approximately two and a half times the highest estimate for the current data (i.e. 40%). More recently, the HSC launched a web-based “ready reckoner” to assist businesses to work out costs of work-related accidents and ill health(126). Although the method takes a similar approach to the research methodology (i.e. calculating costs as a product of time spent and hourly rate), the range of categories is wider and therefore

29 Information on course contents for UK training courses for health and safety advisers available from NEBOSH (http://www.nebosh.org.uk/) and IOSH (http://www.iosh.co.uk/). 30 Source http://www.doh.gov.uk 31 Source http://www.show.scot.nhs.uk/isd

87

costs likely to be higher for a given incident. For example, costs for sanctions and penalties such as fines and legal expenses are included in the ready reckoner but were not included in the current research unless they occurred within the 12 week data collection windows.

The only study to publish costing data that were significantly lower than the current research was a National Audit Office (NAO) study into NHS acute hospitals in England(9). Data wascollected in 30 hospitals over an eight week period in 1995. The immediate costs of accidents were estimated for England at approximately £12million. However, the costs were taken over a shorter time period (eight weeks against twelve weeks) and the data was nearly six years old. Cost ranges are not given and primary care Trusts were not included. It is therefore unlikely that the data is directly comparable. The 2003 review of this work by the NAO(1) concluded that the direct cost of work-related accidents was at least £173 million, but estimated that this was a conservative estimate. The results in this report were just over one third (38%) of the NAO estimate.

However, despite being conservative the research costing protocol was rigorous and applied, via a standardised costing frame, consistently in each Trust. The resultant costs can therefore be regarded as reliable, thus ensuring the validity of making comparisons between Trusts. However, they included only direct costs to the participating Trusts. An illustrative example was a case where a patient was found hanging using a rope suspended from a window catch. The ligature point had been identified as a high risk, but remedial measures had not been actioned by the time of the suicide. The incident was allocated a cost of £19.67, to cover the direct cost of staff time dealing with the immediate actions necessary. The costing protocol did not include the undoubtedly higher cost to the Trust of longer-term issues such as legal action and any official enquiry. The cost to the individual and their family are also not included, nor was any account taken of emotional issues.

In addition, the method did not lend itself to inclusion of cases of ill health, unless they were diagnosed within the data collection windows. For example, most needlestick incidents were costed at between £20 and £25. Obviously the long-term consequences of some of these incidents could result in ill health and the cost obviously does not account in any way for anxiety or distress suffered by the individual.

It is therefore apparent that a possible reason for the difference in cost estimates could be the rigorous inclusion criteria for the current study. It has been suggested that one reason for using extrapolated national costs has been to persuade senior business managers to invest in prevention(112). It has also been suggested that the larger these sums are, the more persuasive they are likely to be. However, it has already been suggested that UK business may not be fully persuaded of this argument(127). Koopmanschap and Rutten identified a body of opinion supporting the exclusion of indirect costs in healthcare-based economic evaluations unless compelling evidence for their inclusion was present(128). An alternative paradigm is therefore proposed, which concentrates more on attempting to change boardroom attitudes to costs of health and safety. This is regarded as a potential key to success.

In the experience of the author it can be more effective to take a conservative incident cost estimate to a board of Directors. Their natural, and correct, response when presented with financial data is to question their validity. When the estimates are high senior managers’ tend to look for reasons why and then to revise the estimates downward. When they are conservative they are more likely to interpret that the true loss to the business may be worse than estimated and are therefore, ironically, more likely to invest in prevention. It is therefore suggested that the results, far from being non-persuasive because of their lower magnitude relative to other estimates, could be more influential since they are obviously pragmatic, understandable by senior managers, and still represent significant losses to employers. It is recommended that

88

further behavioural based research should be conducted to test the potential usefulness of this phenomenon.

In any case the current data undoubtedly represent a significant underestimate of the true cost of occupational health and safety to the healthcare sector. Estimation of this cost represents a considerable challenge. In the participating Trusts sickness absence was not studied, but it is known that they represented approximately 6% of salary costs. If it were assumed that half of this was related to workplace occupational health and safety32 then this represents approximately £1.2billion for the NHS in England and approximately £2.5million in Scotland.

However, these estimates are, at best, speculative, given there is, as yet, no agreed standard for cost estimation or incident reporting within the NHS. This will continue to limit progress in that the debate is currently focused on measuring the magnitude of cost, rather than on means to control them.

It is not suggested that Trusts routinely cost incidents, since the cost-benefit of this action is not likely to be favourable. In any case, there is also sufficient data in the literature to convince that costs, especially of occupational ill health, are considerable. It is however recommended that costing exercises are carried out on an occasional basis in order to check progress and to reaffirm the presence of sources of high cost.

As already noted by others(128), there was no doubt that some of the costs recorded in the data collection were borne by a Trust other than the one in which the incident occurred. This was particularly noticeable with incidents occurring in primary care Trusts. If treatment was needed this was usually done via the local casualty department of an adjacent acute Trust, which would then incur the cost of treatment and potential extended stay of the patient. None of these transferred costs were estimated but they represented a significant group. This transfer of burden of cost from the primary to the secondary care establishments could represent, particularly for the primary care sector, a disincentive to accident prevention on the basis of business management. Although the cost of treatment would usually ultimately be borne by the same health authority/board, the accounting centers would deduct the cost of treatment from the budget of the relevant clinical group in the acute Trust. Additionally, the acute Trust would not recognise cases of this sort as relevant to its accident management strategy because it would regard them as externalities and not in their interests to control.

That the NHS can be seen to routinely transfer costs within its own organisation is a minor issue once consideration is given to the massive financial burden on acute Trusts from every other industry who also transfer costs via their accident and ill-health victims. This has been estimated at hundreds of millions of pounds each year(129). Proposals to recoup these costs have been included in the debate for changes to employers liability insurance. Proposals exist for a “no fault” scheme, which would ensure that the proportion of costs borne by the employer rose to a level that acted as a driver for improved health and safety management(129). Proposals such as this have the potential to assist with the difficulties already discussed with regard to enforcement versus business planning and may be more effective at changing attitudes than the current arguments.

32 This is based on the Revitalising Health and Safety Strategy, described in Chapter 1, which seeks to reduce the number of working days lost from work-related injuries and occupational ill health by 30% by 2010. If it is thought possible to reduce the rate by such a figure it must be assumed that it is currently higher than this.

89

4.1.5 Health and safety culture

The decision to use a previously extensively validated climate questionnaire(58, 59) has been vindicated by the robust nature of the data obtained, as shown by the confirmatory factor analysis (Chapter 3). This data could then be used to gain insight into the safety opinion of the participants, and ultimately an assessment of the safety climate(66). The use of a longitudinal design yielded matched pairs of responses from each individual and reduced the effect of confounders on the data while the sample size also meant that statistically significant results were obtained. The issue of triangulation of the questionnaire responses with the other performance measurement methods is dealt with in Section 3.8.7.

Although the overall questionnaire response rate at 52% was regarded as good, especially given each respondent had to provide two valid questionnaires with a year in between, consideration should be given to additional methods that could have optimised this rate further. For example, the Trusts could have encouraged staff to complete the questionnaire; perhaps by giving them protected time in which to do so. Also, the period in which the respondents were given to respond could have been extended and the non-responders followed up again. A major disadvantage to this approach was that the time period during which the questionnaire was completed might therefore not coincide with the incident data collection phase. This was something that was regarded as important to achieve if the triangulation was to be valid. Therefore, neither of these options was judged to be a cost-effective option because of the additional resources required and the statistical power of the results within the existing response rate.

The job title profile of questionnaire responders was shown in Chapter 3 to be representative of all staff groups. It can therefore be concluded that the non-responders were also representative of all staff groups. However, it is not possible with the current data set to conclude more about the non-responders. It is likely that those who responded had more positive opinion about health and safety but this risk was controlled for in the research design methodology with the same individuals providing a valid questionnaire in the two benchmarking phases to establish whether there was any change in opinion.

The most significant difference found in the questionnaire responses was at Trust level with the between Trust variation larger than the within Trust variation (p<0.05). It had been assumed at the start of the research that grouping of the data from individual units, such as NHS Trusts, would be a valid strategy. However, the apparent similarities between NHS organisations which can be experienced, at a macro level, by those who work in it appear to be over-ridden by the local culture which, presumably, has developed as a result of devolved management to the Trusts and health authorities. Whether this is something that has changed over the lifetime of the NHS is not known. However, the finding was consistent between both data collection phases and therefore represents a snapshot of the situation between the years 2000 and 2002, when the NHS was less devolved than it was during, say, the early 1990’s.

Whether there exists a demographic aspect is also unclear since none of the participating Trusts were from the same geographical area. There was a mixture of secondary and tertiary type of Trusts but the Trusts were all so significantly different from each other that it was not valid to combine them so that they could be analysed with any grouping. In retrospect it may have been useful to have included Trusts from the same health authority area. However, personal experience of tertiary, acute hospitals (in this research, Trusts E and F) is that they care for different client groups and provide different services from those in the secondary care setting, such as primary care Trusts (i.e. Trusts A, B, C, D, & G). In addition their cultures are different. It therefore seems unlikely that if this grouping had been available that the findings would have changed.

90

The other factor that yielded significant differences in questionnaire response, and therefore safety climate, were the staff categories. With the exception of Trust F, none of the other Trusts showed any significant differences between the staff groups, again indicating that the Trust unit was the grouping that exerted the greatest influence on staff opinion.

When the difference in Trust F was examined further it was only present in the first data collection phase. This effect was so strong that when the initial data, pooled from all Trusts, were examined it appeared that the medical & dental grouping represented a group with opinion on health and safety that was significantly different (lower) from the others.

The most obvious unique difference between Trust F and the other Trusts, which might explain the finding was that it formed part of a large medical school and, as such, would have a large number of student doctors and also doctors whose role included a large teaching component. This may have influenced the results, given the knowledge that health and safety was not part of the curriculum of this medical school33. The timing of Phase 1 (1st August 200 to 23rd October 2000) coincided with the annual intake of junior doctors to the Trust. Phase 2 (1st January 2002 to 25th March 2002) was at a more stable time of the year for this Trust. It seems likely that doctors could have been under greater workload pressure during the first Phase as they balanced the needs of their job with the need to ensure adequate induction and supervision of a large number of inexperienced students.

Nevertheless, it is reassuring that the method and sample size was sensitive enough to detect statistically significant differences. This creates a strong message for other researchers and highlights the danger of pooling data without first checking whether it is valid to do so. It also gives a clear addition to the body of research knowledge that it is likely that this phenomenon applies to all NHS establishments and research designs can be tailored accordingly.

4.1.6 Triangulation of health and safety measurement methods

Triangulation was a feature of the research design and is regarded as strengthening the design. Its use has been described and discussed in Chapter 1 and identified as having many advocates, although no identified proof of its effectiveness. The research design made possible effective triangulation across paradigms, leading to the identification of a number of significant correlations.

The first significant correlation was that those tests Trusts whose staff had a higher overall mean questionnaire dimension score (i.e. more positive opinion of health and safety management) were significantly correlated with Trusts with lower incident capture rates. When the control Trusts were included the only questionnaire dimension to be significantly associated with lower incident capture rate was that of safety rules & procedures. Intuitively this seems logical, since having clear safety procedures is indicative of a strong health and safety management system, which is supported by having fewer serious incidents (if the captured incidents are regarded as the most serious). This conclusion is reinforced by the consistent finding that rank of reported incident rate (i.e. all incidents), was not significantly correlated with any questionnaire dimension. In the second phase, these significant findings were repeated, thus increasing confidence in the results.

Also in the second phase was available the workbook monitoring data for the test trusts. Higher reported use of the workbook was significantly correlated with both lower captured incident rate 33 Personal communication regarding an ongoing project between HSE Health Services Unit and the Council of Heads of Medical Schools (CHMS)

91

and higher overall mean questionnaire dimension of working environment. This is an important finding but needs more work to establish the mechanism behind the correlation.

For example, it would be tempting to suggest that more use of the workbook had led to lower numbers of serious incidents and better opinion on the working environment by staff. However, there is no proof of cause and effect with the current data set, especially given that no change in performance was detected following the use of the intervention (discussed further in Section 4.3). It may be that Trusts with better commitment to health and safety might encourage more the use of the workbook and other factors were responsible for the lower serious incident rate and higher opinion. The inclusion of a behavioural aspect may have helped interpretation of these results, already identified as a possible limitation of the intervention.

Nevertheless, the results of the triangulation add to the knowledge base by providing supportive evidence for the paradigm that having a positive health and safety management system is associated with better health and safety performance(18, 25, 130-137).

The main limitation to these significant findings is that, because the control Trusts were not included in the monitoring scheme, there is no data on which to confirm further, or otherwise, the findings. This is an acknowledged limitation of the research design and future work using research designs of this type should ensure that the control received the same monitoring as the test Trusts. The reason that this was not done was that, although the control Trusts had established health and safety management systems supported by workbooks, there was incomplete data on workbook holders, on which to base telephone interviews. This was because the workbooks had been in place for some time and gradually the original list of holders had changed.

4.2 HEALTH AND SAFETY MANAGEMENT INTERVENTION

The workbook was based on the main principles underpinning HSE’s guidance to health and safety management(74, 99). Rather than a detailed information resource, it was intended to be a pragmatic tool that could be used by line managers to ensure that they had addressed all of the aspects of the management system.

It has been suggested that a possible limitation of impact evaluations is that interventions may produce a small effect that is too weak to be detected(138). In the case of this intervention, the field data can provide evidence as to whether the effect was small.

Firstly, the maximum effect that would not be detected can be defined from the power of each aspect of the performance evaluation:

�� For the incident reporting data, based on a standard deviation of 0.014 units in average reported incident rate per person working days, a sample size of seven Trusts would have 98% power in detecting a 2% shift in reported incident rate;

�� For the questionnaire responses (standard deviation of 0.83) units in dimension score measurements sample size of 200 would have 96% power in detecting a 0.2 shift in dimension score;

�� For the workbook usage data (based on an estimated standard deviation of 0.49 units in answers to telephone interview questions) a sample size of 100 would have 98% power in detecting a 15% shift in answer response.

92

The power of the study was therefore regarded as high. It can be assumed that, if there was any change in performance brought about by the use of the workbook, its effect was below these thresholds.

There are two factors that may have impacted on effectiveness of the intervention in this study:

�� The degree of use of the workbook; �� The time interval between the benchmarking phases.

These will be considered in turn.

4.2.1 The degree of use of the workbook

The five Trusts that participated in this phase of the research were free to manage deployment and use, as they saw fit, within the general guidance given by the project team. This meant that there would be a natural variability of attitude and tactics of use within Trusts.

However, as has been shown in Chapter 3, approximately one third of workbook holders had ceased using the workbook by the end of the 12 months implementation phase. It was also shown that, of those who remained, their use of the workbook decreased significantly over the period.

This data on workbook use was obtained from the telephone interviews and inspections of workbooks. As has been described in Chapter 3, many of those who dropped out of the telephone interviews were either unavailable for comments or declined to take further part in the research. It is likely that this group were not using the workbook. There was therefore likely to have been a self-selection bias within the group who provided telephone interview data, probably those who had the most positive views. Although it is not possible to conclude this with certainty it should be borne in mind when interpreting the results. If anything, it indicates that workbook usage could have been worse than measured.

This serves to reinforce the conclusion that there was low implementation of the workbook. It cannot therefore be concluded that the workbook produced an effect that was too weak to be detected. Had the workbook been more fully used as intended, there might have been an effect on health and safety performance.

The decision to allow Trusts relative freedom to manage implementation of use of the workbook was deliberate in that, with the limited resources available to the project, it was necessary to have the co-operation of the participating Trusts. This would have been difficult to secure if participation meant the imposition of a rigorous common standard across all Trusts, since Trusts worked in a culture of establishing procedures, which took account of local factors to broad national standards.

However, this limitation of the research design should be controlled for in future research. Attempting to secure greater commitment from Trusts could achieve this, possibly by securing advance agreement to predetermined measures designed to encourage use.

4.2.2 The time interval between the benchmarking phases

The second factor is the length of time between the first and second benchmarking, which, at approximately 12 months, may have been too short for change to become established. In other

93

words, the process of change is complex and, in a complex organisation like the NHS, likely to take time.

For example the European Foundation for Quality Management (EFQM)34 suggest that the process of implementation of their Business Excellence Model (BEM) takes typically between three and five years. Wright et al(139) adopted the model as part of a Safety culture improvement matrix and advocated that improvement was multi-factorial, therefore likely to take a similar time as the BEM to establish. Ogden(140) described the stages of change model as a five-stage dynamic process which takes place over time. Actively engaging in a new behavior is the fourth stage, preceded by pre-contemplation contemplation and preparation. The author contended that individuals might move between preparation and contemplation several times before taking any action. At group level Houghton et al(141) introduced the concept of a “group mind” in their five-stage cognitive model of team behaviour sustainability. This took account of the group dynamic that undoubtedly exists in the NHS.

The initial benchmarking data was designed to provide, not only a measure of health and safety performance, but also, via the HSE root cause analysis methodology, a baseline measure of the health and safety management system. This was because it was used to track causes of incidents to their root in the health and safety management system of the organisation and had been validated against audit, the more commonly used method to assess health and safety management systems(34).

Further benchmarking exercises with the same Trusts and individuals would be a way to test whether a longer time interval would result in a greater effect. However, the use of the workbook has already been shown to be low, decreasing significantly over time. Unless the Trusts had put greater effort into encouraging use the greater time period would be unlikely to show a difference. Nevertheless, if detailed information about the measures taken by Trusts to encourage use could be obtained and controlled for it might be a worthwhile exercise.

An alternative possibility may be via the use of certification standards, such as BS8800:1996 or OHSAS18001:1999(30), to support the use of workbooks of this type, used as a form of quality manual. This type of practice is already well established and, the benefit of independent inspections of the workbooks, tied into retention of certification, is likely to be a major motivating factor in their use. Unless and until such a scheme is introduced, consistency of use within individual Trusts will continue to be dependent on the motivation and enthusiasm of the senior and line management of the organisation. Consistency will be much more likely to be attributed to chance in this scenario. The continuing UK policy of self-regulation is more likely to continue with the use of standards of this sort on a voluntary basis, for the foreseeable future.

Nevertheless in its relatively crude format, the workbook did meet an apparent need, illustrated by the degree of usefulness reported by users in the telephone interviews conducted to assess feedback on its use. With further refinement, possibly involving some of the individuals who used it during the project, the workbook and its briefing schedule could be made freely available via the Internet. This could be done via the HSE or Scottish Executive sites.

However, it is suggested that the workbook be regarded as only a part of any strategy to improve health and safety performance and that for significant change to occur it should be used as part of a multi-factorial model, such as the ones described above. This would offer a potentially fruitful avenue for further research.

34 http://www.efqm.org

94

4.3 HEALTH AND SAFETY MANAGEMENT PERFORMANCE CHANGE MEASUREMENT METHODOLOGY

A critical evaluation of the research design used in this report was carried out to the strategy identified in Chapter 1. Seven main headings were used so that a relative analysis of the strengths and weaknesses of the research can be made. These are:

�� Intervention objectives; �� Conceptual basis; �� Research format and design; �� External validity; �� Threats to internal validity; �� Outcome measurement; �� Other limitations.

Each will be considered in turn.

4.3.1 Intervention objectives

The aim of the intervention was stated and supported by three relevant objectives (reviewed in Section 4.4). The aim was, broadly, to measure change in health and safety performance in the healthcare sector. The maximum effect that could be detected by the method was established.

4.3.2 Conceptual basis

Although the literature base was diverse in terms of subject base, it mainly comprised non-experimental or quasi-experimental approaches, from a positivist perspective. Although currently popular, behavioural research and that based on management principles, such as qualitative research from a phenomenological standpoint (for example (78, 105)) were relatively less well established in the health and safety field. More lateral thinking is needed as to the foundations of health and safety research, perhaps using other models more normally associated with business(91) or organisational development(142).

The research design included both positivist and phenomenological aspects(68). The objective incident analysis and properties sought data to explain findings. Descriptive data was obtained from the questionnaire and were used to try to understand why individuals, and the Trusts for which they worked, had different experiences within a similar environment of provision of healthcare. An assessment of the baseline health and safety system, as well as performance, was established with the first benchmarking phase. The detection of change was made possible by the second benchmarking phase.

This combined approach was largely successful in that there was more than one indicator of effectiveness successfully measured by triangulation.

4.3.3 Research format and design

So far as the positivist paradigm is concerned, the definitive source of guidance on research design within occupational health and safety was that of NIOSH(44). However, other disciplines, such as medical research and the work of the Cochrane Collaboration(76), bears close scrutiny by occupational health and safety researchers. Whether there is sufficient momentum for an independent health and safety collaboration, which could seek to establish best practice and guidance for research and to form a focus for prioritising research strategies is

95

unclear. The Cochrane system provides a validated pre-existing methodology for research reviews and meta-analyses but the primary research on which it is based must also be of high quality, possibly based on the features of randomised controlled trials.

To attempt to mirror, so far as possible, this best-practice guidance, the research format was that of primary research of a health and safety impact evaluation, the design of which was informed by a review of relevant literature. It was a quasi-experimental, longitudinal design, which included a comparison group. It included introduction of an intervention that was staggered over a three-month period and multiple measures of outcome were used. It was not feasible within existing project budgetary and time constraints to consider the use of a time series design by including a third data collection period (say a further twelve months after the second), or for reversal of the intervention.

However, as has already been discussed, if the intervention produced an effect, it was below the limit of detection of the method. This may have been related to low implementation or the time period that may have been too short for change to become established. These are regarded as two limitations of the research design.

4.3.4 Threats to external validity

The experimental design included a comparison group in that two of the seven participating Trusts did not receive the workbook intervention because they already had a health and safety management system supported by the use of a workbook. This meant that there were no ethical issues associated with not issuing a workbook to these two Trusts. This is a positive feature and strengthens the research design in that if, the intervention produced an effect it should have been already present in the comparison Trusts and that change would therefore occur only in the test Trusts. The main limitation to this was that the degree of use of the comparison Trusts workbooks was not established.

Another limitation was that the selection of the participating Trusts was non-random. This was because the intervention had not only to be of interest to the participating organisations but also for them to be committed to working within the research resources and framework. This limitation was a compromise that was necessary to ensure the continuing participation of the Trusts.

The initial selection of the individuals to receive a safety climate questionnaire was a stratified random selection. The same individuals completed two questionnaires, approximately twelve months apart. This is a strong feature of the research design.

4.3.5 Threats to internal validity

There was potential for the research results to be sensitive to a history threat. The research took place over a twenty-two month window. The NHS is characterised by constant changes in both management and organisational practices and, although formal major reorganisation did not occur during the research period, there were still extraneous influences. For example, two of the Trusts were involved in planning and implementing major construction projects. Also, major governmental initiatives, such as the Revitalising Health and Safety Strategy were launched and publicised. At the outset of the research these possibilities were recognised. The choice of length of the intervention phase was largely influenced by the need to minimise this risk, while still giving sufficient time for any effect to become established. The choice therefore represented a balance of these risks. The use of control Trusts also strengthened the strategy.

96

By the time of the second data collection period the participants had become familiar with the root causation methodology. This represented a potential instrumentation threat (i.e. the measurement method altered between the two project phases). An example to illustrate this is that the number of times a root cause was cited reduced in the second phase. This could be interpreted as an improvement in performance but was identified at an early stage as a consequence of increased familiarity with the methodology. This measure was therefore not used as an indicator of performance (the identification of the root cause itself was used instead). It is therefore unlikely that an instrumentation bias existed in the research results.

Extreme values were detected in the results. In particular, the medical and dental staff of Trust F was identified as having a significantly different opinion within the questionnaire responses. This represents a regression-to-the-mean threat. This was overcome by the analysis strategy, which avoided pooling data once it became clear that results from within individual Trusts were significantly different from each other. This meant that the extreme values were identified and classified as being from within a single Trust and present only in the first data collection phase. Had this strategy not been adopted it is likely that a wrong conclusion would have been reached that medical and dental staff across the whole of the NHS had significantly different views from other staff.

The benchmarking events themselves will have had an effect on the outcome. The inclusion of a comparison group should have ensured that testing and placebo threats were controlled. So far as the potential for an effect to occur because of the involvement of the research team (i.e. Hawthorne effect(143)) is concerned, the comparison group was treated as close as possible to the same as the intervention group, with the exception of the intervention and monitoring of workbook holders. This was also the reason why contact with the intervention group was kept to an absolute minimum during the intervention period. For example the telephone monitoring of workbook holders within the test Trusts was restricted to three times in twelve months (rather than the more traditional three-monthly grouping) and also only a single visual inspection of workbooks was carried out. Apart from these times contact was avoided with the workbook holders. It is therefore thought unlikely that these threats have exerted undue influence on the research outcomes.

The presence of a maturation threat could have influenced the results, particularly given the increasing emphasis within the NHS on risk management and their related insurance schemes (CNST(144)/CNORIS(17)). These schemes were being publicised at the time of the research and Trusts were being encouraged to develop measures to ensure effective management of risks. However, the presence of the control group should have minimised this risk.

Dropout threats were similarly controlled in that none of the participating Trusts left the study once the first benchmarking data was obtained. Within the random sample of questionnaire respondents all staff groups were represented equally in both data collection phases. However, although the initial selection of participants was random, and efforts were made to encourage completion of the questionnaire, replacing non-responders with others selected at random from within the relevant staff group, it is likely that a selection bias still exists within the data. This is because the questionnaires were likely to have been returned by individuals who were more motivated to do so. This is demonstrated by the increase in response rate between the first and second data collection phases. However, the mean response rate from the first phase was satisfactory (55%) and not regarded as a major threat to the results.

It is unlikely that a contamination (or diffusion) threat existed. This is because NHS Trusts tend to work autonomously from each other and, so far as the author is aware, were not in communication with or influenced by each other. For the same reason, it is unlikely that a

97

resentment threat existed between those Trusts in the control group and those implementing the intervention.

4.3.6 Outcome measurement

Outcome measurements were carried out to a rigorous protocol. The incident data comprised of a combination of measurement scales, which were analysed using descriptive methods, including bivariate analyses and cross-tabulations of rates. Ordinal data, such as system failure and management root causes, were analysed using non-parametric tests. Parametric tests such as correlation and T-testing were used for interval and ratio data such as types and costs of incidents.

However, a limitation of the incident data analysis was the small sample size of five test and two control Trusts. This limitation did not exist with the questionnaire response data. Parametric testing (such as ANOVA) was only used once it had been established that it was appropriate to do this with the ordinal data from the questionnaire responses. The analysis strategy was robust in that it avoided pooling of data from individual trusts once it was identified that it was inappropriate to do so. Multiple comparison testing (Scheffe Test) was used to identify where the differences lay. This strategy was robust and gave reliable results.

The workbook usage telephone interview monitoring data was useful in that it clearly established the low degree of implementation of the workbook. The rank order data was also useful for use in the triangulation exercise and were shown to correlate significantly with captured incident rate and the questionnaire dimension of Working Environment.

4.3.7 Other limitations

The potential for selection threats has been identified. This is because any effect from the intervention could be due to differences between the Trusts rather than the intervention itself. The inclusion of the two comparison Trusts was intended to control for these threats. However, in practice the influence of individual Trust units proved to be the overriding finding and differences between the comparison and test group could not be detected. It is suspected that this may have been due to the low use of the intervention workbook. An intervention that sought to control this effect would be useful although, because this would require more intervention from external researchers, this may lead to an increased risk of Hawthorne effects.

A number of issues were not captured by the method. These included cases of occupational ill health; hospital acquired infections; the considerable potential for impact by civil litigation.

4.4 ASSESSMENT OF THE DEGREE TO WHICH THE RESEARCH AIM AND OBJECTIVES HAVE BEEN ACHIEVED

In Chapter 1 it was identified that there was a need to establish best practice in terms of research methodology to aim to evaluate whether it is possible to reliably measure change in health and safety performance when a formal health and safety intervention is introduced to the healthcare sector. This was done by a combination of a review of the literature and primary research. Three research objectives were set, which are reviewed against the outcomes of the research below.

Objective 1: The design of a methodology for evaluation of health and safety performance. This was informed by the literature review described in Chapter 1, which allowed the selection of an appropriate research design. The objective was achieved with the selection of the incident analysis methodology that allowed numbers, costs and management system root causes (which

98

also was used as an assessment of the health and safety management system) to be measured plus the safety climate questionnaire that allowed the measurement of NHS staff opinion on health and safety; Objective 2: The design of a health and safety management instrument, which could be used as an intervention tool for the improvement of health and safety performance. This was informed by the literature review and achieved with the development, by the author, of the managers workbook; Objective 3: The design of a research methodology to measure change in health and safety performance. This objective was achieved with the implementation of an appropriately designed field study, which allowed the evaluation methodology to be used in a quasi-experimental before-and-after longitudinal design, which included a comparison group.

The research has added to the knowledge base by identifying a number of important concepts and issues that can help inform future research.

Despite the fact that minimal change in performance was detected, the research design was able to show significant correlations between three of the performance evaluation aspects (i.e. degree of workbook use, rate of captured incidents and staff opinion on the questionnaire dimension of working environment). The method was robust and possessed an ability to detect change in health and safety performance. It is highly likely that, if a change in health and safety performance had occurred, the method would have detected it, although the cause of the change would not be identified. It seems likely that the small improvement in climate that was detected was at the limit of detection of the method.

This is contrary to claims of dramatic changes in performance cited by others(86, 145) (and reviewed in Chapter 1) but supportive of claims by others who have also failed to detect impressive changes(80, 84, 146). This provides evidence that claims for improvements in health and safety performance should be treated with caution, unless the research design is robust.

Future research strategies for health and safety research programmes now have evidence that priorities for action must include measures to ensure the improvement in research quality. The research in this report concentrated on the healthcare sector but the findings are likely to be of relevance of findings to other industry sectors.

It seems likely that, given the sensitivity of the research method, a simple approach to health and safety management is unlikely to improve performance in the long term. Concerns have been raised earlier in this Chapter that, despite the effort of HSE over two decades, effective management of risk remains elusive. Their systems based approach would benefit from a more in-depth critical evaluation, such as how to positively affect safety behaviour – a challenging task(147).

It is nevertheless concluded that the three project research objectives, set at the start of the project have all been achieved, and that the research aim has been fulfilled.

4.5 SUMMARY OF CONCLUSIONS AND RECOMMENDATIONS

4.5.1 Health and Safety Management Performance Evaluation Methodology

Conclusions �� The conceptual basis for the research was robust in that it included both positivist and

phenomenological aspects. Triangulation across these aspects was achievable and showed statistically significant correlations;

99

�� Although captured incident rates offered potential, as a performance evaluation method, the definition of an incident should be made clearer across the NHS. This would facilitate the production of better benchmarking data;

�� Incident reporting in the NHS has been largely used as a performance indicator. The process of learning from adverse events is comparatively under-developed, especially for non-clinical incidents. The HSE root cause analysis method could reliably identify health and safety management system weaknesses and therefore offered strong potential to be used by organisation to learn lessons from their incident data;

�� Estimated costs ranged between 0.06% and 1.44% of the running costs of the NHS. These are low costs compared to other published data (between 20% and 50%). The costing protocol is therefore regarded as very conservative (cost for ill health, infection and litigation not reliably captured). However, it was rigorous and reproducible;

�� The use of the questionnaire yielded useful, robust data; �� The results of the triangulation added supportive evidence that efforts in health and safety

management are associated with better health and safety performance.

Recommendations �� A common standard for incident reporting and costing should be developed for the NHS,

supported by guidelines. This would ensure a clearer purpose for incident reporting and facilitate improved communication and collaboration between departments currently responsible for individual aspects of incident reporting;

�� Priority should be given to ensuring the effective management of the risk control system of risk assessment, in particular the areas of planning and implementing. If this were done it is likely that health and safety performance would improve;

�� The potential for many incidents in the NHS to be unpreventable (the special category incidents in this research) should be debated and investigated further. This should be done to establish if a clearer definition of reasonably practicable could be established for the NHS;

�� The HSE root cause analysis method should be further developed before it is suitable for use as a routine method;

�� Training providers and accreditation bodies should look at their course content for training of safety advisers to ensure that the skills of behavioural safety and change management are given a higher priority than at present;

�� Further behavioural-based research should be conducted to investigate whether a conservative costing estimate protocol is likely to convince senior managers and budget controllers to take action to control health and safety risks

4.5.2 Health and safety management intervention

Conclusions �� The workbook intervention failed to produce a detectable effect on health and safety

performance; �� The maximum effect detectable by the evaluation method was a 2% shift in reported

incident rate, 0.2 point shift in questionnaire dimension score, and 15% shift in users answers to telephone monitoring questions;

�� Use of the workbook decreased significantly over the 12 months implementation period,indicating low implementation.

�� It cannot be concluded that the workbook produced an effect that was too weak to be detected. Had the workbook been more fully used, there might have been a detectable effect;

�� The time between the two phases might have been too short for change to occur.

100

Recommendations�� Future research should ensure a standardised approach in individual Trust units to

encourage use of the intervention; �� The use of certification standards as a motivating factor in workbook usage should be

further explored; �� Further research into the impact of workbook-type interventions should include other

aspects of multi-factorial change models; �� Since it is likely that other organisations could benefit from the workbook it should be

refined and adapted for wider use, possibly via the Internet.

4.5.3 Health and safety management performance change measurement methodology

Conclusions The main limitations of the method were: �� The intervention did not produce an effect that could be detected (this may have been

caused by low implementation or too short an intervention period). The comparison Trusts did not receive follow-up monitoring, which might have assisted with interpretation of this finding;

�� Non random selection of participating Trusts; �� History threats that may have occurred over the project duration; �� Small sample size in the incident reporting and root cause analysis aspects of the method; �� Cases of ill health, infection, or litigation were not easily included.

The main strengths of the method were: �� Intervention objectives that were stated and defined; �� The use of more than one indicator of effectiveness; �� The use of a comparison group; �� Questionnaire sampling strategy which produced a robust sample, sensitive to statistical

testing and identified the need to avoid pooling data from individual Trusts; �� Effective use of triangulation.

Recommendations:

Further debate suggested exploring whether an independent health and safety collaboration could be established to identify examples of best practice and form a focus for prioritising research strategies.

101

APPENDIX 1 HSE ROOT CAUSE ANALYSIS METHODOLOGY

Figure 1 Process for screening and investigating incidentsIncident Occurs

Incident Report Form isCompleted and Submitted

Harm orPotential for

Harm?

Outwith Power ofTrust to Control? Discuss with RA

Greater than £5 or 15minutes of costed staff

time?

Exclude fromStudy

Allocate Investigation Pack

Events & Causal FactorsChart

Conduct Interviews

Complete Forms 1A & 1B No ApparentSystem Failures

Discuss with RA

Place In SpecialCategory File

Transfer System FailuresFrom Form 1A to Form 2A

Refer to Root CausingTables & Complete Form 3

Transfer Root Causes & Codes to Form2B & Enter Evidence

Complete Project Checklist

END POINT

END POINT

END POINT

NO

YES

NO

NO

YES

NO

YES

NONE FOUND

Process for Screening & Investigating Incidents

102

Incident investigation example

Agitated patient referred to medical admissions ward as a potential suicide risk

Two incident forms were submitted for this patient. These were regarded as the same incident. In both cases a “red alert” was activated. This meant staff attended from other areas of the hospital to assist in a restraint manoeuvre, therefore there was more than 15 minutes of costed staff time involved. Incident was screened – INCLUDED IN STUDY.

Incident form submitted at 18:15 �� 26 year old male. Was supposed to be lying on bed eating �� Located in side room of medical admissions unit �� Description of incident: “Threw meal tray onto floor. Staff nurse in room at time tried

to restrain patient. Became physically aggressive towards social worker. Trying to get out of room. Nursing staff assisted to restrain patient. Red alert put out. Self poisoning – suicidal intent. Major social problems. Patient nursed in single room.”

Incident form submitted at 19:30 �� Patient trying to leave side room �� Description of Incident: “Staff were waiting for psychiatrist to come from Psychiatric

Facility, as patient was to be sectioned. Patient burst from room with social worker trying to prevent him from leaving. Patient became physically aggressive – red alert put out. Self-poisoning – suicidal intent. Major social problems. Police had arrested girlfriend on ward. Patient had been and continued to be nursed in single room.”

What were the crucial questions?

Was there potential for harm?Although no one was actually hurt, there was potential for harm to both patient and staff.

Was there a cost to the Trust? Staff were called out from other areas of the hospital to attend the incident therefore this incident should be included due to the opportunity cost of staff who did not work in the ward where the incident took place.

Was the incident within the power of the Trust to control?It was not reasonable to expect an acute NHS Trust to have measures in place to avoid disturbed/aggressive patients. However elements of management failure may emerge in the way in which such incidents are dealt with. There should be systems in place to reduce the risk to staff/patients so that the risk is as low as is reasonably practicable.

An assessment of the control measures supported by appropriate risk control systems must be made. The following would be needed to do this:

1. What events led to the occurrence of the incidents? 2. What are the Trust-wide and local policies and procedures for dealing with violence

and aggression and detaining patients, regarded as a suicide risk? 3. What actions are necessary to legally detain a patient against their will?

103

Interviews conducted

�� A member of staff on the ward who was not involved in the incident regarding the procedures for emergency assistance

�� The staff nurse who reported the incident/restrained the patient prior to arrival of the “fast action response team”

�� The senior staff nurse on the ward �� A nurse manager who is involved in training of the response team �� The health & safety co-ordinator �� The consultant in charge of Accident & Emergency Dept �� The clinical director of medicine �� A violence & aggression adviser

Questions to Consider in Identifying Management System Root Causes

Policy Reference was made to local or Trust wide policies. The Health & Safety Policy may have been a good starting point, but many other systems may have had their own policies, e.g. risk assessment, manual handling, sharps handling etc.

Implementing Were there means to operate the system? (e.g. within the system of risk assessment, there may have been trained risk assessors, forms disseminated, specialist advisers etc.)

Had the system been implemented partially or not at all? What was the position regarding risk assessment (the risk may have been documented but appropriate control measures may have not been in place, or not utilised).

Control Was there a standard that defined responsibilities? This may have been laid out in the relevant policy, or within individual employment contracts. In most cases it will have been assigned through the line management chain.

The main thing that should have been considered was whether responsibility was adequately assigned not whether the individual or group concerned met it. Not meeting responsibilities should have been regarded as equivalent to not implementing the system. Awareness of responsibility could have been a factor.

Planning Was there a plan to implement the system in the organisation (resources, training, procedures etc.). Also was there local planning, which might have involved task management, supervision, communication etc.

CompetenceWhat standard of competency was required by the organisation? What was the skill knowledge and experience of the individual or group involved? Were they trained appropriately to implement this system? Was their training up to date?

104

Communication What was the standard for communication? This may have been written information in policy files, on Trust intranets, on staff notice-boards, verbal information via departmental meetings, induction training, visits by specialist advisers’ etc. When interviewing individuals, who might have breached the policy, try to ascertain whether they were aware of this? Was this a one-off slip up – people might have said, “I don’t usually work this way but…” or was the person adopting a method that was used by everyone in their workplace? This may have helped to assess the effectiveness of the communication methods employed.

Co-operationWhat were the consultation mechanisms that were in place? What ways were staff involved in the development and planning of the system? Were they given the opportunity to work together to develop systems that can work in their particular environment? This root cause should have been associated with team issues, not whether or not someone implemented a system.

Measuring How was the system measured? This may have been written into a policy or individuals employment contract. Consider how the effectiveness of the system was determined. Were measurements made at a regular frequency?

Reviewing This involved review of the system, not of individual risk assessment etc. Was there a formal process for review of the system? If so had the system been reviewed when it should have been and was the outcome from the review been acted upon.

Now we know WHAT HAPPENED we need to establish the ROOT CAUSES

(i.e. Identification of system failures & management root causes)

What systems were involved in handling an incident such as this? In this case the investigator chose three relevant systems to explore.

1. Training – Management of violence and aggression 2. Procedures – Detention for mental health assessment 3. Emergency procedures – Dealing with aggressive patients.

1. Training

What were the “active failures”?�� Improper restraint of the patient prior to the arrival of the red alert team. �� Inadequate training provided, particularly with regard to de-escalation techniques and

breakaway techniques.

What was the evidence? �� Only individuals who were on the response team rota were trained.�� Training took place over one day, with a half-day refresher on a yearly basis. This is

not enough time to learn both de-escalation and control and restraint techniques.�� Internal staff who had attended a course delivered training themselves, not by

experienced trainers.�� The training programme was targeted on the basis of strength, i.e. mostly male staff

trained, rather than areas where violent incidents occurred most frequently.

105

What were the management root causes? �� Policy was inadequate

2. Procedures

What were the “active failures”?�� The patient was detained (illegally) for four hours against his will.

What was the evidence? �� Three red alerts were called over an extended period. No record of detention of the

patient under the Mental Health Act during that period.

What were the management root causes? �� There was no Trust-wide or departmental policy in place regarding the procedure to

adopt for detention of patients requiring psychiatric assessment (based on the Mental Health Act).

3. Emergency procedures

What were the “active failures”? �� Improper restraint of the patient prior to the arrival of the red alert team

What was the evidence? �� Staff nurse involved stated in interview that himself, the social worker and the doctor

were wrestling with the patient on the ground attempting to detain him on the floor. Only the staff nurse had received any training on control and restraint techniques. Therefore this group manoeuvre put the patient/staff at risk.

What were the management root causes? �� Implementation - Lack of implementation of management of violence & aggression

policy – Two individuals restraining the patient were untrained. �� Planning - Inadequate resources allocated to the development, implementation and

maintenance of emergency procedures. �� Competency - Standard of competency was inadequate for two reasons

- Did not specify evaluation of training content - Standard did not specify the provision of competent advice e.g. from a

psychiatrically trained nurse – staff had very few psychiatric skills to draw from. �� Communication – Not all relevant staff groups were informed of the system. �� Co-operation – Not all relevant staff groups were consulted and few were involved in

the planning, measuring and reviewing of the system.

The information obtained from the investigation has been inserted into the project forms below. The codes obtained in Form 3, the “System failure form” should be entered into the project database for subsequent analysis.

