topic 7: introduction to quality improvement · pdf file26.01.2009 · study...

165

Why students need to know aboutquality improvement methods Students will be familiar with the termevidence-based medicine and the randomizedcontrolled trial, which has enabled medicine toestablish if a particular treatment is validated byevidence or merely one that rests on belief of thepractitioners. Research methods such as therandomized controlled trial measure clinicaleffectiveness. But such methods do not measurecontextual components or the process of care.The problems in health systems are significantlyaffected by the processes of care and arandomized controlled trial will not measure theproblems nor fix them. Students need to be awarethat process measures require different methods.The science of improvement is new to medicineand as a result there has been debate aboutwhether the measures of quality improvement arerigorous enough. Berwick has captured thisdebate in a paper published in 2008 where heargued that both research methods are necessaryto improve health-care research for improvingclinical evidence and research for improvingprocesses of care [1].

Quality improvement methods are designed tostudy processes and have been successfully usedfor decades in other industries. In health care,students will be familiar with the goal of scientificresearch, which is to discover new knowledge butthey will be less familiar with quality improvement,which is to change performance [2]. Students areencouraged to observe or join a team undertakingan improvement activity. Medical students canbegin to understand the role of qualityimprovement methods by:

• asking about measures that improve qualityand safety;

• recognizing that good ideas can come fromanyone;

• being aware that the situation in the localenvironment is a key factor in trying to

make improvements;• being aware that the way people think and

react is as important as the structures andprocesses in place;

• realizing that the spread of innovativepractices is a result of people