106

Accident Ref: Leave blank Date: 28/09/00 Time 1815

INITIAL FOLLOW UP

1. Follow up incident form with person who reported it �

2. Does the incident cost more than £5 or take up more than 15 minutes of costed staff time?

No end Yes ��

3. Was there harm or potential for harm? �

______________________________________________________________

INVESTIGATION

4. Allocate investigation pack and mark date/time of incident on forms �

5. Obtain details about costs and incident. Identify failed systems. Forms 1A and 1B

�

6. Draw up Events and Causal factors chart to identify further issues �

7. Transfer details to Evidence table Form 2A �

8. Examine root causes Form 3 �

9. Complete Evidence Table Form 2B �

10. Verify that no costs are outstanding �

COMPLETED �

Date Actions complete / Outstanding / Notes Initials 27/11/00 All incident investigation/root cause analysis complete. Logging of

incident outstanding.

Figure 2 HSE Project Checklist

107

Figure 3 Event and causal factors chart

108

Department where incident occurred: Medical assessment Main Department* where management failure arose: Medical assessment

ACDC Dep’t Code*: 64

Accident Ref: Leave blank Date: 28/09/00 Time: 1815 Describe the sequence of events and consequences (What, When, Where, Who, Which, How, Why?)

Interviewed

26 year old male patient admitted as ?overdose to medical assessment ward, accompanied by social worker. Psychiatrist called from local primary care hospital to assess patient but takes four hours to arrive. During this time the patient attempts to leave the ward and becomes aggressive upon being held there against his will. Three V&A alerts were called and the patient was restrained.

Incident Reporter�Injured Party Line Manager �Senior Nurse�Nursing Manager�Director (Med) �Doctor �Witness �Others (please state)Health & Safety Co-ordinator A & E Consultant V & A Adviser

Identify the physical controls: Were they used? Were they effective? Elimination Substitution Enclosure Time 72 hr detention SpaceContainment design Insulation Guards Interlocks Other: (Please state)

LEV Safety valves Two handed control PPE Access/Egress Workstation layout Man/machine interface Work environment Equipment design

No, a registered medical practitioner could have used this to legally detain patient

Not applicable

Identify organisational controls: Were they used? Were they effective? Responsibilities Risk assessment Training �(1)Supervision procedures�(2)DetentionWork organisation Emergency procedure�(3)Work environment Maintenance Disposal Management of violence & aggression Control of contractors Procurement

Security Handling sharps Recruitment Installation Commission Specification Inspections Decommission Change management Design (Humanfactor) Other (please state)

(1) YES (2) NO (3) YES (V&A alert was activated)

(1)NO(2)Not applicable (3)Partially, due to increased numbers of staff present to deal with the situation

Identify who was present (1) social worker (2) nursing staff (3) doctor (4) police

Were they supposed to be there? (1) No (2,3,4) Yes

Identify who was absent Psychiatrist

Where were they? Not established

Figure 4 Data capture form – Form 1A

109

Figure 5 Accident/incident costs – Form 1B

110

Acc

iden

t Ref

: Le

ave

blan

k Su

mm

ary:

Vio

lent

inci

dent

invo

lvin

g su

icid

al

patie

nt a

wai

ting

psyc

hiat

rist

Dat

e:

28/0

9/00

Tim

e: 1

815

FOR

M 2

A –

leve

l 2

FOR

M 2

B –

leve

l 1

SYST

EM

Act

ive

failu

re

Evid

ence

M

anag

emen

t sys

tem

fa

ilure

(roo

t cau

se)

Roo

tca

use

code

Evid

ence

TRA

ININ

G

1. P

atie

nt w

as im

prop

erly

rest

rain

ed

befo

re th

e re

spon

se te

am a

rriv

ed.

2. In

adeq

uate

trai

ning

pro

vide

d,

parti

cula

rly w

ith re

gard

to d

e-es

cala

tion

tech

niqu

es a

nd b

reak

away

, but

als

o fo

r co

ntro

l & re

stra

int t

echn

ique

s.

1. T

he p

atie

nt w

as

wre

stle

d to

the

floor

by

a st

aff n

urse

, do

ctor

and

soci

al

wor

ker.

Onl

y th

e st

aff n

urse

had

any

tra

inin

g in

safe

re

stra

int t

echn

ique

. 2.

Mos

t sta

ff

mem

bers

on

the

war

d ha

ve h

ad n

o de

-esc

alat

ion

train

ing.

Tho

se th

at

are

train

ed to

re

stra

in h

ave

rece

ived

min

imal

train

ing.

POLI

CY

1.

2 1.

Tra

inin

g on

ly fo

r 1 d

ay in

itial

ly

+ ha

lf da

y re

fres

her y

early

. 2.

Tra

inin

g is

del

iver

ed b

y in

tern

al

staf

f who

hav

e at

tend

ed a

cou

rse,

no

t by

expe

rienc

ed tr

aine

rs.

3. T

rain

ing

prog

ram

me

was

not

ta

rget

ed to

thos

e ar

eas w

here

vi

olen

t inc

iden

ts o

ccur

mos

t fr

eque

ntly

.

111

Acc

iden

t Ref

: Le

ave

blan

k Su

mm

ary:

Vio

lent

inci

dent

invo

lvin

g su

icid

al

patie

nt a

wai

ting

psyc

hiat

rist

Dat

e:

28/0

9/00

Tim

e: 1

815

FOR

M 2

A –

leve

l 2

FOR

M 2

B –

leve

l 1

SYST

EM

Act

ive

failu

re

Evid

ence

M

anag

emen

t sys

tem

fa

ilure

(roo

t cau

se)

Roo

tca

use

code

Evid

ence

PRO

CE

DU

RE

S (T

O L

EG

ALL

Y

DET

AIN

PA

TIEN

TS

AG

AIN

ST T

HEI

R

WIL

L)

Patie

nt w

as d

etai

ned

for 4

hou

rs u

ntil

a ps

ychi

atris

t cou

ld a

rriv

e an

d m

ake

an

asse

ssm

ent.

As i

t was

the

patie

nt’s

will

to

leav

e th

e ho

spita

l the

det

entio

n w

as

illeg

al a

nd a

dded

to th

e pa

tient

’s st

ate

of

agita

tion.

Had

he

been

lega

lly d

etai

ned

chem

ical

rest

rain

t wou

ld h

ave

been

an

optio

n th

at m

ay h

ave

redu

ced

the

risk

to

staf

f. A

llow

ing

the

patie

nt to

dis

char

ge

him

self

wou

ld a

lso

have

redu

ced

the

risk

to st

aff.

1. O

n in

terv

iew

the

staf

f nur

se p

rese

nt

was

una

war

e of

the

cond

ition

s of

dete

ntio

n.

2. N

o re

cord

exi

sts

of d

eten

tion

of th

e pa

tient

und

er th

e m

enta

l hea

lth a

ct.

3. 3

V&

A a

lerts

w

ere

calle

d ov

er a

n ex

tend

ed p

erio

d.

POLI

CY

1.

1 N

o Tr

ust-w

ide

or d

epar

tmen

tal

polic

y in

pla

ce (c

omm

unic

ated

in

writ

ing

or v

erba

lly) r

egar

ding

the

proc

edur

es to

ado

pt fo

r det

entio

n of

pat

ient

s req

uirin

g ps

ychi

atric

as

sess

men

t (ba

sed

on th

e m

enta

l he

alth

act

). C

onfir

med

by

the

clin

ical

dire

ctor

of m

edic

ine.

EMER

GE

NC

Y

PRO

CE

DU

RE

Im

prop

er re

stra

int o

f the

pat

ient

prio

r to

the

arriv

al o

f the

resp

onse

team

. Th

e pa

tient

was

w

rest

led

to th

e flo

or

by a

staf

f nur

se,

doct

or a

nd so

cial

w

orke

r. O

nly

the

staf

f nur

se h

ad a

ny

train

ing

in sa

fe

rest

rain

t tec

hniq

ue.

IMPL

EMEN

TIN

G

7.4

Res

train

t of t

he p

atie

nt sh

ould

be

carr

ied

out,

only

by

mem

bers

of

the

resp

onse

team

trai

ned

to

rest

rain

safe

ly.

PL

AN

NIN

G

6.2

The

reso

urce

s allo

cate

d to

the

deve

lopm

ent,

impl

emen

tatio

n an

d m

aint

enan

ce o

f the

em

erge

ncy

proc

edur

e fo

r vio

lent

inci

dent

s w

ere

inad

equa

te.

112

Acc

iden

t Ref

: Le

ave

blan

k Su

mm

ary:

Vio

lent

inci

dent

invo

lvin

g su

icid

al

patie

nt a

wai

ting

psyc

hiat

rist

Dat

e:

28/0

9/00

Tim

e: 1

815

FOR

M 2

A –

leve

l 2

FOR

M 2

B –

leve

l 1

SYST

EM

Act

ive

failu

re

Evid

ence

M

anag

emen

t sys

tem

fa

ilure

(roo

t cau

se)

Roo

tca

use

code

Evid

ence

C

OM

PETE

NC

E 3.

2 1.

Sta

ndar

ds o

f com

pete

ncy

did

not s

peci

fy e

valu

atio

n of

trai

ning

co

nten

t. 2.

Sta

ndar

ds o

f com

pete

ncy

did

not s

peci

fy th

e pr

ovis

ion

of

com

pete

nt a

dvic

e –

staf

f hav

e ve

ry

few

psy

chia

tric

skill

s to

draw

fr

om.

C

OM

MU

NIC

AT

ION

4.

4 N

ot a

ll re

leva

nt st

aff g

roup

s are

in

form

ed o

f the

syst

em.

C

O-O

PER

AT

ION

5.

2 N

ot a

ll re

leva

nt st

aff g

roup

s wer

e co

nsul

ted

and

few

wer

e in

volv

ed

in th

e pl

anni

ng, m

easu

ring

and

revi

ewin

g of

the

syst

em.

Figu

re 6

Evi

denc

e –

Form

2A

and

2B

114

Instructions: (1) Enter a cross for NO (2) Where there is no policy in place STOP root causing completely.

(3) Once a cross is entered within a series of four move onto the next set. (4) Upon completion transfer all root causes & codes into evidence form

(Form 2B) and complete the evidence column.

Accident Ref: LEAVE BLANK Date: 28/09/00 Time: 1815 SYSTEM (from Form 2a)

Training Proced-ures

Emer. Proc.

1.1 Was a policy in place regarding the system?

X

1.2 Was the policy adequate? X IMPLEMENTING 6.3 Was a standard for the system

specified?

6.4 Was the standard for the system adequate?

7.3 Was the system implemented? 7.4 Was the system implemented to

the required standard? X

CONTROL 2.1 Was a standard specified defining

responsibilities?

2.2 Was the definition of responsibilities adequate?

2.3 Was responsibility for the system assigned?

2.4 Were responsibilities assigned according to the standard?

PLANNING 6.1 Was there a plan to implement the

system?

6.2 Was the plan adequate? X 7.1 Was the plan put into operation? 7.2 Was the plan put into operation

adequately?

COMPETENCE 3.1 Were standards of competency

specified?

3.2 Were the standards for competency adequate?

X

3.3 Was the person/persons competent?

3.4 Was competency assessed against the standard?

COMMUNICATION 4.1 Was a standard for communication

of the system specified?

4.2 Was the standard for communication adequate?

4.3 Was the system communicated? 4.4 Was the communication in line

with the standard? X

115

CO-OPERATION 5.1 Was a standard for co-operation

specified?

5.2 Was the standard for ensuring co-operation adequate?

X

5.3 Was there co-operation? 5.4 Was the co-operation in line with

the standard?

MEASURING 2.5 Was a standard for measuring the

effectiveness specified?

2.6 Was the standard for measuring effectiveness adequate?

8.1 Was the system measured for effectiveness?

8.2 Were these measures in line with the standard?

REVIEWING 2.7 Was a standard for reviewing the

system specified?

2.8 Was the standard for reviewing the system adequate?

9.1 Was the system reviewed? 9.2 Were these reviews in line with the

standard?

Figure 7 System failure - Form 3

116

APPENDIX 2 COST INCLUSIONS AND EXCLUSIONS

Non staff costs

Include

1. Cost of disposable materials used to treat staff or patient injuries that occur as a result of an incident.

1.1 NOTE: In the case of needlestick injuries this will depend on whether the Trust provides its own services or has a service level agreement (SLA) with an Occupational Health Service (OHS) run by another Trust. It may also depend on the location of the member of staff and the time that the injury occurs, as treatment may be delivered at the A&E department instead of the OHS Department.

2. Cost of disposable materials used to make the area safe again.

2.1 EXAMPLE: To mop up a spillage.

3. Cost of an extended patient stay that is either the direct or indirect result of an incident.

4. Cost of replacing damaged equipment.

5. Cost of drugs that are administered as a result of the incident.

6. Cost of raw materials used to repair damage to equipment or property.

6.1 EXAMPLE: Glass and other materials used to repair a broken window.

Exclude

7. Cost of equipment that is in-situ and is re-useable.

8. Cost of equipment that has been damaged but will not be replaced.

9. Costs that are associated with improvement/upgrading during replacement.

10. Patient facilities that are unavailable due to damage i.e. short-term reduction in the patient capacity of a ward.

11. Vaccination and disposable materials costs when they are covered by a service level agreement.

12. Cost of materials that are purchased to prevent further harm (as part of risk control measure) following the incident.

12.1. EXAMPLE: Hip protector pads purchased following patient fall.

Staff costs

Include

13. Time spent receiving or administering first aid to a member of staff, visitor, contractor involved in an incident.

14. Time spent during a shift visiting A&E or OHS.

15. Sick leave.

117

16. Time spent performing duties, not normally part of job, as a result of an incident.

16.1 EXAMPLE: A needle goes missing during eye surgery, theatre staff & porters etc. spend time searching for the needle.

16.2 EXAMPLE: A needlestick injury is sustained from a needle sticking out of a general waste bag. The bag is passed onto a member of nursing staff & they have to carefully dispose of the needle & search through the contents of the bag to ensure there are no more needles.

17. Time spent performing regular duties outwith normal hours.

17.1. EXAMPLE: Engineer called out to reset a fire alarm system.

18. Time spent outwith designated ward areas managing violence & aggression.

19. Time spent outwith designated ward areas searching for absconded patients.

20. Time spent outwith designated ward areas escorting a patient involved in an incident.

21. Replacement labour costs if additional staff are brought in to cover absence due to an incident.

22. Contractor’s fees for services required following an incident such as repairs to equipment etc.

Exclude

23. Time spent by doctors & nurses administering first aid to a patient on their ward. The situation becomes slightly more complex if the patient is injured outwith their ward area.

23.1. EXAMPLE: If a radiographer provides first aid to a patient their time should be costed.

24. Staff time to care for patients, where the provision of care forms part of their everyday role.

24.1. EXAMPLE: If a patient requires an x-ray, do not cost for the radiographers’ time.

25. Time spent being interviewed & completing forms that are part of normal reporting & investigation procedures following an incident. Thus, incident reporting, police interviews & internal investigations involving staff associated with the incident or staff with an investigative role in the Trust are excluded from the costing process.

26. Additional staff required following the incident as a result of a review of the risk assessment following the incident.

26.1. EXAMPLE: Increased staff numbers to allow for constant observations of a patient who is considered high risk following an incident.

118

APPENDIX 3 COVERING LETTERS, QUESTIONNAIRE AND QUALITY CONTROL SYSTEM FOR INPUTTING

QUESTIONNAIRE DATA

Dear Colleague,

Safety Questionnaire

Tayside Occupational Health and Safety Service is undertaking a research project in collaboration with the NHS Management Executive and the Health and Safety Executive, aimed at improving the management of health and safety in the NHS and reducing the costs of accidents at work. As part of this endeavour it is our intention to obtain some indication of employee perception of health and safety by conducting a confidential survey of 10% of staff working at all levels in the NHS.

You have been randomly selected to participate in the survey and we would be very grateful for your co-operation. This entails completion of the enclosed questionnaire and a further identical questionnaire which will be issued to you in 12 to 14 months. It is important that you complete and return the tear-off slip at the bottom of this letter if you do not wish to participate or are unlikely to be available to complete the second questionnaire. All responses will be treated in strict confidence, however you are required to put your name on the questionnaire so that I can monitor changes in your perception of health and safety.

The questionnaire is straightforward to complete and asks about your attitudes to safety issues as well as any suggestions you might have to improve things. Please try to answer all of the questions. The findings will be fed back to you on completion of the study.

Should you have any queries relating to the nature of the study or the questionnaire please do not hesitate to contact me. I appreciate that it is often difficult to find time to respond to the requests of researchers but I would urge you to take this rare opportunity to express a personal opinion on an important issue which affects everyone in the NHS. I will contact you again if I do not hear from you in 10 days.

Many thanks for your assistance.

Yours sincerely,

----------------------------------------------------------------------------------------------------------

I do not wish to participate in the study.

Name ____________________________________________________________ Trust ____________________________________________________________ Department____________________________________________________________ Job Function____________________________________________________________

Figure 1 Covering letter (Phase 2)

119

Dear Colleague,

Safety Questionnaire

You may recall being contacted just over a year ago by Tayside Occupational Health & Safety Service to ask for your help in an externally funded study aimed at improving the management of health and safety in the NHS and reducing the costs of accidents at work. Since the occupational health services that cover Tayside and Fife have now merged, the name of our organisation has changed to the Occupational Health & Safety Advisory Service (OHSAS). However, the remit of the research team is unaltered and I am writing to thank you for returning a completed questionnaire in the first round and to ask for your cooperation again so that we can determine whether the overall opinion of staff regarding health and safety issues has changed over the last year.

As before, the survey is strictly confidential and will be processed by a member of the research team in OHSAS. We are issuing the questionnaire to 1547 people across 7 NHS Trusts and in some cases the reply envelope may bear an internal address. This is purely to reduce postage costs, the collated envelopes will be collected unopened by a member of the research team and only the overall opinion of the sampled group will be fed back to the Trust.

10% of staff working at all levels of the Trust were randomly selected to take part in this survey which is aimed at assessing the effectiveness of a new health and safety management system that has been introduced in 5 of the 7 NHS Trusts participating. Only those that responded in the first round of the survey have been contacted again as we feel it is important to compare the opinions, over time, of the same group of people. This means that your cooperation is even more valuable to us at this crucial stage of the project and we would be very grateful if you would complete and return the enclosed questionnaire.

The questionnaire is identical to the last one you completed. It asks about your attitudes to health and safety issues and the importance assigned to them by the Trust. Please try to answer all of the questions. The findings of the survey will be fed back to you once we have collated the responses from this second round and have established whether staff opinion of the management of health and safety has changed for the better, worse or stayed the same over the period of the study.

The results from the whole programme of research, of which this is the only opinion based aspect, will be fed back to line managers, health and safety committee members and Trust executives. I would therefore urge you to take this rare opportunity to express a personal opinion on an important issue which affects everyone in the NHS.

Should you have any queries relating to the nature of the study or the questionnaire please do not hesitate to contact me. I will be in touch again if I do not hear from you in 14 days.

Many thanks for your assistance.

Yours sincerely,

Figure 2 Covering letter (Phase 4)

120

SAFETY SURVEY

We would like to find out how you feel about safety practices and principles in the NHS and in order to do this we would like you to complete this questionnaire.

It is important for you to be completely honest about your feelings. All responses will be treated in strict confidence and will be processed by the Tayside Occupational Health & Safety Service.

It should take 15 to 20 minutes to complete this questionnaire.

We would like you to enter your name (so that we can link two questionnaires administered at different time points in the study), NHS Trust, Department and Job function to assist us with the interpretation of the results.

Thank you for your co-operation.

Name ___________________________________________________

Trust ___________________________________________________

Department ___________________________________________________

Job Function ___________________________________________________

You will be presented with a series of statements on the following pages about health and safety. You should indicate your response by ticking the appropriate box.

For example, if you agreed with the following statement you would tick under the ‘I agree’ category, thus:

Strongly agree

Agree Neither agree nor disagree

Disagree Strongly disagree

1. Health & safety issues are very important �

121

Please tick the appropriate box to indicate your level of agreement

Strongly agree



1. Management operates an open door policy on safety issues

2. Safety is the number one priority in my mind when completing a job

3. Co-workers often give tips to each other on how to work safely

4. Safety rules and procedures are carefully followed

5. Management clearly considers the safety of employees of great importance

6. I am sure it is only a matter of time before I am involved in an accident

7. Sometimes I am not given enough time to get the job done safely

8. I am involved in informing management of important safety issues

9. Management acts decisively when a safety concern is raised

10. There is good communication here about safety issues which affect me

11. I understand the safety rules for my job

12. It is important to me that there is a continuing emphasis on safety

13. I am involved with safety issues at work

14. This is a safer place to work than other organisations* I have worked for

(*“organisations” changed from “companies”)

15. I am strongly encouraged to report unsafe conditions

16. In my workplace management turn a blind eye to safety issues

17. Some safety rules and procedures do not need to be followed to get the job done safely

123


Strongly agree



35. Sometimes it is necessary to depart from safety requirements in order to meet the demands of my job*

(* “to meet the demands of my job” changed from “for productions sake”.)

36. A safe place to work has a lot of personal meaning to me

37. There are always enough people available to get the job done safely

38. In my workplace managers/ supervisors show interest in my safety

39. I am never involved in the ongoing review of safety

40. Management considers safety to be equally as important as getting the job done

41. A no-blame approach is used to persuade people acting unsafely that their behaviour is inappropriate

42. Managers and supervisors express concern if safety procedures are not adhered to

43. I cannot always get the equipment I need to do the job safely

Do you have any other comments about health and safety in your workplace? ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Thank you for completing this questionnaire.

Figure 3 Safety survey

124

Quality control system for inputting questionnaire data

1. Input a “batch” of questionnaires into the database – a batch should be done in one sitting so there may be a number of batches input in one day.

2. Make a note of the day, date, batch/sample number, number of questionnaires entered, the time data inputting commenced and the time data inputting finished in the data inputting table.

3. Randomly sample five questionnaires from the pile of questionnaires entered – ensure there are some questionnaires sampled from the latterly input half of the pile.

4. Check inputting on the database, count and correct any errors found for each questionnaire (only check errors for multiple choice statements).

5. Plot the total number of errors found in the five questionnaires sampled on the control chart.

6. Monitor the control chart plots checking for:

�� Non-random patterns �� Eight consecutive plots above the mean line �� Two/three plots between the upper warning limit and the upper control limit �� Any plot above the upper control limit

7. In the event of the above, return to the batches concerned and re-check the entire batch, correct the entries and then conduct the quality control exercise again, re-plotting the “corrected” error rate.

125

APPENDIX 4 MANAGEMENT SYSTEM WORKBOOK

Introduction

The Trust aims to achieve a safe and healthy environment for staff, patients and their relatives and visitors.

Clinical leaders/managers have a vital role to play in achieving this because health and safety is NOT an “add-on” to your job. It is a fact that in departments and companies where health and safety is positively managed performance, efficiency, morale, and cost-effectiveness all improve.

A safe and healthy workplace can only be achieved if health and safety is INTEGRATED with the work that we do. Therefore health and safety must be MANAGED. This is where you come in.

Clinical leaders/managers are responsible for managing the health and safety of their staff and the premises in which they work. This is a legal requirement of the Health and Safety at Work Act 1974 and the Management of Health and Safety at Work Regulations 1999.

This workbook has been designed to help you manage health and safety. You are encouraged to use it fully.

On the next page is a FORWARD PLANNER, which you can use to record summary information of your actions to ensure health and safety is managed adequately in your area of responsibility. Use it to plan your year but REMEMBER it is a summary sheet and there are other actions that you will need to take which are contained in each section of this workbook.

A WORKED EXAMPLE of the planner is also included to help you.

REMEMBER: HEALTH AND SAFETY MUST BE MANAGED. YOU WILL FIND THIS WORKBOOK FORMS A CONTINUOUS LOOP AND THAT YOU WILL END BACK AT THE START AGAIN, EACH TIME REFINING AND IMPROVING YOUR SYSTEM. IF YOU ARE NOT SURE ASK FOR HELP (see “Where to find more help & information” at the end of the Workbook)

YOU DO NOT HAVE TO USE EVERY SECTION IN THE WORKBOOK IF YOU ALREADY HAVE SOMETHING THAT PERFORMS THE SAME FUNCTION. IF YOU DO - MAKE A NOTE IN THE RELEVANT SECTION AS TO WHERE THE INFORMATION CAN BE OBTAINED

122


Strongly agree



18. I am rarely worried about being injured on the job

19. Management acts only after accidents have occurred

20. I believe that safety issues are not assigned a high priority

21. Some health and safety rules and procedures are not really practical

22. Employees are not encouraged to raise safety concerns

23. Personally I feel that safety issues are not the most important aspect of my job

24. In my workplace the chances of being involved in an accident are quite large

25. I do not receive praise for working safely

26. Corrective action is always taken when management is told about unsafe practices

27. Operational targets often conflict with safety measures

28. My line manager/supervisor does not always inform me of current concerns and issues

29. I can influence health and safety performance here

30. Sometimes conditions here hinder my ability to work safely

31. Safety information is always brought to my attention by my line manager/supervisor

32. When people ignore safety procedures here, I feel it is none of my business

33. In my workplace management acts quickly to correct safety problems.

34. I am clear about what my responsibilities are for health and safety

126

Departmental Health & Safety Management System: FORWARD PLANNER (WORKED EXAMPLE)

Financial year1999-2000

Policy Indiv-idual roles & respons-ibilities

Incident report forms

Training records

Safety meet-ings

Risk assess-ments

Action plan

Safety inspec-tions

Audit Review

April Review �

Produce plan �

General �

�

May Assign �

Manual handling �

�

June � COSHH �

�

July � DSE � � Aug Other � � Sept Review

�

� Produce �

Submit to TEG

�

Oct � � � Nov � Dec � � Jan � � �

Feb � �

Mar Review� � �

WHERE TO FIND MORE HELP & INFORMATION The first person to approach if you have any queries is your clinical leader/manager. However, the following individuals are also responsible for health and safety within the Trust:

Name Position Area of responsibility �or Extension

127

POLICY

Introduction

Policy documents should have three main sections:

STATEMENT OF INTENT - which outlines briefly an organisation or departments position on a certain subject (like health & safety). It should also give a broad description of the strategy to be adopted which will put the statement into practice (e.g. a series of objectives). ORGANISATION - which sets out the roles and responsibilities of the policy. ARRANGEMENTS - which sets out how the policy aims and objectives will be achieved.

(N.B. “PROCEDURES” OR “RISK CONTROL MEASURES”, on the other hand, are detailed descriptions of standardised or understood ways of doing tasks. You probably have many in use in your department).

This section is split into 2 parts:

National/Area/Trust policies (e.g. Trust A Area Infection Control Policy, Trust Health and Safety Policy etc.)

Departmental policies

You should now:

Insert a sheet indicating where all relevant documents are located if they are stored elsewhere

OR

file copies of the documents in these parts.

128

There is a blank index sheet at the start of each part for you to record the document titles (if you don’t already have one).

National/Area/Trust/Policy index

Title of policy Date of policy Comments

Departmental policy index

Title of policy Date of policy Comments

ORGANISING

Introduction

To make your health and safety management system really effective you need to get organised

This is often referred to as a “HEALTH AND SAFETY CULTURE”.

There are 4 aspects to achieving this:

1. Control (this is often maintained by ensuring that all staff know and understand their roles and responsibilities) 2. Co-operation (this comes from consulting staff and involving them in planning, reviewing performance, writing procedures & problem solving) 3. Communication (this involves discussing health and safety regularly and providing information about hazards, risks and preventative measures) 4. Competence (this means ensuring adequate information, instruction, training and supervision are available)

Each of the 4 aspects of organising listed above contributes to a positive health and safety culture.

You should record the roles and responsibilities of individuals within the department in this part of the workbook.

129

Control

To help you understand how the Trust assigns roles & responsibilities, a diagram of the TRUST HEALTH AND SAFETY MANAGEMENT SYSTEM is included overleaf. This identifies how health and safety is managed and supported at Trust level. A description of the roles and responsibilities of the individuals identified in the diagram is given below:

Chief Executive: has overall responsibility for health & safety within the Trust.

Clinical Director/General Manager: is responsible for the health & safety of all staff working within their “patch”.

Managers: are responsible for the health & safety of their staff. Should draw up and implement an action plan, if necessary, following risk assessment. Responsible to the Clinical Director / General Manager. Staff: all staff have a responsibility for their own health & safety. Director of Human Resources: the executive director responsible for co-ordination of health & safety management. Responsible to the Trust Board.

Health & Safety Co-ordinators: act on behalf of their Clinical Director/General Manager to ensure that managers fulfil their legal requirements and to act as a communication channel and local source of advice.

Risk Assessors: responsible for conducting risk assessments as required by their manager (N.B. Not for implementing the recommendations of assessments).

Safety “Reps”: represent a group of staff on health & safety issues. Specialist Advisers: act as a source of advice and guidance on health & safety (N.B. Not for managing health & safety). A blank sheet is provided in this section to allow you to list your specialist advisers.

To help you,

There is a sheet for recording the roles and responsibilities of yourself and your staff.

You should make sure that this sheet is kept up to date and that any gaps are entered into the Record of training.

130

KEY: Line Management

Health & Safety

Support for line management

Specialist advisors

131

List of Trust specialist advisers

Position Name Contact number Safety Officer Occupational Hygienist Infection Control Radiation Protection Adviser Manual Handling Adviser Occupational Health Dept Fire Adviser

List of individual roles & responsibilities

Role Name(s) Date Trained

Date re-training required

General Risk Assessor COSHH Assessor Manual Handling Risk Assessor Display Screen Equipment Assessor Other Risk Assessor Infection Control Link Person Manual Handling Key Worker Safety Inspectors Staff Safety Rep First Aider

Remember:

This list is not exhaustive. You may have specialised roles & responsibilities that are not mentioned. Include them in the spaces provided

You may not need all the types of role described. For example, you might have a general risk assessor but do not need a COSHH assessor. Just use the headings you need

132

Co-operation

Participation, commitment and involvement in health and safety activities at all levels in your department is essential because: You have a LEGAL DUTY to consult your staff Pooling knowledge and experience is a KEY ASPECT OF RISK CONTROL This can be done in a number of different ways such as: Holding regular departmental meetings to:

�� Plan �� Establish departmental policies �� Produce a departmental action plan �� Measure, audit and review performance �� Routinely revise policies, procedures and the action plan �� Introduce improvement initiatives �� Promote health & safety awareness �� Raise awareness of health and safety within the department �� Be the focal point of efforts to prevent accidents and manage risks �� Promote a reduction in environmental emissions and the efficient use of paper,

energy and water �� Ensure action is taken on health and safety matters �� Communicate and manage information �� Co-ordinate the responsibilities of management and staff for health and safety to

improve the health, safety and welfare of staff at work �� Identify training needs

Involving staff in working groups (to develop policies, procedures etc).

Encouraging the nomination of and providing support for staff safety representatives.

Operating suggestion schemes.

Record of consultation with staff.

Measure taken to achieve co-operation Dates of meetings Comments/actions necessary

You should now:

Begin by thinking about how you can improve co-operation within your area of responsibility. Consider the suggestions given above. You DO NOT have to introduce ALL of the suggestions above. They are given as examples. Make a plan for implementation of the measures that you decide upon. Record your actions on the sheet provided.

133

Communication

Effective communication is essential. It may help to think of communication in terms of FLOW both into and out of a department or organisation:

Record of health & safety information

Document title Location

Record of departmental safety meetings

Date of meeting Action taken

You should now record your communication measures. This part of the workbook contains 2 dividers for you to record:

Health and safety information (i.e. your “library” of information)

Departmental safety meetings (i.e. minutes etc.)

134

Competence

Arrangements need to be made to ensure the competence of all employees (including managers) if they are to make the maximum contribution to health and safety.

These should include:

�� Identifying health and safety training needs arising from recruitment, changes in staff, plant, processes or working practices.

�� Maintaining or enhancing competence by refresher training �� Managing contractors �� Ensuring relevant information, instruction, training and supervision are provided �� Making arrangements to ensure cover for staff absences �� Re-organising staff roles in accordance with changes in competence which arise as a

result of external factors such as a recent injury, illness or bereavement

Record of training needs

Name Area of responsibility Training need Date completed

Date re-training due

PLANNING & IMPLEMENTATION

Introduction

Planning, and then ensuring that these plans are implemented, is the key to ensuring that your health & safety efforts really work.

You must make sure you have WRITTEN assessments of risk for all the procedures in your area of responsibility that have SIGNIFICANT health and safety risks associated with them.

A summary sheet for your training records is provided below.

This is a bit like a loop leading to an update of the procedure after a risk assessment.

ProcedureRisk assessment

Planning involves: setting objectives identifying hazards assessing risks implementing standards of performance in your operational procedures.

135

Begin by making a list of all the TASKS your staff undertake. If you have operational procedures (or equivalent) these will be ideal. Do this with your risk assessor(s).

Identify, with your risk assessor(s), which tasks do not have SIGNIFICANT health and safety risks (If you need help with this, a decision matrix can be found at the end of this section).

Remove the tasks that do not need action from your list (if in doubt -include the task), RETAIN these and review them annually or sooner if the task changes in some way.

Now make a new prioritised list of tasks with significant health & safety risks (the decision matrix will also help with this).

Ask your risk assessor(s) to conduct written assessments for the remaining tasks.

Start with GENERAL RISK ASSESSMENTS for compliance with the Management of Health and Safety at Work Regulations, 1999.

If necessary, arrange for a more detailed risk assessment to be carried out (e.g. COSHH, Manual Handling, DSE.

136

Your risk assessments form the foundation of your RISK CONTROL SYSTEM

When preparing your action plan you MUST take account of the hierarchy of risk control principles:

1. Eliminate risks by substituting the dangerous with the less dangerous (e.g. using a less hazardous substance or substituting a better guarded machine)

2. Combat risks at source by engineering controls and giving collective protective measures priority. (e.g. enclose the process, provide guards, suppress or contain air-borne hazards, remote/automatic operation)

3. Minimise risk by designing operational procedures so that they are safer (e.g. reducing exposure processes by using competent or specialist contractors)

4. Use personal protective equipment (remember this is a LAST resort)

Review the assessments once they are complete. Make a written ACTION PLAN based on the conclusions. The plan may consist of a number of different types of action:

If the solution to the problem is obvious, and it is achievable within your range of authority, IMPLEMENT THE SOLUTION; If the solution is obvious but you cannot implement the measure(s) because of constraints, you must IMPLEMENT SHORT-TERM CONTROL MEASURES IMMEDIATELY (to protect staff in the meantime) and then PREPARE A BUSINESS CASE for the relevant member of the Corporate Team (a model business case layout is included) If the solution to the problem is not obvious, identify possible solutions by carrying out an OPTION APPRAISAL (a model for this is included). If the preferred solution is achievable, implement it. If it is not then IMPLEMENT SHORT-TERM CONTROL MEASURES immediately and prepare a BUSINESS CASE for the relevant member of the Corporate Team. Remember to contact your specialist advisers if you need more helpMake sure that you implement your action plan - It’s no good making a lot of good plans and then doing nothing about them.

Remember to record your findings. Blank copies of the forms to use are in each of the labelled sections. You should store OR record where you have stored your completed risk assessments and action plans in these sections too.

Assign a date on your planner when the assessments are due for review. It is your job to instruct your risk assessor to repeat or re-assess risk assessments.

Check that progress is being made with your action plan from time to time. It is easy to let things slip!

137

The process is summarised in the following flow diagram:

138

HAZARD�

RISK�

Low Medium High Very high

Highly likely Action required Urgent action Urgent action Suspend work

Likely Action required Action soon Action soon Urgent action

Possible Action required Action required Action soon Action soon

Not likely No action Action required Action required Action required

Action planning

You should prepare an action plan if any of your risk assessments show action to control risk is necessary. The flow diagram in the previous Section shows that there are 3 levels of action plan. The solution is obvious and achievable. You should implement it. The solution is obvious but you are unable to implement it for some reason (can’t authorise the expenditure, for example). In this case you should ensure that the task can be done safely by some means (which may not be ideal but you must make the task safe). You should then prepare a business case and submit it to your head of department. The solution is not obvious. In this case you should make the task safe in the short-term and carry out an option appraisal before preparing your business case. You may also need to seek help with this from a specialist adviser.

A blank sheet for recording your action plans, model business case and option appraisal are all included in this section.

Record of action plans

Task Action needed Person responsible for action Target date

Use this matrix to help you decide which of your tasks have significant health & safety risks (i.e. those where action is necessary). Remove those where no action is indicated (remember to review them annually or if they change)Make sure detailed, written risk assessments are carried out for the remainder. Prioritise these, tackling the most urgent needs first.

REMEMBER, TACKLE THE MOST SIGNIFICANT RISKS FIRST

139

MODEL BUSINESS CASE

Provide the following information and submit to the Head of Department or equivalent

Signed………………………………………………Date………………… Designation…………………………………………Department…………

1. Description of support requested (attach quotations etc. if appropriate)

2. Evidence for significant health and safety risk (summarise outcome of risk assessment)

3. Are other options available? (If so include an option appraisal with this document

4. Potential consequences should support not be available (attempt a realistic assessment)

140

Operational procedures

Completed risk assessments (see the following Sections) should be used by you to develop changes in your operational procedures.

Remember, it’s like a LOOP:

If you wish you can store your written operational procedures in this workbook. Alternatively, you can store a list of them with their locations using the sheet provided overleaf.

141

List of operational procedures

Title of procedure Date Location

General risk assessment

Written general risk assessments are needed to comply with the legal requirements of the Management of Health and Safety at Work Regulations 1999.

A risk assessment has four main parts:

A blank risk assessment form is included in this part. You should also store your risk assessments here or indicate below where they are stored.

Remember to seek guidance from your specialist advisers (you can refer to the list at the beginning of the workbook) if you need help.

Identify the significant hazards within the task (i.e. aspects of the task with the potential to cause harm such as slipping hazards, electrical hazards, manual handling hazards etc.).

Evaluate the risk (i.e. the probability that things will go wrong. For example, how likely it is that someone will trip in the circumstances of the task? You can use previous experience to help you here. For example has anyone already tripped? Make sure you consider who could be affected and what control measures already exist as part of this process. Also ask yourself before you decide. Is this a foreseeable risk?)

Identify measures needed to control any risks that are unacceptable. (This may need to be a package of measures. It may also include asking for another, more specific risk assessment to be done before you can decide. Examples of this are COSHH, Manual handling, and DSE assessments as well as asking your specialist adviser for an opinion).

Review the risk assessment. (If the risks are controlled to your satisfaction you only need review in a year to check nothing has changed. If there are actions that are needed you should review more frequently to make sure progress is being made.)

142

Control of Substances Hazardous to Health Regulations 1999 (COSHH) Risk Assessments

COSHH assessments concentrate on the use and production of hazardous substances. This means in practice looking for the potential within the task for exposure and harm to occur by:

�� Inhalation; �� Skin contact; �� Skin absorption; �� Injection via a sharps injury; �� Eye contact; �� Ingestion.

Grounds for concluding that exposure is not a significant risk to health are:

Quantities or rate of use/production of the substance(s) are too small to constitute any risk to health under foreseeable circumstances of use, even if control measures break down;

OR

There is no available route of exposure.

If you are not sure if exposure is significant or not, contact your special adviser (refer to the list at the beginning of the workbook), who will be able to give you an opinion or arrange for measurements to be taken to help you decide.

A blank risk assessment form is included in this part. You should store your risk assessments here or indicate below where they are stored

The Manual Handling Operations Regulation 1992 Risk Assessment

The numerical guidelines on the next page provide an initial filter which can help to identify those manual handling operations deserving more detailed examination. The guidelines set out an approximate boundary within which operations are unlikely to create a risk of injury sufficient to warrant more detailed assessment. This should enable assessment work to be concentrated where it is most needed.

There is no threshold below which manual handling operations may be regarded as "safe". Even operations lying within the boundary mapped out by the guidelines should be avoided or made less demanding wherever it is reasonably practicable to do so.

There is a wide range of individual physical capability, even among those fit and healthy enough to be at work. For the working population the higher guideline figures will give reasonable protection to about 95% of men and between one half and two thirds of women. The lower guideline figures will give reasonable protection to about 95% of women.

It is important to understand that the guideline figures are not limits. They may be exceeded where a more detailed assessment shows that it is appropriate to do so, having regard always to the employer's duty to avoid or reduce risk of injury where this is reasonably practicable. However, even for fit, well-trained individuals working under favourable conditions ANY MANUAL HANDLING OPERATION SHOULD NOT EXCEED TWO TIMES THE GUIDELINE FIGURES.

143

Basic guideline figures (kilograms) for manual handling operations involving lifting and lowering are set out overleaf. They assume that the load is readily grasped with both hands and that the operation takes place in reasonable working conditions with the handler in a stable body position.

The guideline figures take into consideration the vertical and horizontal position of the hands as they move the load during the handling operation, as well as the height and reach of the individual handler. It will be apparent that the capability to lift or lower is reduced significantly if, for example, the load is held at arm's length or the hands pass above shoulder height.

CARRYING

Guideline figures for carrying are the same as for lifting and lowering (overleaf), provided: the hands are not below knuckle height the load is held against the body the load is carried no further than 10 metres without resting.

When the load can be carried securely on the shoulder without first having to be lifted (e.g. unloading sacks from a lorry) a more detailed assessment may show that it is acceptable to exceed the guideline figure.

PUSHING AND PULLING

Start / stop: 25 kg force (men); 17 kg force (women)

Steady motion: 10 kg force (men); 7 kg force (women)

These figures assume that the force is applied between knuckle and shoulder height.

HANDLING WHILE SEATED

The figures in the diagram assume: No handling outside the box No twisting

144

Twisting Frequency

0o

<30/hour 0o

<2/min 0o

<12/min 0o

>12/min

Twisting Frequency

45o

<30/hour

45o

<2/min 45o

<12/min 45o

>12/min

Twisting Frequency

90o

<30/hour

90o

<2/min

90<12/min

90o

>12/min

(men)(women

(men)(women)

(men)(wome

145

WEIGHT IN KG A blank risk assessment form is included next. You can store your risk assessments here or indicate below where they are stored

The Health & Safety (Display Screen Equipment) Regulations 1992 Risk Assessment

You are responsible for carrying out workstation assessments for those staff who are display screen equipment (DSE) “users”. A “user” is someone who uses DSE as a significant part of their normal work. Therefore any person who works with DSE more or less continuously on most days should be regarded as a user. If you are not sure whether someone is a “user “ or not the following guide may help.

A DSE “user” will: Normally use DSE for continuous spells of an hour or more at a time; andUse it in this way more or less daily; andHave to transfer information quickly to or from the screen; and also:Need to apply high levels of attention or concentration; orAre highly dependent on DSE or have little choice about using them; orNeed special training or skills to use the equipment.

The DSE assessment has two forms that should be completed for each workstation:

A workstation assessment. This should be used by someone who has attended DSE risk assessor training; A self-assessment checklist. Each person identified as a DSE “user” should complete this. It will probably be best to ensure that any training in setting up a workstation is given to “users” BEFORE they complete this checklist.

A checklist for “users” to follow when setting up their workstation is included on the next page. This can be used as part of departmental training.

For further advice on training contact your specialist Health and Safety Adviser

As a manger of a DSE “user” you are responsible for:

The workstation layout and assessment; The daily work routine; General health & safety training on the use of DSE; Provision of information. This should include entitlement to a free eye test (check with personnel for detailed arrangements for accessing this service).

146

Display Screen Equipment - Checklist for setting up a workstation

Whenever you sit down at a workstation you should go through these 10 steps before you start work. 1. Sit back well in the seat, adjust the angle (and height if appropriate) of the back rest so

your back is well supported. 2. Make sure that the small of your back is well supported. 3. Adjust the height of your chair so that your forearms are approximately horizontal when

you place your hands to use the keyboard (see position 7 in the diagram. 4. Check that you don’t have too much pressure on the underside of your thighs and backs

of your knees or that your feet are not dangling. If you answer YES to any of these carry out 5. If NO go to 6.

5. Support the feet with a foot rest of suitable height. 6. Make sure that you have enough space under the desk with no obstacles which may

restrict your posture. 7. Check again that your forearms are approximately horizontal when using the keyboard

and then, 8. Check that your wrists are in a natural position. 9. Now look at the screen and make sure its height and angle allow you to hold your head

comfortably. You may need a document holder if you do a lot of copy typing. 10. Check the screen for glare. (You may need to close blinds, dim lights, reposition the

screen to do this). 11. Check the characters on the screen are bright enough (but not too bright or they will

become “fuzzy”). 12. Check that you have enough space in front of the keyboard to support your hands/wrists

during times when you are not using the keyboard.

Remember to take breaks away from the screen.

147

MEASURING PERFORMANCE

Introduction

Just like finance or quality of service, you need to measure your health and safety performance to find out if you are being successful.

There are two sorts of ways to measure your performance:

ACTIVE methods i.e. BEFORE things go wrong REACTIVE methods i.e. AFTER things go wrong

It is suggested that you adopt 3 different methods of measuring your performance:

Safety inspections (an “active” method) Safety tours (another “active” method) Accident/incident data (a “reactive” method)

Space has been made in this workbook for you to record relevant information for all three parts.

Safety inspections

These should be carried out regularly (e.g. monthly). A blank safety inspection form is included so you can make a start. A safety inspection is different to a risk assessment and usually concentrates on hazard spotting which produces a list of actions. A safety inspection typically identifies a lot of “maintenance” and “housekeeping” issues which must be managed if accidents are to be avoided. Allocate the duty of carrying out an inspection to a member of staff (it does not HAVE to be a trained risk assessor). When the inspection is carried out make sure you take action on those problems that have been identified.

148

Health & safety inspection checklist

LOCATION: ______________DATE OF INSPECTION:___________ INSPECTED BY:___________________________________________

HAZARD PRESENT �

HAZARD IDENTIFIED BEFORE �

DETAILS / COMMENTS / ACTIONREQUIRED

Relating to both workplaces and activities. Identify only hazards which you could reasonably expect to result in significant harm under the conditions in your workplace, otherwise the list becomes excessive. Use the following examples as a guide. 1. Slipping/tripping hazards 2. Infective agents (e.g. sharps) 3. People falling (e.g. working at height) 4. Falling materials 5. Confined spaces 6. Fire (e.g. flammable materials, exits blocked) 7. Moving parts of machinery (e.g. blades) 8. Ejection of material (e.g. material grindstones) 9. Pressure systems (e.g. boilers, autoclaves) 10.Vehicles (e.g. trolleys, fork lift trucks, wheelchairs) 11. Electricity (e.g. poor wiring) 12. Manual handling 13. Display screen equipment 14. Noise 15. Vibration 16. Poor lighting 17. Radiation (e.g. x-ray, UV, microwave, laser) 18. Inhalation of dusts, fumes, mists, gases, vapours, infective agents 19. Skin contact with chemicals 20. Assault 21. Working with animals 22. Poor housekeeping (e.g. untidy and dirty) 23. Extremes of temperature 24. Inadequate ventilation (e.g. no control, draughty) 25. Overcrowding 26. Poor conditions of floors and traffic routes 27. Inadequate or dirty toilets and rest areas 28.29.30.31.32.

149

Safety tours

These are your opportunity to “walk the job”. As managers we often become isolated from our staff and one way of improving this is to take an informal walk through your “patch” on a regular basis. The benefits are considerable but not always obvious (e.g. improved morale, becoming more “in touch” etc.). Don’t arrive with a notebook and announce your arrival. Be informal and talk to your staff. You will be amazed at what you find. Don’t be put off if staff find it strange at first. Begin by assuring them that you want to listen to their opinion and take it from there. Make sure that you book a regular time in your diary to do a tour and DON’T cancel it when something “more important” comes up - it should have a HIGH PRIORITY.

Accident/incident data

This should be reported to the Trust on the official incident report form. The Trust may issue reports about trends in accidents and incidents. Keep an eye on what is happening in your area and investigate any occurrences, which highlight a problem.

REMEMBER: FOR EVERY ACCIDENT THAT OCCURS ABOUT 30 “NEAR MISSES” WILL ALSO HAPPEN. YOU CAN GET VALUABLE INFORMATION IF YOU ENCOURAGE STAFF TO REPORT THESE TOO.

Record of accident/incident reporting (Incident Report Form & RIDDOR)

Date of incident

Name of reporter

Report form completed

Retained incident record pad reference

Action taken

Reviewing performance

Introduction

This section is about LEARNING FROM YOUR EXPERIENCES and consists of two parts:

Audit Review

You will find a master copy of a departmental audit in this part. Use it every year to evaluate how you are doing and where gaps still exist. You can either conduct the audit yourself or get another member of staff to do it.

Take the findings of the audit and use them to make an ACTION PLAN for the coming year. You can also incorporate this plan into a revision of your DEPARTMENTAL HEALTH AND SAFETY POLICY.

150

This brings us back to the beginning again.

NOW GO TO THE BEGINNING OF THE WORKBOOK AND BEGIN AGAIN BY MAKING (OR REVIEWING) YOUR PLAN!

MANAGERS HEALTH AND SAFETY AUDIT

Note: A= Essential B= Good Practice

Policy

There is a written health and safety policy which has a statement of intent, organisation and arrangement sections. It should be signed and dated by the person in charge. (A)

Yes No (please circle answer)

Comments

Written departmental health and safety policies and procedures exist and are implemented consistent with the Trust health and safety policy. (A)


Comments

Health and safety policies are subject to continuous review. (A)

Yes No (please circle answer) Comments

151

Organisational development

There is an individual at senior management level who has overall responsibility for formulating, developing and implementing health and safety policy. (A)


Comments

Health and safety responsibilities of managers are clearly defined within their job descriptions. (A)

(Note: these managers have the necessary authority and competence to carry out their duties effectively and are held accountable for their actions, health and safety objectives are set and reviewed annually as part of the performance review process)


Comments

Arrangements are in place for obtaining competent safety advice. (A)


Comments

There is a multidisciplinary safety committee (or committees). (A)

(Note: meets regularly includes senior management, staff and staff representation and is consulted on the development, implementation and monitoring of health and safety policy, is actively involved in the setting and monitoring of performance standards for health and safety.)


Comments

The committee reports to the Corporate Team regularly. (B)


152

Comments

A Trust annual health and safety report is produced. (A)

(Note: presented to the executive management team, made available to all staff)


Comments

Local arrangements for consultation with staff are in place and staff are provided with the training necessary to make an informed contribution to health and safety issues. (A)


Comments

First aid arrangements are in place and are in accordance with the Health and Safety (First Aid) Regulations 1997. (A)


Comments

Mechanisms are in place to promote awareness of the Trust and Departmental health and safety policies and health and safety issues (e.g. notice boards, newsletters, etc). (B)


Comments

There is a written departmental safety education programme. (A)

(Note: includes orientation of new employees to safety practices, is reviewed at least annually to determine its effectiveness)


Comments

153

All local orientation and induction programmes include an introduction to the overall health and safety policy and any necessary health and safety instruction. (A)


Comments

Arrangements are in place for identifying and providing on-going health and safety instruction and training (for example, when changes in staff or working practices occur). All instruction and training is recorded. (A)


Comments

Temporary workers on fixed or short-term contracts are provided with information concerning health and safety issues which may be encountered in their work. (A)


Comments

Planning and implementation

There is an up-to-local date plan, which identifies health and safety objectives, targets and timescales and is developed in consultation with staff. (A)


Comments

Departmental hazards are identified and written risk assessments have been carried out in accordance with the Management of Health and Safety at Work Regulations 1999 or other relevant health and safety legislation. (A)

154

Such as: �� Control of Substances Hazardous to Health Regulations 1999 �� Electricity at Work Regulations 1989 �� Health and Safety (Display Screen Equipment) Regulations 1992 �� Manual Handling Operations Regulations 1992 �� Noise at Work Regulations 1989 �� Personal Protective Equipment at Work Regulations 1992 �� Provision and Use of Work Equipment Regulations 1998 �� Lifting Operations and Lifting Equipment Regulations 1998 �� Workplace (Health, Safety and Welfare) Regulations 1992


Comments

Where necessary preventive and protective measures (control measures) are implemented. (A)


Comments

Departmental risk assessments are reviewed and updated on a systematic basis or when circumstances change. (A)


Comments

Measuring performance

Regular departmental safety inspections are carried out in hazardous areas. (A)


Comments

There is a clear reporting procedure in place recording, investigating, reporting and taking action on accidents, incidents, hazards and defects. (A)


155

Comments

The Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 1995 (RIDDOR) are complied with. (A)


Comments

There is a system in place for disseminating safety action notices and hazard notices. (A)


Comments

Is health and safety on the agenda of departmental meetings so that health and safety objectives and effectiveness of arrangements are evaluated annually and modified as required. (B)


Comments

Audit and review

Departmental audit and review systems are established, operated and maintained. (A)


Comments

(Note: These are designed to assess the following elements of the health and safety management system:

�� policy �� organisation �� planning and policy implementation �� measuring systems �� reviewing systems)

156

Review of management system and setting new objectives

Those questions that you answered “NO” in the Managers health and safety audit represent the areas of your departmental health and safety management system that you need to develop. They therefore represent your objectives for the future.

Write your new objectives (with timescales and how you plan to achieve them) below.

Quarterly reports

Each quarter you should prepare a report on the progress you have made developing, implementing and refining your health and safety management system.

The purpose of reports of this kind is not for you to let everyone know how well things are going but rather to state what you planned to do in the relevant period, what you achieved and, in the light of this, what you plan to do next. If you are having problems you should say so. At least this can then be used as an indication to yourself that you need advice and guidance.

It is suggested that you use the “5-steps” format that this workbook follows to structure your report. A blank report is included. (Don’t feel obliged to use it if it doesn’t fit with your style of reporting though).

REMEMBER: NEVER BE AFRAID TO ASK FOR HELP - START WITH YOUR OWN MANAGER, HEALTH & SAFETY CO-ORDINATOR OR SPECIALIST ADVISER.

(Worked Example)

157

QUARTERLY REPORT FOR PERIOD April 2000 TO June 2000

Section of health & safety management system

Policy Organising Planning & implementing

Measuring performance

Audit/ review

Planned activities Distribute new policy

Roles & responsibilities clarified

Undertake general, COSHH & manual handling risk assessments

Review incident forms

None till January

Achievements Done Mostly complete Most done Highlighted trend of staff tripping. Modified floor covering. All OK now

N/A

Outstanding items

None First-aider still to be trained

Need Assistance from Occ.Health with COSHH Assessment for dust exposure

None None

Future plans Arrange seminar with staff to ensure all aware of new policy

Arrange first aid training course with occupational health

Appointment for COSHH survey made for July

Review in July Audit in January

Comments May not get training till August

All assessments will be complete after COSHH survey

Safety inspections done monthly

Any other items

158

APPENDIX 5 PROTOCOL FOR BRIEFING OF WORKBOOK HOLDERS

Aim of the Workbook Briefing: To introduce the workbook to managers, explain its purpose and ensure that manager’s understand how they can use the workbook to maximum effect.

1. Introductions –Project Coordinator, Project Research Assistant (both in attendance at all workbook briefings).

2. Background of the research project – The philosophy behind it, testing the theory, “Is good management of health & safety in the workplace good business?” Who is funding the work and why they are interested in looking at these issues. The various aspects of the project, i.e. data collection, the issue of questionnaires, operating the workbook for a period of a year and trying to detect change. Also why the research fits in so well with the internal health & safety objectives of the participating Trusts and the need for the NHS to improve its health & safety performance and implement appropriate management systems that are user-friendly to the managers who have to use them.

3. Legal requirement to manage health & safety in the workplace – Workbook holders are referred to the Health & Safety at Work Act (1974) and the Management of Health & Safety at Work Regulations (1992/1999) and how these pieces of legislation define the role of managers. Emphasis is placed on the obligation to manage health & safety through the line management chain and the need for acceptance of the responsibility to do this as opposed to the delegation of that responsibility to other individuals in the department who perform a health & safety role. Workbook holders are referred to the HSE guidance on how managers achieve compliance with the law, “Successful Health & Safety Management” (HSG65).

4. Health & Safety Management System – The five steps of the health & safety system are outlined and explained briefly.

�� Policy – A distinction is made between Trust-wide and local policy. �� Organising – Introduction of the “four Cs” of a positive health & safety culture

and a brief explanation of what is meant by each of these. �� Planning & implementing – This is largely tied up with risk assessment, the

importance of conducting risk assessments that feed in directly to the procedures and tasks that are being conducted in the workplace is explained and the concept that these should form a continuous loop. In order for the risk assessments to do any good they must be actioned – therefore an action plan is essential.

�� Measuring performance – Divided into reactive and active measures. Reactive being learning from experience, e.g. trends identified in the incident reports. Also the importance of “near misses” is mentioned and workbook holders are referred to the accident pyramid to demonstrate this.

�� Audit & review – It is made clear that this is not the same as measuring performance. It is an assessment of the management system itself and should feed back into the start of the management system so that it operates as a continuous loop.

5. Going through the workbook page by page – The workbook itself becomes the focus of the rest of the briefing. Although the workbooks have all been tailored to suit each of the participating Trusts the framework remains the same. Any particular sections that relate

159

specifically to the individual Trust e.g. their own business case application rather than the one featured in the general issue are referred to in passing and questions relating to the processes involved forwarded to the Trust representative who is present at the briefing. Copies of the workbook disk are given to the Trust contact who is responsible for setting up a system for providing workbook holders with copies of blank tables etc as required.

(i) Log Page

This is usually completed at the beginning of the briefing. The need for the project team to track the workbooks is explained and the importance of comprehensive coverage of the Trust by the workbook system. Thus workbook holders are asked to inform the Trust contact if their role in the Trust changes or they leave the Trust, so that the workbook can be passed onto someone else who can cover their “patch”. Attention is drawn to the contact details of the Project Research Assistant, for the purposes of answering workbook-related questions.

(ii) Contents, Introduction & Further information

�� Contents Pages – These are bypassed. �� Introduction – Attention is focussed on the bold print at the bottom of the

introduction page which asks workbook holders not to “reinvent” their working methods to adapt to the workbook but to use the workbook so it fits in with what they have in place already. It is anticipated that the adoption of this approach will encourage people to use the workbook and there is nothing to be gained by asking managers to invest more time and energy than is required. The point is made that it is possible to do “too much” in attempting to manage health & safety and managers must prioritise and be sensible in their assessment of the gains to be made from investing effort.

�� Forward planner – It is recommended that managers incorporate their health & safety management system into their forward planning efforts, either by using the table provided, adapting the format if necessary, or by incorporating it into their existing organiser or wall planner etc. However, they are asked to make a note on the page referring to where the information is kept.

�� Where to find more help & information – The last part of this section is highlighted as a source of information that may prove useful to managers. (It usually includes the main individuals responsible for health & safety in the Trust, addresses for organisations like the HSE or Occupational Health and sometimes information on the training courses available in the Trust).

(iii) Policy

�� Workbook holders are given the option to list their main policies in the two tables provided and are asked to make a reference on these pages to where copies of the policies can be found.

(iv) Organising

�� Control – The structure of the line management system in each particular Trust is examined briefly by reference to the list of Trust roles and the flow diagram. The flow diagram is used to reinforce the message of the legal obligation of managers to take responsibility but to use the network of health & safety support that is available. Reference is made to the list of specialist advisers that is provided in this

160

section. Managers are asked to consider the individual roles and responsibilities within their departments and enter them in the table provided, adding other roles to the list if necessary.

�� Co-operation – A number of suggestions to achieve co-operation are listed, attention is drawn to these. Workbook holders are asked to either enter a record of consultation with staff into the table provided or refer to the appropriate documentation e.g. the minutes of a departmental health & safety meeting, stating where these can be found. The need to consult staff on pro-active health & safety issues, like policy development for example, is highlighted rather than limiting consultation to the discussion of solving hazards identified through the occurrence of incidents.

�� Communication – It is recommended that some reference be made to the “stock” of health & safety advisory documents held by the manager, either by listing these in the table and/or stating where they are kept. The importance of documenting actions taken following departmental safety meetings is emphasised, either a note of the main points in the workbook or reference to the minutes of the meetings are required.

�� Competence – The point is made that managers should be identifying the training needs within their department, including their own! Feedback regarding training needs is more effective than sending staff on whatever courses are available. The various factors to consider to ensure staff are competent are listed by bulletpoint and are discussed very briefly. Managers are encouraged to keep some record of staff training needs that they can refer to in order to ensure their staff are competent in their roles and responsibilities (particularly as the centralised recording of training requirements rarely includes the issue of re-training reminders to individuals).

(v) Planning & Implementing

�� Risk assessment/Procedures loop – The need to assess risk in the context of the tasks that are performed within the department is emphasised again. This should be a dynamic process whereby the risk assessments performed should be reviewed on a regular basis, or perhaps following some change in the method, equipment or individual associated with the task.

�� Working through the processes of a “risk control system” (shaded text box) - �� A method for controlling risk is outlined, it is based on the assumption that

managers have done little or nothing to tackle this issue prior to the briefing. It is recommended that workbook holders make a list of the tasks performed in their department. They are then required to assess the tasks (together with their risk assessors and the people who conduct the task if necessary) to determine whether they present a “significant” health & safety risk. At this point reference is made to the decision matrix (in some workbooks replaced by the Trust’s own version) which allows assessors to combine the factors of the severity of the hazard with the likelihood that an incident will occur. It is recommended that tasks that do not represent a significant risk are removed from the list and set aside for annual review. The risk assessor is instructed by the manager to conduct written risk assessments for the remaining tasks and the outcomes of these risk assessments are discussed with the manager on completion. It is stressed that at this point the responsibility is placed back on the manager to decide on suitable action to take to eliminate or reduce the level of risk.

161

�� Three possible outcomes following risk assessment are discussed, as well as the role that a business case and option appraisal (both included in the workbook) can play when managers find themselves constrained or unsure of the best action to take. An important point that is made is the need to prioritise assessment and action plans according to the degree of risk. This will allow managers to make the best use of the time they have available to concentrate on health & safety issues.

�� To assist with these processes tables are provided to record operational procedures and action plans, although workbook holders can refer elsewhere e.g. procedures manual, minutes of health & safety committee, as long as the important facts are documented.

�� Model business case/Option appraisal – The model business case is referred to very briefly as the form has a simplistic design. Workbook holders are encouraged to consider completing this form as a means of taking positive action under circumstances that are beyond their level of authority to action. At least five minutes is spent taking workbook holders through the worked example of the option appraisal (based on a manual handling of patients scenario). The use of this case to demonstrate the method is stressed, as one of the options involves increasing staffing levels and this approach is less up-to-date than the purchase of additional hoists to mechanise lifting. The difference between this “option appraisal” and traditional NHS option appraisal is emphasised and the inclusion of the option appraisal in the workbook in order to field test this new method with what is expected to be a minority of workbook holders. Attention is drawn to the weaknesses of the method – the subjective aspect of the scoring system, and for this reason anyone interested in using it is advised to tackle the analysis as a group. The basis of the scoring method is outlined – low score is good, reversal of the scale for costs and benefits aspects and the need for ranking of the three aspects of the analysis in order to combine the scores in a meaningful way. Finally the importance of making a written statement to support the findings of the analysis and the potential “next stage” of the process, being a full financial appraisal of the favoured options are discussed.

�� Risk assessment regulations/Form inclusion – Workbook holders are referred to the guidance available on the regulations which govern general risk assessment and the more specific risk assessment types such as COSHH and DSE. Copies of the Trust’s own forms are included in these sections and therefore the role of the manager in completion of the forms is again made clear. They should know what the forms look like and understand their content, but they are expected to delegate the task of completing the forms to a trained risk assessor.

(vi) Measuring Performance

�� Active/reactive measurement – These are defined. �� Safety inspections – Many departments have something in place to ensure regular

formal safety inspections. The benefits of conducting relatively informal safety inspections to pick up maintenance of housekeeping issues that may be going unnoticed in the department is highlighted. Workbook holders are referred to the hazard spotting checklist included. The distinction between inspection and task-based risk assessment is made clear.

�� Safety tours – Managers are encouraged to improve communication regarding health & safety issues by conducting informal tours of their department.

�� Accident/incident data – A brief discussion of the systems in place to record incident reports submitted takes place. Depending on the numbers of incidents received and the systems in place, workbook holders are made aware that a table is provided in the

162

workbook for recording the types of incidents that occur within the department so that any local trends might become apparent.

(vii) Reviewing performance

�� Manager’s audit – It is recommended that Managers conduct the audit that is included in the workbook as an entry point to the “loop system” that the workbook operates. This should help them to identify their health & safety objectives. They are asked to document these on a space that is left blank in the workbook and this should lead them to the completion of the forward planner at the beginning of the workbook.

�� Quarterly reporting – Workbook holders’ attention is drawn to the quarterly report format that is provided at the end of the workbook. It is recommended that they complete the reports using brief notes that state clearly their current position and any future plans. They are instructed to channel the reports through their risk management department/the project Trust contact rather than forwarding directly to the project team so that the process can be adopted by the Trusts upon completion of the research.

163

APPENDIX 6 PROTOCOL FOR FOLLOW-UP OF WORKBOOK HOLDERS

Feedback obtained via telephone interviews

1. All workbook holders within each test Trust will be contacted for feedback regarding their use of the workbook in accordance with the telephone interview protocol attached.

2. Each workbook holder will be contacted on three separate occasions and asked the same questions. This should occur within the following designated “windows”.

�� Between 1st April 2001 and 31st July 2001. �� Between 1st August 2001 and 30th November 2001. �� Between 1st December 2001 and 31st March 2002.

3. At first contact the workbook holder will be asked if it is convenient to answer questions regarding the workbook, if not the researcher will make an appointment to call back at an agreed time. The researcher will make it clear that the questions should only take around 10 minutes to answer.

4. Actions to be taken during questioning:

5. Q1. If no, end of questions but try to establish why. 6. Q2. If no end of questions but try to establish why. 7. Q3. List the sections used, try to establish if the workbook holder has completed

the manager’s audit, used the workbook generally or used particular sections of the workbook.

8. Q4. Ask workbook holder to justify their answer. 9. Q5. Record any risk control measures implemented, regardless of cost. Try to

ascertain the cost to the Trust. 10. Q6. If yes obtain a copy of the option appraisal form. 11. Q7. If yes note suggestions. 12. Some of those individuals who are using their workbooks will be selected at

random for examination of their workbooks.

164

1st April 2001 – 31st July 2001

ISSUE NUMBER 1: TRUST A PRIMARY CARE NHS TRUST

Contact details

Workbook code

Number______Name________________________________________

Location_______________________________________________________________Question 1

Are you still the holder of the workbook? YES NO

If not WHY?Question 2

Have you used the workbook? YES NO

If not WHY?

1.1.1 Question 3

If yes WHICH SECTIONS?

Question 3

Have you found the workbook helpful or not? YES NO

REASONS?Question 4

Have you implemented any control measures as a result of any risk assessment? YES NO

If yes, what were MEASURES?Question 5

Question 6

Have you any suggestions for improvements? YES NO

Date of interview______________

166

APPE

ND

IX 7

INC

IDEN

T D

ESC

RIP

TIVE

DAT

A

PHA

SE 1

Ta

ble

1 Tr

ust A

(20

Inci

dent

s)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

peD

escr

iptio

n

1 £3

3.13

O

ther

O

ther

O

ther

Fa

lse

fire

alar

m –

faul

ty sm

oke

dete

ctor

(det

ecto

r req

uire

d cl

eani

ng).

The

re w

as a

m

aint

enan

ce sc

hedu

le b

ut E

stat

es D

ept w

as b

ehin

d w

ith im

plem

enta

tion.

2

£689

.00

Staf

f Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t N

urse

hea

d-bu

tted

by p

atie

nt (a

lthou

gh p

atie

nt w

as o

n cl

ose

obse

rvat

ion

prio

r to

inci

dent

as

it ha

d be

en id

entif

ied

they

wer

e be

com

ing

aggr

essi

ve).

The

pat

ient

was

rest

ricte

d to

the

war

d.3

£130

.27

Staf

f A

ccid

enta

l in

jury

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Two

nurs

es c

ut w

ith p

oten

tially

infe

cted

gla

ss w

hile

tryi

ng to

take

it fr

om a

pat

ient

who

had

al

read

y cu

t her

self

with

it.

The

patie

nt w

as in

tent

on

self-

harm

ing

and

did

not w

ant s

taff

to

rem

ove

the

glas

s, ho

wev

er th

e in

jurie

s to

the

nurs

es w

ere

mor

e ac

cide

ntal

than

inte

ntio

nal.

4 £4

.00

Patie

nt

Nea

r mis

s O

ther

Pa

tient

det

aine

d un

der S

ectio

n 26

MH

A, l

eft w

ard

with

out p

erm

issi

on.

Foun

d by

staf

f in

an

adja

cent

car

par

k. D

ue to

pat

ient

’s h

ostil

ity a

nd a

ggre

ssio

n, p

olic

e as

sist

ance

was

requ

ired

to re

turn

her

to h

ospi

tal.

5 £3

4.44

Pa

tient

N

ear m

iss

Oth

er

A d

etai

ned

patie

nt a

bsco

nded

from

hos

pita

l war

d, ju

mpe

d in

to c

anal

and

refu

sed

to c

ome

out.

Eve

ntua

lly re

mov

ed fr

om th

e ca

nal b

y th

e po

lice.

6

£14.

00

Staf

f A

ccid

enta

l in

jury

H

ot o

r col

d co

ntac

t N

ursi

ng A

ssis

tant

unl

oadi

ng lu

nch

trolle

y tra

nsfe

rrin

g fo

od to

hot

ove

n. S

oup

cont

aine

r slid

to

floo

r as d

oor o

pene

d. F

oot s

cald

ed w

ith so

up.

Off

dut

y fo

llow

ing

first

aid

. 7

£18.

96

Patie

nt

Nea

r mis

s O

ther

Pa

tient

abs

cond

ed fr

om h

ospi

tal.

Pur

sued

by

staf

f bec

ause

of h

isto

ry o

f sui

cide

atte

mpt

s.

Ret

urne

d to

hos

pita

l with

min

imal

rest

rain

t. 8

£493

.31

Staf

f A

ccid

enta

l in

jury

O

ther

Tw

o st

aff n

urse

s pur

sued

an

absc

ondi

ng p

atie

nt w

ho st

rugg

led

whe

n re

stra

ined

. N

urse

was

pu

shed

off

pav

emen

t and

wen

t ove

r on

her a

nkle

. 9

£112

.00

Staf

f A

ccid

enta

l in

jury

St

ruck

aga

inst

so

met

hing

(fur

nitu

re

etc)

Nur

se w

alki

ng th

roug

h an

unl

it co

rrid

or k

nock

ed le

ft ha

nd a

gain

st w

all,

sust

aini

ng in

jury

.

10

£431

.20

Staf

f Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t A

n au

xilia

ry n

urse

was

ass

aulte

d w

hile

tryi

ng to

retri

eve

tabl

ets f

rom

the

floor

bec

ause

a

patie

nt h

ad sp

at o

ut th

e ta

blet

s. 11

£2

80.6

4 St

aff

Phys

ical

vi

olen

ce

Pers

on to

per

son

assa

ult

A st

aff n

urse

ass

aulte

d du

ring

drug

roun

d.

167

12

£88.

00

Staf

f Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t Pa

tient

with

lear

ning

dis

abili

ties a

ssau

lted

fello

w p

atie

nt.

Nur

sing

Ass

ista

nt w

as k

icke

d w

hils

t atte

mpt

ing

to d

efus

e th

e in

cide

nt.

Patie

nt w

as re

turn

ed to

his

room

by

forc

e w

here

he

proc

eede

d to

cau

se d

amag

e to

the

room

and

its c

onte

nts.

13

£413

.13

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

Nur

se to

iletin

g re

side

nt w

ho d

idn’

t wei

ght-b

ear.

Sus

tain

ed b

ack

inju

ry.

14

£52.

18

Staf

f Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t St

aff n

urse

ass

aulte

d tw

ice

by p

atie

nt a

fter t

hey

beca

me

aggr

essi

ve.

Req

uire

d su

ppor

t of 2

ot

her s

taff

to se

ttle

situ

atio

n.

15

£5.1

0 Pa

tient

N

ear m

iss

Oth

er

Patie

nt w

ith se

nile

dem

entia

left

war

d un

notic

ed.

Staf

f sea

rche

d lo

cal a

rea,

pat

ient

foun

d ab

out 1

½ m

iles a

way

and

retu

rned

to w

ard

unha

rmed

. 16

£6

57.4

5 St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/han

dlin

g St

aff n

urse

and

aux

iliar

y nu

rse

wer

e tra

nsfe

rrin

g pa

tient

from

bed

to w

heel

chai

r. T

he

patie

nt u

nabl

e/un

will

ing

to w

eigh

t-bea

r. A

uxili

ary

nurs

e su

stai

ned

back

inju

ry.

17

£7.7

2 Pa

tient

N

ear m

iss

Oth

er

Patie

nt n

oted

to b

e ab

sent

from

war

d. D

etai

ned

unde

r Sec

tion

26 M

HA

. Fo

llow

ing

a se

arch

, pat

ient

foun

d ou

twith

hos

pita

l gro

unds

aro

und

¾ m

ile a

way

. N

o ap

pare

nt in

jury

. 18

£1

2.88

Pa

tient

N

ear m

iss

Oth

er

Patie

nt a

bsco

nded

, sea

rch

inst

igat

ed.

Patie

nt fo

und

in lo

ng g

rass

in g

roun

ds.

Had

falle

n an

dw

as u

nabl

e to

get

up.

No

appa

rent

inju

ry n

oted

. 19

£1

0.12

Pa

tient

A

ccid

enta

l in

jury

O

ther

Pa

tient

abs

cond

ed fr

om w

ard.

Fou

nd a

bout

½ m

ile a

way

. Pa

tient

had

falle

n, su

stai

ned

smal

l abr

asio

ns.

20

£5.1

0 Pa

tient

N

ear m

iss

Oth

er

Patie

nt n

otifi

ed m

issi

ng.

Spot

ted

by o

ff-d

uty

mem

ber o

f sta

ff w

ho re

turn

ed h

im to

the

war

d.

Tabl

e 2

Trus

t B (1

5 In

cide

nts)

No

Cost

Staf

f/ pa

tient

/ ot

her

Cate

gory

Ty

peD

escr

iptio

n

1 £2

5.16

St

aff

Oth

er

Oth

er

Com

mun

ity p

sych

iatri

c nu

rse

bitte

n by

pup

py in

pat

ient

’s h

ome.

2 £3

0.87

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Nee

dles

tick

inju

ry su

stai

ned

by d

istri

ct n

urse

in p

atie

nt’s

hom

e. D

urin

g ad

min

istra

tion

of

inje

ctio

n, p

atie

nt b

ecam

e an

xiou

s and

mov

ed a

way

from

nur

se, c

ausi

ng a

shar

ps in

jury

to h

er

finge

r. A

t the

tim

e of

the

inci

dent

it w

as n

ot p

ossi

ble

to p

rocu

re re

tract

able

nee

dles

for

inje

ctio

ns c

arrie

d ou

t in

the

com

mun

ity.

This

has

sinc

e be

en c

orre

cted

. 3

£16.

00

Staf

f O

ther

O

ther

A

you

ng p

erso

n’s a

ddic

tion

wor

ker s

usta

ined

a d

og b

ite in

a c

lient

’s h

ouse

. 4

£173

.48

Staf

f A

ccid

enta

l in

jury

O

ther

D

urin

g a

cont

rol a

nd re

stra

int m

anoe

uvre

a n

ursi

ng a

ssis

tant

was

pre

ssed

aga

inst

ano

ther

m

embe

r of s

taff

and

an

alar

m o

n he

r bel

t pus

hed

into

her

ribs

. A

rib

inju

ry su

stai

ned.

5

£4,4

11.8

8 St

aff

Acc

iden

tal

Patie

nt li

fting

/han

dlin

g A

n en

rolle

d nu

rse

was

rolli

ng a

pat

ient

in b

ed in

ord

er to

was

h an

d ch

ange

him

. A

s she

rolle

d

168

inju

ry

the

patie

nt o

ver,

she

felt

a sh

arp

pain

. La

ter d

evel

oped

a p

ainf

ul le

ft hi

p.

6 £2

5.16

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Enro

lled

nurs

e su

stai

ned

shar

ps in

jury

thro

ugh

a cl

inic

al w

aste

bag

into

whi

ch 5

gla

ss

ampo

ules

and

gla

ss to

ps h

ad b

een

plac

ed.

Late

r dis

cove

red

to h

ave

been

dis

pose

d of

by

a do

ctor

dur

ing

the

adm

inis

tratio

n of

an

inje

ctio

n of

Vita

min

K to

a p

atie

nt.

7 £9

,849

.60

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

A G

rade

E st

aff n

urse

mov

ing

a co

nfus

ed p

atie

nt in

bed

, tw

iste

d he

r nec

k. L

ift h

ad b

een

cond

ucte

d by

the

nurs

e on

her

ow

n.

8 £3

12.1

8 St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

N

urse

slip

ped

dow

n 2

stai

rs w

hils

t run

ning

to a

ttend

an

emer

genc

y al

arm

(vio

lenc

e an

d ag

gres

sion

). D

urin

g sl

ip, l

eft a

nkle

turn

ed in

war

ds a

nd a

nkle

inju

ry w

as su

stai

ned.

9

£525

.00

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

An

auxi

liary

nur

se w

as a

ssis

ting

a pa

tient

to st

and

and

trans

ferr

ing

him

ont

o hi

s whe

elch

air

whe

n sh

e ex

perie

nced

a su

dden

pai

n in

her

mid

bac

k.

10

£1,3

39.2

2 St

aff

Acc

iden

tal

inju

ry

Stru

ck a

gain

st

som

ethi

ng (f

urni

ture

etc

) N

urse

chi

pped

a b

one

in a

rm d

urin

g a

cont

rol a

nd re

stra

int m

anoe

uvre

.

11

£358

.88

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

Supp

ort w

orke

r atte

mpt

ed to

bre

ak th

e fa

ll of

a p

atie

nt w

ho sl

ippe

d du

ring

a w

alk.

The

y su

stai

ned

a fr

actu

red

thum

b an

d cu

t to

right

han

d.

12

£141

.96

Staf

f A

ccid

enta

l in

jury

Pe

rson

to p

erso

n as

saul

t N

ursi

ng a

ssis

tant

sust

aine

d ha

nd in

jury

bei

ng k

icke

d by

pat

ient

whi

lst a

ttem

ptin

g to

adm

inis

ter

emer

genc

y m

edic

atio

n.

13

£2,3

10.0

0 St

aff

Acc

iden

tal

inju

ry

Stru

ck b

y an

obj

ect

Nur

sing

ass

ista

nt st

ruck

on

foot

by

flip

char

t, w

hich

was

pus

hed

over

by

a pa

tient

. Fl

ip C

hart

shou

ld n

ot h

ave

been

stor

ed in

that

are

a.

14

£2,2

05.0

0 St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/han

dlin

g N

ursi

ng a

uxili

ary

sust

aine

d tw

iste

d ba

ck a

ssis

ting

patie

nt to

stan

d.

15

£560

.81

Staf

f A

ccid

enta

l in

jury

M

anua

l lift

ing/

hand

ling

Tech

nica

l ins

truct

or su

stai

ned

back

inju

ry h

andl

ing

timbe

r whe

n th

ey w

ere

push

ing

timbe

r th

roug

h a

pane

l saw

ben

ch.

Tabl

e 3

Trus

t C (1

6 In

cide

nts)

No

Cost

Staf

f/ pa

tient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £2

2.54

St

aff

Acc

iden

tal i

njur

y Pa

tient

lifti

ng/h

andl

ing

Whi

le a

ssis

ting

in m

ovin

g a

dece

ased

pat

ient

from

the

bed

to a

trol

ley,

the

trolle

y sh

elf

mov

ed a

nd a

n au

xilia

ry n

urse

sust

aine

d an

inju

ry to

her

bac

k. T

he tr

olle

y w

as in

appr

opria

te

for t

his k

ind

of tr

ansf

er b

ut it

was

cho

sen

beca

use

it w

as p

lann

ed to

be

used

to tr

ansf

er th

e bo

dy in

to th

e fr

idge

. 2

£41.

44

Oth

er

Van

dalis

m

Oth

er

Win

dow

s bro

ken

by u

nkno

wn

pers

on(s

). N

o ot

her d

amag

e. R

epai

red

sam

e da

y.

3 £1

2.50

O

ther

V

anda

lism

O

ther

W

indo

w d

isco

vere

d sm

ashe

d. R

epor

ted

to p

olic

e an

d Es

tate

s Dep

t. N

o w

itnes

ses.

4 £4

1.72

O

ther

V

anda

lism

O

ther

Tw

o bo

ys tr

espa

ssin

g th

roug

h a

bath

room

win

dow

, bre

akin

g it.

Pol

ice

wer

e ca

lled

to th

e

169

scen

e. W

indo

w re

paire

d sa

me

even

ing.

5

£52.

11

Oth

er

Van

dalis

m

Oth

er

An

alar

m w

as a

ctiv

ated

and

was

hea

rd b

y st

aff o

n a

near

by w

ard.

Bot

h Po

lice

and

Esta

tes

Dep

t wer

e no

tifie

d. A

win

dow

was

bro

ken

and

the

build

ing

had

been

ent

ered

and

are

as

dist

urbe

d, b

ut n

othi

ng w

as st

olen

. 6

£1,0

68.5

0 O

ther

Th

eft

Oth

er

Bre

ak-in

and

thef

t of c

ompu

ter e

quip

men

t. 7

£102

.94

Oth

er

Thef

t O

ther

Th

e w

indo

w in

the

fron

t off

ice

was

bro

ken

durin

g th

e ni

ght a

nd a

n un

iden

tifie

d in

trude

r st

ole

a co

mpu

ter k

eybo

ard.

No

witn

esse

s. 8

£7.0

6 Pa

tient

A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Pa

tient

fell

whi

le w

alki

ng to

toile

t. P

atie

nt w

as w

alki

ng in

stoc

king

feet

with

out s

lippe

rs o

r Zi

mm

er.

Patie

nt d

id n

ot c

all f

or a

ssis

tanc

e an

d su

bseq

uent

ly fe

ll ba

ckw

ards

, stri

king

hea

d on

floo

r sus

tain

ing

lace

ratio

n to

hea

d.

9 £2

09.4

1 St

aff

Acc

iden

tal i

njur

y Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Enro

lled

nurs

e he

ard

a pa

tient

cho

king

. R

ushe

d to

hel

p, a

nd fa

iled

to n

otic

e a

haza

rd c

one,

an

d sl

ippe

d on

wet

floo

r. 10

£7

0.16

St

aff

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t St

aff n

urse

ass

aulte

d by

pat

ient

with

out w

arni

ng.

11

£8.7

2 Pa

tient

A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Pa

tient

uno

bser

ved

on c

orrid

or.

Fell,

frac

turin

g hi

p. T

rans

ferr

ed to

Acu

te T

rust

for

treat

men

t. 12

£7

.80

Patie

nt

Acc

iden

tal i

njur

y Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Patie

nt b

eing

nur

sed

on fl

oor s

usta

ined

frac

ture

d hi

p. T

rans

ferr

ed to

Acu

te H

ospi

tal f

or

treat

men

t. 13

£2

07.8

0 Pa

tient

A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Pa

tient

slip

ped

whi

le u

nobs

erve

d, su

stai

ning

an

inju

ry to

the

left

foot

. N

o fr

actu

res

dete

cted

. 14

£3

.55

Staf

f A

ccid

enta

l inj

ury

Nee

dles

tick/

shar

ps

Nur

sing

ass

ista

nt su

stai

ned

lace

ratio

n w

hils

t sha

ving

pat

ient

. 15

£3

0.27

O

ther

A

ccid

enta

l pr

oper

ty

loss

/dam

age

Oth

er

Gut

terin

g bl

ocke

d w

ith le

aves

ove

rflo

wed

, cau

sing

dam

age

to th

e ce

iling

whi

ch fe

ll in

pl

aces

cau

sing

floo

ding

to th

e flo

or.

16

£71.

89

Patie

nt

Ver

bal a

buse

or

thre

aten

ing

beha

viou

r

Oth

er

Patie

nt a

rriv

ed o

n th

e w

ard

acco

mpa

nied

by

2 fr

iend

s but

with

out a

refe

rral

lette

r fro

m th

eir

GP.

The

pat

ient

was

adv

ised

by

staf

f he

coul

d no

t be

asse

ssed

with

out a

lette

r and

su

gges

ted

he g

o to

A&

E or

bac

k to

his

GP.

As t

he p

atie

nt le

ft th

e bu

ildin

g, st

aff h

eard

a

win

dow

bre

akin

g. T

he p

olic

e di

d no

t wan

t to

take

act

ion

until

the

patie

nt w

as a

sses

sed.

Th

e pa

tient

retu

rned

to th

e w

ard

late

r for

ass

essm

ent.

170

Tabl

e 4

Trus

t D (2

8 In

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £2

02.1

0 St

aff

Acc

iden

tal i

njur

y C

ut w

ith sh

arp

mat

eria

l/obj

ect

A c

ater

ing

assi

stan

t sus

tain

ed a

dee

p la

cera

tion

to h

is h

and

from

a b

roke

n m

ilk ju

g w

hils

t was

hing

dis

hes.

Not

wea

ring

glov

es.

2 £3

2.40

O

ther

N

ear m

iss

Oth

er

Elec

tric

grill

in th

e ki

tche

n of

staf

f res

iden

ce w

as le

ft on

. Th

e ki

tche

n do

or w

as

open

and

smok

e tra

velle

d in

to th

e co

rrid

or a

nd a

ctiv

ated

the

smok

e al

arm

. Fi

re

Brig

ade

atte

nded

but

no

actio

n w

as re

quire

d. A

larm

s wer

e re

set b

y th

e on

cal

l el

ectri

cian

. 3

£17.

28

Staf

f A

ccid

enta

l inj

ury

Nee

dles

tick/

shar

ps

A c

omm

unity

staf

f nur

se w

as ta

king

a b

lood

sam

ple

from

a p

atie

nt.

On

rem

oval

of

the

need

le fr

om th

e pa

tient

’s a

rm, t

he n

eedl

e sl

ippe

d an

d pr

icke

d th

e nu

rse’

s lef

t th

umb.

4

£9.4

0 St

aff

Acc

iden

tal i

njur

y C

ut w

ith sh

arp

mat

eria

l/obj

ect

Dom

estic

was

doi

ng d

ishe

s fol

low

ing

a fu

nctio

n in

her

dep

artm

ent.

Not

wea

ring

yello

w k

itche

n du

ty g

love

s pro

vide

d. W

hils

t dry

ing

a fin

e-st

emm

ed g

lass

, it

slip

ped

and

brok

e in

her

han

d. S

ent t

o th

e A

cute

Tru

st fo

r tre

atm

ent.

5 £1

8.80

St

aff

Acc

iden

tal i

njur

y St

ruck

aga

inst

som

ethi

ng

(fur

nitu

re e

tc)

Elec

trici

an st

ruck

hea

d on

pip

e w

ork

whi

le c

heck

ing

batte

ries i

n ge

nera

tor r

oom

. N

o PP

E (b

ump

cap)

was

wor

n, a

lthou

gh th

ese

wer

e av

aila

ble

in th

e w

orks

hop.

In

jure

d pa

rty a

ttend

ed A

cute

Tru

st A

&E

Dep

t for

trea

tmen

t. 6

£66.

27

Staf

f A

ccid

enta

l inj

ury

Stru

ck b

y an

obj

ect

Mai

nten

ance

ass

ista

nt w

as a

djus

ting

the

cloc

ks b

y cl

imbi

ng o

nto

a ch

air.

The

cl

ock

fell

off w

all a

nd st

ruck

mai

nten

ance

ass

ista

nt o

n br

idge

of n

ose.

7

£238

.96

Staf

f A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

El

ectri

cian

wor

king

in a

n at

tic sp

ace

slip

ped

off a

jois

t. T

o av

oid

falli

ng th

roug

h th

e ce

iling

, fel

l ont

o hi

s lef

t sid

e in

jurin

g hi

s lef

t sho

ulde

r and

arm

. 8

£14.

60

Staf

f A

ccid

enta

l inj

ury

Stru

ck a

gain

st so

met

hing

(f

urni

ture

etc

) A

mem

ber o

f the

Est

ates

Dep

t was

esc

ortin

g a

cont

ract

or to

an

unde

r flo

or a

rea

to

cond

uct e

lect

rical

repa

irs.

Nei

ther

man

was

wea

ring

a ha

rd h

at.

The

cont

ract

or

wal

ked

into

a lo

w b

eam

, stri

king

his

fore

head

. Su

stai

ned

a la

cera

tion

and

suff

ered

co

ncus

sion

and

a b

rief l

oss o

f con

scio

usne

ss.

Esco

rted

to A

cute

Tru

st A

&E

Dep

t fo

r tre

atm

ent.

9 £2

38.0

0 O

ther

A

ccid

enta

l pr

oper

ty

loss

/dam

age

Con

tact

with

eq

uipm

ent/m

achi

nery

Po

rter d

amag

ed a

utom

atic

doo

rs b

y st

rikin

g th

em w

ith a

food

trol

ley

whi

ch h

e w

as

push

ing.

10

£806

.40

Staf

f A

ccid

enta

l inj

ury

Patie

nt li

fting

/han

dlin

g W

hils

t giv

ing

a pa

tient

a b

ath,

the

patie

nt d

ecid

ed to

subm

erge

him

self.

A

Hea

lthca

re A

ssis

tant

sust

aine

d ba

ck in

jury

whi

le, t

oget

her w

ith a

seni

or c

harg

e nu

rse,

atte

mpt

ing

to su

ppor

t the

pat

ient

’s h

ead

abov

e w

ater

. Th

e st

aff h

ad d

ecid

ed

that

he

shou

ld b

e gi

ven

a ba

th, d

espi

te h

is c

are

plan

stat

ing

he b

e sh

ower

ed.

171

11

£21.

02

Patie

nt

Nea

r mis

s O

ther

A

lega

lly d

etai

ned

patie

nt w

as se

en le

avin

g th

e w

ard

by st

aff.

At t

he ti

me,

the

absc

onde

d pa

tient

was

und

er g

ener

al o

bser

vatio

n. T

hree

mem

bers

of s

taff

pur

sued

th

e pa

tient

and

retu

rned

him

to th

e w

ard

unha

rmed

. 12

£2

1.02

Pa

tient

N

ear m

iss

Oth

er

Patie

nt a

bsco

nded

from

an

open

war

d w

here

he

was

in g

ener

al o

bser

vatio

n.

Pers

on re

sidi

ng n

earb

y to

the

hosp

ital p

hone

d to

not

ify o

f the

ir w

here

abou

ts a

nd

the

patie

nt w

as c

olle

cted

by

a st

aff m

embe

r and

retu

rned

to th

e w

ard.

13

£4

3.40

St

aff

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t En

rolle

d nu

rse

head

-but

ted

by p

atie

nt w

hile

ass

istin

g w

ith a

vio

lent

out

burs

t. 14

£6

.60

Patie

nt

Nea

r mis

s O

ther

Pa

tient

det

aine

d un

der M

enta

l Hea

lth A

ct u

nder

con

stan

t obs

erva

tion

was

take

n by

se

nior

hou

se o

ffic

er to

be

inte

rvie

wed

in a

n ex

tern

al in

terv

iew

room

. Pa

tient

re

turn

ed to

the

war

d w

ith so

me

forc

e re

quire

d.

15

£67.

51

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Nur

sing

ass

ista

nt in

terv

ened

to se

para

te p

atie

nts w

ho w

ere

fight

ing

and

sust

aine

d in

jurie

s to

her n

eck

and

elbo

w.

The

nurs

e se

en b

y th

e A

cute

Tru

st A

&E

Dep

t. 16

£1

0.78

Pa

tient

N

ear m

iss

Oth

er

Abs

cond

ing

patie

nt.

Sear

ch in

itiat

ed b

ut p

atie

nt la

ter r

etur

ned

to th

e w

ard

of h

is

own

free

will

, unh

arm

ed.

17

£5.5

0 Pa

tient

N

ear m

iss

Oth

er

Abs

cond

ed p

atie

nt re

turn

ed to

the

war

d af

ter a

sear

ch.

18

£19.

47

Staf

f A

ccid

enta

l inj

ury

Stru

ck a

gain

st so

met

hing

(f

urni

ture

etc

) G

ener

al a

dmin

ass

ista

nt w

as in

jure

d w

hile

car

ryin

g a

flip

char

t and

mar

ker p

ens

thro

ugh

doub

le d

oors

. Tr

eate

d by

Acu

te T

rust

A&

E D

ept.

19

£95.

20

Staf

f A

ccid

enta

l inj

ury

Patie

nt li

fting

/han

dlin

g A

Hea

lthca

re A

ssis

tant

was

ass

istin

g an

eld

erly

pat

ient

to w

alk

with

the

aid

of a

Zi

mm

er.

Patie

nt le

t go

of th

e Zi

mm

er a

nd fe

ll. T

he H

ealth

care

ass

ista

nt su

ffer

ed

carp

et b

urns

atte

mpt

ing

to c

ontro

l the

fall.

20

£1

2.40

Pa

tient

N

ear m

iss

Oth

er

Dem

entia

pat

ient

abs

cond

ed.

Foun

d 30

min

s lat

er u

nhar

med

, by

nurs

ing

auxi

liary

an

d re

turn

ed to

war

d.

21

£28.

00

Staf

f A

ccid

enta

l inj

ury

Patie

nt li

fting

/han

dlin

g H

ealth

care

Ass

ista

nt tr

ansf

errin

g pa

tient

from

bed

to w

heel

chai

r whe

n pa

tient

st

umbl

ed a

nd tr

appe

d H

CA

’s h

and

betw

een

him

self

and

lock

er.

Nig

htsh

ift st

aff

used

hoi

st to

tran

sfer

pat

ient

due

to h

is d

eter

iora

tion

in h

ealth

, but

this

had

not

be

en c

omm

unic

ated

to d

ay st

aff.

22

£26.

75

Staf

f A

ccid

enta

l inj

ury

Nee

dles

tick/

shar

ps

Staf

f nur

se o

btai

ned

ster

ile u

rine

sam

ple

usin

g a

syrin

ge a

nd n

eedl

e fr

om a

cath

eter

por

t at t

he p

atie

nt’s

bed

side

. Pl

aced

the

used

nee

dle

back

on

the

tray

and

took

to th

e sl

uice

are

a fo

r dis

posa

l. D

istra

cted

whe

n so

meo

ne sp

oke

to h

er, s

he

pick

ed u

p th

e ne

edle

by

the

shar

p en

d an

d w

as in

jure

d.

23

£7.0

7 St

aff

Acc

iden

tal i

njur

y C

onta

ct w

ith e

xpos

ure

to

harm

ful s

ubst

ance

s A

22½

-litre

ble

ach

barr

el b

urst

whi

le b

eing

tran

spor

ted

into

the

laun

dry.

The

bl

each

spill

ed o

nto

the

clot

hing

and

skin

of a

laun

dere

tte su

perin

tend

ent,

caus

ing

her t

o su

stai

n irr

itatio

n to

her

legs

and

che

st.

Mem

ber o

f Est

ates

dec

ante

d th

e bl

each

into

oth

er c

onta

iner

s and

dilu

ted

the

spilt

ble

ach.

24

£3

7.12

O

ther

O

ther

O

ther

R

esid

ents

wer

e sm

okin

g in

the

sitti

ng ro

om, w

hich

act

ivat

ed th

e fir

e al

arm

whi

ch

172w

as se

t to

“sm

oke

dete

ctio

n”.

25

£260

.00

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Res

iden

t gra

bbed

an

enro

lled

nurs

e by

her

clo

thin

g as

she

ente

red

a fla

t. T

he

pinp

oint

ala

rm w

as a

ctiv

ated

but

did

n’t w

ork.

The

resi

dent

ass

aulte

d th

e nu

rse

and

bit h

er ri

ght f

orea

rm.

26

£12.

37

Patie

nt

Nea

r mis

s O

ther

D

emen

tia p

atie

nt a

bsco

nded

from

war

d. P

atie

nt e

vent

ually

loca

ted

at a

loca

l ch

emis

t sho

p.

27

£1,2

08.7

0 St

aff

Acc

iden

tal i

njur

y M

anua

l lift

ing/

hand

ling

Porte

r rem

ovin

g a

parc

el fr

om th

e ba

ck o

f a h

igh

top

van

sust

aine

d a

pulle

d ha

mst

ring.

Atte

nded

A&

E fo

r tre

atm

ent a

nd w

as o

n si

ck le

ave

follo

win

g th

e in

cide

nt.

28

£338

.40

Staf

f A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Po

rter f

ell o

ver c

ardb

oard

box

es w

hile

pic

king

up

blac

k ba

gs fr

om th

e ca

rdbo

ard

stor

age

area

in th

e ho

spita

l. S

usta

ined

a sp

rain

ed a

nkle

and

sore

shou

lder

.

Tabl

e 5

Trus

t E (2

2 In

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £5

.83

Staf

f A

ccid

enta

l in

jury

C

onta

ct w

ith

equi

pmen

t/mac

hine

ry

Porte

r was

mov

ing

trest

le ta

ble

and

was

wou

nded

by

a sp

linte

r in

his l

eft t

hum

b.

Atte

nded

A&

E.

2 £1

4.10

St

aff

Acc

iden

tal

inju

ry

Hot

or c

old

cont

act

Seas

onal

wor

ker (

stud

ent)

was

was

hing

cut

lery

. R

outin

e w

ashe

r bro

ken.

On

usin

g w

ashe

r in

kitc

hen

area

, sca

lded

righ

t han

d/w

rist a

fter i

mm

ersi

ng in

wat

er a

bove

gl

ove

leve

l. 3

£14.

10

Staf

f A

ccid

enta

l in

jury

C

onta

ct w

ith

equi

pmen

t/mac

hine

ry

A c

ater

ing

assi

stan

t sus

tain

ed a

n an

kle

inju

ry a

fter p

ullin

g (tr

aine

d to

pus

h ra

ther

th

an to

pul

l) pl

ate

hold

ers t

rolle

y to

stor

age

area

. Th

e ca

terin

g as

sist

ant c

lippe

d le

ft an

kle

on w

heel

bra

kes.

4 £3

2.90

St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

C

ater

ing

assi

stan

t, w

earin

g sa

fety

shoe

s, sl

ippe

d on

kitc

hen

floor

(non

slip

surf

ace)

in

jurin

g rig

ht e

lbow

and

coc

cyx.

5

£2.3

5 St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/han

dlin

g A

por

ter s

usta

ined

a b

ack

inju

ry tr

ansf

errin

g an

ana

esth

etis

ed p

atie

nt fr

om a

thea

tre

trolle

y to

a b

ed u

sing

pat

slid

e in

rece

ptio

n/re

cove

ry a

rea.

The

stra

in w

as su

stai

ned

as a

resu

lt of

poo

r pos

ition

ing

of th

e pa

tient

/bed

/trol

ley.

6

£9.3

5 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Porte

r sus

tain

ed a

nee

dles

tick

inju

ry fr

om a

clin

ical

was

te b

ag.

The

need

lest

ick

inju

ry w

as su

stai

ned

to th

e rig

ht le

g w

hen

mov

ing

the

bag.

7

£11.

39

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

Fo

llow

ing

vena

pun

ctur

e, re

mov

ed c

anul

a an

d pl

aced

in p

ulp

tray.

On

trans

fer f

rom

tra

y to

shar

ps d

ispo

sal b

ox, s

usta

ined

nee

dles

tick

inju

ry.

8 £5

28.7

5 St

aff

Acc

iden

tal

inju

ry

Man

ual l

iftin

g/ha

ndlin

g A

dom

estic

dis

posi

ng o

f rub

bish

into

a c

entra

l cor

e ar

ea.

Rec

epta

cle

alre

ady

over

full,

and

whe

n th

row

ing

rubb

ish

onto

top

of ru

bbis

h pi

le, d

omes

tic p

ulle

d a

173

mus

cle

in h

er b

ack.

9

£4.7

0 St

aff

Acc

iden

tal

inju

ry

Stru

ck a

gain

st so

met

hing

(f

urni

ture

etc

) D

omes

tic se

cond

ed to

out

patie

nts d

epar

tmen

t. H

igh

dust

ing

in c

onfin

ed a

rea

on

step

s. B

ange

d he

ad o

n lig

ht fi

tmen

t.

10

£709

.70

Staf

f A

ccid

enta

l in

jury

C

onta

ct w

ith

equi

pmen

t/mac

hine

ry

Dom

estic

com

men

cing

floo

r cle

anin

g w

ith b

uffin

g m

achi

ne in

adve

rtent

ly tr

od o

n el

ectri

c ca

ble

caus

ing

her t

o je

rk b

ackw

ards

, lea

ding

to a

mus

culo

-ske

leta

l inj

ury.

11

£1

0.39

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Nur

se w

as c

uttin

g ox

ygen

tubi

ng w

hen

her s

ciss

ors s

lippe

d, c

ausi

ng a

pun

ctur

e w

ound

in th

e th

umb

pad

of h

er le

ft ha

nd.

12

£11.

39

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

N

urse

dis

posi

ng o

f nee

dle

into

ove

rful

l sha

rps b

ox su

stai

ned

need

lest

ick

inju

ry fr

om

prot

rudi

ng n

eedl

e.

13

£6,8

55.0

0 Pa

tient

A

ccid

enta

l in

jury

Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Patie

nt le

ft al

one

on a

com

mod

e sl

ippe

d an

d fe

ll. F

ract

ured

nec

k of

fem

ur su

stai

ned.

14

£6.5

8 Pa

tient

A

ccid

enta

l in

jury

Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Patie

nt fe

ll as

leep

in c

hair

whi

le w

atch

ing

tele

visi

on.

On

getti

ng u

p to

go

to b

ed,

slip

ped

and

fell

(no

slip

pers

on

due

to sw

olle

n to

e), h

ittin

g he

ad o

f doo

r and

su

stai

ning

a la

cera

tion

to h

is le

ft ea

r. 15

£1

0.54

Pa

tient

A

ccid

enta

l in

jury

Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

An

81 y

ear o

ld fe

mal

e pa

tient

in a

hig

hly

conf

used

stat

e at

tem

pted

to v

isit

the

toile

t un

aide

d by

staf

f. F

ell i

n th

e to

ilet a

rea.

Inj

urie

s to

head

and

shin

wer

e su

stai

ned.

16

£5

0.90

St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/han

dlin

g A

uxili

ary

nurs

e su

stai

ned

knee

inju

ry d

urin

g m

anua

l han

dlin

g of

pat

ient

.

17

£17.

00

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

C

linic

al su

perv

isor

ove

rsee

ing

proc

edur

e, g

ivin

g in

stru

ctio

n on

cor

rect

pos

ition

ing

of

need

le d

urin

g su

turin

g, re

ceiv

ed n

eedl

estic

k in

jury

. 18

£2

6.70

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Dom

estic

cle

anin

g be

d in

labo

ur w

ard

afte

r clin

ical

pro

cedu

re.

Una

war

e us

ed n

eedl

e w

as m

issi

ng.

Lost

in fo

ld o

f bed

. Su

bseq

uent

ly sc

ratc

hed

her r

ight

arm

, sus

tain

ing

need

lest

ick

inju

ry.

19

£94.

50

Staf

f Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t A

vio

lent

inci

dent

invo

lvin

g su

icid

al p

atie

nt in

war

d.

20

£23.

99

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

N

urse

was

rem

ovin

g a

butte

rfly

nee

dle

from

pat

ient

’s sk

in.

On

rem

ovin

g th

e ne

edle

, sh

e st

abbe

d he

r rig

ht th

umb,

nee

dle

pene

trate

d he

r glo

ve.

21

£7

.69

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

M

edic

al te

chni

cian

sust

aine

d cu

t fro

m S

tanl

ey k

nife

, whi

ch h

e w

as u

sing

to c

ut

suct

ion

tubi

ng.

Kni

fe sl

ippe

d, c

uttin

g fin

ger o

n le

ft ha

nd.

22

£21.

88

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

R

adio

grap

her’

s ass

ista

nt su

stai

ned

need

lest

ick

inju

ry c

lear

ing

away

pro

cedu

res

trolle

y. W

hils

t fol

ding

dis

card

ed p

ad u

sed

by d

octo

r dur

ing

proc

edur

e, n

eedl

e pr

otru

ded

and

pric

ked

left

ring

finge

r.

174

Tabl

e 6

Trus

t F (c

ontro

l) (1

6 In

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £4

.62

Staf

f A

ccid

enta

l inj

ury

Man

ual l

iftin

g/ha

ndlin

g C

ater

ing

Supe

rvis

or su

stai

ned

a sh

ould

er in

jury

sorti

ng m

eat o

n tra

ys b

ecau

se th

e w

rong

ord

er o

f mea

t was

del

iver

ed to

the

Cat

erin

g D

ept (

bags

had

bee

n or

dere

d in

4

kg lo

ts).

2 £1

90.3

5 St

aff

Acc

iden

tal i

njur

y St

ruck

aga

inst

som

ethi

ng

(fur

nitu

re e

tc)

Cat

erin

g A

ssis

tant

cle

anin

g in

staf

f cof

fee

loun

ge tu

rned

to se

rve

a cu

stom

er a

nd

caug

ht h

er le

ft fo

ot o

n a

low

cof

fee

tabl

e, c

ausi

ng h

er to

lose

her

bal

ance

and

fall,

tw

istin

g he

r rig

ht k

nee.

3

£236

.71

Staf

f A

ccid

enta

l inj

ury

Cut

with

shar

p m

ater

ial/o

bjec

t A

coo

k w

as c

uttin

g br

ead

into

cro

uton

s usi

ng a

bre

ad k

nife

. B

ecam

e di

stra

cted

and

cu

t fin

ger o

n le

ft ha

nd.

4 £8

0.18

St

aff

Acc

iden

tal i

njur

y St

ruck

aga

inst

som

ethi

ng

(fur

nitu

re e

tc)

Staf

f Nur

se su

stai

ned

head

inju

ry in

col

lisio

n w

ith a

cei

ling

mou

nted

mic

rosc

ope

in

thea

tre.

5 £1

67.5

5 St

aff

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t St

aff N

urse

sust

aine

d fa

cial

/nec

k in

jurie

s in

viol

ent i

ncid

ent w

ith a

con

fuse

d pa

tient

(p

ost o

pera

tive

hallu

cina

tions

/par

anoi

a).

6 £1

57.4

5 St

aff

Acc

iden

tal i

njur

y M

anua

l lift

ing/

hand

ling

A d

omes

tic a

ssis

tant

sust

aine

d ba

ck in

jury

lifti

ng a

grid

from

a sh

ower

bas

e.

7 £1

49.6

0 St

aff

Acc

iden

tal i

njur

y Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Dom

estic

ass

ista

nt sp

rain

ed a

nkle

trip

ping

ove

r vac

uum

hos

e w

hile

wor

king

in

thea

tre.

Did

not

tidy

aw

ay v

acuu

m c

lean

er in

dom

estic

serv

ices

room

. W

ent t

o do

an

othe

r tas

k an

d su

bseq

uent

ly tr

ippe

d ov

er th

e ho

se o

f the

vac

uum

cle

aner

. 8

£40.

00

Oth

er

Acc

iden

tal

prop

erty

lo

ss/d

amag

e

Oth

er

A re

vers

ing

van

driv

er st

ruck

ext

erna

lly p

roje

ctin

g w

ard

win

dow

.

9 £8

1.93

St

aff

Acc

iden

tal i

njur

y St

ruck

by

an o

bjec

t St

aff N

urse

was

dis

conn

ectin

g an

airl

ine

from

the

wal

l in

the

HD

U u

nit.

Bui

ld u

p of

pre

ssur

e w

as su

ffic

ient

to p

ush

the

conn

ecto

r off

its s

eatin

g an

d w

as h

it on

the

chin

by

the

met

al p

oint

. 10

£9

0.56

St

aff

Acc

iden

tal i

njur

y M

anua

l lift

ing/

hand

ling

Cle

rical

Offi

cer s

usta

ined

a b

ack

inju

ry re

cove

ring

note

s fro

m a

hig

h sh

elf.

11

£14.

37

Staf

f A

ccid

enta

l inj

ury

Man

ual l

iftin

g/ha

ndlin

g A

n au

xilia

ry n

urse

was

pus

hing

an

incu

bato

r tro

lley

alon

g th

e co

rrid

or to

a c

lean

ing

room

in th

e N

eona

tal U

nit.

The

nur

se p

ushe

d th

roug

h fir

e do

ors,

mis

judg

ed th

e an

gle

of th

e in

cuba

tor a

nd c

augh

t one

of t

he la

rge

whe

els o

n th

e ed

ge o

f the

doo

r.

The

incu

bato

r rec

oile

d an

d th

e au

xilia

ry n

urse

hit

her h

ead

on th

e Pe

rspe

x ho

od.

12

£265

.68

Staf

f A

ccid

enta

l inj

ury

Patie

nt li

fting

/han

dlin

g St

aff N

urse

sust

aine

d ne

ck in

jury

tran

sfer

ring

a pa

tient

in a

whe

elch

air.

13

£17.

63

Staf

f A

ccid

enta

l inj

ury

Stru

ck a

gain

st so

met

hing

(f

urni

ture

etc

) Po

rter s

usta

ined

hea

d in

jury

bum

ping

aga

inst

incu

bato

r box

han

dle,

whi

ch w

as

sitti

ng o

n a

tabl

e in

the

porte

r’s o

ffic

e.

14

£592

.75

Staf

f A

ccid

enta

l inj

ury

Con

tact

with

Fi

nanc

e D

ept w

as m

ovin

gpr

emis

es a

nd th

e Fi

nanc

ial A

ssis

tant

was

ass

istin

g in

the

175

equi

pmen

t/mac

hine

ry

co-o

rdin

atio

n of

the

mov

ing

of fi

les a

nd e

quip

men

t bet

wee

n of

fices

. Su

stai

ned

a fo

ot c

rush

inju

ry o

n va

n ta

ilgat

e.

15

£17.

01

Staf

f A

ccid

enta

l inj

ury

Man

ual l

iftin

g/ha

ndlin

g Su

perin

tend

ent R

adio

grap

her s

usta

ined

leg

inju

ry m

ovin

g le

ad a

pron

trol

ley.

16

£2

,091

.20

Patie

nt

Fata

lity

Fall

from

hei

ght

Patie

nt su

stai

ned

head

inju

ry in

fall

from

nar

row

scre

enin

g ta

ble.

Tabl

e 7

Trus

t G (c

ontro

l) (1

0 In

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £4

50.8

0 St

aff

Acc

iden

tal

inju

ry

Stru

ck b

y an

obj

ect

Staf

f Nur

se su

stai

ned

head

inju

ry w

hen

stru

ck b

y fa

lling

ligh

t fitt

ing.

Fol

low

ing

the

inci

dent

, Sta

ff N

urse

giv

en T

etan

us in

ject

ion,

whi

ch le

d to

an

adve

rse

phys

iolo

gica

l re

actio

n, re

sulti

ng in

ext

ende

d si

ck le

ave.

2

£22.

00

Staf

f Ph

ysic

Pe

rson

to p

erso

n as

saul

t C

harg

e N

urse

sust

aine

d a

hum

an b

ite to

arm

dur

ing

cont

rol a

nd re

stra

int p

roce

dure

.Tr

ansf

erre

d to

Acu

te T

rust

A&

E D

ept f

or tr

eatm

ent.

3 £1

2.13

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Enro

lled

Nur

se d

ispo

sing

of I

V fl

uids

into

the

sink

of t

he T

reat

men

t Roo

m.

Nee

dle

fell

into

the

sink

and

, on

pick

ing

it up

, the

nur

se su

stai

ned

a ne

edle

stic

k in

jury

to h

er

finge

r. 4

£263

.10

Staf

f A

ccid

enta

l in

jury

C

onta

ct w

ith

equi

pmen

t/mac

hine

ry

Enro

lled

Nur

se su

stai

ned

arm

inju

ry w

hen

a co

mm

ode

topp

led,

afte

r a sc

rew

cam

e ou

t of a

cas

tor a

nd le

ft fr

ont w

heel

snap

ped

off.

5 £1

2.13

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

An

auxi

liary

nur

se w

as fe

edin

g a

patie

nt in

bed

whi

le a

Sta

ff N

urse

was

si

mul

tane

ousl

y ad

min

iste

ring

a flu

vac

cine

. Th

e au

xilia

ry n

urse

’s a

rm w

as ra

ised

an

d, a

s the

nee

dle

was

with

draw

n it

punc

ture

d he

r lef

t for

earm

. 6

£40.

00

Patie

nt

Oth

er

Oth

er

Lear

ning

dis

abili

ties p

atie

nt b

roke

a w

indo

w d

urin

g ra

mpa

ge.

A d

rug

erro

r fol

low

ed

in th

e en

suin

g co

nfus

ion.

7

£2,8

71.7

5 St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/han

dlin

g St

aff N

urse

sust

aine

d in

jury

to b

ack

whi

le a

ssis

ting

patie

nt.

Patie

nt c

ould

n’t b

e tra

nsfe

rred

usi

ng h

oist

/slin

gs a

s she

was

a st

roke

vic

tim a

nd v

ery

limite

d in

abi

lity,

an

d te

nded

to g

rab

and

thro

w h

erse

lf fo

rwar

d.

8 £6

60.0

2 St

aff

Acc

iden

tal

inju

ry

Pers

on to

per

son

assa

ult

Nur

sing

Ass

ista

nt st

ruck

on

brid

ge o

f nos

e by

pla

te th

row

n by

a p

atie

nt.

9 £3

9.36

Pa

tient

A

ccid

enta

l in

jury

Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Patie

nt su

stai

ned

hip

inju

ry a

fter f

allin

g w

hils

t bei

ng e

scor

ted

to th

e to

ilet.

10

£2,4

50.0

0 St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Te

leph

onis

t sus

tain

ed fr

actu

red

ankl

e tri

ppin

g on

a h

ole

in p

rote

ctiv

e vi

nyl m

at.

176

PHA

SE 2

Tabl

e 8

Trus

t A (2

7 In

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £2

1.42

St

aff

Ver

bal a

buse

or

thre

aten

ing

beha

viou

r O

ther

C

omm

unity

Nur

se th

reat

ened

by

son

of p

atie

nt.

The

nurs

e be

ing

thre

aten

ed v

erba

lly d

e-es

cala

ted

the

situ

atio

n, b

ut th

e vi

sit w

as e

xten

ded

by a

n ho

ur a

s a re

sult.

2

£16.

08

Patie

nt

Fire

Ex

posu

re to

fire

Ps

ychi

atric

pat

ient

set f

ire to

pap

er to

wel

s, th

us c

ausi

ng fi

re a

larm

to so

und.

3

£280

.72

Patie

nt

Van

dalis

m

Expo

sure

to fi

re

Patie

nt se

t fire

to b

eddi

ng a

nd m

attre

ss in

thei

r roo

m.

Fire

ala

rm a

ctiv

ated

by

smok

e.

4 £1

37.1

0 Pa

tient

A

ccid

enta

l inj

ury

Oth

er

An

old

wou

nd o

pene

d on

fore

arm

dur

ing

stru

ggle

with

pat

ient

. 5

£19.

67

Patie

nt

Fata

lity

Oth

er

Patie

nt fo

und

hang

ing,

usi

ng a

rope

susp

ende

d fr

om a

win

dow

cat

ch.

Iden

tifie

d as

a

high

risk

, but

had

not

yet

bee

n ac

tione

d.

6 £1

9.04

St

aff

Ver

bal a

buse

or

thre

aten

ing

beha

viou

r O

ther

N

urse

Tea

m L

eade

r ver

bally

abu

sed

and

phys

ical

ly th

reat

ened

whe

n he

ask

ed tw

o m

ale

inpa

tient

s to

hand

ove

r alc

ohol

they

wer

e fo

und

to b

e co

nsum

ing.

7

£6.7

2 Pa

tient

O

ther

O

ther

Pa

tient

det

aine

d un

der M

HA

but

hou

sed

in a

n ac

ute

psyc

hiat

ric w

ard,

bec

ame

a nu

isan

ce to

the

neig

hbor

ing

war

d du

e to

thei

r con

tinua

lly le

avin

g th

e w

ard.

De-

esca

latio

n m

easu

res f

aile

d, re

sulti

ng in

pat

ient

rest

rain

t by

3 st

aff a

nd re

turn

ed to

war

d.

8 £1

5.47

Pa

tient

Ph

ysic

al v

iole

nce

Oth

er

A 5

3 ye

ar o

ld fe

mal

e pa

tient

with

lear

ning

dis

abili

ties l

eft w

ard

to w

alk

cons

ider

able

dist

ance

hom

e. R

etur

ned

to w

ard

with

staf

f. A

ggre

ssiv

e on

retu

rn, r

equi

red

to b

e re

stra

ined

. 9

£11.

68

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t A

53

year

old

fem

ale

patie

nt le

ft w

ard

thre

aten

ing

suic

ide.

Hou

sed

in a

cute

psy

chia

tric

war

d w

here

staf

f do

not r

outin

ely

wor

k w

ith th

is c

lient

gro

up.

Foun

d he

r pro

blem

atic

to

man

age.

10

£9

8.88

Pa

tient

V

erba

l abu

se o

r th

reat

enin

g be

havi

our

Oth

er

Patie

nt w

ith le

arni

ng d

isab

ilitie

s abs

cond

ed fr

om a

cute

psy

chia

tric

war

d. N

o st

aff w

ith

train

ing

in le

arni

ng d

isab

ilitie

s. 11

£9

.13

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t A

53-

year

old

fem

ale

patie

nt w

ith le

arni

ng d

isab

ilitie

s det

aine

d in

acu

te p

sych

iatri

c w

ard.

12

£23.

80

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Nur

sing

aux

iliar

y bi

tten

by p

atie

nt.

Had

rece

ived

no

viol

ence

and

agg

ress

ion

train

ing.

13

£429

.65

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Nur

sing

aux

iliar

y le

ft on

her

ow

n in

unf

amili

ar a

rea

assa

ulte

d by

pat

ient

. N

ursi

ng

auxi

liary

not

trai

ned

in v

iole

nce

& a

ggre

ssio

n.

14

£2.6

1 St

aff

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t M

ale

patie

nt w

ith le

arni

ng d

iffic

ultie

s bec

ame

phys

ical

ly a

busi

ve to

war

ds fe

llow

pa

tient

. W

hen

staf

f int

erve

ned

patie

nt b

ecam

e ph

ysic

ally

abu

sive

tow

ards

them

. A

ssis

tanc

e re

quire

d to

rest

rain

the

patie

nt fo

llow

ing

the

atta

ck.

177

15

£5.1

5 St

aff

Phys

ical

vio

lenc

e O

ther

A

pat

ient

requ

ired

rest

rain

ing

afte

r thr

eate

ning

phy

sica

l vio

lenc

e. A

larm

act

ivat

ed fo

ras

sist

ant t

o in

crea

se n

umbe

rs fo

r res

train

t. 16

£3

.85

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Occ

upat

iona

l The

rapi

st b

itten

by

patie

nt w

ith le

arni

ng d

isab

ilitie

s.

17

£1,2

85.8

9 St

aff

Oth

er

Oth

er

Wor

k re

late

d st

ress

in n

ursi

ng a

uxili

ary

thou

ght t

o be

rela

ted

to u

nsta

ble

man

agem

ent

arra

ngem

ents

ove

r ens

uing

per

iod

follo

win

g m

ove

to n

ew p

rem

ises

. O

n in

terv

iew

, ot

her s

taff

wor

king

in th

e ar

ea e

xpre

ssed

feel

ings

of w

ork

rela

ted

stre

ss.

18

£3,3

85.0

2 St

aff

Ill h

ealth

O

ther

St

aff a

bsen

t fro

m w

ork

due

to p

rolo

nged

wor

k re

late

d st

ress

. U

nabl

e to

cop

e an

y lo

nger

w

ith p

atie

nt w

ith se

vere

beh

avio

ural

pro

blem

s, ru

nnin

g al

ongs

ide

staf

fing

prob

lem

s pr

ecip

itate

d by

dis

cipl

inar

y ac

tion

resu

lting

in d

ism

issa

ls/re

sign

atio

ns a

s wel

l as a

co

mpl

ete

chan

ge in

env

ironm

ent f

or p

atie

nts a

nd st

aff o

n ne

w p

rem

ises

a y

ear

prev

ious

ly.

19

£3.8

8 St

aff

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t St

aff r

equi

ring

assi

stan

ce to

sepa

rate

2 le

arni

ng d

isab

ilitie

s pat

ient

s fig

htin

g.

20

£7,2

62.7

8 St

aff

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t St

aff N

urse

kic

ked

by p

atie

nt w

hile

tryi

ng to

pro

tect

ano

ther

(Sta

ff N

urse

had

not

had

vi

olen

ce &

agg

ress

ion

train

ing)

. 21

£3

.60

Oth

er

Oth

er

Oth

er

Fire

ala

rm a

ctiv

ated

by

burn

t toa

st.

Kitc

hen

door

hel

d op

en, s

o sm

oke

able

to e

scap

e fr

om k

itche

n. P

atie

nt u

nsup

ervi

sed,

des

pite

pol

icy

that

all

toas

t mak

ing

shou

ld b

e su

perv

ised

. 22

£4

.59

Patie

nt

Oth

er

Oth

er

A 4

5 ye

ar o

ld fe

mal

e de

tain

ed p

atie

nt a

bsco

nded

from

war

d, h

ad to

be

rest

rain

ed to

stop

he

r fro

m g

ettin

g ou

t of m

ovin

g ca

r.

23

£1,5

94.3

2 St

aff

Phys

ical

vio

lenc

e Pa

tient

lif

ting/

hand

ling

Nur

se in

jure

d ne

ck a

nd sh

ould

er d

urin

g pa

tient

rest

rain

t.

24

£120

.00

Patie

nt

Phys

ical

vio

lenc

e O

ther

A

41

year

old

det

aine

d pa

tient

trie

d to

bre

ak d

own

fire

door

with

a fi

re e

xtin

guis

her.

25

£5.4

8 Pa

tient

N

ear m

iss

Oth

er

An

80 y

ear o

ld p

atie

nt a

bsco

nded

from

psy

cho-

geria

tric

war

d.

26

£2.4

3 St

aff

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t St

aff N

urse

ass

aulte

d in

caf

é by

a m

an w

ho se

lf pr

esen

ted,

ask

ing

to se

e a

doct

or.

27

£10.

08

Staf

f A

ccid

enta

l inj

ury

Cut

with

shar

p m

ater

ial/o

bjec

t N

ursi

ng A

uxili

ary

drop

ped

plat

e w

hen

clea

ning

tabl

e af

ter p

atie

nt’s

lunc

h. T

he p

late

w

as b

roke

n. N

ursi

ng A

uxili

ary

cut f

inge

rs w

hen

pick

ing

up b

roke

n pi

eces

.

178

Tabl

e 9

Trus

t B (1

1 In

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £5

.50

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

N

urse

suff

ered

nee

dles

tick

whi

le a

ttend

ing

patie

nt a

t hom

e. P

atie

nt m

oved

whi

lst t

ryin

g to

inse

rt ne

edle

and

pun

ctur

ed m

iddl

e fin

ger o

f non

-dom

inan

t lef

t han

d th

at w

as

posi

tione

d on

top

of th

e pa

tient

’s a

rm to

stea

dy it

. 2

£1,5

39.7

8 St

aff

Acc

iden

tal

inju

ry

Man

ual l

iftin

g/ha

ndlin

g Se

nior

Den

tal M

anag

er tr

ansf

errin

g eq

uipm

ent o

ut o

f the

bac

k of

his

car

whi

lst o

n da

y vi

sits

stra

ined

bac

k.

3 £4

24.5

5 St

aff

Acc

iden

tal

inju

ry

Stru

ck b

y an

obj

ect

Spee

ch a

nd L

angu

age

Ther

apis

t hit

on h

ead

by lo

ose

proj

ecto

r scr

een.

4 £9

57.0

0 St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

St

aff N

urse

resp

ondi

ng to

em

erge

ncy

alar

m ra

n in

to c

orrid

or a

rea.

As n

urse

ran

alon

g co

rrid

or fa

iled

to se

e w

arni

ng si

gn a

nd sl

ippe

d on

wet

floo

r, su

stai

ning

a tw

iste

d an

kle.

5

£12.

76

Staf

f Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t O

ccup

atio

nal T

hera

py st

uden

t ass

aulte

d w

ith p

ool c

ue b

y cl

ient

with

his

tory

of

aggr

essi

ve b

ehav

iour

, alth

ough

ther

e w

as n

o w

arni

ng th

at th

e ou

tbur

st w

as a

bout

to ta

ke

plac

e. T

he te

nden

cy o

f the

clie

nt to

beh

ave

aggr

essi

vely

was

not

com

mun

icat

ed to

the

stud

ent O

T.

6 £7

,078

.50

Staf

f A

ccid

enta

l in

jury

Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Staf

f Nur

se w

ashi

ng h

er h

ands

in tr

eatm

ent r

oom

. W

ater

spill

age

durin

g ha

nd w

ashi

ng

led

to w

ater

on

floor

. N

urse

slip

ped

on fl

oor,

sust

aini

ng in

jury

to k

nee.

7

£3.7

9 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Enro

lled

Nur

se a

ttem

ptin

g to

dis

pose

of a

use

d sh

arp

follo

win

g ad

min

istra

tion

of in

sulin

. A

s she

trie

d to

ope

n th

e lid

to th

e sh

arps

box

it st

uck.

She

tran

sfer

red

the

need

le fr

om

her r

ight

han

d to

left

hand

so sh

e co

uld

open

the

lid w

ith h

er ri

ght h

and

and

in d

oing

so

sust

aine

d ne

edle

stic

k in

jury

to h

er le

ft ha

nd.

8 £1

3.16

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Whi

lst a

dmin

iste

ring

inje

ctio

n to

pat

ient

, nur

se re

ceiv

ed a

shar

ps in

jury

.

9 £1

92.5

0 St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

St

aff N

urse

fell

on st

eps r

espo

ndin

g to

em

erge

ncy

alar

m.

10

£1,0

38.9

8 St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/land

ling

Hea

lthca

re A

ssis

tant

atte

mpt

ed to

bre

ak a

pat

ient

’s fa

ll, su

stai

ning

a b

ack

inju

ry.

11

£5.6

0 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

GP

taki

ng b

lood

sam

ple

from

pat

ient

. O

n co

mpl

etio

n, p

asse

d us

ed n

eedl

e to

aux

iliar

y nu

rse

to d

ispo

se o

f in

shar

ps b

in.

She

caug

ht h

er 3

rd fi

nger

with

the

need

le.

179

Tabl

e 10

Tru

st C

(13

Inci

dent

s)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £1

08.7

5 St

aff

Acc

iden

tal

inju

ry

Hot

or c

old

cont

act

An

assi

stan

t coo

k w

as sc

alde

d by

soup

whe

n it

sudd

enly

bub

bled

up

and

spla

shed

on

to

her r

ight

han

d, sc

aldi

ng tw

o fin

gers

. 2

£9.7

4 St

aff

Acc

iden

tal

inju

ry

Stru

ck a

gain

st so

met

hing

(f

urni

ture

etc

) D

omes

tic w

as p

ullin

g tro

lley

from

din

ing

room

into

kitc

hen.

Whi

le g

oing

thro

ugh

door

way

she

caug

ht h

er ri

ght h

and

betw

een

the

door

fram

e an

d th

e tro

lley.

Thi

s res

ulte

d in

pai

n to

3rd

and

4th

fing

ers.

3 £4

00.0

0 St

aff

Nea

r mis

s C

onta

ct w

ith e

lect

ricity

C

harg

er fo

r den

tal t

rolle

y w

as n

ot w

orki

ng so

ano

ther

was

bor

row

ed fr

om su

rger

y ne

xt

door

. D

espi

te h

avin

g a

loos

e pi

n it

was

bei

ng ro

utin

ely

used

in th

e ot

her s

urge

ry.

Whe

n th

e de

ntal

nur

se p

lugg

ed in

the

char

ger t

he e

arth

pin

faile

d an

d th

e ch

arge

r ble

w o

ut o

f th

e so

cket

with

a lo

ud b

ang.

RC

D h

ad b

low

n an

d re

set.

4 £1

3.06

St

aff

Acc

iden

tal

inju

ry

Stru

ck b

y an

obj

ect

Secr

etar

y in

staf

f kitc

hen

open

ed a

cup

boar

d (u

pper

uni

t) do

or a

nd a

pla

te sl

ippe

d ou

t, st

rikin

g he

r on

the

uppe

r lip

. 5

£655

.84

Patie

nt

Acc

iden

tal

inju

ry

Fall

from

hei

ght

Nin

ety

year

old

pat

ient

fell

out o

f bed

.

6 £5

.80

Patie

nt

Acc

iden

tal

inju

ry

Fall

from

hei

ght

Stro

ke p

atie

nt o

verc

ame

cot s

ides

and

fell

to fl

oor.

7 £1

,684

.24

Staf

f A

ccid

enta

l in

jury

Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Nur

sing

Ass

ista

nt sl

ippe

d on

wet

car

pet,

sust

aini

ng a

hea

d in

jury

.

8 £9

8.40

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Staf

f Nur

se su

stai

ned

need

lest

ick

inju

ry w

hile

dis

posi

ng o

f lan

cet a

fter c

heck

ing

patie

nt’s

gl

ucos

e.

9 £3

28.8

8 St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/han

dlin

g En

rolle

d N

urse

sust

aine

d lo

wer

bac

k in

jury

ben

ding

to a

ssis

t pat

ient

who

had

bee

n si

tting

on

com

mod

e.

10

£10.

28

Patie

nt

Acc

iden

tal

inju

ry

Fall

from

hei

ght

Patie

nt fe

ll to

floo

r fro

m c

hair,

sust

aini

ng h

ead

inju

ry.

Ref

erre

d to

Acu

te T

rust

A&

E fo

r tre

atm

ent.

11

£6.9

9 Pa

tient

O

ther

O

ther

Pa

tient

cho

ked

on si

ngle

text

ure

mea

l (tu

na a

nd v

eg).

12

£9.3

1 Pa

tient

N

ear m

iss

Oth

er

Patie

nt lo

cked

in to

ilet a

fter l

ocki

ng d

oor,

and

coul

d no

t unl

ock

door

. 13

£2

60.4

5 St

aff

Acc

iden

tal

inju

ry

Man

ual l

iftin

g/ha

ndlin

g Po

rter i

njur

ed b

ack

pick

ing

up ta

ilgat

e of

trai

ler.

180

Tabl

e 11

Tru

st D

(34

Inci

dent

s)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £1

0.08

St

aff

Acc

iden

tal i

njur

y N

eedl

estic

k/sh

arps

D

omes

tic A

ssis

tant

was

mop

ping

the

floor

in c

asua

lty ro

om.

She

notic

ed a

pi

ece

of th

read

on

the

floor

and

pic

ked

it up

. Sh

e di

d no

t rea

lise

that

a su

ture

ne

edle

was

atta

ched

and

this

pie

rced

her

glo

ve a

nd h

er sk

in.

2

£91.

12

Staf

f A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

C

ook

slip

ped

on w

et fl

oor s

usta

inin

g ba

ck in

jury

. Th

e w

ater

may

hav

e co

me

from

a n

earb

y di

shw

ashi

ng a

rea.

3

£2.4

0 St

aff

Acc

iden

tal i

njur

y C

onta

ct w

ith

equi

pmen

t/mac

hine

ry

Coo

k su

stai

ned

cut f

inge

r try

ing

to re

mov

e m

etal

stor

age

bin

from

rack

.

4 £1

9.17

St

aff

Acc

iden

tal i

njur

y C

onta

ct w

ith

equi

pmen

t/mac

hine

ry

Ass

ista

nt c

ook

sust

aine

d bu

rn w

hils

t lig

htin

g de

ep fa

t fry

er.

5 £2

.80

Patie

nt

Acc

iden

tal i

njur

y St

ruck

aga

inst

som

ethi

ng

(fur

nitu

re e

tc)

Elde

rly c

onfu

sed

patie

nt b

eing

take

n ho

me

by b

us fr

om th

e da

y ho

spita

l. S

eat

belts

not

fitte

d to

the

bus.

Whe

n bu

s sto

pped

at t

raff

ic li

ghts

, pat

ient

stoo

d up

an

d as

bus

mov

ed o

ff a

gain

pat

ient

fell

back

war

ds, s

triki

ng th

eir h

ead

of th

e lif

t at

the

back

of t

he b

us.

6 £6

7.42

St

aff

Acc

iden

tal i

njur

y C

ut w

ith sh

arp

mat

eria

l/obj

ect

Nur

sing

Ass

ista

nt su

stai

ned

cut t

o fin

ger o

peni

ng ti

n.

7 £3

7.60

St

aff

Acc

iden

tal i

njur

y St

ruck

aga

inst

som

ethi

ng

(fur

nitu

re e

tc)

Dom

estic

Ass

ista

nt w

as v

acuu

min

g in

an

offic

e an

d th

e va

cuum

cle

aner

cau

ght

agai

nst a

fold

ing

tabl

e th

at w

as p

ropp

ed u

p ag

ains

t the

wal

l and

the

tabl

e fe

ll on

to th

e do

mes

tic c

ausi

ng in

jury

to h

er le

g an

d an

kle.

8

£2.3

4 St

aff

Acc

iden

tal i

njur

y C

onta

ct w

ith

equi

pmen

t/mac

hine

ry

Dom

estic

inju

red

thum

b ch

angi

ng b

rush

es o

n “r

otaw

ash”

.

9 £7

.76

Oth

er

Nea

r mis

s O

ther

W

ater

leak

into

off

ice

due

to ta

nk c

lean

ing

proc

ess.

10

£48.

56

Oth

er

Oth

er

Oth

er

Vis

itor d

rivin

g a

high

-topp

ed v

an a

ttem

pted

to d

rive

unde

r an

arch

way

lead

ing

to a

n in

ner c

ourty

ard

of th

e ho

spita

l. T

he v

ehic

le st

ruck

the

over

head

arc

hway

. 11

£8

.72

Patie

nt

Oth

er

Oth

er

A le

gally

det

aine

d pa

tient

abs

cond

ed to

a lo

cal b

ar.

Two

mem

bers

of s

taff

co

llect

ed h

im a

nd d

rove

him

bac

k in

the

hosp

ital c

ar.

12

£9.2

0 Pa

tient

O

ther

O

ther

Le

gally

det

aine

d pa

tient

abs

cond

s to

loca

l bar

aga

in.

13

£4.7

0 St

aff

Acc

iden

tal i

njur

y N

eedl

estic

k/sh

arps

D

omes

tic A

ssis

tant

was

in tr

eatm

ent r

oom

cle

anin

g th

e si

nk.

She

did

not

notic

e a

need

le st

uck

in th

e dr

ain

and

it pu

nctu

red

her r

ight

mid

dle

finge

r th

roug

h th

e gl

ove.

14

£9

.84

Patie

nt

Ver

bal a

buse

or

thre

aten

ing

Oth

er

A p

sych

otic

teen

ager

acc

esse

d a

roof

via

a b

roke

n w

indo

w in

the

bedr

oom

.

181

beha

viou

r 15

£2

7.40

St

aff

Acc

iden

tal i

njur

y N

eedl

estic

k/sh

arps

H

ealth

Vis

itor s

usta

ined

nee

dles

tick

inju

ry w

hen

dist

ract

ed b

y pa

tient

’s

sibl

ings

. 16

£2

,688

.00

Staf

f A

ccid

enta

l inj

ury

Patie

nt li

fting

/han

dlin

g N

ursi

ng A

uxili

ary

sust

aine

d el

bow

inju

ry u

sing

hyd

raul

ic h

oist

(ele

ctric

hoi

st

awai

ting

parts

for r

epai

rs).

17

£42.

00

Staf

f A

ccid

enta

l inj

ury

Patie

nt li

fting

/han

dlin

g N

ursi

ng A

uxili

ary

sust

aine

d ba

ck in

jury

ass

istin

g pa

tient

to u

ndre

ss.

Giv

en th

e hi

gh d

epen

denc

y le

vel o

f the

pat

ient

, a se

cond

nur

se sh

ould

hav

e be

en p

rese

nt

to a

ssis

t. 18

£7

98.0

0 St

aff

Acc

iden

tal i

njur

y Pa

tient

lifti

ng/h

andl

ing

Hea

lthca

re A

ssis

tant

sust

aine

d ne

ck in

jury

dur

ing

patie

nt h

oist

tran

sfer

. Pa

tient

be

cam

e un

coop

erat

ive

durin

g th

e tra

nsfe

r and

the

HA

inte

rven

ed to

supp

ort t

he

patie

nt to

pre

vent

him

from

falli

ng to

the

floor

. 19

£1

43.5

5 St

aff

Acc

iden

tal i

njur

y Pa

tient

lifti

ng/h

andl

ing

Patie

nt b

eing

ass

iste

d in

the

toile

t by

2 m

embe

rs o

f sta

ff.

Patie

nt b

ecam

e w

eak

at th

e kn

ees a

nd w

as su

ppor

ted

by a

mem

ber o

f sta

ff.

Whi

lst a

ttem

ptin

g to

br

eak

the

patie

nt’s

fall,

the

Reh

ab A

ssis

tant

twis

ted

her b

ack.

20

£1

27.6

0 St

aff

Acc

iden

tal i

njur

y Pa

tient

lifti

ng/h

andl

ing

Nur

sing

Ass

ista

nt su

stai

ned

back

inju

ry a

ssis

ting

a pa

tient

out

of t

he b

ath.

The

pa

tient

turn

ed a

nd sl

ippe

d ca

usin

g bo

th n

urse

s to

supp

ort h

is w

eigh

t. 21

£8

.52

Staf

f A

ccid

enta

l inj

ury

Nee

dles

tick/

shar

ps

Nur

sing

Ass

ista

nt sc

ratc

hed

by u

sed

razo

r bla

de, u

sed

by a

n el

derly

stro

ke

patie

nt w

ho w

as sh

avin

g hi

mse

lf. T

he p

atie

nt a

ttem

pted

to c

ontin

ue sh

avin

g as

th

e N

ursi

ng A

ssis

tant

was

dry

ing

his f

ace

with

a to

wel

.

22

£4.7

2 St

aff

Acc

iden

tal i

njur

y N

eedl

estic

k/sh

arps

H

otel

Ser

vice

s Ass

ista

nt su

stai

ned

need

lest

ick

inju

ry fr

om n

eedl

e in

bin

bag

rem

oved

from

pat

ient

’s ro

om.

GP

had

take

n bl

ood

from

the

patie

nt a

nd le

ft sy

ringe

and

nee

dle

on a

tabl

e. T

he p

atie

nt h

ad p

ut th

ese

into

the

bin.

23

£5

66.9

1 Pa

tient

O

ther

O

ther

V

acci

ne a

dmin

istra

tion

erro

r to

infa

nt d

urin

g im

mun

isat

ion

clin

ic.

Staf

f ab

senc

e du

e to

stre

ss o

ver t

he e

rror

in a

dmin

iste

ring

MM

R v

acci

ne in

stea

d of

an

othe

r. 24

£6

04.8

0 St

aff

Acc

iden

tal i

njur

y Pa

tient

lifti

ng/h

andl

ing

Nur

sing

Aux

iliar

y in

jure

d ba

ck w

hils

t fitt

ing

inco

ntin

ence

pad

on

patie

nt.

25

£3.7

9 Pa

tient

N

ear m

iss

Con

tact

with

/exp

osur

e to

ha

rmfu

l sub

stan

ces

Lear

ning

dis

abili

ties p

atie

nt c

over

ed th

emse

lves

with

pai

nt.

26

£1,7

12.1

5 St

aff

Acc

iden

tal i

njur

y Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Cle

rical

Offi

cer s

lippe

d on

pol

ishe

d flo

or.

27

£19.

70

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Psyc

hotic

pat

ient

pun

ched

Sta

ff N

urse

dur

ing

asse

ssm

ent i

nter

view

.

28

£3.5

9 Pa

tient

O

ther

O

ther

Pa

tient

abs

cond

ed fr

om h

ospi

tal g

roun

ds, r

efus

ed to

retu

rn w

ith st

aff.

Pol

ice

wer

e no

tifie

d an

d re

turn

ed th

e pa

tient

to th

e w

ard

one

hour

late

r. 29

£5

.12

Patie

nt

Oth

er

Oth

er

Det

aine

d pa

tient

abs

cond

ed fr

om fi

re e

scap

e do

or le

ft un

lock

ed.

30

£5

6.80

Pa

tient

A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

Patie

nt fe

ll in

cor

ridor

.

182

leve

l or s

tairs

31

£9

0.51

Pa

tient

A

ccid

enta

l inj

ury

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Pa

tient

fall

due

to Z

imm

er g

ettin

g ca

ught

on

cont

amin

ated

was

te b

in.

32

£134

.40

Staf

f A

ccid

enta

l inj

ury

Hot

or c

old

cont

act

Hea

lthca

re A

ssis

tant

scal

ded

hand

dur

ing

porr

idge

spill

age

follo

win

g se

rvin

g br

eakf

ast o

n th

e w

ards

from

the

trolle

y.

33

£33.

60

Staf

f A

ccid

enta

l inj

ury

Man

ual l

iftin

g/ha

ndlin

g N

ursi

ng A

uxili

ary

park

ing

a fo

od tr

olle

y ja

mm

ed th

eir f

inge

r bet

wee

n th

e w

all

and

the

trolle

y.

34

£317

.94

Staf

f Ill

hea

lth

Oth

er

Enro

lled

Nur

se c

ontra

cted

infe

ctio

n in

elb

ow fr

om sp

lash

of u

rine.

Tabl

e 12

Tru

st E

(16

Inci

dent

s)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £2

4.97

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Doc

tor s

usta

ined

nee

dles

tick

inju

ry st

retc

hing

to re

ach

shar

ps d

ispo

sal b

ox.

2 £1

8.70

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Nur

sing

Aux

iliar

y tid

ying

up

pick

ed u

p sh

arps

box

to c

heck

if it

was

full

and

sust

aine

d a

need

lest

ick

inju

ry fr

om n

eedl

e st

icki

ng o

ut to

p of

box

. 3

£2.8

9 St

aff

Acc

iden

tal

inju

ry

Stru

ck a

gain

st so

met

hing

(f

urni

ture

etc

) St

aff N

urse

retri

evin

g dr

essi

ng fr

om b

ase

unit

cupb

oard

hit

head

on

wor

ktop

as s

he

stoo

d up

. 4

£30.

12

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

A

n an

aest

hetis

t sus

tain

ed n

eedl

estic

k du

e to

a c

ollis

ion

with

a c

olle

ague

wor

king

in

clos

e pr

oxim

ity.

5 £1

9.18

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

A S

enio

r Hou

se O

ffic

er a

ssis

ting

durin

g su

turin

g, su

stai

ned

a ne

edle

stic

k, w

orki

ng in

cl

ose

prox

imity

to th

e su

rgeo

n w

ho w

as st

itchi

ng.

6 £3

4.00

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Con

sulta

nt su

turin

g sk

in in

EN

T th

eatre

sust

aine

d ne

edle

stic

k in

jury

to p

alm

of l

eft

hand

.7

£19.

18

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

Se

nior

Hou

se O

ffic

er a

ssis

ting

surg

eon

durin

g su

turin

g, re

ceiv

ed n

eedl

estic

k in

jury

.

8 £2

0.63

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Juni

or H

ouse

Off

icer

sust

aine

d sh

arps

inju

ry w

hils

t sw

abbi

ng w

ound

, ass

istin

g su

rgeo

n.

9 £9

62.5

2 Pa

tient

Fa

talit

y Sl

ip/tr

ip o

r fal

l on

the

sam

e le

vel o

r sta

irs

Con

fuse

d pa

tient

got

up

durin

g th

e ni

ght t

o vi

sit t

oile

t, an

d fe

ll by

her

bed

side

on

retu

rnin

g to

bed

. X

-ray

s mis

plac

ed, 9

day

s bef

ore

radi

olog

ist r

epor

ts o

n X

-ray

s and

re

ports

frac

ture

. 10

£6

.18

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

Inju

ry to

fing

er su

stai

ned

by st

aff n

urse

mov

ing

patie

nt u

p be

d.

11

£9.9

0 Pa

tient

Ill

hea

lth

Oth

er

Pres

sure

sore

s ide

ntifi

ed in

eld

erly

pat

ient

. N

o pr

essu

re re

lief m

easu

res.

183

12

£4.4

9 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Nur

sing

Aux

iliar

y cu

t thu

mb

whi

lst w

ashi

ng ta

ble

knife

.

13

£7.9

6 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Mid

wife

sust

aine

d su

ture

nee

dles

tick.

14

£9.0

8 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Nee

dle

left

on tr

eatm

ent c

ouch

in O

utpa

tient

s. N

ursi

ng A

uxili

ary

pulle

d th

read

and

ne

edle

spra

ng u

p an

d w

ent i

nto

the

Nur

sing

Sis

ter s

tand

ing

in th

e vi

cini

ty.

15

£20.

19

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

N

ursi

ng A

uxili

ary

sust

aine

d la

cera

tion

clea

ning

a k

nife

.

16

£17.

00

Staf

f A

ccid

enta

l in

jury

N

eedl

estic

k/sh

arps

St

aff N

urse

sust

aine

d ne

edle

stic

k du

ring

canu

latio

n of

col

leag

ue.

Tabl

e 13

Tru

st F

(con

trol)

(20

Inci

dent

s)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £2

,679

.84

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

Staf

f Nur

se su

stai

ned

back

inju

ry.

Lack

of s

lide

shee

t.

2 £1

9.03

St

aff

Phys

ical

vi

olen

ce

Pers

on to

per

son

assa

ult

Patie

nt sl

appe

d St

aff N

urse

in c

onfu

sed/

aggr

essi

ve o

utbu

rst.

3 £3

.82

Staf

f A

ccid

enta

l in

jury

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Tech

nici

an su

stai

ned

a ha

nd in

jury

atte

mpt

ing

to o

pen

a ja

mm

ed d

oor.

4 £2

57.2

6 St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

St

erile

Ser

vice

s Tec

hnic

ian

Supe

rvis

or fe

ll du

e to

uns

tabl

e ha

ndra

il.

5 £5

,480

.55

Patie

nt

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Tr

ansp

lant

pat

ient

trip

ped

in S

how

er R

oom

, sus

tain

ed a

frac

ture

d hi

p.

6 £8

56.9

2 St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

St

aff N

urse

sust

aine

d fr

actu

red

finge

r cha

ngin

g w

ater

can

iste

r on

wat

er c

oole

r mac

hine

.

7 £1

,932

.26

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

Clin

ical

Sup

port

Wor

ker s

usta

ined

shou

lder

inju

ry m

ovin

g pa

tient

up

bed.

8 £1

0.58

Pa

tient

O

ther

Fa

ll fr

om h

eigh

t B

aria

tric

patie

nt fa

ll. F

ire B

rigad

e re

quire

d to

lift

patie

nt b

ack

into

bed

. Pa

tient

lifte

d ba

ck to

bed

by

4 fir

emen

and

3 p

orte

rs w

ith th

e ai

d of

a sa

lvag

e sh

eet.

A h

oist

cap

able

of

lifti

ng th

e pa

tient

’s w

eigh

t cou

ld n

ot b

e lo

cate

d in

the

hosp

ital.

Wei

ght e

stim

ated

at

>200

kgs

. 9

£504

.05

Staf

f A

ccid

enta

l in

jury

Fa

ll fr

om h

eigh

t B

aria

tric

patie

nt fe

ll ou

t of b

ed.

Nur

se su

stai

ned

frac

ture

d w

rist.

6 F

irem

en a

nd 2

po

rters

man

ually

lifte

d th

e pa

tient

bac

k in

to b

ed.

184

10

£3.1

6 St

aff

Phys

ical

vi

olen

ce

Pers

on to

per

son

assa

ult

Agg

ress

ive

outb

urst

from

pat

ient

suff

erin

g fr

om a

lcoh

ol w

ithdr

awal

.

11

£5.6

0 St

aff

Phys

ical

vi

olen

ce

Pers

on to

per

son

assa

ult

Ban

k N

ursi

ng A

uxili

ary

assa

ulte

d by

pat

ient

suff

erin

g fr

om a

lcoh

ol w

ithdr

awal

.

12

£3.2

5 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Nur

se su

stai

ned

need

lest

ick

inju

ry fr

om u

sed

lanc

et.

13

£14.

20

Staf

f A

ccid

enta

l in

jury

M

anua

l lift

ing/

hand

ling

Clin

ical

Sup

port

Wor

ker s

usta

ined

bac

k in

jury

pul

ling

out b

ack

rest

to m

ake

patie

nt

mor

e co

mfo

rtabl

e du

ring

the

nigh

t. 14

£5

.95

Patie

nt

Oth

er

Oth

er

Con

fuse

d pa

tient

goe

s mis

sing

from

med

ical

war

d.

15

£164

.70

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/h

andl

ing

Trai

nee

Clin

ical

Sup

port

Wor

ker s

usta

ined

bac

k in

jury

dur

ing

patie

nt tr

ansf

er.

16

£735

.90

Patie

nt

Fata

lity

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

C

onfu

sed

patie

nt su

stai

ned

frac

ture

d hi

p in

cor

ridor

of w

aitin

g ar

ea.

Follo

win

g su

rger

y to

the

frac

ture

, pat

ient

did

not

reco

ver f

rom

ope

ratio

n.

17

£4.9

2 St

aff

Phys

ical

vi

olen

ce

Pers

on to

per

son

assa

ult

Nur

se su

stai

ned

shou

lder

inju

ry d

urin

g m

ovin

g an

d ha

ndlin

g of

agg

ress

ive

patie

nt.

18

£5.7

1 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Staf

f Nur

se su

stai

ned

need

lest

ick

inju

ry fr

om u

sed

need

le in

foil

tray.

19

£2.3

5 St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Trai

nee

Clin

ical

Sup

port

Wor

ker s

usta

ined

nee

dles

tick

inju

ry fr

om la

ncet

dis

card

ed o

n flo

or.

20

£43.

85

Staf

f A

ccid

enta

l in

jury

St

ruck

by

an o

bjec

t St

aff N

urse

sust

aine

d he

ad in

jury

from

falli

ng d

rip st

and.

Tabl

e 14

Tru

st G

(con

trol)

(10

Inci

dent

s)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £4

4.71

Pa

tient

A

ccid

enta

l in

jury

Fa

ll fr

om h

eigh

t Pa

tient

fell

out o

f bed

sust

aini

ng fr

actu

res.

Tra

nsfe

rred

to A

cute

Tru

st fo

r tre

atm

ent.

2 £7

5.20

St

aff

Acc

iden

tal

inju

ry

Slip

/trip

or f

all o

n th

e sa

me

leve

l or s

tairs

Po

rter s

lippe

d on

dar

k st

eps.

3 £1

9.60

St

aff

Acc

iden

tal

inju

ry

Cut

with

shar

p m

ater

ial/o

bjec

t In

the

Fore

nsic

Psy

chia

tric

War

d a

knife

had

gon

e m

issi

ng fo

llow

ing

was

hing

and

N

ursi

ng A

ssis

tant

thou

ght i

t had

falle

n un

der t

he d

ishw

ashe

r. P

ut h

is fi

nger

s und

er th

e m

achi

ne to

feel

if th

e kn

ife w

as th

ere.

Whi

lst d

oing

this

, sus

tain

ed a

cut

fing

er o

n a

shar

p lip

. 4

£116

.00

Patie

nt

Oth

er

Oth

er

Patie

nt a

bsco

nded

thro

ugh

a w

indo

w.

185

5 £1

2.13

St

aff

Acc

iden

tal

inju

ry

Nee

dles

tick/

shar

ps

Staf

f Nur

se su

stai

ned

a ne

edle

stic

k in

jury

dis

posi

ng o

f nee

dle

in fu

ll sh

arps

bin

.

6 £7

3.80

St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/han

dlin

g St

aff N

urse

on

light

dut

ies d

etai

led

to e

scor

t pat

ient

for X

-ray

. O

n tra

nsfe

rrin

g pa

tient

, St

aff N

urse

stra

ined

her

wea

ker a

rm.

7 £2

.46

Patie

nt

Oth

er

Oth

er

Patie

nt is

a sm

oker

and

hab

itual

ly le

aves

war

d to

do

this

. St

aff h

ave

to se

arch

for

patie

nt w

ho is

forg

etfu

l and

con

fuse

d, b

eing

trea

ted

for a

lcoh

ol a

buse

. 8

£37.

00

Patie

nt

Oth

er

Oth

er

Abs

cond

ed p

atie

nt fo

und

in ri

ver.

9 £1

,469

.46

Staf

f Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t Fr

actu

red

wris

t app

lyin

g co

ntro

l and

rest

rain

t.

10

£63.

72

Patie

nt

Phys

ical

vi

olen

ce

Oth

er

Patie

nt re

stra

ined

on

grou

nds a

wai

ting

doct

or a

sses

smen

t.

186

APPE

ND

IX 8

SPE

CIA

L C

ATEG

OR

Y IN

CID

ENT

DES

CR

IPTI

VE D

ATA

PHA

SE 2

Tabl

e 1

Trus

t A (2

1inc

iden

ts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £2

7.94

St

aff

Oth

er

Oth

er

Car

was

bei

ng d

riven

by

a C

omm

unity

Psy

chia

tric

Nur

se (C

PN) o

n a

road

, whi

ch w

as

cove

red

in sn

ow, b

ut h

ad re

cent

ly b

een

salte

d. C

PN a

ttem

pted

to tu

rn c

ar b

ut d

idn’

t re

spon

d. B

rake

s wor

ked

but t

he ty

res d

id n

ot g

rip a

nd c

ar w

ent o

ff th

e ro

ad.

Spec

ial

cate

gory

: out

with

rea

sona

ble

prac

ticab

ility

– T

rust

has

som

e co

ntro

l of d

rivin

g ab

ility

of

staf

f but

not

ove

r roa

d co

nditi

ons.

2 £7

.26

Patie

nt

Van

dalis

m

Oth

er

Thirt

y-tw

o ye

ar o

ld m

ale

patie

nt w

ith le

arni

ng d

isab

ilitie

s act

ivat

ed fi

re a

larm

out

side

his

ro

om –

no

fire.

Pat

ient

was

und

er g

ener

al o

bser

vatio

ns a

t the

tim

e. S

peci

al c

ateg

ory:

ou

twith

rea

sona

ble

prac

ticab

ility

– u

nder

gen

eral

obs

erva

tion

at ti

me,

whi

ch w

as

reas

onab

le, a

s the

pat

ient

had

not

bee

n pr

oble

mat

ic.

3 £7

.26

Patie

nt

Van

dalis

m

Oth

er

Patie

nt w

ith le

arni

ng d

isab

ilitie

s act

ivat

ed fi

re a

larm

for t

he se

cond

tim

e in

a d

ay.

Spec

ial

cate

gory

: out

with

rea

sona

ble

prac

ticab

ility

– O

nly

2 in

cide

nts o

f thi

s nat

ure

had

prev

ious

ly o

ccur

red

over

the

prev

ious

12

wee

k pe

riod.

Not

dee

med

by

Trus

t to

be

suff

icie

nt g

roun

ds to

war

rant

cos

ts o

f pur

chas

ing

alar

m c

over

s in

this

are

a.

4 £1

3.96

Pa

tient

O

ther

O

ther

Fi

fty-th

ree

year

old

fem

ale

patie

nt w

ith le

arni

ng d

isab

ilitie

s lef

t the

war

d an

d ho

spita

l gr

ound

s. P

olic

e w

ere

info

rmed

. St

aff s

earc

hed

for h

er in

car

, whe

n fo

und

she

initi

ally

re

fuse

d to

retu

rn b

ut w

as p

ersu

aded

. Sp

ecia

l cat

egor

y: C

linic

al ju

dgem

ent –

Clin

icia

n ju

dged

the

patie

nt’s

con

ditio

n w

ith re

fere

nce

to th

e de

gree

of s

uper

visi

on a

nd re

stric

tion

of

free

dom

. 5

£351

.63

Patie

nt

Van

dalis

m

Expo

sure

to fi

re

Patie

nt se

t fire

to a

shee

t in

show

er a

rea.

Als

o se

t fire

to sh

eet a

nd m

attre

ss o

n be

d.

Patie

nt h

ad h

er o

wn

light

er a

nd c

igar

ette

s – u

sed

light

er to

star

t fire

. Pa

tient

had

his

tory

of

self-

harm

but

no

hist

ory

of fi

re ra

isin

g. S

peci

al c

ateg

ory:

ris

k un

fore

seea

ble

– N

o hi

stor

y of

fire

rais

ing.

187

6 £3

,886

.48

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/

hand

ling

Nur

sing

Aux

iliar

y (N

A) w

as p

repa

ring

a re

side

nt fo

r bed

. Th

e cl

othi

ng th

e pa

tient

was

w

earin

g w

as ti

ght a

roun

d th

e ne

ck c

ausi

ng d

iffic

ulty

get

ting

it ov

er th

e pa

tient

’s h

ead.

Th

is c

ause

d th

e pa

tient

som

e di

stre

ss a

nd h

e su

dden

ly je

rked

cau

sing

the

NA

to su

stai

n a

brok

en fi

nger

. N

A w

as a

bsen

t fro

m w

ork

with

the

inju

ry.

Spec

ial c

ateg

ory:

lack

of

info

rmat

ion

– N

A in

pro

cess

of c

ivil

clai

m a

gain

st T

rust

so d

eclin

ed to

pro

vide

furth

er

info

rmat

ion

to p

roje

ct re

sear

cher

s. 7

£20.

00

Patie

nt

Acc

iden

tal

inju

ry

Stru

ck a

gain

st

som

ethi

ng

(fur

nitu

re e

tc)

Sixt

y-fiv

e ye

ar-o

ld m

ale

patie

nt st

umbl

ed a

nd fe

ll ba

ckw

ards

bre

akin

g th

e sa

fety

gla

ss in

th

e do

or.

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– p

atie

nt in

depe

nden

tly

mob

ile, n

o hi

stor

y of

falls

, alre

ady

unde

r gen

eral

obs

erva

tion.

8

£42.

00

Oth

er

Van

dalis

m

Oth

er

Bro

ken

win

dow

s and

forc

ed w

indo

ws i

n su

mm

erho

use

notic

ed b

y si

te p

orte

r. E

stat

es

depa

rtmen

t boa

rded

up

and

secu

red

sum

mer

hous

e. G

lass

not

repl

aced

as h

ospi

tal d

ue to

be

rebu

ilt la

ter i

n ye

ar.

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– a

s site

was

be

ing

dem

olis

hed

and

rede

velo

ped,

dec

isio

n w

as m

ade

to w

ork

with

pol

ice

to re

duce

risk

s by

pro

mpt

ly c

lean

ing

up a

nd se

curin

g ra

ther

than

incr

ease

secu

rity

sinc

e a

cost

/ben

efit

anal

ysis

had

sugg

este

d th

is w

as n

ot w

orth

whi

le.

9 £3

1.50

Pa

tient

Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t Tw

o pa

tient

s on

acut

e ps

ychi

atric

war

d. F

emal

e pa

tient

thou

ght m

ale

patie

nt w

as la

ughi

ng

at h

er.

She

then

pro

ceed

ed to

atta

ck m

ale

patie

nt ri

ppin

g hi

s shi

rt in

the

proc

ess.

Spe

cial

cate

gory

: out

with

rea

sona

ble

prac

ticab

ility

– a

dequ

ate

cont

rol m

easu

res i

n pl

ace.

10

£5

.80

Patie

nt

Oth

er

Oth

er

Eigh

teen

-yea

r-ol

d fe

mal

e pa

tient

with

susp

ecte

d pa

race

tam

ol o

verd

ose.

Pat

ient

agr

eed

she

had

take

n ov

erdo

se a

nd b

lood

sam

ple

take

n to

labs

by

taxi

for a

naly

sis.

Spe

cial

cat

egor

y:

Clin

ical

judg

emen

t – st

aff h

ad b

een

cons

ider

ing

her d

isch

arge

and

she

was

bei

ng ta

ught

co

ping

stra

tegi

es a

nd h

ad b

een

allo

wed

out

for a

wal

k, a

s par

t of t

his r

ehab

ilita

tion.

11

£1

4.76

Pa

tient

Ph

ysic

al

viol

ence

Pe

rson

to p

erso

n as

saul

t Tw

o m

ale

patie

nts s

harin

g a

flat i

n a

reha

bilit

atio

n un

it. P

atie

nt A

bec

ame

anno

yed

by

Patie

nt B

and

rais

ed th

is w

ith h

im.

Patie

nt B

then

ges

ture

d an

d sw

ore

at P

atie

nt A

mak

ing

him

agi

tate

d an

d ag

gres

sive

. Pa

tient

A th

en lu

nged

at P

atie

nt B

hitt

ing

him

acr

oss t

he fa

ce

and

knoc

king

his

spec

tacl

es o

ff.

Inci

dent

unw

itnes

sed

by st

aff.

Spe

cial

cat

egor

y:

Clin

ical

judg

emen

t – n

ot h

eavi

ly su

perv

ised

, as t

his w

as a

reha

b. u

nit.

No

prev

ious

kn

owle

dge

of p

robl

em b

etw

een

patie

nts.

12

£14.

76

Patie

nt

Oth

er

Oth

er

Thirt

y-on

e-ye

ar-o

ld m

enta

l hea

lth p

atie

nt fo

und

to h

ave

take

n ov

erdo

se o

f par

acet

amol

by

nigh

t-dut

y nu

rse.

Pat

ient

had

his

tory

of s

elf-

harm

and

alc

ohol

abu

se.

Spec

ial c

ateg

ory:

C

linic

al ju

dgem

ent –

risk

ass

essm

ent i

ndic

ated

low

risk

. Th

e pa

tient

seem

ed w

ell t

hat

day

and

was

ther

efor

e on

ly u

nder

gen

eral

obs

erva

tion.

13

£6

.27

Patie

nt

Oth

er

Oth

er

Sixt

y-th

ree

year

old

mal

e pa

tient

adm

itted

2 d

ays p

revi

ousl

y fo

und

to h

ave

mul

tiple

la

cera

tions

infli

cted

with

his

ow

n ra

zor.

Spe

cial

cat

egor

y: r

isk

unfo

rese

eabl

e –

asse

ssed

as

not

bei

ng su

icid

al a

nd fa

mily

repo

rted

no h

isto

ry o

f sel

f har

m.

188

14

£4.2

0 Pa

tient

O

ther

O

ther

Pa

tient

det

aine

d un

der M

enta

l Hea

lth A

ct (M

HA

) abs

cond

ed w

hils

t on

esco

rted

wal

k.

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– a

ll pr

oced

ures

follo

wed

, pho

ned

war

d im

med

iate

ly.

15

£4.0

2 O

ther

Fi

re

Oth

er

Patie

nt w

as sm

okin

g ci

gare

tte in

his

bed

room

, whi

ch tr

igge

red

the

fire

alar

m.

Patie

nt w

as

rem

inde

d ab

out w

ard

smok

ing

polic

y. S

peci

al c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– c

lient

men

tally

ill b

ut u

nder

stoo

d he

was

not

allo

wed

to sm

oke

in ro

om.

At n

ight

hou

rly c

heck

s wer

e do

ne b

ut p

atie

nt o

ther

wis

e ou

t of s

ight

dur

ing

nigh

t. 16

£1

1,47

7.71

Pa

tient

A

ccid

enta

l in

jury

Sl

ip, t

rip o

r fal

l on

leve

l or s

tairs

N

inet

y-on

e ye

ar o

ld fe

mal

e pa

tient

fell

retu

rnin

g to

bed

sust

aini

ng a

bro

ken

hip.

Pat

ient

ha

d an

ext

ende

d st

ay o

f six

ty-th

ree

days

. Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

no

hist

ory

of fa

lls, w

as to

ld a

bout

usi

ng b

uzze

r whe

n ne

edin

g as

sist

ance

bu

t pat

ient

did

not

use

it.

17

£6.2

5 Pa

tient

V

anda

lism

O

ther

Pa

tient

not

ified

staf

f he

witn

esse

d an

othe

r pat

ient

bre

akin

g th

e gl

ass o

n th

e fir

e al

arm

, ac

tivat

ing

it. P

atie

nt w

as d

istu

rbed

at t

he ti

me

and

psyc

hotic

/ de

lusi

onal

. Sp

ecia

lca

tego

ry: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

. Occ

urre

nce

of o

nly

2 in

cide

nts o

ver t

he

prev

ious

12

wee

k pe

riod

not d

eem

ed to

be

suff

icie

nt to

war

rant

cos

ts o

f pur

chas

ing

alar

m

cove

rs in

this

are

a.

18

£6.0

0 Pa

tient

O

ther

O

ther

Pa

tient

cam

e ba

ck to

war

d af

ter h

avin

g a

pass

allo

win

g he

r to

be o

ut.

War

d st

aff w

ere

enga

ged

for a

n ho

ur in

war

d re

port.

Pat

ient

info

rmed

staf

f hal

f an

hour

afte

r ret

urn

to w

ard

that

she

had

take

n an

ove

rdos

e of

med

icat

ion

and

para

ceta

mol

she

had

boug

ht w

hen

out o

f w

ard.

Cos

t for

blo

ods t

o be

take

n by

taxi

to la

bs fo

r ana

lysi

s. S

peci

al c

ateg

ory:

Clin

ical

ju

dgem

ent –

the

risk

asse

ssm

ent i

ndic

ated

that

the

patie

nt w

as su

itabl

e fo

r lea

ve a

nd

disc

harg

e w

as b

eing

con

side

red.

19

£2

.91

Patie

nt

Oth

er

Oth

er

Fifty

-two

year

old

pat

ient

with

Alz

heim

er’s

Dis

ease

. Pa

tient

act

ivat

ed fi

re a

larm

. Sp

ecia

lca

tego

ry: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

pat

ient

und

er g

ener

al o

bser

vatio

n.

Diff

icul

t to

cont

rol s

uch

beha

viou

r but

fire

ala

rm n

eces

sary

. 20

£5

0.00

Pa

tient

Ph

ysic

al

viol

ence

O

ther

Pa

tient

, ver

y el

ated

, kic

ked

door

off

hin

ges i

n be

droo

m.

Spec

ial c

ateg

ory:

out

with

re

ason

able

pra

ctic

abili

ty –

pol

icy

was

to st

and

back

as l

ong

as n

o ris

k to

pat

ient

/oth

ers.

Pa

tient

in e

xcita

ble

stat

e, la

ngua

ge p

oor,

reas

onin

g di

ffic

ult,

ther

efor

e ap

prop

riate

act

ion

was

take

n.

21

£4.8

4 Pa

tient

O

ther

O

ther

45

-yea

r-ol

d pa

tient

det

aine

d un

der M

HA

on

clos

e 30

min

ute

obse

rvat

ions

atte

mpt

ed

suic

ide

from

cur

tain

rail.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

m

easu

res i

n pl

ace

adeq

uate

and

all

follo

wed

.

189

Tabl

e 2

Trus

t B (5

inci

dent

s)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £6

.53

Staf

f A

ccid

enta

l in

jury

Sl

ip, t

rip o

r fa

ll on

leve

l or

stai

rs

Mem

ber o

f adm

inis

tratio

n st

aff t

rippe

d on

the

stai

rs su

stai

ning

a c

ut to

the

right

kne

e. S

peci

alca

tego

ry: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

con

ditio

n on

step

s goo

d, li

ghtin

g ad

equa

te,

non-

slip

lam

inat

ed fl

oorin

g. S

he sl

ippe

d on

met

al e

dgin

g of

stai

r, an

d no

oth

er c

ontri

buto

ry

fact

or c

ould

be

iden

tifie

d 2

£60.

50

Patie

nt

Phys

ical

vi

olen

ce

Pers

on to

pe

rson

ass

ault

Whe

n as

ked

to ti

dy ro

om p

atie

nt g

ot a

ngry

and

atta

cked

mem

ber o

f sta

ff.

Two

mem

bers

of

staf

f res

pond

ed to

pan

ic a

larm

and

pat

ient

was

con

trolle

d us

ing

cont

rol &

rest

rain

t (C

&R

) te

chni

ques

. D

urin

g th

e ph

ysic

al in

terv

entio

n th

e cl

ient

dam

aged

the

ribs o

f one

of t

he m

embe

rs

of st

aff.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

all

prot

ocol

s wer

e ad

here

d to

an

d cl

ient

was

trea

ted

acco

rdin

g to

pol

icy.

3

£16.

50

Staf

f A

ccid

enta

l in

jury

Pe

rson

to

pers

on a

ssau

lt Pa

tient

in w

ard

beca

me

viol

ent t

owar

ds st

aff.

The

pat

ient

had

to b

e re

stra

ined

, whi

lst

perf

orm

ing

this

a st

aff n

urse

sust

aine

d a

wris

t inj

ury.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

risk

ass

essm

ent i

n pl

ace,

all

staf

f tra

ined

to a

ppro

pria

te st

anda

rd.

4 £1

9.68

St

aff

Phys

ical

vi

olen

ce

Pers

on to

pe

rson

ass

ault

Patie

nt in

war

d be

cam

e vi

olen

t and

ass

aulte

d a

mem

ber o

f sta

ff.

Ano

ther

nur

se re

spon

ded

and

unde

rtook

C&

R to

con

trol t

he p

atie

nt.

Whi

lst d

oing

this

the

nurs

e su

stai

ned

an in

jury

to h

er

finge

r and

requ

ired

treat

men

t at A

&E.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

ris

k as

sess

men

t had

bee

n un

derta

ken,

ala

rm a

ctiv

ated

as p

er p

roto

col a

nd a

ll st

aff w

ere

train

ed

to th

e in

-hou

se st

anda

rd.

5 £1

53.1

2 St

aff

Phys

ical

vi

olen

ce

Pers

on to

pe

rson

ass

ault

Patie

nt in

war

d be

cam

e ho

stile

and

phy

sica

lly a

ggre

ssiv

e, h

avin

g to

be

rest

rain

ed.

Whi

lst

Nur

sing

Ass

ista

nt w

as c

arry

ing

out C

&R

man

oeuv

re p

atie

nt m

anag

ed to

elb

ow h

er in

jurin

g rib

ca

ge o

n rig

ht si

de.

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– a

ll th

ree

staf

f tra

ined

in C

&R

. A

lthou

gh p

atie

nt h

ad a

his

tory

of v

iole

nce,

an

appr

opria

te ri

sk a

sses

smen

t had

be

en c

arrie

d ou

t and

all

othe

r pro

cedu

res w

ere

follo

wed

app

ropr

iate

ly.

190

Tabl

e 3

Trus

t C (1

3 in

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £1

,537

.20

Staf

f A

ccid

enta

l in

jury

Pa

tient

lifti

ng/

hand

ling

Nur

sing

Ass

ista

nt c

ompl

aine

d sh

e ha

d hu

rt he

r bac

k w

hils

t lea

ning

ove

r bed

tend

ing

to

patie

nt, r

esul

ting

in si

ck le

ave.

Inc

iden

t unw

itnes

sed.

Spe

cial

cat

egor

y: la

ck o

f in

form

atio

n –

No

witn

esse

s to

inci

dent

and

cha

rge

nurs

e di

dn’t

ask

for f

urth

er in

fo w

hen

NA

repo

rted

in si

ck.

2 £7

8.93

St

aff

Acc

iden

tal

inju

ry

Patie

nt li

fting

/ ha

ndlin

g En

rolle

d nu

rse

settl

ing

a pa

tient

for t

he n

ight

was

adj

ustin

g th

e pa

tient

’s p

illow

whe

n sh

e fe

lt a

sudd

en sh

arp

pain

in h

er m

id to

upp

er b

ack.

Spe

cial

cat

egor

y: la

ck o

f inf

orm

atio

n –

nigh

t sis

ter c

ould

not

pro

vide

any

mor

e in

form

atio

n. E

N h

ad le

ft th

e Tr

ust.

3 £6

0.00

Pa

tient

V

anda

lism

O

ther

Pa

tient

bec

ame

aggr

essi

ve a

nd sm

ashe

d a

stoo

l off

a w

indo

w, c

rack

ing

inne

r pan

e of

w

indo

w.

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– th

is in

cide

nt w

as c

ause

d by

frus

tratio

n on

the

part

of th

e pa

tient

. Diff

icul

t to

prev

ent w

hen

patie

nts h

ave

own

room

s. 4

£6.8

0 Pa

tient

Ph

ysic

al

viol

ence

St

ruck

aga

inst

so

met

hing

Tw

enty

-sev

en y

ear o

ld m

ale

patie

nt w

as to

ld h

e w

as to

be

deta

ined

und

er th

e M

HA

and

be

cam

e an

gry;

he

punc

hed

a w

all a

nd a

win

dow

(win

dow

did

n’t b

reak

). S

peci

al c

ateg

ory:

C

linic

al ju

dgem

ent –

dec

isio

n ta

ken

to le

t him

ven

t his

frus

tratio

n ra

ther

than

inte

rven

e ex

cept

as a

last

reso

rt.

5 £1

4.37

Pa

tient

O

ther

O

ther

Pa

tient

in M

enta

l Hea

lth A

cute

war

d ab

scon

ded.

Sta

ff N

urse

& N

A sp

ent o

ne h

our l

ooki

ng

for t

he p

atie

nt.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – a

form

al se

arch

pro

cedu

re e

xist

s bu

t was

app

lied

to so

me

patie

nts a

nd n

ot to

oth

ers,

base

d on

clin

ical

judg

emen

t of t

he st

aff.

6 £8

.69

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

Patie

nt w

ho w

as in

depe

nden

tly m

obile

with

his

rola

tor f

ell b

ackw

ards

in c

orrid

or fa

lling

to

floor

. Pa

tient

sust

aine

d a

frac

ture

to h

is sh

ould

er.

Spec

ial c

ateg

ory:

risk

unf

ores

eeab

le –

ris

k as

sess

men

t was

app

ropr

iate

. Th

ere

wer

e no

sign

s tha

t the

pat

ient

was

uns

tead

y, it

had

ne

ver h

appe

ned

befo

re.

7 £7

.39

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

Patie

nt g

ot o

ut o

f bed

and

fell

to th

e flo

or su

stai

ning

a la

cera

tion

to h

is h

ead,

hae

mat

oma

to

head

and

com

plai

ned

of p

ain

to h

is ri

ght h

ip.

X-r

ay c

onfir

med

frac

ture

to ri

ght n

eck

of

fem

ur.

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– a

ll th

at c

ould

be

done

had

be

en im

plem

ente

d.

8 £3

5.00

Pa

tient

V

anda

lism

O

ther

El

derly

pat

ient

with

dem

entia

bec

ame

frus

trate

d an

d w

ante

d to

go

hom

e. S

taff

hea

rd

bang

ing

in th

e w

ard.

Whe

n N

A a

ttend

ed, p

atie

nt w

as st

rikin

g a

smal

l win

dow

pane

with

her

zi

mm

er, c

ausi

ng it

to c

rack

. Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

eve

nts

appe

ared

to b

e as

soci

ated

with

pat

ient

’s m

enta

l/beh

avio

ural

pro

blem

s and

staf

f too

k pr

ompt

ac

tion.

191

9 £1

77.4

3 Pa

tient

N

ear m

iss

Oth

er

90 y

ear o

ld p

atie

nt o

n a

soft

food

die

t was

bei

ng in

trodu

ced

to m

ore

text

ured

food

s. W

hile

th

e pa

tient

was

eat

ing

her l

unch

she

chok

ed o

n he

r foo

d. P

atie

nt h

ad to

be

take

n to

A&

E fo

r tre

atm

ent,

as a

irway

cou

ld n

ot b

e cl

eare

d in

war

d. S

peci

al c

ateg

ory:

Clin

ical

judg

emen

t –

clin

ical

dec

isio

n m

ade

to in

trodu

ce m

ore

text

ured

food

s to

diet

. 10

£6

0.00

O

ther

V

anda

lism

O

ther

Pr

actic

e m

anag

er re

porte

d th

at fi

ve w

indo

ws h

ad b

een

brok

en.

Polic

e an

d es

tate

s off

icer

at

tend

ed.

Roo

ms c

heck

ed a

nd g

lazi

er fi

tted

new

win

dow

s. S

peci

al c

ateg

ory:

out

with

re

ason

able

pra

ctic

abili

ty –

ade

quat

e se

curit

y m

easu

res i

n pl

ace

(i.e.

ala

rm sy

stem

and

se

curit

y ca

mer

as).

11

£30.

87

Oth

er

Oth

er

Oth

er

Dis

able

d to

ilet i

n H

ealth

Cen

tre so

iled

with

exc

rem

ent.

So

bad

that

one

NA

refu

sed

clea

n up

mes

s. S

peci

al c

ateg

ory:

risk

unf

ores

eeab

le –

firs

t tim

e th

is h

as h

appe

ned

on th

is si

te.

12

£36.

11

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

One

pat

ient

trie

d to

ass

ist a

noth

er p

atie

nt w

ho w

as fa

lling

. B

oth

the

patie

nts’

zim

mer

s be

com

e en

tang

led,

resu

lting

in b

oth

patie

nts f

allin

g. P

atie

nt w

ho w

as a

ssis

ting

sust

aine

d a

frac

ture

d fe

mur

and

lace

ratio

ns to

righ

t han

d. S

peci

al c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– ri

sk a

sses

smen

ts a

nd c

are

plan

s wer

e in

pla

ce fo

r bot

h pa

tient

s. A

dequ

ate

mea

sure

s in

plac

e.

13

£2,5

10

Patie

nt

Phys

ical

vi

olen

ce

Oth

er

Var

ious

epi

sode

s ove

r 12-

wee

k pe

riod,

of p

atie

nt w

ho w

as e

asily

ups

et, t

hrow

ing

and

brea

king

item

s. S

trate

gy w

as to

let h

im th

row

the

item

s the

n ca

lm d

own

rath

er th

en

inte

rven

e. S

peci

al c

ateg

ory:

Clin

ical

judg

emen

t – st

rate

gy to

let p

atie

nt th

row

item

s ra

ther

than

inte

rven

e.

Tabl

e 4

Trus

t D (9

1 in

cide

nts)

No

Cost

Staf

f/pat

ient

/ ot

her

Cate

gory

Ty

pe

Des

crip

tion

1 £1

,796

.74

Oth

er

Nea

r mis

s O

ther

C

lient

, liv

ing

at h

ome,

had

tem

pora

ry lo

an o

f airb

ed w

hils

t ow

n w

as in

for r

epai

r.

Req

uest

ed te

mpo

rary

bed

was

left

perm

anen

tly a

s it w

as q

uiet

er th

an h

is o

ld o

ne.

This

was

dec

lined

. Sm

ell o

f bur

ning

repo

rted

from

uni

t on

its fi

rst n

ight

afte

r re

turn

from

repa

ir. T

empo

rary

bed

bro

ught

bac

k un

til n

ew re

plac

emen

t de

liver

ed.

Spec

ial c

ateg

ory:

Lac

k of

info

rmat

ion

– no

evi

denc

e to

supp

ort

susp

icio

n of

crim

inal

dam

age.

2

£54.

31

Patie

nt

Acc

iden

tal i

njur

y Fa

ll fr

om h

eigh

t In

depe

nden

t 89

year

old

mal

e co

mpl

eted

was

hing

han

ds in

toile

t, tu

rned

roun

d to

re

ach

for p

aper

tow

el a

nd fe

ll. A

n ag

ency

nur

se w

as su

perv

isin

g pa

tient

. Sp

ecia

lca

tego

ry: C

linic

al ju

dgem

ent –

NA

cou

ld d

o no

thin

g to

pre

vent

fall

as it

ha

ppen

ed su

dden

ly.

192

3 £1

1.20

Pa

tient

O

ther

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Patie

nt fo

und

to b

e st

agge

ring

on w

ard,

smel

ling

of a

lcoh

ol.

Lock

ed h

imse

lf in

th

e to

ilet.

Re-

appe

ared

hav

ing

cut r

ight

fore

arm

and

sayi

ng th

at h

e ha

d sw

allo

wed

the

blad

e (n

ot b

elie

ved)

. Es

corte

d to

A&

E, b

ecam

e ag

gres

sive

and

A

&E

refu

sed

to tr

eat.

Spe

cial

cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

staf

f ab

ided

by

drug

& a

lcoh

ol p

olic

y.

4 £6

0.20

St

aff

Phys

ical

vio

lenc

e O

ther

Pa

tient

kic

ked

dow

n pa

rtitio

n do

or in

to fe

mal

e en

d of

war

d (u

nder

rede

cora

tion)

. A

rmed

with

pol

e th

reat

ened

staf

f. A

larm

act

ivat

ed a

nd p

olic

e ca

lled.

Spe

cial

cate

gory

: rea

sona

ble

prac

ticab

ility

– p

atie

nt w

as a

lread

y un

der i

ncre

ased

ob

serv

atio

ns.

5 £2

4.45

Pa

tient

V

anda

lism

O

ther

Pa

tient

smas

hed

door

into

fem

ale

end

of w

ard

(und

er re

deco

ratio

n).

Thre

w p

aint

at

win

dow

. A

larm

act

ivat

ed, s

taff

nur

se d

ecid

ed d

e-es

cala

tion

mos

t app

ropr

iate

ac

tion

cons

ider

ing

envi

ronm

ent.

Spe

cial

cat

egor

y: C

linic

al ju

dgem

ent –

car

e pl

an fo

r pat

ient

did

not

incl

ude

incr

ease

obs

erva

tion

regi

me.

6

£295

.20

Staf

f A

ccid

enta

l inj

ury

Patie

nt

liftin

g/ha

ndlin

g Pa

tient

am

pute

e, e

xcep

tiona

lly h

eavy

, rai

sed

from

bed

usi

ng a

hoi

st.

Two

staf

f un

derta

king

pro

cedu

re. S

taff

Nur

se p

ulle

d ho

ist a

nd su

ffer

ed p

ain

in h

er n

eck.

O

ff w

ork

for o

ne w

eek.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

al

l pol

icy

mea

sure

s in

plac

e.

7 £4

01.5

9 Pa

tient

Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Patie

nt re

ceiv

ed in

sulin

in tr

eatm

ent r

oom

, bec

ame

verb

ally

, the

n vi

olen

tly

aggr

essi

ve g

rabb

ing

staf

f nur

se b

y ar

ms.

Nur

se u

nabl

e to

act

ivat

e al

arm

. N

urse

gr

abbe

d ag

ain

and

thro

wn

agai

nst s

ink.

Pat

ient

atte

mpt

ed to

stra

ngle

nur

se a

nd

forc

ed h

er o

ver h

is k

nee.

Nur

se m

anag

ed to

rais

e al

arm

and

pat

ient

rest

rain

ed.

Staf

f Nur

se o

ff w

ork

for s

ix w

eeks

. Sp

ecia

l cat

egor

y: r

easo

nabl

e pr

actic

abili

ty–

gene

ral p

reve

ntat

ive

mea

sure

s in

plac

e, n

urse

was

alo

ne w

ith p

atie

nt b

ut h

ad

not b

een

asse

ssed

as l

ikel

y to

ass

ault.

8

£30.

70

Patie

nt

Oth

er

Cut

with

shar

p m

ater

ial/o

bjec

t Pa

tient

in b

edro

om a

ctiv

ated

em

erge

ncy

buzz

er.

Foun

d to

hav

e cu

t bot

h fo

rear

ms.

Tak

en to

A&

E by

am

bula

nce

for s

utur

ing

then

retu

rned

to w

ard.

R

azor

bla

des f

ound

in ro

om, p

atie

nt h

ad se

lf-ha

rmed

twic

e be

fore

. Sp

ecia

lca

tego

ry: C

linic

al ju

dgem

ent –

car

e pl

an d

id n

ot re

gard

clo

ser o

bser

vatio

n or

ot

her m

easu

res t

o co

ntro

l ris

k ne

cess

ary.

9

£25.

00

Patie

nt

Van

dalis

m

Oth

er

Patie

nt re

peat

edly

slam

med

seat

on

toile

t, sm

ashi

ng it

. Pa

tient

dis

turb

ed, h

earin

g vo

ices

from

the

devi

l in

his h

ead.

Spe

cial

cat

egor

y: C

linic

al ju

dgem

ent –

de

cisi

on to

kee

p pa

tient

on

gene

ral o

bser

vatio

ns d

ue to

pre

viou

s exp

erie

nce.

193

10

£23.

16

Patie

nt

Phys

ical

vio

lenc

e O

ther

D

elus

iona

l pat

ient

pun

ched

a w

all d

urin

g ni

ght.

Sta

ff u

naw

are

until

bre

akfa

st.

Seen

by

doct

or a

nd se

nt fo

r x-r

ay w

ith e

scor

t. N

o fr

actu

re c

onfir

med

and

re

turn

ed to

war

d. S

peci

al c

ateg

ory:

rea

sona

ble

prac

ticab

ility

– n

o in

dica

tion

that

cha

nge

to c

are

plan

was

nee

ded.

11

£20.

00

Patie

nt

Phys

ical

vio

lenc

e O

ther

Pa

tient

hav

ing

brea

kfas

t in

dini

ng ro

om w

hen

loud

noi

se h

eard

. Pa

tient

had

th

row

n pl

ate

and

kniv

es a

t wal

l. P

atie

nt su

ffer

ing

audi

tory

hal

luci

natio

ns a

nd

para

noid

del

usio

ns.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – u

nder

gen

eral

ob

serv

atio

n at

the

time,

no

staf

f mem

bers

pre

sent

. 12

£1

1.59

Pa

tient

O

ther

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Staf

f ale

rted

to n

oise

in m

ale

toile

t. T

oile

t loc

ked

and

patie

nt re

fuse

d en

try to

st

aff.

Ala

rm a

ctiv

ated

, sta

ff p

reve

nted

furth

er la

cera

tion

to a

rm.

Firs

t aid

giv

en.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – se

cond

vio

lent

inci

dent

of t

he d

ay,

reac

ting

to th

e pr

evio

us d

ays a

bsco

ndin

g.

13

£11.

59

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t Pa

tient

, on

seei

ng N

A, g

rabb

ed h

im b

y th

e ar

m a

nd p

unch

ed h

im o

n th

e fo

rehe

ad.

NA

bro

ke fr

ee a

nd a

ctiv

ated

ala

rm.

Spec

ial c

ateg

ory:

risk

not

fore

seea

ble

– in

cide

nt h

appe

ned

“out

of b

lue”

. 14

£2

3.80

Pa

tient

O

ther

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Patie

nt d

ism

antle

d hi

s raz

or a

nd in

flict

ed a

15c

m la

cera

tion

to a

bdom

en.

Iden

tical

inci

dent

4 w

eeks

pre

viou

sly.

Atte

mpt

at c

losi

ng w

ound

, but

had

to b

e se

nt to

A&

E fo

r tre

atm

ent.

Spe

cial

cat

egor

y: C

linic

al ju

dgem

ent –

bas

ed o

n st

rate

gy to

incr

ease

free

dom

as p

art o

f reh

abili

tatio

n st

rate

gy.

15

£23.

80

Patie

nt

Oth

er

Cut

with

shar

p m

ater

ial/o

bjec

t Pa

tient

requ

este

d an

d w

as g

iven

his

safe

ty ra

zor.

In to

ilet p

atie

nt d

ism

antle

s ra

zor a

nd se

lf in

flict

s 15c

m la

cera

tion

to h

is a

bdom

en.

Take

n to

A&

E fo

r sut

ures

th

en re

turn

ed to

war

d. S

peci

al c

ateg

ory:

rea

sona

ble

prac

ticab

ility

– in

cide

nt

not f

ores

eeab

le (n

o pr

evio

us h

isto

ry).

16

£11.

59

Patie

nt

Phys

ical

vio

lenc

e O

ther

Pa

tient

foun

d sm

okin

g in

dor

mito

ry.

NA

ask

ed p

atie

nt to

ext

ingu

ish

ciga

rette

. Pa

tient

bec

ame

viol

ent.

Cha

rge

Nur

se re

stra

ined

pat

ient

. Sp

ecia

l cat

egor

y:

reas

onab

le p

ract

icab

ility

– a

ppro

pria

te sm

okin

g po

licy

of w

hich

pat

ient

was

aw

are

and

usua

lly c

ompl

ied.

17

£1

1.59

Pa

tient

Ph

ysic

al v

iole

nce

Oth

er

Patie

nt o

vertu

rned

lade

n ta

ble.

Ala

rm a

ctiv

ated

pat

ient

rest

rain

ed a

nd m

edic

atio

n gi

ven.

Spe

cial

cat

egor

y: r

isk n

ot fo

rese

eabl

e –

inci

dent

hap

pene

d w

ithou

t w

arni

ng.

18

£11.

59

Patie

nt

Van

dalis

m

Oth

er

Patie

nt b

ecam

e an

gry

afte

r rec

eivi

ng v

erba

l abu

se fr

om a

noth

er p

atie

nt, t

hen

pick

ed u

p a

chai

r and

beg

an h

ittin

g it

off a

win

dow

in th

e di

ning

room

. A

larm

ac

tivat

ed, p

atie

nt re

mov

ed fr

om d

inin

g ro

om, s

ituat

ion

de-e

scal

ated

qui

ckly

. Sp

ecia

l cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

all

reas

onab

le c

ontro

l mea

sure

s in

plac

e.

194

19

£11.

59

Patie

nt

Van

dalis

m

Oth

er

Patie

nt in

sitti

ng ro

om p

icke

d up

cha

ir an

d at

tem

pted

to sm

ash

a w

indo

w.

Patie

nt

said

he

was

hea

ring

voic

es te

lling

him

to d

o so

. A

larm

act

ivat

ed a

nd p

atie

nt

rest

rain

ed.

Spec

ial c

ateg

ory:

ris

k un

fore

seea

ble

– in

cide

nt h

appe

ned

with

out

war

ning

. 20

£2

3.16

Pa

tient

O

ther

O

ther

Pa

tient

on

cons

tant

obs

erva

tion,

but

this

did

not

ext

end

to th

e to

ilet f

or p

atie

nt

priv

acy.

Pat

ient

ope

ned

up p

revi

ous a

bdom

inal

wou

nd.

Esco

rted

to a

cute

ho

spita

l for

trea

tmen

t. P

atie

nt c

onst

antly

self-

harm

ing

but n

ot se

lf-th

reat

enin

g.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – P

olic

y on

obs

erva

tion

regi

me

allo

wed

fo

r tim

e fo

r per

sona

l hyg

iene

in p

rivac

y so

no

brea

ch o

f pol

icy.

21

£4

0.00

Pa

tient

Ph

ysic

al v

iole

nce

Oth

er

Patie

nt si

tting

qui

etly

in si

tting

room

pic

ked

up c

hair

and

brok

e a

win

dow

, thi

s w

as re

peat

ed 5

min

utes

late

r. U

ncom

mun

icat

ive

thro

ugho

ut.

Spec

ial c

ateg

ory:

ri

sk u

nfor

esee

able

– b

oth

inci

dent

s hap

pene

d su

dden

ly w

ithou

t any

trig

ger.

22

£40.

00

Patie

nt

Phys

ical

vio

lenc

e O

ther

Pa

tient

in d

orm

itory

und

er g

ener

al o

bser

vatio

n w

as re

stle

ss in

bed

. Pa

tient

di

scon

nect

ed e

nd o

f bed

and

smas

hed

win

dow

. Pa

tient

rem

oved

to si

tting

are

a to

ve

nt a

nxie

ties.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

pat

ient

on

gen

eral

obs

erva

tion.

23

£4

6.69

Pa

tient

O

ther

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Patie

nt, f

or th

e th

ird ti

me,

re-o

pene

d pr

evio

us se

lf in

flict

ed w

ound

. W

ound

cl

eane

d an

d sk

in c

losu

res a

pplie

d. W

ound

bec

ame

infe

cted

, dre

ssin

gs a

pplie

d an

d a

cour

se o

f ant

ibio

tics g

iven

. Sp

ecia

l cat

egor

y: C

linic

al ju

dgem

ent -

re

gard

ing

care

pla

n.

24

£11.

59

Patie

nt

Phys

ical

vio

lenc

e O

ther

Pa

tient

ask

ed to

use

the

offic

e ph

one.

Req

uest

refu

sed.

Pat

ient

pun

ched

cha

rge

nurs

e in

the

face

. A

larm

act

ivat

ed, p

atie

nt re

stra

ined

and

take

n to

sitti

ng ro

om.

Spec

ial c

ateg

ory:

rea

sona

ble

prac

ticab

ility

– p

ayph

one

avai

labl

e fo

r pat

ient

us

e.25

£1

1.20

Pa

tient

O

ther

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Patie

nt in

mal

e do

rmito

ry in

serte

d ho

ok o

f coa

t han

ger i

nto

rece

nt a

bdom

inal

w

ound

. U

nwitn

esse

d, 3

rd v

iole

nt in

cide

nt o

f day

. Pa

tient

sent

to A

&E

for

treat

men

t. S

peci

al c

ateg

ory:

lack

of i

nfor

mat

ion

– un

able

to o

btai

n in

form

atio

n re

gard

ing

obse

rvat

ion

requ

ired.

26

£6

7.64

O

ther

O

ther

C

onta

ct

with

/exp

osur

e to

ha

rmfu

l sub

stan

ces

Staf

f nur

se sm

elle

d ga

s. O

n ca

ll en

gine

er tr

aced

to re

nted

LPG

stor

age

tank

s.

Engi

neer

cal

led

to re

pair

leak

ing

fittin

g. S

peci

al c

ateg

ory:

lack

of i

nfor

mat

ion

–tec

hnic

al re

cord

s una

vaila

ble.

195

27

£26.

31

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t St

aff n

urse

alo

ne in

trea

tmen

t roo

m w

hen

patie

nt e

nter

ed u

nann

ounc

ed.

Staf

f nu

rse

mov

ed to

war

ds a

skin

g pa

tient

wha

t he

wan

ted.

Pat

ient

repl

ied

that

he

wan

ted

to g

o ho

me.

Whe

n ex

plai

ned

to h

im th

at h

e co

uldn

’t pa

tient

gra

bbed

staf

f nu

rse

by th

e th

roat

resu

lting

in h

er fa

lling

to th

e gr

ound

. Sp

ecia

l cat

egor

y:

outw

ith r

easo

nabl

e pr

actic

abili

ty –

nur

se re

spon

ded

wel

l to

the

situ

atio

n. N

o al

arm

fact

ors t

hat c

ould

hav

e pr

even

ted

inci

dent

. 28

£4

47.2

7 St

aff

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Staf

f nur

se tr

ippe

d on

pav

ing

slab

at d

oorw

ay.

No

obvi

ous d

efec

t. R

esul

ting

inju

ries w

ere

bang

on

head

, spr

aine

d w

rist a

nd c

ut fi

nger

s. A

bsen

t fro

m w

ork.

Sp

ecia

l cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

not

hing

foun

d to

con

tribu

te to

fa

ll.

29

£137

.40

Patie

nt

Oth

er

Oth

er

Patie

nt (d

etai

ned

unde

r MH

A) o

n pa

role

leav

e fa

iled

to re

turn

to w

ard.

Par

ent

tele

phon

ed to

say

he w

as a

t hom

e. T

eam

was

ass

embl

ed a

nd p

icke

d up

pat

ient

fr

om h

ome

and

retu

rned

to w

ard.

Spe

cial

cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

In

cide

nt d

ealt

with

as b

est a

s it c

ould

hav

e be

en, i

n th

e ju

dgem

ent o

f the

rese

arch

te

am.

30

£4.3

9 Pa

tient

O

ther

C

ut w

ith sh

arp

mat

eria

l/obj

ect

Staf

f wer

e pr

esen

t in

patie

nt’s

room

due

to h

is m

enta

l sta

te.

Patie

nt su

dden

ly

took

hol

d of

gla

ss tu

mbl

er sm

ashe

d it

agai

nst t

he w

all c

uttin

g hi

s han

d in

the

proc

ess.

Stru

ggle

d vi

olen

tly b

efor

e be

ing

rest

rain

ed b

y st

aff.

Spe

cial

cat

egor

y:

reas

onab

le p

ract

icab

ility

– a

ll ris

k co

ntro

l mea

sure

s in

plac

e.

31

£770

.17

Patie

nt

Oth

er

Oth

er

Patie

nt, g

rant

ed 3

0 m

inut

es le

ave

from

war

d, a

bsco

nded

. Po

lice

foun

d pa

tient

at

hom

e w

ith e

vide

nce

of o

verd

ose

– pa

tient

sent

by

ambu

lanc

e to

A&

E an

d de

tain

ed fo

r app

rox.

28

hour

s with

two

staf

f in

atte

ndan

ce o

n sp

ecia

l nur

sing

.Sp

ecia

l cat

egor

y: C

linic

al ju

dgem

ent –

clin

ical

judg

emen

t to

gran

t lea

ve fr

om

war

d.32

£6

.58

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t Pa

tient

ran

tow

ards

ano

ther

, pus

hing

him

bac

kwar

ds a

nd h

ittin

g hi

s hea

d of

f cu

pboa

rd, s

usta

inin

g a

lace

ratio

n to

hea

d. G

iven

firs

t aid

then

sent

to A

&E.

Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

reas

onab

le m

easu

res i

n pl

ace,

no

hist

ory

of b

ad fe

elin

g be

twee

n pa

tient

s and

no

hist

ory

of v

iole

nce

from

ag

gres

sor.

33

£1,1

28.0

0 St

aff

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

On

open

ing

door

to c

orrid

or, n

urse

stum

bled

on

a st

one

whe

n pu

tting

ligh

t on.

H

e hi

t his

hea

d of

f fac

ing

wal

l and

was

kno

cked

unc

onsc

ious

. W

as ta

ken

to A

&E

by a

mbu

lanc

e an

d w

as su

bseq

uent

ly o

ff w

ork.

Spe

cial

cat

egor

y: o

utw

ith

reas

onab

le p

ract

icab

ility

– n

o ev

iden

ce to

sugg

est i

ncid

ent w

as c

ause

d by

lig

htin

g le

vels

and

no

way

of k

now

ing

whe

re st

one

cam

e fr

om.

196

34

£36.

40

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t N

ursi

ng a

ssis

tant

was

bat

hing

pat

ient

. Pa

tient

hit

out a

t NA

cau

sing

her

left

thum

b to

be

bent

bac

k. N

A ta

ken

to A

&E

by ta

xi b

ut n

o fr

actu

re –

retu

rned

to

war

d.Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

risk

s red

uced

to

as lo

w a

s pos

sibl

e, p

atie

nt h

it ou

t im

puls

ivel

y.

35

£355

.19

Patie

nt

Acc

iden

tal i

njur

y O

ther

Pa

tient

got

out

of b

ed a

nd w

alke

d to

war

ds d

oor a

nd n

urse

. Pa

tient

sudd

enly

je

rked

hea

d up

war

ds a

nd st

ruck

nur

se o

n ja

w.

Nur

se o

ff w

ork

for o

ne w

eek

with

w

hipl

ash.

Spe

cial

cat

egor

y: r

isk

not f

ores

eeab

le –

alth

ough

pat

ient

alw

ays

wal

ked

awkw

ardl

y th

e in

cide

nt h

appe

ned

very

qui

ckly

. 36

£3

0.31

Pa

tient

O

ther

O

ther

Pa

tient

agi

tate

d, d

octo

r exp

lain

ed to

him

that

he

was

not

per

mitt

ed to

leav

e th

e w

ard.

Pat

ient

lock

ed w

ard

door

s, pu

lled

dow

n cu

rtain

rails

and

bar

ricad

ed

him

self

in si

tting

room

. A

larm

act

ivat

ed a

nd si

tuat

ion

de-e

scal

ated

. Sp

ecia

lca

tego

ry: C

linic

al ju

dgem

ent –

let p

atie

nt v

ent f

rust

ratio

n an

d ch

ose

to a

dopt

de

-esc

alat

ion

rath

er th

an re

stra

in p

atie

nt im

med

iate

ly.

37

£25.

52

Staf

f V

erba

l abu

se o

r th

reat

enin

g be

havi

our

Oth

er

Supp

ort w

orke

r on

hom

e vi

sit.

Clie

nt b

ecam

e ve

rbal

ly a

busi

ve a

nd p

hysi

cally

th

reat

enin

g. S

W m

ade

deci

sion

to w

ithdr

aw a

nd re

turn

ed to

off

ice

for

coun

selli

ng.

Wen

t hom

e ea

rly.

Spec

ial c

ateg

ory:

risk

not

fore

seea

ble/

outw

ith

reas

onab

le p

ract

icab

ility

– C

lient

had

not

show

n a

pred

ispo

sitio

n to

vio

lenc

e be

fore

. SW

follo

wed

pro

cedu

re b

y w

ithdr

awin

g fr

om si

tuat

ion.

38

£2

56.7

7 St

aff

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Gar

dene

r pru

ning

bor

der s

hrub

s tur

ned

to c

ut a

noth

er sh

rub

whe

n he

stum

bled

to

grou

nd tw

istin

g an

kle.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

G

arde

ner w

as fo

llow

ing

corr

ect p

roce

dure

and

usi

ng c

orre

ct e

quip

men

t (fo

otw

ear

etc.

). 39

£1

1.51

Pa

tient

Ph

ysic

al v

iole

nce

Oth

er

Frus

trate

d pa

tient

smas

hed

cup

in m

ain

kitc

hen.

C&

R te

chni

ques

use

d to

reso

lve

situ

atio

n an

d pa

tient

was

put

to b

ed w

ith m

edic

atio

n. S

peci

al c

ateg

ory:

Clin

ical

ju

dgem

ent –

staf

f fel

t med

icat

ion

wou

ld c

alm

pat

ient

dow

n bu

t clo

ser

obs/

supe

rvis

ion

also

nee

ded.

40

£1

9.90

Pa

tient

O

ther

O

ther

Pa

tient

abs

cond

ed a

fter b

eing

adm

itted

nig

ht b

efor

e in

to p

sych

iatri

c as

sess

men

t w

ard.

Abs

cond

ed w

hen

othe

r pat

ient

s wer

e do

wns

tairs

at m

ealti

me

(din

ing

room

on

floo

r bel

ow) –

Spe

cial

cat

egor

y: C

linic

al ju

dgem

ent –

Pat

ient

was

adm

itted

on

a v

olun

tary

bas

is.

41

£11.

20

Patie

nt

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Patie

nt w

alki

ng in

to d

ay ro

om tr

ippe

d ov

er a

noth

er p

atie

nt’s

zim

mer

fram

e. N

o br

uisi

ng e

vide

nt a

t tim

e al

thou

gh p

atie

nt h

ad a

lot o

f pai

n in

left

wris

t. S

peci

alca

tego

ry: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

pre

senc

e of

zim

mer

fram

e un

avoi

dabl

e.

197

42

£50.

00

Oth

er

Oth

er

Oth

er

Fire

ala

rm a

ctiv

ated

for n

o ap

pare

nt re

ason

. C

all o

ut fo

r mem

ber o

f sta

ff to

rese

t fir

e al

arm

. Sp

ecia

l cat

egor

y: in

suff

icie

nt in

form

atio

n –

no fu

rther

info

rmat

ion

or w

itnes

ses a

vaila

ble.

43

£9

.79

Patie

nt

Oth

er

Oth

er

Patie

nt fo

rced

ope

n lo

cked

inte

rnal

doo

r at t

he sa

me

time

as a

join

er w

as o

peni

ng

an a

djac

ent e

xter

nal d

oor.

Pat

ient

pus

hed

past

join

er a

nd a

bsco

nded

. Sp

ecia

lca

tego

ry: C

linic

al ju

dgem

ent –

pat

ient

not

judg

ed to

requ

ire a

ny p

artic

ular

ob

serv

atio

n re

gim

e.

44

£18.

20

Patie

nt

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Patie

nt fo

und

lyin

g on

floo

r in

corr

idor

and

take

n to

A&

E by

taxi

. W

as re

turn

ed

to w

ard

by ta

xi a

fter x

-ray

. Sp

ecia

l cat

egor

y: r

isk n

ot fo

rese

eabl

e –

first

tim

e pa

tient

had

falle

n.

45

£6.1

4 Pa

tient

O

ther

O

ther

A

bsco

ndin

g pa

tient

det

aine

d in

gro

unds

but

cou

ld n

ot b

e pe

rsua

ded

to re

turn

to

war

d. N

o po

wer

to d

etai

n an

d pa

tient

retu

rned

hom

e bu

t cam

e ba

ck to

war

d tw

o ho

urs l

ater

apo

logi

sing

. Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

ap

prop

riate

con

trol m

easu

res i

n pl

ace.

46

£6

7.16

Pa

tient

O

ther

O

ther

Pa

tient

info

rmed

staf

f she

had

inge

sted

her

hea

ring

aid

and

was

take

n fo

r x-r

ay.

It w

as c

onfir

med

but

no

treat

men

t nec

essa

ry a

nd re

turn

ed to

war

d. P

atie

nt h

as

hist

ory

of m

isch

ievo

us se

lf-ha

rm.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – n

ot

judg

ed to

nee

d sp

ecia

l obs

erva

tion

regi

me.

47

£3

.25

Patie

nt

Oth

er

Oth

er

Patie

nt, g

iven

15

min

utes

una

ccom

pani

ed p

arol

e, fa

iled

to re

turn

. D

octo

r pho

ned

war

d to

info

rm o

f sig

htin

g, c

harg

e nu

rse

then

wen

t to

sear

ch lo

cal a

rea

in c

ar.

Patie

nt p

hone

d 1½

hou

rs la

ter t

o gi

ve lo

catio

n. P

olic

e re

turn

ed p

atie

nt to

war

d.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – a

lthou

gh p

atie

nt u

nder

goin

g pe

rson

al

chan

ges a

nd q

uite

stre

ssed

it w

as ju

dged

app

ropr

iate

to g

rant

par

ole.

48

£1

6.80

Pa

tient

O

ther

O

ther

Pa

tient

faile

d to

retu

rn fr

om O

ccup

atio

nal T

hera

pist

. Pa

tient

seen

boa

rdin

g a

bus

into

tow

n. T

wo

nurs

ing

auxi

liarie

s fol

low

in c

ar b

ut fa

il to

loca

te p

atie

nt in

tow

n or

at m

othe

r’s h

ome.

The

y re

turn

to w

ard

and

call

polic

e to

repo

rt m

issi

ng

pers

on.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – p

atie

nt ju

dged

suita

ble

for

unac

com

pani

ed v

isit

to O

T.

49

£15.

00

Patie

nt

Van

dalis

m

Oth

er

Patie

nt w

ith se

vere

dem

entia

bro

ke fi

re a

larm

gla

ss b

ut a

larm

did

n’t a

ctiv

ate.

Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

pat

ient

diff

icul

t to

man

age

and

refu

sed

to ta

ke m

edic

atio

n.

50

£15.

00

Patie

nt

Van

dalis

m

Oth

er

Lear

ning

dis

abili

ties p

atie

nt se

t off

fire

ala

rm b

y br

eaki

ng g

lass

. Pa

tient

s and

st

aff e

vacu

ated

and

fire

brig

ade

atte

nded

. Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

no

prev

ious

his

tory

of v

anda

lism

so a

dequ

ate

mea

sure

s in

plac

e.

51

£53.

56

Oth

er

Oth

er

Oth

er

Fals

e fir

e al

arm

in c

otta

ge h

ospi

tal.

Ala

rm si

lenc

ed o

n ve

rifyi

ng n

o fir

e in

198in

dica

ted

zone

and

on-

call

engi

neer

atte

nded

to re

set a

nd e

xam

ine

syst

em.

No

faul

t was

foun

d, b

elie

ved

to b

e ca

used

by

pow

er b

eing

off

dur

ing

that

day

. Sp

ecia

l cat

egor

y: r

isk n

ot fo

rese

eabl

e –

all o

ther

con

trol m

easu

res i

n pl

ace.

52

£1

07.1

2 O

ther

O

ther

O

ther

Fi

re a

larm

act

ivat

ed fo

r no

appa

rent

reas

on.

Ala

rm si

lenc

ed a

nd e

ngin

eer c

alle

d ou

t. F

aulty

sens

or re

plac

ed a

nd s

yste

m re

set.

Spe

cial

cat

egor

y: o

utw

ith

reas

onab

le p

ract

icab

ility

– 2

5% o

f sen

sors

che

cked

eve

ry 3

-mon

ths.

Yea

rly

cont

ract

mai

nten

ance

che

ck.

53

£25.

00

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t Pa

tient

pun

ched

ano

ther

kno

ckin

g of

f the

ir gl

asse

s. F

ram

es re

plac

ed b

y ho

spita

l op

ticia

n. S

peci

al c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– n

o hi

stor

y of

an

tago

nism

bet

wee

n pa

tient

so a

dequ

ate

mea

sure

s wer

e in

pla

ce.

54

£50.

00

Patie

nt

Van

dalis

m

Oth

er

Patie

nt sm

ashe

d w

all m

irror

in b

athr

oom

for n

o ap

pare

nt re

ason

. Sp

ecia

lca

tego

ry: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

not

com

mon

for m

irror

s in

war

d to

be

brok

en, g

laze

d w

ith fi

lm-b

acke

d gl

ass t

o pr

even

t sca

tter o

f gla

ss.

55

£60.

00

Oth

er

Oth

er

Oth

er

Win

dow

in c

ubic

le d

orm

itory

foun

d to

be

brok

en (n

ot h

eard

or w

itnes

sed)

. Pa

tient

in b

ed h

ad h

isto

ry o

f ban

ging

on

win

dow

. Sp

ecia

l cat

egor

y: in

suff

icie

nt

info

rmat

ion

– in

cide

nt w

asn’

t hea

rd o

r witn

esse

d.

56

£85.

00

Oth

er

Van

dalis

m

Oth

er

Pane

of g

lass

foun

d to

be

shat

tere

d in

sing

le b

edde

d ro

om.

No

nois

e an

d in

cide

nt

unw

itnes

sed.

Pat

ient

repe

ated

inci

dent

2 w

eeks

late

r. S

peci

al c

ateg

ory:

out

with

re

ason

able

pra

ctic

abili

ty –

no

prev

ious

his

tory

and

gla

ss ju

dged

by

Trus

t sta

ff

to b

e of

an

appr

opria

te ty

pe fo

r typ

e of

war

d.

57

£2,6

52.9

3 St

aff

Acc

iden

tal i

njur

y O

ther

St

aff n

urse

, with

hel

p fr

om N

A, w

as c

hang

ing

inco

ntin

ence

pad

on

patie

nt.

Patie

nt ro

lled

over

on

side

and

new

pad

put

into

pos

ition

in th

e pr

oces

s sta

ff n

urse

be

nt b

ack

mid

dle

finge

r. S

taff

nurs

e su

stai

ned

a fr

actu

red

finge

r. S

peci

alca

tego

ry: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

staf

f wer

e fo

llow

ing

corr

ect

proc

edur

es.

58

£13.

89

Patie

nt

Oth

er

Oth

er

Patie

nt u

nder

con

stan

t obs

erva

tion

wen

t int

o en

-sui

te to

ilet w

hils

t sta

ff re

mai

ned

outs

ide.

Sta

ff e

nter

ed to

ilet w

hen

nois

e of

key

s hea

rd.

Patie

nt sa

id th

at sh

e ha

d sw

allo

wed

a k

ey a

nd w

as ta

ken

to A

&E

for x

-ray

then

retu

rned

to w

ard.

Spe

cial

cate

gory

: rea

sona

ble

prac

ticab

ility

– u

nder

con

stan

t obs

and

pro

cedu

re o

f pa

tient

priv

acy.

Pat

ient

thou

ght n

ot to

be

in p

osse

ssio

n of

key

s as t

hey

had

been

re

mov

ed.

59

£12.

60

Oth

er

Acc

iden

tal

prop

erty

lo

ss/d

amag

e

Oth

er

Win

dow

in st

aff t

ea-r

oom

ope

ned.

Stro

ng w

ind

caug

ht w

indo

w o

peni

ng it

fully

re

sulti

ng in

the

win

dow

smas

hing

off

inte

rnal

wal

l. S

peci

al c

ateg

ory:

risk

not

fo

rese

eabl

e –

win

dow

not

exp

ecte

d to

be

caug

ht b

y w

ind.

199

60

£324

.24

Staf

f A

ccid

enta

l inj

ury

Patie

nt li

fting

/ ha

ndlin

g Pa

tient

requ

ired

toile

t dur

ing

nigh

t so

com

mod

e br

ough

t to

beds

ide.

Not

qui

te in

po

sitio

n w

hen

patie

nt tr

ied

to si

t on

it. N

A in

stin

ctiv

ely

stre

tche

d ou

t lef

t arm

to

prev

ent p

atie

nt fa

lling

and

sust

aine

d an

inju

ry d

oing

so.

NA

was

off

wor

k fo

r tw

o ni

ghts

with

susp

ecte

d pu

lled

mus

cle.

Spe

cial

cat

egor

y: r

isk

unfo

rese

eabl

e–

patie

nt h

ad n

o hi

stor

y of

falls

, mis

judg

ed d

ista

nce.

61

£1

5.00

Pa

tient

O

ther

O

ther

M

ale

patie

nt w

ith d

emen

tia a

ctiv

ated

a fi

re a

larm

. Pa

tient

is o

vera

ctiv

e an

d th

inks

he

is in

the

arm

y. W

hen

he re

ads “

in c

ase

of fi

re p

ress

bel

l” h

e do

es it

. Fi

re b

rigad

e at

tend

but

no

evac

uatio

n. S

peci

al c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– p

atie

nt h

ad o

nly

done

this

onc

e be

fore

but

not

thou

ght l

ikel

y to

do

it a

gain

. Tr

ust p

lann

ing

inst

all a

larm

cov

ers.

62

£26.

70

Patie

nt

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Patie

nt h

ad ju

st b

een

adm

itted

and

was

wal

king

with

zim

mer

fram

e to

see

doct

or,

a st

aff m

embe

r alo

ng si

de.

Patie

nt lo

st b

alan

ce a

nd fe

ll to

floo

r. T

rans

ferr

ed to

ac

ute

hosp

ital.

Spe

cial

cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

pat

ient

wal

king

w

ell w

ith st

aff m

embe

r. 63

£1

1.87

Pa

tient

O

ther

O

ther

Pa

tient

last

seen

on

war

d ar

ound

10:

30, f

ound

to b

e m

issi

ng a

t 12:

00.

A se

arch

of

the

grou

nds a

nd su

rrou

ndin

g ar

ea, i

n tw

o ca

rs, w

as c

arrie

d ou

t. P

atie

nt fo

und

at

13:5

0 by

pol

ice

and

retu

rned

to th

e w

ard.

Spe

cial

cat

egor

y: r

isk

unfo

rese

eabl

e–

Sect

ion

18 p

atie

nt, b

ehav

iour

nor

mal

. W

ard

alw

ays u

nloc

ked.

64

£2

.80

Patie

nt

Oth

er

Oth

er

Patie

nt fi

t and

abl

e, w

aitin

g on

pla

ce in

car

e ho

me,

left

war

d un

obse

rved

. Fo

llow

ing

exte

nsiv

e se

arch

pat

ient

was

foun

d in

loca

l soc

ial c

lub

and

retu

rned

to

war

d by

the

polic

e. S

peci

al c

ateg

ory:

risk

unf

ores

eeab

le –

no

prev

ious

in

dica

tion

of ri

sk o

f abs

cond

ing.

65

£2,1

04.8

0 Pa

tient

O

ther

O

ther

Pa

tient

out

on

unac

com

pani

ed le

ave.

Whe

n sh

e re

turn

ed in

form

ed st

aff s

he h

ad

take

n an

ove

rdos

e of

par

acet

amol

. B

lood

s tak

en b

y du

ty d

octo

r con

firm

ed

over

dose

. Ta

ken

to a

cute

hos

pita

l for

trea

tmen

t. R

etur

ned

to w

ard

five

days

la

ter.

Spe

cial

cat

egor

y: C

linic

al ju

dgem

ent –

not

judg

ed to

be

a su

icid

e ris

k.

66

£7.9

9 Pa

tient

V

erba

l abu

se o

r th

reat

enin

g be

havi

our

Oth

er

Thirt

y ye

ar o

ld m

ale

patie

nt, w

ith d

epre

ssiv

e co

nditi

on.

Trig

gerin

g fa

ctor

usu

ally

ph

onin

g hi

s mot

her.

Afte

r pho

ne c

all h

e de

man

ded

to g

o ho

me.

Pat

ient

then

be

cam

e ag

gres

sive

, ala

rm a

ctiv

ated

, pat

ient

rest

rain

ed.

Spec

ial c

ateg

ory

of

Clin

ical

judg

emen

t – d

ecis

ion

take

n no

t to

with

hold

acc

ess t

o te

leph

one

to sp

eak

to m

othe

r. 67

£5

.60

Patie

nt

Oth

er

Oth

er

Patie

nt d

ue fo

r dis

char

ge b

y bo

ardi

ng o

ut.

Nex

t day

pat

ient

esc

orte

d ba

ck to

ow

n w

ard,

app

rox.

One

hou

r lat

er st

aff f

ound

an

empt

y as

pirin

pac

ket a

t bed

side

. Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

unl

ikel

y th

at st

aff c

ould

be

exp

ecte

d to

ant

icip

ate

risk

of o

verd

ose.

200

68

£2.1

9 Pa

tient

O

ther

O

ther

Pa

tient

atte

nded

Occ

upat

iona

l The

rapy

(OT)

dep

t une

scor

ted.

OT

phon

ed w

ard

right

aw

ay in

form

ing

them

of p

atie

nt’s

dep

artu

re.

Staf

f Nur

se, i

n ca

r, se

arch

ed

loca

l roa

ds a

nd p

olic

e in

form

ed.

Patie

nt re

turn

ed o

f ow

n fr

ee w

ill 2

½ h

ours

late

r.

Spec

ial c

ateg

ory;

Clin

ical

judg

emen

t - ju

dgem

ent t

o al

low

una

ccom

pani

ed v

isit

to O

T de

partm

ent.

69

£336

.00

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Patie

nt a

ttem

pted

to le

ave

war

d. H

isto

ry o

f vio

lenc

e ag

ains

t sta

ff a

nd w

as b

eing

es

corte

d du

e to

pre

viou

s abs

cond

ing

atte

mpt

s. P

atie

nt a

ssau

lted

staf

f and

was

br

ough

t to

floor

by

two

staf

f mem

bers

. In

the

proc

ess,

NA

suff

ered

an

inju

ry

whe

n pu

shed

into

a w

ardr

obe.

Spe

cial

cat

egor

y: C

linic

al ju

dgem

ent –

ju

dgem

ent o

n de

cisi

on to

allo

w p

atie

nt to

leav

e w

ard.

70

£4

6.86

Pa

tient

O

ther

O

ther

Pa

tient

with

long

his

tory

of a

bsco

ndin

g w

as g

iven

par

ole

to a

ttend

ther

apy

depa

rtmen

t. A

gree

men

t was

mad

e w

ith p

atie

nt to

go

unes

corte

d to

bui

ld u

p Tr

ust.

Pat

ient

faile

d to

arr

ive

at th

erap

y de

partm

ent a

nd a

sear

ch w

as in

itiat

ed,

polic

e in

form

ed.

Polic

e re

ceiv

ed c

all s

ayin

g pa

tient

was

at h

er so

licito

rs a

nd h

ad

told

them

she

had

take

n an

ove

rdos

e. P

atie

nt w

as th

en ta

ken

to A

&E

for

treat

men

t. S

peci

al c

ateg

ory

of C

linic

al ju

dgem

ent-

judg

emen

t on

reha

bilit

atio

n re

gim

e.

71

£23.

10

Patie

nt

Oth

er

Oth

er

Patie

nt a

bsco

nded

whe

n sh

ould

hav

e be

en a

ttend

ing

OT

depa

rtmen

t. S

he

pres

ente

d he

rsel

f at p

olic

e H

Q st

atin

g sh

e ha

d ta

ken

an o

verd

ose

so w

as ta

ken

to

A&

E. P

atie

nt th

en a

bsco

nded

from

A&

E an

d w

as se

en in

the

grou

nds o

f ho

spita

l. W

as e

vent

ually

retu

rned

by

car.

Spe

cial

cat

egor

y of

Clin

ical

ju

dgem

ent –

clin

icia

n in

cha

rge

of c

are

wis

hes t

o bu

ild T

rust

by

cont

inui

ng to

al

low

una

ccom

pani

ed v

isits

to O

T de

partm

ent.

72

£8.2

4 Pa

tient

O

ther

O

ther

Pa

tient

on

esco

rted

wal

k in

hos

pita

l gro

unds

abs

cond

s. S

earc

h in

itiat

ed w

ith

polic

e an

d tw

o st

aff a

lso

sear

ch in

car

. Po

lice

loca

ted

the

patie

nt a

nd re

turn

ed

him

to w

ard.

Spe

cial

cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty -

all

reas

onab

le m

easu

res i

n pl

ace.

73

£3

4.71

Pa

tient

O

ther

O

ther

Pa

tient

was

on

two

day

pass

for 1

7th b

irthd

ay.

Thre

e m

embe

rs o

f sta

ff w

ere

requ

ired

to c

olle

ct p

atie

nt fr

om h

ome,

to re

turn

to w

ard,

bec

ause

pat

ient

had

be

com

e in

toxi

cate

d w

ith a

lcoh

ol.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t –

judg

emen

t to

allo

w b

irthd

ay p

arol

e.

74

£11.

58

Patie

nt

Oth

er

Oth

er

Patie

nt o

bser

ved

on w

ard

15 m

inut

es b

efor

e te

a bu

t by

teat

ime

patie

nt h

ad

absc

onde

d un

witn

esse

d. A

sear

ch o

f the

war

d an

d gr

ound

s was

car

ried

out.

Po

lice

calle

d to

info

rm w

ard

of p

ossi

ble

sigh

tings

. A

sear

ch in

car

car

ried

out.

Po

lice

loca

ted

patie

nt a

nd re

turn

ed h

im to

the

war

d. S

peci

al c

ateg

ory:

Clin

ical

ju

dgem

ent –

judg

emen

t as t

o le

vel o

f obs

erva

tion.

201

75

£1.4

0 Pa

tient

O

ther

O

ther

Pa

tient

gra

nted

15-

min

utes

tim

eout

but

faile

d to

retu

rn.

No

hist

ory

of

absc

ondi

ng.

Polic

e in

form

ed, a

s the

re w

as c

once

rn fo

r pat

ient

’s sa

fety

. A

m

embe

r of s

taff

saw

pat

ient

in to

wn

and

advi

sed

him

to re

turn

to th

e w

ard,

whi

ch

he d

id.

Spec

ial c

ateg

ory

of C

linic

al ju

dgem

ent –

judg

emen

t to

allo

w p

arol

e.76

£8

.11

Patie

nt

Oth

er

Oth

er

Patie

nt a

bsco

nded

from

unl

ocke

d w

ard.

Not

long

afte

rwar

ds w

ard

rece

ived

a c

all

advi

sing

them

pat

ient

was

on

a br

idge

in to

wn.

Uns

ucce

ssfu

l sea

rch

carr

ied

out

in h

ospi

tal c

ar, p

olic

e w

ere

info

rmed

and

they

retu

rned

pat

ient

to w

ard.

Spe

cial

cate

gory

of C

linic

al ju

dgem

ent –

judg

emen

t tha

t pat

ient

was

suite

d to

an

unlo

cked

war

d.

77

£4.1

5 Pa

tient

O

ther

O

ther

Pa

tient

adm

itted

to w

ard

with

long

his

tory

of a

lcoh

ol a

buse

. G

iven

som

ethi

ng to

ea

t and

seem

ed c

omfo

rtabl

e. S

hortl

y af

ter t

he p

atie

nt a

bsco

nded

. A

sear

ch o

f ho

spita

l and

gro

unds

was

car

ried

out b

ut n

ot fo

und.

Pat

ient

was

foun

d at

his

flat

. Sp

ecia

l cat

egor

y: r

isk n

ot fo

rese

eabl

e –

patie

nt w

as c

omfo

rtabl

e at

tim

e of

ab

scon

ding

. 78

£5

.10

Patie

nt

Oth

er

Oth

er

Patie

nt a

gita

ted,

giv

en m

edic

atio

n an

d pu

t on

15 m

inut

es o

bser

vatio

ns (w

ard

unlo

cked

). Pa

tient

abs

cond

ed a

nd se

arch

car

ried

out i

n ho

spita

l car

. Pa

tient

lo

cate

d an

d re

turn

ed to

war

d. S

peci

al c

ateg

ory:

Clin

ical

judg

emen

t – w

ard

unlo

cked

whe

re p

revi

ousl

y ha

d be

en.

79

£4.3

9 Pa

tient

O

ther

O

ther

Pa

tient

abs

cond

ed w

hile

on

gran

ted

paro

le.

Patie

nt h

ad a

his

tory

of a

bsco

ndin

g,

usua

lly to

loca

l pub

. Se

arch

car

ried

out i

n ho

spita

l car

and

pat

ient

retu

rned

to

war

d.Sp

ecia

l cat

egor

y: C

linic

al ju

dgem

ent –

pat

ient

had

his

tory

but

gra

nted

15

min

ute

paro

le.

80

£51.

73

Patie

nt

Phys

ical

vio

lenc

e Pe

rson

to p

erso

n as

saul

t M

ale

patie

nt u

nder

con

stan

t obs

erva

tion

jum

ped

in a

ir ki

ckin

g fe

mal

e pa

tient

w

ithou

t war

ning

. A

larm

act

ivat

ed.

Patie

nt re

stra

ined

and

giv

en m

edic

atio

n.

Spec

ial c

ateg

ory:

rea

sona

ble

prac

ticab

ility

– a

ll pr

actic

able

mea

sure

s in

plac

e,

inci

dent

hap

pene

d ve

ry q

uick

ly.

81

£20.

31

Patie

nt

Phys

ical

vio

lenc

e O

ther

Pa

tient

mad

e th

reat

enin

g ge

stur

e to

NA

twic

e al

so k

icke

d ou

t at w

indo

w ra

diat

or.

Ala

rm a

ctiv

ated

. Te

am re

stra

ined

pat

ient

and

giv

en m

edic

atio

n. S

peci

alca

tego

ry: r

easo

nabl

e pr

actic

abili

ty –

ade

quat

e m

anag

emen

t of t

he si

tuat

ion.

82

£1

83.0

1 Pa

tient

O

ther

O

ther

Pa

tient

with

a su

spec

ted

over

dose

pho

ned

war

d w

hile

out

on

a pa

ss.

Patie

nt

retu

rned

to w

ard

and

had

bloo

d ta

ken.

Ove

rdos

e co

nfirm

ed a

nd p

atie

nt se

nt to

ac

ute

hosp

ital f

or tr

eatm

ent.

Pat

ient

requ

ired

spec

ial n

ursi

ng w

hils

t aw

ay fr

om

war

d.Sp

ecia

l cat

egor

y: C

linic

al ju

dgem

ent –

judg

emen

t to

allo

w p

arol

e.

202

83

£7.0

0 Pa

tient

O

ther

O

ther

Pa

tient

retu

rned

to w

ard

afte

r tw

o ho

urs p

arol

e in

form

ing

staf

f she

had

take

n 34

pa

race

tam

ol.

Bloo

d an

alys

is c

onfir

med

a la

rge

over

dose

. Pa

tient

tran

sfer

red

to

acut

e ho

spita

l, es

corte

d on

jour

ney

by N

A, f

or tw

o da

ys th

en re

turn

ed to

war

d.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t - ju

dgem

ent t

o al

low

par

ole.

84

£4

48.8

5 St

aff

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Porte

r par

ked

a va

n go

t out

. H

is fo

ot sl

ippe

d of

f ste

p su

stai

ning

ank

le in

jury

. H

e re

turn

ed to

dep

ot so

ano

ther

driv

er c

ould

com

plet

e m

ail d

eliv

erie

s. W

ent t

o A

&E

afte

r shi

ft w

here

it w

as c

onfir

med

that

he

had

pulle

d lig

amen

ts.

Was

off

for a

pe

riod

of si

ck le

ave.

Spe

cial

cat

egor

y: r

isk u

nfor

esee

able

– fo

llow

ed a

ll pr

oced

ures

. 85

£2

8.20

St

aff

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Dom

estic

mop

ping

bat

hroo

m fl

oor s

lippe

d an

d fe

ll, in

jurin

g he

r ank

le.

Spec

ial

cate

gory

: ins

uffic

ient

info

rmat

ion

– no

t eno

ugh

info

rmat

ion

avai

labl

e on

in

cide

nt.

86

£559

.76

Staf

f Ph

ysic

al v

iole

nce

Pers

on to

per

son

assa

ult

Patie

nt a

ssis

ted

from

wet

bed

to c

omm

ode

by 2

staf

f. P

atie

nt b

ecam

e ph

ysic

ally

an

d ve

rbal

ly a

busi

ve a

nd sp

at a

t nur

se.

Nur

se o

ff tw

o da

ys w

ith st

ress

, dut

ies

cove

red

by o

ther

nur

ses.

Spe

cial

cat

egor

y: r

isk u

nfor

esee

able

– fi

rst m

ajor

ag

gres

sive

inci

dent

. 87

£8

.77

Patie

nt

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

Patie

nt w

alki

ng d

own

corr

idor

turn

ed a

nd st

umbl

ed to

floo

r. E

scor

ted

to A

&E

whe

re fr

actu

red

to ri

ght n

eck

of fe

mur

con

firm

ed.

Frac

ture

pin

ned

in th

eatre

, pa

tient

rem

aine

d in

acu

te h

ospi

tal u

ntil

she

died

thre

e da

ys la

ter.

Spe

cial

cate

gory

: rea

sona

ble

prac

ticab

ility

– n

o pr

ior e

vide

nce

of ri

sk o

f fal

ling.

88

£9

.84

Patie

nt

Acc

iden

tal i

njur

y Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r sta

irs

82 y

ear o

ld fe

mal

e pa

tient

with

nig

ht d

emen

tia lo

st b

alan

ce in

toile

t and

fell

hitti

ng h

er h

ead

on d

oorf

ram

e. P

atie

nt h

ad tw

o kn

ee re

plac

emen

ts b

ut

inde

pend

ently

mob

ile w

ith m

obila

tor.

Pat

ient

seen

by

duty

doc

tor a

nd se

nt to

A

&E

for s

titch

es.

Spec

ial c

ateg

ory:

Clin

ical

judg

emen

t – ju

dgem

ent t

o en

cour

age

mob

ilisa

tion.

89

£5

.60

Patie

nt

Acc

iden

tal i

njur

y Fa

ll fr

om h

eigh

t 90

yea

r old

mal

e pa

tient

, reg

iste

red

blin

d, a

wai

ting

disc

harg

e to

nur

sing

hom

e ha

d fa

ll fr

om b

ed.

Cot

side

s in

posi

tion.

Inc

iden

t see

n by

two

staf

f but

they

wer

e to

o fa

r aw

ay to

inte

rven

e. S

uffe

red

frac

ture

d ne

ck o

f fem

ur, a

nd d

ied

with

in o

ne

to tw

o w

eeks

. Sp

ecia

l cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

pat

ient

und

erst

ood

conc

ept o

f buz

zer f

or h

elp,

cot

side

s in

plac

e.

90

£2.8

0 Pa

tient

A

ccid

enta

l inj

ury

Slip

, trip

or f

all o

n sa

me

leve

l or s

tairs

82

yea

r old

pat

ient

fell

in to

ilet,

unw

itnes

sed.

Pat

ient

had

his

tory

of b

eing

ra

tiona

l and

safe

with

zim

mer

then

like

“a

switc

h” c

ould

sudd

enly

bec

ome

angr

y,

aggr

essi

ve a

nd d

isor

ient

ated

& a

t the

sam

e tim

e un

stea

dy.

Spec

ial c

ateg

ory

of

outw

ith r

easo

nabl

e pr

actic

abili

ty –

all

reas

onab

le c

ontro

l mea

sure

s in

plac

e.

203

91

£8.5

8 Pa

tient

A

ccid

enta

l inj

ury

Slip

, trip

or f

all o

n sa

me

leve

l or s

tairs

Pa

tient

stoo

d up

from

cha

ir w

ith fo

ot tw

iste

d ro

und

chai

r leg

and

fell

to th

e flo

or.

Frac

ture

d fe

mur

. In

itial

ly re

fuse

d su

rger

y th

en h

as o

ther

com

plic

atio

n. S

pend

s 31

day

s in

acut

e ho

spita

l bef

ore

bein

g re

turn

ed to

war

d. S

peci

al c

ateg

ory:

re

ason

able

pra

ctic

abili

ty –

all

reas

onab

le c

ontro

l mea

sure

s in

plac

e.

Tabl

e 5

Trus

t E (4

inci

dent

s)

No

Cost

Staf

f/pat

ient

/oth

er

Cate

gory

Ty

pe

Des

crip

tion

1 £3

,833

.28

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or

fall

on sa

me

leve

l or

stai

rs

Patie

nt w

as b

eing

trea

ted

in su

rgic

al w

ard

but b

ed w

as n

eede

d so

was

tran

sfer

red

to a

n or

thop

aedi

c w

ard

whe

re sh

e w

as to

stay

ove

rnig

ht th

en b

e di

scha

rged

in th

e m

orni

ng.

The

next

da

y pa

tient

was

foun

d ly

ing

on th

e flo

or h

avin

g st

umbl

ed a

nd fa

llen.

Pat

ient

requ

ired

12 d

ay

exte

nded

stay

in o

rthop

aedi

c w

ard.

Spe

cial

cat

egor

y: r

isk n

ot fo

rese

eabl

e –

patie

nt w

as

inde

pend

ently

mob

ile a

nd d

ue to

go

hom

e, th

eref

ore

she

was

con

side

red

capa

ble

of v

isiti

ng th

e to

ilet a

lone

. 2

£5.3

1 Pa

tient

A

ccid

enta

l in

jury

Sl

ip, t

rip o

r fa

ll on

sam

e le

vel o

r st

airs

84 y

ear o

ld m

ale

patie

nt h

ad b

een

adm

itted

to th

e re

habi

litat

ion

unit

havi

ng b

eing

trea

ted

for a

fr

actu

red

neck

of f

emur

. Pa

tient

trie

d to

rise

from

a c

hair

unas

sist

ed a

nd o

verb

alan

ced

and

fell,

su

stai

ning

lace

ratio

ns to

the

face

. Sp

ecia

l cat

egor

y: o

utw

ith r

easo

nabl

e pr

actic

abili

ty –

pa

tient

kno

wn

to re

quire

ass

ista

nce

and

had

been

co-

oper

ativ

e in

the

past

. St

aff w

ere

clos

e by

bu

t cou

ld n

ot p

reve

nt th

e fa

ll.

3 £5

.12

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or

fall

on sa

me

leve

l or

stai

rs

Patie

nt re

ques

ted

hot d

rink.

Giv

en d

rink

and

seat

ed in

her

recl

iner

with

a b

edsi

de ta

ble

and

the

nurs

e ca

ll be

ll pl

aced

to h

and.

Fiv

e m

inut

es la

ter t

he st

aff h

eard

a n

oise

and

on

inve

stig

atio

n th

ey fo

und

the

patie

nt o

n th

e flo

or b

etw

een

the

bed

and

the

recl

iner

. Pa

tient

sust

aine

d tw

o la

cera

tions

on

her f

oreh

ead

and

the

back

of h

er sk

ull a

nd w

as se

nt fo

r x-r

ays.

Spe

cial

cat

egor

y:

outw

ith r

easo

nabl

e pr

actic

abili

ty –

pat

ient

was

kno

wn

to b

e hi

gh ri

sk, b

ut sh

e w

as n

ot

conf

used

and

refu

sed

to g

o to

bed

. St

aff a

ttend

ed fi

ve m

inut

es b

efor

e in

cide

nt.

4 £7

.84

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or

fall

on sa

me

leve

l or

stai

rs

Into

xica

ted

patie

nt a

dmitt

ed to

A&

E. P

atie

nt le

ft th

e cu

bicl

e, b

ecom

ing

enta

ngle

d in

the

curta

in

as h

e di

d so

, fal

ling

to th

e gr

ound

bef

ore

staf

f cou

ld re

ach

him

. Sp

ecia

l cat

egor

y: o

utw

ith

reas

onab

le p

ract

icab

ility

– m

easu

res w

ere

in p

lace

to p

reve

nt in

cide

nt, b

ut w

as la

rgel

y du

e to

in

toxi

cate

d st

ate

of p

atie

nt.

204

Tabl

e 6

Trus

t F (9

inci

dent

s)

No

Cost

Staf

f/pat

ient

/oth

er

Cate

gory

Ty

pe

Des

crip

tion

1 £1

,289

.32

Oth

er

Nea

r mis

s St

ruck

by

an

obje

ct

Cei

ling

had

been

fitte

d w

ithin

the

last

2 y

ears

. O

n da

y of

inci

dent

ther

e w

as h

igh

win

ds a

nd

wea

ther

con

ditio

ns w

ere

exce

ptio

nally

seve

re.

A w

indo

w in

the

staf

f roo

m c

ould

not

be

prop

erly

clo

sed

and

whe

n th

e w

ind

blew

in it

was

thou

ght t

hat a

cur

rent

of a

ir go

t int

o ce

iling

spac

e. T

he c

eilin

g co

llaps

ed in

secr

etar

ies’

off

ice

brin

ging

dow

n lig

ht fi

tting

s.

Spec

ial c

ateg

ory:

lack

of i

nfor

mat

ion

– no

info

rmat

ion

coul

d be

gai

ned

on th

e st

anda

rds

for f

ittin

g ce

iling

s. H

ad b

een

fitte

d by

con

tract

ors,

who

wer

e no

long

er c

onta

ctab

le.

2 £1

,012

.80

Patie

nt

Acc

iden

tal

inju

ry

Slip

77

yea

r old

pat

ient

foun

d ly

ing

on th

e flo

or.

X-r

ay sh

owed

frac

ture

to h

ip.

Patie

nt

trans

ferr

ed to

orth

opae

dic

depa

rtmen

t for

hip

repl

acem

ent r

esul

ting

in a

n up

grad

ed st

ay o

f ei

ght d

ays (

durin

g st

udy

perio

d).

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

–

patie

nt n

orm

ally

inde

pend

ently

mob

ile.

3 £3

,798

.24

Staf

f A

ccid

enta

l in

jury

Sl

ip

Clin

ical

Sup

port

Wor

ker s

lippe

d on

urin

e on

floo

r nea

r bed

spac

e. S

he w

as se

nt to

A&

E an

d th

en h

ome.

Sic

k le

ave

for a

ppro

x. th

ree

mon

ths a

nd w

as re

plac

ed b

y an

age

ncy

nurs

e.

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– th

e w

ettin

g on

the

floor

was

un

avoi

dabl

e an

d ac

cord

ing

to th

e in

form

atio

n av

aila

ble

no o

ne k

new

whe

n th

e flo

or b

ecam

e w

et so

pro

cedu

re fo

r dea

ling

with

spill

age

was

not

initi

ated

. D

im li

ghtin

g w

as a

lso

esse

ntia

l for

pat

ient

s to

slee

p.

4 £7

.10

Patie

nt

Acc

iden

tal

inju

ry

Slip

83

yea

r old

mal

e pa

tient

got

up

to u

se th

e bo

ttle.

Sta

ff h

eard

a lo

ud b

ang

and

they

foun

d th

e pa

tient

lyin

g on

the

floor

by

his b

ed.

He

was

ass

iste

d ba

ck to

bed

by

hois

t. T

he S

HO

st

itche

d la

cera

tion

and

sent

pat

ient

for s

kull

x-ra

y (e

scor

ted

by D

gra

de st

aff n

urse

).

Spec

ial c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– p

atie

nt’s

con

fusi

on w

as a

n in

crea

sing

pro

blem

and

he

pers

iste

d on

tryi

ng to

get

out

of b

ed b

y cl

imbi

ng o

ver c

ot si

des

or g

oing

to b

otto

m o

f bed

. 5

£42.

85

Oth

er

Oth

er

Nee

dles

tick/

sh

arps

W

hile

gra

fting

pat

ient

in c

ardi

ac th

eatre

the

surg

eon

notic

ed th

at o

ne o

f the

sutu

re n

eedl

es

was

mis

sing

. A

thor

ough

sear

ch w

as c

arrie

d ou

t and

the

rout

ine

ches

t x-r

ay w

as re

view

ed

as p

er h

ospi

tal p

olic

y. N

eedl

e w

as n

ot o

bser

ved

on x

-ray

. Sp

ecia

l cat

egor

y: la

ck o

f in

form

atio

n –I

mpo

ssib

le to

asc

erta

in o

n th

e ba

sis o

f the

info

rmat

ion

avai

labl

e.

6 £6

.00

Oth

er

Oth

er

Oth

er

ENT

Con

sulta

nt re

quire

d an

alte

rnat

ive

trach

eoto

my

tray

be b

roug

ht b

y ta

xi fr

om a

noth

er

hosp

ital t

o pe

rfor

m e

mer

genc

y pr

oced

ure.

Spe

cial

cat

egor

y: r

isk u

nfor

esee

able

– th

e st

erili

sing

cen

tre p

repa

red

inst

rum

ent t

rays

in a

ccor

danc

e w

ith n

orm

al re

ques

ts a

nd w

ere

not a

war

e th

at th

is c

onsu

ltant

requ

ired

a sp

ecia

l ins

trum

ent t

o co

nduc

t the

pro

cedu

re.

205

7 £7

.45

Patie

nt

Acc

iden

tal

inju

ry

Patie

nt li

fting

/ H

andl

ing

Patie

nt su

stai

ned

a sh

ould

er in

jury

whi

lst b

eing

rolle

d on

bed

into

hoi

st sl

ing.

Spe

cial

cate

gory

: risk

unf

ores

eeab

le –

clin

ical

con

ditio

n of

pat

ient

not

fully

app

reci

ated

at t

he

time

of th

e in

cide

nt.

Had

staf

f kno

wn

that

the

patie

nt’s

can

cer h

ad m

etas

tasi

sed

to b

ones

th

ey w

ould

n’t h

ave

hois

ted.

8

£238

.65

Oth

er

Fire

Ex

posu

re to

fire

Fr

idge

in w

ard

caug

ht fi

re.

Poss

ible

cau

se sp

ilt m

ilk ru

nnin

g fr

om b

ase

of c

abin

et b

etw

een

case

and

fasc

ia c

over

. Fi

re b

rigad

e w

as c

alle

d al

thou

gh tw

o st

aff n

urse

s wer

e ab

le to

ex

tingu

ish

the

fire.

Spe

cial

cat

egor

y of

risk

unf

ores

eeab

le –

staf

f una

war

e th

at m

ilk h

ad

been

spilt

. 9

£200

.00

Oth

er

Oth

er

Oth

er

Win

dow

was

blo

wn

in b

y w

ind

smas

hing

ove

r an

unoc

cupi

ed b

ed a

nd th

e flo

or in

war

d.

Esta

tes w

ere

cont

acte

d to

secu

re w

indo

w sa

fely

unt

il co

ntra

ctor

s arri

ved

to re

plac

e gl

ass.

Sp

ecia

l cat

egor

y: r

isk u

nfor

esee

able

– b

ased

on

the

fact

s kno

wn

risk

coul

d no

t be

fore

seen

.

Tabl

e 7

Trus

t G (1

6 in

cide

nts)

No

Cost

Staf

f/pat

ient

/oth

er

Cate

gory

Ty

pe

Des

crip

tion

1 £2

9.52

Pa

tient

A

ccid

enta

l in

jury

Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r st

airs

Patie

nt in

depe

nden

tly m

obile

with

zim

mer

usu

ally

man

aged

to g

et a

roun

d on

her

ow

n.

She

was

foun

d on

floo

r at h

er b

edsi

de e

ntan

gled

in h

er z

imm

er fr

ame

suff

erin

g fr

om

pain

in h

er h

ip.

X-r

ay c

onfir

med

frac

ture

. Sp

ecia

l cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

pat

ient

, alth

ough

eld

erly

, was

inde

pend

ently

mob

ile w

ith h

er z

imm

er

fram

e.2

£15.

78

Patie

nt

Oth

er

Oth

er

Patie

nt le

ft un

lock

ed w

ard

with

out p

erm

issi

on.

Foun

d in

car

par

k an

d re

turn

ed to

war

d by

firm

per

suas

ion.

Pat

ient

had

epi

leps

y/br

ain

dam

age/

chal

leng

ing

beha

viou

r. S

peci

alca

tego

ry: C

linic

al ju

dgem

ent –

pat

ient

not

judg

ed to

nee

d co

nsta

nt o

bser

vatio

n.

3 £4

.92

Patie

nt

Oth

er

Oth

er

Patie

nt tr

ied

to le

ave

war

d bu

t per

suad

ed to

retu

rn.

Patie

nt ra

n of

f whi

le a

ccom

pani

ed a

t fr

ont d

oor (

to sm

oke

a ci

gare

tte).

Ala

rm ra

ised

, pat

ient

foun

d on

road

and

per

suad

ed to

re

turn

to w

ard.

Spe

cial

cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

all

appr

opria

te m

easu

res

in p

lace

to c

ontro

l ris

k, c

omm

ensu

rate

with

pat

ient

car

e.4

£16.

80

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

Patie

nt fe

ll ou

t of b

ed w

rapp

ed in

bed

ding

. Pr

one

to w

ande

ring

and

did

not u

nder

stan

d us

e of

buz

zer d

ue to

men

tal s

tate

. Sp

ecia

l cat

egor

y: r

easo

nabl

e pr

actic

abili

ty.

Ris

k as

sess

men

t con

clud

ed th

at th

ere

was

no

need

for c

ot si

des o

r ext

ra su

perv

isio

n as

pat

ient

ha

d no

his

tory

of f

allin

g fr

om b

ed.

5 £4

,130

.08

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

Patie

nt d

ue to

be

disc

harg

ed fe

ll at

bed

side

. C

omm

ode

tang

led

up in

falle

n kn

icke

rs.

Frac

ture

con

firm

ed.

Add

ition

al 3

4 da

y st

ay o

n w

ard.

Spe

cial

cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

pat

ient

was

inde

pend

ently

mob

ile &

fit.

206

6 £3

00.0

0 St

aff

Phys

ical

V

iole

nce

Pers

on to

Pe

rson

Ass

ault

Nur

se w

as p

unch

ed b

y re

side

nt w

hen

she

was

tryi

ng to

get

the

resi

dent

to d

ress

her

self.

Th

is c

ause

d da

mag

e to

the

nurs

e’s s

pect

acle

s. S

peci

al c

ateg

ory:

rea

sona

ble

prac

ticab

ility

– P

atie

nt la

shed

out

with

out w

arni

ng.

7 £2

1.50

Pa

tient

V

erba

l abu

se o

r th

reat

enin

g be

havi

our

Oth

er

Doc

tor a

dvis

ed p

atie

nt th

at, b

ecau

se h

is c

ondi

tion

had

dete

riora

ted

he w

as to

be

trans

ferr

ed to

an

inte

nsiv

e ps

ychi

atric

car

e un

it (I

PCU

). P

atie

nt re

acte

d to

the

deci

sion

w

ith h

ostil

ity a

nd a

ggre

ssio

n re

quiri

ng th

e st

aff t

o se

cure

the

situ

atio

n w

ith C

&R

te

chni

ques

. Sp

ecia

l cat

egor

y: r

easo

nabl

e pr

actic

abili

ty. A

ppro

pria

te ri

sk a

sses

smen

t ha

d be

en c

arrie

d ou

t. 8

£150

.17

Patie

nt

Phys

ical

vi

olen

ce

Oth

er

Dec

isio

n ha

d be

en ta

ken

to m

ove

a pa

tient

to a

secu

re w

ard.

On

bein

g to

ld th

is a

nd

whi

lst b

eing

inje

cted

with

med

icat

ion,

pat

ient

star

ted

to th

row

him

self

abou

t and

be

com

e ve

ry v

iole

nt.

Mem

ber o

f sta

ff in

jure

d hi

s han

d w

hils

t try

ing

to re

stra

in p

atie

nt.

Spec

ial c

ateg

ory:

rea

sona

ble

prac

ticab

ility

– R

isk

was

reco

gnis

ed, p

atie

nt w

as in

pr

oces

s of b

eing

tran

sfer

red.

9

£17.

54

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

Patie

nt (s

troke

reha

b.) a

ssis

ted

to b

edsi

de c

omm

ode

and

left

with

buz

zer.

Whe

n st

aff

cam

e ba

ck to

che

ck fo

und

patie

nt ly

ing

on th

e flo

or.

Frac

ture

con

firm

ed re

sulti

ng in

an

exte

nded

stay

on

war

d. S

peci

al c

ateg

ory:

out

with

rea

sona

ble

prac

ticab

ility

– b

uzze

r to

han

d, p

atie

nt c

apab

le o

f usi

ng it

but

app

ears

she

may

hav

e st

ood

hers

elf u

p an

d th

en

lost

bal

ance

. 10

£3

5.08

Pa

tient

A

ccid

enta

l in

jury

Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r st

airs

Patie

nt re

cove

ring

from

stro

ke w

alke

d un

aide

d to

toile

t to

prac

tice

wal

king

. Se

en to

be

falli

ng a

nd g

ently

low

ered

to fl

oor b

y tw

o m

embe

rs o

f sta

ff.

Patie

nt su

stai

ned

inju

ry to

an

kle

and

was

sent

for x

-ray

. Sp

ecia

l cat

egor

y: C

linic

al ju

dgem

ent –

pat

ient

bei

ng

reha

bilit

ated

inju

red

ankl

e w

hile

wal

king

inde

pend

ently

for s

hort

dist

ance

und

er

obse

rvat

ion

of st

aff.

11

£2.8

9 Pa

tient

O

ther

O

ther

79

yea

r old

pat

ient

with

dem

entia

left

war

d un

obse

rved

by

staf

f. P

atie

nt w

as fo

und

in

grou

nds a

nd re

turn

ed to

war

d. S

peci

al c

ateg

ory:

risk

not

fore

seea

ble

– un

char

acte

ristic

beh

avio

ur d

ue to

exc

item

ent o

f goi

ng h

ome.

12

£4

,515

.00

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

Patie

nt m

obile

with

use

of z

imm

er w

alke

d al

ong

corr

idor

and

left

zim

mer

out

side

toile

t.

Whe

n fin

ishe

d st

ood

up a

nd fe

ll to

floo

r. T

aken

to a

cute

hos

pita

l and

retu

rned

to w

ard

for e

xten

ded

stay

. Sp

ecia

l cat

egor

y: r

isk n

ot fo

rese

eabl

e/cl

inic

al ju

dgem

ent –

pat

ient

di

d no

t req

uire

ass

ista

nce,

mob

ile w

ith z

imm

er.

Staf

f did

not

fore

see

patie

nt le

avin

g zi

mm

er o

utsi

de to

ilet.

13

£21.

93

Patie

nt

Acc

iden

tal

inju

ry

Slip

, trip

or f

all

on sa

me

leve

l or

stai

rs

Patie

nt w

ith A

lzhe

imer

’s, c

onst

antly

shou

ting

rose

from

bed

and

fell

to fl

oor.

Pat

ient

di

d no

t com

preh

end

use

of b

uzze

r. S

ent f

or x

-ray

con

firm

ing

frac

ture

d hu

mer

us a

nd

retu

rned

to w

ard.

Spe

cial

cat

egor

y: r

easo

nabl

e pr

actic

abili

ty –

pat

ient

und

er g

ener

al

obs.

Inde

pend

ently

mob

ile w

ith z

imm

er b

ut re

quire

d as

sist

ance

get

ting

in &

out

of b

ed.

207

14

£89.

40

Patie

nt

Van

dalis

m

Expo

sure

to fi

re

Patie

nt se

ts li

ght t

o he

r bed

line

n an

d ac

tivat

es fi

re a

larm

. Pa

tient

adm

its to

star

ting

fire

to g

ain

atte

ntio

n. F

ire b

rigad

e at

tend

but

fire

was

ext

ingu

ishe

d by

NA

with

fire

ex

tingu

ishe

r. E

ngin

eer c

alle

d in

to re

set f

ire a

larm

. Sp

ecia

l cat

egor

y: r

isk

unfo

rese

eabl

e –

first

tim

e pa

tient

had

set l

ight

to a

nyth

ing.

15

£9

96.8

0 Pa

tient

A

ccid

enta

l in

jury

Sl

ip, t

rip o

r fal

l on

sam

e le

vel o

r st

airs

Patie

nt ro

se fr

om b

ed w

ith m

inim

um a

ssis

tanc

e. W

alke

d to

toile

t una

ided

, tur

ned

to

clos

e do

or, l

ost b

alan

ce a

nd fe

ll. T

rans

ferr

ed to

acu

te h

ospi

tal f

or x

-ray

, fra

ctur

e co

nfirm

ed a

nd d

etai

ned

2 da

ys b

efor

e re

turn

ing

to w

ard

for e

xten

ded

stay

. Sp

ecia

lca

tego

ry: r

easo

nabl

e pr

actic

abili

ty –

pat

ient

inde

pend

ently

mob

ile a

nd n

o in

dica

tion

of in

crea

sed

risk

of fa

lling

. 16

£3

9.69

Pa

tient

V

anda

lism

O

ther

Pa

tient

retu

rned

to ro

om a

fter l

unch

. N

oise

hea

rd fr

om ro

om, w

indo

w fo

und

to b

e br

oken

. Pa

tient

agi

tate

d fo

r no

appa

rent

reas

on.

Patie

nt w

as re

stra

ined

, rem

oved

from

be

droo

m a

nd m

edic

atio

n gi

ven.

Spe

cial

cat

egor

y: C

linic

al ju

dgem

ent –

no

indi

catio

n to

alte

r obs

erva

tion

regi

me.

208

APPENDIX 9 INDIVIDUAL TRUSTS BREAKDOWN OF TELEPHONE INTERVIEW RESPONSES

Figure 1 All Trusts – workbook holder telephone interviews feedback from three monitoring periods

0

10

20

30

40

50

60

70

80

90

100Ar

e yo

u st

ill th

e ho

lder

of

the

wor

kboo

k?

Hav

e yo

u us

ed th

ew

orkb

ook?

Hav

e yo

u fo

und

the

wor

kboo

k he

lpfu

l?

Hav

e yo

u us

ed th

em

anag

er's

aud

it?

Hav

e yo

u us

ed th

ew

orkb

ook

gene

rally

?

Hav

e yo

u fo

cuss

ed in

on

parti

cula

r sec

tions

?

Hav

e yo

u im

plem

ente

dan

y co

ntro

l mea

sure

s as

are

sult

of a

risk

asse

ssm

ent?

Hav

e yo

u us

ed th

e op

tion

appr

aisa

l?

Hav

e yo

u an

y su

gges

tions

for i

mpr

ovem

ents

?

Question Posed

Perc

enta

ge o

f Wor

kboo

k H

olde

rs th

at R

eplie

d Ye

s (o

ut o

f 177

issu

ed)

1st April to 31st July 2001 1st August to 30th November 2001 1st December 2001 to 31st March 2002

209

Figure 2 Trust B

0

20

40

60

80

100

Are

you

still

the

hold

er o

f the

wor

kboo

k?

Hav

e yo

u us

ed th

e w

orkb

ook?

Hav

e yo

u fo

und

the

wor

kboo

k he

lpfu

l?

Hav

e yo

u us

ed th

e m

anag

er's

aud

it?

Hav

e yo

u us

ed th

e w

orkb

ook

gene

rally

?

Hav

e yo

u us

ed p

artic

ular

sec

tions

?

Hav

e yo

u im

plem

ente

d an

y co

ntro

l mea

sure

sfo

llow

ing

risk

asse

ssm

ent?

Hav

e yo

u us

ed th

e op

tion

appr

aisa

l?

Hav

e yo

u an

y su

gges

tions

for i

mpr

ovem

ents

?Question Posed

Perc

enta

ge o

f Wor

kboo

k H

olde

rs S

urve

yed

that

Rep

lied

Yes

(Out

of 2

1, 2

1 &

11

resp

ectiv

ely)


210

Figure 3 Trust C

0

20

40

60

80

100Ar

e yo

u st

ill th

eho

lder

of t

hew

orkb

ook?

Hav

e yo

u fo

und

the

wor

kboo

khe

lpfu

l?

Hav

e yo

u us

edth

e w

orkb

ook

gene

rally

?

Hav

e yo

uim

plem

ente

dan

y co

ntro

lm

easu

res

Hav

e yo

u an

ysu

gges

tions

for

impr

ovem

ents

?

Question Posed

Perc

enta

ge o

f Wor

kboo

k H

olde

rs th

at R

eplie

d Y

es (o

ut o

f 19,

25

& 1

9 re

spec

tivel

y)


211

Figure 4 Trust D

0

20

40

60

80

100

Are

you

still

the

hold

er o

fth

e w

orkb

ook?

Hav

e yo

u us

ed th

ew

orkb

ook?

Hav

e yo

u fo

und

the

wor

kboo

k he

lpfu

l?

Hav

e yo

u us

ed th

em

anag

er's

aud

it?

Hav

e yo

u us

ed th

ew

orkb

ook

gene

rally

?

Hav

e yo

u us

edpa

rticu

lar s

ectio

ns?

Hav

e yo

u im

plem

ente

dan

y co

ntro

l mea

sure

sfo

llow

ing

risk

asse

ssm

ent?

Hav

e yo

u us

ed th

eop

tion

appr

aisa

l?

Hav

e yo

u an

ysu

gges

tions

for

impr

ovem

ents

?

Question Posed

Perc

enta

ge o

f Wor

kboo

k H

olde

rs th

at R

eplie

d Ye

s(o

ut o

f 38,

37

& 2

8 re

spec

tivel

y)


212

Figure 5 Trust E

0102030405060708090

100

Are

you

still

the

hold

er o

f the

wor

kboo

k?

Hav

e yo

u us

ed th

e w

orkb

ook?

Hav

e yo

u fo

und

the

wor

kboo

k he

lpfu

l?

Hav

e yo

u us

ed th

e m

anag

er's

aud

it?

Hav

e yo

u us

ed th

e w

orkb

ook

gene

rally

?

Hav

e yo

u us

ed p

artic

ular

sec

tions

?

Hav

e yo

u im

plem

ente

d an

y co

ntro

lm

easu

res

follo

win

g ris

k as

sess

men

t?

Hav

e yo

u us

ed th

e op

tion

appr

aisa

l?

Hav

e yo

u an

y su

gges

tions

for

impr

ovem

ents

?

Question Posed

Perc

enta

ge o

f Wor

kboo

k H

olde

rs

that

Rep

lied

Yes


213

REFERENCES

1. NHS Scotland Partnership Forum. Managing health at work. Edinburgh: Partnership Information Network (PIN) Guideline; 2003.

2. Health and Safety Executive. http://www.hse.gov.uk. In; 2003. 3. Health and Safety Executive. Successful Health and Safety Management. first ed: HSE

Books; 1991. 4. Health and Safety Executive. Successful Health and Safety Management. Second ed.

Sudbury: Health and Safety Executive; 1997. 5. Health and Safety Executive. Reducing error and influencing behaviour. Sudbury: HSE

Books; 1999. 6. Gay AS, New NH. Auditing health and safety management systems: a regulator's view.

Occup. Med. 1999;49(7):471-473. 7. Health and Safety Executive. The costs of accidents at work. Sudbury: HSE Books;

1993. 8. Livingston AD, Jackson G, Priestley K. Root causes analysis: Literature review.

Sudbury: HSE Books; 2001. 9. Niven KJM. Accident Costs in the NHS. The Safety & Health Practitioner

1999;September:34-38. 10. National Audit Office. Health and safety in NHS acute hospital Trusts in England.

Report by the Comptroller and Auditor General. London; 1996. 11. Revitalising Health and Safety. http://www.hse.gov.uk/revitalising/. In; 2000. 12. Securing Health Together. http://www.ohstrategy.net/. 2000. 13. Health and Safety Commission. Health and safety targets: How are we doing? A

supplement to the HSC annual report and HSC/E accounts 2001/02. Sudbury: HSE Books; 2002.

14. Department of Health. An organisation with a memory: report of an expert group on learning from adverse events in the NHS. London: The Stationary Office; 2000.

15. Department of Health. Building a safer NHS for patients. www.doh.gov.uk/buildsafenhs/;2001.

16. Department of Health. Doing less harm: improving the safety and quality of care through reporting, analysing and learning from adverse incidents involving NHS patients - key requirements for health care providers: National Patient Safety Agency; 2001.

17. Department of Health. Preventing accidental injury - Priorities for action. Norwich: The Stationary Office ISBN 0-11-322477-X; 2002.

18. Department of Health. The effective management of occupational health and safety services in the NHS. London: The Stationary Office; 2002.

19. Scottish Executive. Towards a safer healthier workplace: http://www.scotland.gov.uk/library2/doc08/shwm-00.htm; 1999.

20. NHS Executive. NHS Indemnity - Arrangement for handling clinical negligence claims against NHS staff. HSG(96)48: http://www.info.doh.gov.uk/doh/coin4.nsf/Circulars?ReadForm; 1996.

21. Scottish Executive. Clinical negligence and other risks indemnity scheme (CNORIS). NHS HDL(2001)65: http://www.show.scot.nhs.uk/sehd/mels/hdl2001_65.htm; 2001.

22. National Institute for Occupational Safety and Health. Guide to evaluating the effectiveness of strategies for preventing work injuries: how to show whether a safety intervention really works. Cincinnati: NIOSH; 2001.

23. Zwerling C. A review of a guide to evaluating the effectiveness of strategies for preventing work injuries: How to show whether a safety intervention really works. Journal of Safety Research 2001;32:501-503.

214

24. Clarke M, Oxman AD, editors. Cochrane reviewers handbook 4.1.4 [Updated October 2001]. Oxford: Update Software: Updated Quarterly; 2001.

25. Crombie IK, Davies HTO. Research in healthcare: Design, conduct and interpretation of health services research. Chichester: Wiley; 1996.

26. Fagard RH, Staessen JA, Thijs L. Advantages and disadvantages of the meta-analysis approach. J. Hypertens. Suppl. 1996;14(2):S9-12.

27. Crombie IK. The pocket guide to critical appraisal: a handbook for health care professionals. London: BMJ Publishing Group; 1996.

28. Rakel H, Gerrard S, Langford I, Shaw K. Evaluating the impact of contact techniques. Sudbury: HSE Books; 1999.

29. Stufflebeam DL. Evaluation checklists: Practical tools for guiding and judging evaluations. American Journal of Evaluation 2001;22(1):71-79.

30. Easterby-Smith M, Thorpe R, Lowe A. Management research: An introduction. London: Sage Publications Ltd.; 1991.

31. Shannon HS, Robson LS, Guastello SJ. Methodological criteria for evaluating occupational safety intervention research. Safety Science 1999;31:161-179.

32. Smallman C. The reality of "Revitalizing Health and Safety". Journal of Safety Research 2001;32:391-439.

33. DeRoo LA, Rautiainen RH. A systematic review of farm safety interventions. Am. J. Prev. Med. 2000;18(4S):51-62.

34. Chilton S, Jones-Lee M, Loomes G, Robinson A, Cookson R, Covey J, et al. Valuing health and safety controls: A literature review. Sudbury: HSE Books; 1998.

35. ACSNI Study Group on Human Factors. Third report: Organising for Safety. Sudbury: HSE Books; 1998.

36. Nivolianitou Z, Oberhagemann D, Lunn G, Markert F, Nessvi K, Rogers R, et al. SAFETYNET - a European network for process safety. Journal of Hazardous Materials 2001;A87:1-10.

37. Oliver D, Hopper A, Seed P. Do hospital fall prevention programs work? A systematic review. J Am Geriatr Soc 2000;48(12):1679-1689.

38. Krause TR, Seymour KJ, Sloat KCM. Long-term evaluation of a behaviour-based method for improving safety performance: a meta-analysis of 73 interrupted time-series replications. Safety Science 1999;32:1-18.

39. Cooper MD. Towards a model of safety culture. Safety Science 2000;36:111-136. 40. Scotney V. Development of a health and safety performance measurement tool.

Sudbury: HSE Books; 2000. 41. Yassi A, Tate R, Cooper JE, Snow C, Vallentyne S, Khokhar JB. Early intervention for

back injured nurses at a large Canadian tertiary care hospital: an evaluation of the effectiveness and cost benefits of a two-year pilot project. Occup. Med. 1995;45(4):209-214.

42. Haslam RA, Bentley TA. Follow-up investigations of slip, trip and fall accidents among postal delivery workers. Safety Science 1998;32:33-47.

43. Ludbrook A, Godfrey C, Wyness L, Parrott S, Haw S, Napper M, et al. Effective and cost-effective measures to reduce alcohol misuse in Scotland: A literature review. Edinburgh: The Stationary Office; 2001.

44. Harper AC, Cordery JL, de Klerk NH, Sevastos P, Geelhoed E, Gunson C, et al. Curtin industrial safety trial: managerial behaviour and program effectiveness. Safety Science 1996;24(3):173-179.

45. Volinn E. Do workplace interventions prevent low-back disorders? If so, why?: a methodologic commentary. Ergonomics 1999;42(1):258-272.

46. Gressel MG. An engineer's perspective of the intervention research workshop. Am J Ind Med 1996;29:382-383.

47. Guldenmund FW. The nature of safety culture: a review of theory and research. Safety Science 2000;34:215-257.

215

48. Hurst NW, Young S, Gibson H, Muyselaar A. Measures of safety performance and attitudes to safety at major hazard sites. Journal of Loss Prevention in the Process Industry 1996;9(2):161-172.

49. Cabrera DD, Isla R, Viledla LD. An evaluation of safety climate in ground handling activities. In: Soekkha HM, editor. Aviation Safety; 1997 27-29 August; Proceedings of the IASC-97 International Aviation Safety Conference, Netherlands; 1997. p. 255-268.

50. Cox SJ, Cheyne AJT. Assessing safety culture in offshore environments. Safety Science 2000;34:111-129.

51. Flin R, Mearns K, O'Connor P, Bryden R. Measuring safety climate: identifying the common features. Safety Science 2000;34:177-192.

52. Mearns K, Whitaker SM, Flin R. Safety climate, safety management practice and safety performance in offshore environments. Safety Science 2002;In Press.

53. St. Vincent M, Tellier C, Petitjean-Roget T. Accidents that occurred in three hospitals in one year. Safety Science 1999;31:197-212.

54. McGrail MP, Tsai SP, Bernacki EJ. A comprehensive initiative to manage the incidence and cost of occupational injury and illness: report of an outcomes analysis. J. Occup. Environ. Med. 1995;37(11):1263-1268.

55. Bracker A, Blumberg J, Hodgson M, Storey E. Industrial hygiene recommendations as interventions: A collaborative model within occupational medicine. Applied Occupational & Environmental Hygiene 1999;14:85-96.

56. Smallman C, John G. British directors perspectives on the impact of health and safety on corporate performance. Safety Science 2001;38:227-239.

57. Rundmo T, Hale AR. Managers' attitudes towards safety and accident prevention. Safety Science 2002;In press.

58. Harper AC, Gunson C, Robinson L, de Klerk NH, Osborn D, Sevastos P, et al. Curtin industrial safety trial: methods and safe practice and housekeeping outcomes. Safety Science 1996;24(3):159-172.

59. Hauer E. Observational before-after studies in road safety. Estimating the effect of highway and traffic engineering measures on road safety. Oxford: Elsevier Science Ltd. ISBN 0-08-043-053-8; 1997.

60. Vassie LH, Lucas WR. An assessment of health and safety management within working groups in the UK manufacturing sector. Journal of Safety Research 2001;32(479-490).

61. Hayes BE, Perander J, Smeko T, Trask J. Measuring perceptions of workplace safety: development and validation of the work safety scale. Journal of Safety Research 1998;29(3):145-161.

62. Lingard H, Rowlinson S. Behaviour-based safety management in Hong Kong's construction industry. Journal of Safety Research 1997;28(4):243-256.

63. Henderson J, Whittington C, Wright K. Accident investigation - the drivers, methods and outcomes. Sudbury: HSE Books; 2001.

64. Scherer RF, Brodzinski JD, Canty AL. An examination of process and outcome differences in health and safety inspections conducted by state and federal agencies. Journal of Safety Research 1997;28(3):203-211.

65. Cowan J. Achieving consistency in grading adverse clinical incidents: does Doing Less Harm do the job? British Journal of Clinical Governance 2002;7(1):63-67.

66. Spangenberg S, Baarts C, Dyreborg J, Jensen L, Kines P, Mikkelsen KL. Factors contributing to the differences in work related injury rates between Danish and Swedish construction workers. Safety Science 2002;in press.

67. Parker SK, Jackson PR, Sprigg CA, Whybrow AC. Organisational interventions to reduce the impact of poor work design. Sudbury: HSE Books; 1998.

68. Niven KJM. A review of the application of health economics to health and safety in healthcare. Health Policy 2002;61(3):291-304.

69. Niven KJM. Economic principles in occupational health and safety. Occupational Health Review 2000;November/December:13-18.

216

70. Health and Safety Commission, editor. Management of Health and Safety at Work Regulations (MHSWR): The Stationary Office; 1999.

71. Health and Safety Commission. Proposals for a new duty to investigate accidents, dangerous occurrences and diseases. Sudbury: HSE Books (www.hes.gov.uk/condocs/);2001.

72. Centre for Hazard and Risk Management. Offshore safety climate assessment toolkit. In. http://www.lboro.ac.uk/departments/ec/JIP/ ed: Loughborough University; 2000.

73. Kelly DL. How to use control charts for healthcare. Milwaukee: American Society for Quality (ASQ).

74. Carey RG, Lloyd RC. Measuring quality improvement in healthcare: A guide to statistical process control applications. Milwaukee: Amercian Society for Quality (ASQ).

75. Duncan AJ. Quality Control and Industrial Statistics. Homewood, Illinois: Irwin; 1986. 76. Statistical Package for the Social Sciences. SPSS version 10. In. version 10 ed. Woking:

http://www.spss.com/uk; 2001. 77. Emory CW, Cooper DR. Chapter 6 Measurement. In: Business Research Methods:

Irwin; 1991. 78. Siegel S. Nonparametric Statistics for the Behavioural Sciences. In: P.32. New York:

McGraw-Hill; 1956. 79. Anderson NH. Scales and Statistics: Parametric and Nonparametric. Psychological

Bulletin 1961;58(4):305-316. 80. Miles MB, Huberman AM. Qualitative data analysis: an expanded source book. 2nd ed.

London: Sage; 1994. 81. Diamantopoulos A, Schlegelmilch BB. Taking the fear out of data analysis. 1st Edition

ed. London: Business Press; 1997. 82. Neter J, Kutner MH, Nachtsheim CJ, Wassweman W. Applied Linear Statistical

Models. 3rd ed: Irwin; 1996. 83. Kerlinger FN, Lee HB. Foundations of Behavioural Research. 4th ed. Fort Worth, TX:

Harcourt College Publishers; 2000. 84. Health and Safety Executive. A Guide to the Reporting of Injuries, Diseases and

Dangerous Occurrences Regulations 1995 (RIDDOR). Sudbury: HSE Books; 1996. 85. Health and Safety Executive. The Costs of Accidents at Work. Second ed. Sudbury:

Health and Safety Executive; 1997. 86. Slack N, Chambers S, Harland C, Harrison A, Johnston R. Operations Management.

London: Pitman Publishing; 1995. 87. Neder, Wasserman, Kutner. Applied Linear Statistical Models (3rd Edition). 88. Emory CW, Cooper DR. Business Research Methods (4th Edition). 4th Edition ed:

Irwin; 1991. 89. http://www.richland.cc.il.us/james/lecture/m113/post_anova.html. In. 90. De Bono E. Six thinking hats. London: Penguin ISBN 0-14-013784-X; 1985. 91. Levenstein C. Policy Implications of intervention research: research on the social

context for intervention. Am J Ind Med 1996;29:358-361. 92. Adams SJ. Projecting the next decade in safety management: A Delphi technique study.

Professional Safety 2001;October:26-29. 93. Wagenaar WA, van der Schrier J. Accident analysis. The goal and how to get there.

Safety Science 1997;26(1/2):25-33. 94. Culvenor J. Comparison of team and individual judgements of solutions to safety

problems. Safety Science 2003;In Press. 95. Falconer L, Hoel H. Occupational safety and health: A method to test the collection of

"grey data" by line managers. Occup. Med. 1997;47(2):81-89. 96. Davies M, Faulkner D. Evaluating health and safety inspection in EU candidate

countries. Journal of the Institution of Occupational Safety and Health 2002;6(2):65-78.

217

97. Swuste P, Arnoldy F. The safety adviser/manager as an agent of organisational change: a new challenge to expert training. Safety Science 2003;41:15-27.

98. Hawkins J, Booth RT. Safety and health management system guidance. Journal of the Institution of Occupational Safety and Health 1998;2(2):7-24.

99. Wynn PA, Aw TC, Williams NR, Harrington M. Teaching of occupational medicine to undergraduates in UK schools of medicine. Medical Education 2002;36:697-701.

100. Whitaker S, Wynn P, Williams N. Occupational health teaching for pre registration nursing students. Nurse Education Today 2002;22:152-158.

101. Health and Safety Commission. Management of health and safety at work regulations; 1992.

102. Hale AR. Occupational health and safety professionals and management: identity, marriage, servitude or supervision? Safety Science 1995;20:233-245.

103. Hale AR. Conditions of occurrence of major and minor accidents. Journal of the Institution of Occupational Safety and Health 2001;5(1):7-21.

104. Karageorgiou A, Jensen PL, Walters D, Wilthagen T. Risk assessment in four member states of the European Union. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 251-284.

105. Dorman P. If safety pays, why don't employers invest in it? In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 351-365.

106. Health and Safety Commission Health Services Advisory Committee. Management of health and safety in the health services - information for directors and managers. Sudbury: HSE Books; 1994.

107. Scottish Executive. Minimum dataset for occupational health and safety in the NHS Scotland. NHS HDL(2001)22 2001;http://www.show.nhs.uk/sehd/mels/HDL2001_22.htm.

108. Hale A. Culture's confusion. Safety Science 2000;34:1-14. 109. Goldenhar LM, Schulte PA. Intervention research in occupational health and safety. J.

Occup. Med. 1994;36(7):763-775. 110. Smith D, Hunt G, Green C. Managing safety the systems way. BS8800 to OHSAS

18001 - Implementing a cost-effective management system for occupational health and safety. Second ed. London: British Standards Institution; 2000.

111. Scottish Executive. http://www.show.scot.nhs.uk/isd/. In; 2002. 112. Health and Safety Executive. The costs to britain of workplace accidents and work-

related ill health in 1995/96 (second edition). Sudbury: HSE Books ISBN 0-7176-1709-2; 1999.

113. Health and Safety Executive. HSE Ready Reckoner "Reduce Risks - Cut Costs". In: http://www.hse.gov.uk/costs/; 2002.

114. Wright M, Marsden S. Changing business behaviour - would bearing the true cost of poor health and safety performance make a difference? Sudbury: HSE Books; 2002.

218

BIBLIOGRAPHY

1. National Audit Office. A safer place to work - Improving the management of health and safety risks to staff in NHS trusts. London: The Stationary Office; 2003.

2. Department of Health. Building a safer NHS for patients. www.doh.gov.uk/buildsafenhs/; 2001.

3. Health and Safety Commission. The health and safety at work etc. act. London: HMSO; 1974: ISBN 010 543774 3.

4. Health and Safety Commission. Management of health and safety at work regulations: The Stationary Office; 1992.

5. Health and Safety Commission. Management of health and safety at work regulations: The Stationary Office; 1999.

6. Karageorgiou A, Jensen PL, Walters D, Wilthagen T. Risk assessment in four member states of the European Union. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 251-284.

7. Department of Health. Working together, securing a quality workforce for the NHS: Managing violence, accidents and sickness absence in the NHS. The performance management process for incidents of violence to staff, accidents and sickness absence. In. www.open.gov.uk.doh/nawnhs/hrstrat.htm; 1999.

8. Scottish Executive. Towards a safer healthier workplace. In: http://www.scotland.gov.uk/library2/doc08/shwm-00.htm; 1999.

9. National Audit Office. Health and safety in NHS acute hospital trusts in England. Report by the Comptroller and Auditor General. London; 1996.

10. Health and Safety Commission Strategic Plan 2001 -2004. http://www.hse.gov.uk/aboutus/plans/hscplans/plan0104.htm. In; 2001.

11. Health and Safety Commission. Health and safety targets: How are we doing? A supplement to the HSC annual report and HSC/E accounts 2001/02. Sudbury: HSE Books; 2002.

12. Health and Safety Executive. A guide to the reporting of injuries, diseases and dangerous occurrences regulations 1995 (RIDDOR). Sudbury: HSE Books: L73; 1996: ISBN 0 7176 1012 8.

13. Department of Health. An organisation with a memory: report of an expert group on learning from adverse events in the NHS. London: The Stationary Office; 2000.

14. Department of Health. Doing less harm: improving the safety and quality of care through reporting, analysing and learning from adverse incidents involving NHS patients - key requirements for health care providers: National Patient Safety Agency; 2001.

15. Department of Health. The effective management of occupational health and safety services in the NHS. London: The Stationary Office; 2002.

16. Department of Health. NHS Indemnity arrangements for handling clinical negligence claims against NHS staff. Health Service Guidelines HSG(96)48. In: http://www.info.doh.gov.uk/doh/coin4.nsf/Circulars/; 1996.

17. Scottish Executive. Clinical negligence and other risks indemnity scheme (CNORIS). In. NHS HDL(2001)65: http://www.show.scot.nhs.uk/sehd/mels/hdl2001_65.htm; 2001.

18. Health and Safety Executive. Successful health and safety management (HSG65). 2nd ed. Sudbury: Health and Safety Executive; 1997.

19. Health and Safety Commission. The health and safety system in Great Britain. 3rd Edition ed. Sudbury: HSE Books; 2002.

20. Wright M, Lancaster R, Jacobson-Maher C, Talwalkar M, Woolmington T. Evaluation of the good health is good business campaign. Sudbury: HSE Books; 2000.

219

21. Monnery N. The costs of accidents and work-related ill-health to a cheque clearing department in a financial services organisation. Safety Science 1999;31:59-69.

22. Revitalising Health and Safety. http://www.hse.gov.uk/revitalising/. In; 2000. 23. Securing Health Together. http://www.ohstrategy.net/. 2000. 24. Wright M, Marsden S. Changing business behaviour - would bearing the true cost of

poor health and safety performance make a difference? Sudbury: HSE Books; 2002. 25. Health and Safety Executive. Successful health and safety management (HSG65). 1st

ed: HSE Books; 1991. 26. Health and Safety Executive. Reducing error and influencing behaviour. Sudbury: HSE

Books; 1999. 27. British Standards Institution. Guide to occupational health and safety management

systems; BS 8800. London: British Standards Institution; 1996. Report No.: BS 8800. 28. British Standards Institution. OHSAS 18001:1999 Occupational health and safety

management systems - Specification. London: British Standards Institution; 1999. 29. British Standards Institution. OHSAS 18002:2000 Occupational health and safety

management systems - Guidelines for the implementation of OHSAS 18001. London: British Standards Institution; 2000.

30. Smith D, Hunt G, Green C. Managing safety the systems way. BS8800 to OHSAS 18001 - Implementing a cost-effective management system for occupational health and safety. Second ed. London: British Standards Institution; 2000.

31. Frick K, Wren J. Reviewing occupational health and safety management - Multiple roots, diverse perspectives and ambiguous outcomes. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 17-42.

32. Nielsen KT. Organisation theoroes implicity in various approaches to OHS management. In: Wilthagen T, editor. Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 99-123.

33. Harper B. A comparison of the national and organisational structures for health and safety in the United Kingdom and Germany. Journal of the Institution of Occupational Safety and Health 2001;5(1):23-42.

34. Niven KJM. Accident costs in the NHS. The Safety & Health Practitioner 1999;September:34-38.

35. Health and Safety Executive. The costs of accidents at work. Sudbury: HSE Books; 1993.

36. Health and Safety Executive. The costs of accidents at work. 2nd ed. Sudbury: Health and Safety Executive; 1997.

37. Davies NV, Teasdale P. The costs to the british economy of work accidents and work-related ill health. Sudbury: HSE Books; 1994.

38. Health and Safety Executive. The costs to Britain of workplace accidents and work-related ill health in 1995/96. 2nd ed. Sudbury: HSE Books; 1999: ISBN 0 7176 1709 2.

39. Kase D, Wiese K. Safety auditing: A management tool. New York: Van Nostrand; 1990.

40. Glendon I. Risk management in the 1990s: Safety auditing. Journal of Occupational Health and Safety, Australia and New Zealand 1995;11(6):569-575.

41. Saunders R. The safety audit: Designing effective strategies. London: Pitman; 1992. 42. Gay AS, New NH. Auditing health and safety management systems: a regulator's view.

Occup. Med. 1999;49(7):471-473. 43. Livingston AD, Jackson G, Priestley K. Root causes analysis: Literature review.

Sudbury: HSE Books; 2001. 44. National Institute for Occupational Safety and Health. Guide to evaluating the

effectiveness of strategies for preventing work injuries: how to show whether a safety intervention really works. Cincinnati: NIOSH; 2001.

220

45. Smallman C. The reality of "Revitalizing Health and Safety". Journal of Safety Research 2001;32:391-439.

46. DeRoo LA, Rautiainen RH. A systematic review of farm safety interventions. Am. J. Prev. Med. 2000;18(4S):51-62.

47. Chilton S, Jones-Lee M, Loomes G, Robinson A, Cookson R, Covey J, et al. Valuing health and safety controls: A literature review. Sudbury: HSE Books; 1998.

48. ACSNI Study Group on Human Factors. Third report: Organising for Safety. Sudbury: HSE Books; 1998.

49. Beahler CC, Sundheim JJ, Trapp NI. Information retreival in systematic reviews - challenges in the public health arena. Am. J. Prev. Med. 2000;18(4S):6-10.

50. Rivara FP, Thompson DC. Systematic reviews of injury-prevention strategies for occupational injuries. Am. J. Prev. Med. 2000;18(4(S)):1-3.

51. Budworth N. Indicators of performance in safety management. The Safety & Health Practitioner 1996;November:23-29.

52. Hale AR, Heming B, Carthey J, Kirwan B. Modelling of safety management systems. Safety Science 1997;26(1/2):121-140.

53. Emslie SV, Lowe RF. Safecode and successful management of health and safety risk in the national health service. Health Informatics 1995:91-100.

54. Health and Safety Commission Health Services Advisory Committee. Management of health and safety in the health services - information for directors and managers. Sudbury: HSE Books; 1994.

55. Chaplin R, Hale AR. An evaluation of the use of the international safety rating system (ISRS) as intervention to improve the organisation of safety. In: Baram M, editor. Safety Management. The Challenge of Change. London: Pergamon; 1998. p. 165-185.

56. Guldenmund FW. The nature of safety culture: a review of theory and research. Safety Science 2000;34:215-257.

57. Hale AR, Hovden J. Management and culture: the third age of safety. A review of approaches to organizational aspects of safety, health and environment. In: Occupational Injury. Risk, Prevention and Intervention P129-165; 1998.

58. Cox SJ, Cheyne AJT. Assessing safety culture in offshore environments. Safety Science 2000;34:111-129.

59. Centre for Hazard and Risk Management. Offshore safety climate assessment toolkit. In. http://www.lboro.ac.uk/departments/ec/JIP/ ed: Loughborough University; 2000.

60. Brown R, Holmes H. The use of a factor-analytic procedure for assessing the validity of an employee safety climate model. Accid. Anal. & Prev. 1986;18(6):455-470.

61. Dedobbeleer N, Beland F. A safety climate measure for construction sites. J. Safety Res. 1991;22:97-103.

62. Niven KJM. A review of the application of health economics to health and safety in healthcare. Health Policy 2002;61(3):291-304.

63. St. Vincent M, Tellier C, Petitjean-Roget T. Accidents that occurred in three hospitals in one year. Safety Science 1999;31:197-212.

64. Rakel H, Gerrard S, Langford I, Shaw K. Evaluating the impact of contact techniques. Sudbury: HSE Books; 1999.

65. Mearns K, Whitaker SM, Flin R. Safety climate, safety management practice and safety performance in offshore environments. Safety Science 2003;41(8):641-680.

66. Cooper MD. Towards a model of safety culture. Safety Science 2000;36:111-136. 67. Flin R, Mearns K, O'Connor P, Bryden R. Measuring safety climate: identifying the

common features. Safety Science 2000;34:177-192. 68. Easterby-Smith M, Thorpe R, Lowe A. Management research: An introduction.

London: Sage Publications Ltd.; 1991. 69. Hurst NW, Young S, Gibson H, Muyselaar A. Measures of safety performance and

attitudes to safety at major hazard sites. Journal of Loss Prevention in the Process Industry 1996;9(2):161-172.

221

70. Cabrera DD, Isla R, Viledla LD. An evaluation of safety climate in ground handling activities. In: Soekkha HM, editor. Aviation Safety; 1997 27-29 August; Proceedings of the IASC-97 International Aviation Safety Conference, Netherlands; 1997. p. 255-268.

71. Jacobs R, Haber S. Organizational processes and nuclear power plant safety. Reliability Engineering and System Safety 1994;45:75-83.

72. Haber S, Metlay D, Crouch D. Influence of organizational factors on safety. Proceedings of the Human Factors Society 1990:871-875.

73. Shannon HS, Mayr J, Haines T. Overview of the relationship between organizational and workplace factors and injury rates. Safety Science 1997;26(3):201-217.

74. Health and Safety Executive. Successful Health and Safety Management. first ed: HSE Books; 1991.

75. Department of Health. Preventing accidental injury - Priorities for action. Norwich: The Stationary Office; 2002: ISBN 0 11 322477 X.

76. Clarke M, Oxman AD, editors. Cochrane reviewers handbook 4.1.4 [Updated October 2001]. Oxford: Update Software: Updated Quarterly; 2001.

77. Elvik R. Assessing the validity of evaluation research by means of meta-analysis. Oslo: Institute of Transport Economics; 1999. Report No.: 430.

78. Crombie IK, Davies HTO. Research in healthcare: Design, conduct and interpretation of health services research. Chichester: Wiley; 1996.

79. Krause TR, Seymour KJ, Sloat KCM. Long-term evaluation of a behaviour-based method for improving safety performance: a meta-analysis of 73 interrupted time-series replications. Safety Science 1999;32:1-18.

80. Oliver D, Hopper A, Seed P. Do hospital fall prevention programs work? A systematic review. J Am Geriatr Soc 2000;48(12):1679-1689.

81. Guastello SJ. Do we really know how well our occupational accident prevention programs work? Safety Science 1993;16:445-463.

82. Rogers B, Goodno L. Evaluation of interventions to prevent needlestick injuries in health care occupations. Am. J. Prev. Med. 2000;18(4(S)):90-98.

83. Rivara FP, Thompson DC. Prevention of falls in the construction industry: evidence for program effectiveness. Am. J. Prev. Med. 2000;18(4S):23-26.

84. Lingard H, Rowlinson S. Behaviour-based safety management in Hong Kong's construction industry. Journal of Safety Research 1997;28(4):243-256.

85. Harper AC, Gunson C, Robinson L, de Klerk NH, Osborn D, Sevastos P, et al. Curtin industrial safety trial: methods and safe practice and housekeeping outcomes. Safety Science 1996;24(3):159-172.

86. Yassi A, Tate R, Cooper JE, Snow C, Vallentyne S, Khokhar JB. Early intervention for back injured nurses at a large Canadian tertiary care hospital: an evaluation of the effectiveness and cost benefits of a two-year pilot project. Occup. Med. 1995;45(4):209-214.

87. Hale AR. Editorial: special issue for the 10th anniversary of the safety science group and the tu Delft. Safety Science 1997;26(1/2):1.

88. Swuste P, Arnoldy F. The safety adviser/manager as an agent of organisational change: a new challenge to expert training. Safety Science 2003;41:15-27.

89. McGuire T, Hanson M, Moody J. Management attitudes throughout Scotland towards the funding, provision and use of mechanical aids for moving and handling purposes. Edinburgh: Lothian NHS Occupational Health Service; 1995.

90. Sheehan JJ. Cost-benefit anaysis: a technique gone awry. In: Rom WN, editor. Legal and ethical dilemmas in occupational health: Ann Arbor Science: The Butterworth Group; 1983. p. 51-215.

91. Gressel MG. An engineer's perspective of the intervention research workshop. Am J Ind Med 1996;29:382-383.

92. Hale A, Swuste P. Safety rules: procedural freedom or action constraint? Safety Science 1998;29:163-177.

222

93. Hale A, Swuste P. Avoiding square wheels: international experience in sharing solutions. Safety Science 1997;25(1-3):3-14.

94. Hillage J, Tyers C, Davis S, Guppy A. The impact of the HSC/E: A review. Sudbury: HSE Books; 2001.

95. Scottish Health Management Efficiency Group. Accidents in the NHS in Scotland. Edinburgh: SCOTMEG; 1993. Report No.: SP54.

96. Bentler PM, Wu E. EQS for Windows: User's Guide. In: Multivariate Software, Inc. Encino, CA.; 1995.

97. Kelly DL. How to use control charts for healthcare. Milwaukee: American Society for Quality (ASQ).

98. Carey RG, Lloyd RC. Measuring quality improvement in healthcare: A guide to statistical process control applications. Milwaukee: Amercian Society for Quality (ASQ).

99. Health and Safety Executive. Successful Health and Safety Management. Second ed. Sudbury: HSE Books; 1997.

100. Niven KJM. Economic principles in occupational health and safety. Occupational Health Review 2000;November/December:13-18.

101. Statistical Package for the Social Sciences. SPSS version 10. In. version 10 ed. Woking: http://www.spss.com/uk; 2001.

102. Emory CW, Cooper DR. Chapter 6 Measurement. In: Business Research Methods: Irwin; 1991.

103. Siegel S. Nonparametric Statistics for the Behavioural Sciences. In: P.32. New York: McGraw-Hill; 1956.

104. Anderson NH. Scales and Statistics: Parametric and Nonparametric. Psychological Bulletin 1961;58(4):305-316.

105. Miles MB, Huberman AM. Qualitative data analysis: an expanded source book. 2nd ed. London: Sage; 1994.

106. Diamantopoulos A, Schlegelmilch BB. Taking the fear out of data analysis. 1st Edition ed. London: Business Press; 1997.

107. Neter J, Kutner MH, Nachtsheim CJ, Wassweman W. Applied Linear Statistical Models. 3rd ed: Irwin; 1996.

108. Kerlinger FN, Lee HB. Foundations of Behavioural Research. 4th ed. Fort Worth, TX: Harcourt College Publishers; 2000.

109. Health and Safety Executive. A Guide to the Reporting of Injuries, Diseases and Dangerous Occurrences Regulations 1995 (RIDDOR). Sudbury: HSE Books; 1996.

110. National Health Service. National Patient Safety Agency. In: http://www.npsa.org.uk/;2001.

111. Slack N, Chambers S, Harland C, Harrison A, Johnston R. Operations Management. London: Pitman Publishing; 1995.

112. Hale AR. Conditions of occurrence of major and minor accidents. Journal of the Institution of Occupational Safety and Health 2001;5(1):7-21.

113. Neder, Wasserman, Kutner. Applied Linear Statistical Models (3rd Edition). 114. Emory CW, Cooper DR. Business Research Methods (4th Edition). 4th Edition ed:

Irwin; 1991. 115. Scottish Executive. Towards a safer healthier workplace:

http://www.scotland.gov.uk/library2/doc08/shwm-00.htm; 1999. 116. Scottish Executive. Minimum dataset for occupational health and safety in the NHS

Scotland. NHS HDL(2001)22 2001;http://www.show.nhs.uk/sehd/mels/HDL2001_22.htm.

117. National Patient Safety Agency. NPSA pilot project evaluation report: National Patient Safety Agency; 2002.

118. Wagenaar WA, van der Schrier J. Accident analysis. The goal and how to get there. Safety Science 1997;26(1/2):25-33.

223

119. Neale G, Woloshynowych M, Vincent C. Exploring the causes of adverse events in NHS hospital practice. Journal of the Royal Society of Medicine 2001;94:322-330.

120. Cox SJ, Janes WH, Walker D, Wenham DG. Tolley's Health and Safety Handbook. Croydon: Tolley Publishing Co. Ltd. ISBN 9 780854 599981; 1995.

121. Culvenor J. Comparison of team and individual judgements of solutions to safety problems. Safety Science 2003;In Press.

122. Falconer L, Hoel H. Occupational safety and health: A method to test the collection of "grey data" by line managers. Occup. Med. 1997;47(2):81-89.

123. Health and Safety Commission. Proposals for a new duty to investigate accidents, dangerous occurrences and diseases. Sudbury: HSE Books (www.hes.gov.uk/condocs/);2001.

124. Scottish Executive. http://www.show.scot.nhs.uk/isd/. In; 2002. 125. Health and Safety Executive. The costs to Britain of workplace accidents and work-

related ill health in 1995/96. 2nd ed. Sudbury: HSE Books; 1999: ISBN 0 7176 1709 2. 126. Health and Safety Executive. HSE Ready Reckoner "Reduce Risks - Cut Costs". In:

http://www.hse.gov.uk/costs/; 2002. 127. Dorman P. If safety pays, why don't employers invest in it? In: Wilthagen T, editor.

Systematic Occupational Health and Safety Management. Oxford: Pergamon; 2000. p. 351-365.

128. Koopmanschap MA, Van Ineveld BM. Towards a new approach for estimating indirect costs of disease. Soc. Sci. Med. 1992;34(9):1005-1010.

129. Wright M, Marsden S. Changing business beahviour - would bearing the true cost of poor health and safety performance make a difference? Sudbury: HSE Books; 2002.

130. Smith M, Cohen H, Cohen A, Cleveland R. Characteristics of successful safety programs. J. Safety Res. 1978;10(1):5-15.

131. Simonds R, Shafai-Sahrai Y. Factors apparently affecting injury frequency in eleven matched pairs of companies. J. Safety Res. 1977;9(3):120-127.

132. Cohen A. Factors in successful occupational safety programs. J. Safety Res. 1977;9(4):168-178.

133. Shafai-Sahrai Y. Determinants of occupational injury experience: A study of matched pairs of companies. East Lansing, Michigan: MSU Business Studies; 1973.

134. Cohen A, Smith M, MH. C. Safety program practices in high versus low accident rate companies - An interim report: National Institute for Occupational Safety and Health (NIOSH); 1975.

135. Hovden J, Tinmannsvik R. Internal control: A strategy for occupational safety and health. Experiences from Norway. J. Occup. Accidents 1990;12:21-30.

136. Ronald L. Identifying the elements of successful safety programs: A literature review. Richmond, British Columbia: Workers' Compensation Board of British Columbia; 1999.

137. Saari J, editor. Successful accident prevention: recommendations and ideas field tested in the Nordic countries. The final report of the Nordic cooperative program of effective accident prevention methods. Helsinki: Institute of Occupational Health; 1987.

138. Goldenhar LM, Schulte PA. Intervention research in occupational health and safety. J. Occup. Med. 1994;36(7):763-775.

139. Wright M, Brabazon P, Tipping A, Talwalkar M. Development of a business excellence model of safety culture: Safety culture improvement matrix. Sudbury: HSE Books; 1999.

140. Ogden J. Health psychology - a textbook. second ed. Buckingham: Open University Press; 2000.

141. Houghton J, Neck C, Manz C. We think we can, we think we can, we think we can: the impact of thinking patterns and self-efficacy on work team sustainability. Team Performance Management 2003;9(1/2):31-41.

142. De Bono E. Six thinking hats. London: Penguin ISBN 0-14-013784-X; 1985.

224

143. Shivers CH. Halos, horns and hawthorne: Potential flaws in the evaluation process. Professional Safety 1998;March:38-41.

144. NHS Executive. NHS Indemnity - Arrangement for handling clinical negligence claims against NHS staff. HSG(96)48: http://www.info.doh.gov.uk/doh/coin4.nsf/Circulars?ReadForm; 1996.

145. McGrail MP, Tsai SP, Bernacki EJ. A comprehensive initiative to manage the incidence and cost of occupational injury and illness: report of an outcomes analysis. J. Occup. Environ. Med. 1995;37(11):1263-1268.

146. Harper AC, Cordery JL, de Klerk NH, Sevastos P, Geelhoed E, Gunson C, et al. Curtin industrial safety trial: managerial behaviour and program effectiveness. Safety Science 1996;24(3):173-179.

147. Chandler DL. Rocket science is not the hard part. New Scientist 2003;20/27 December 2003:14-15.

Printed and published by the Health and Safety ExecutiveC30 1/98

Printed and published by the Health and Safety ExecutiveC1.10 10/04

RR 280

£25.00 9 78071 7 6291 1 4

ISBN 0-7176-2911-2

rr280 - real time evaluation of health and safety ...hse health & safety executive real time...

Documents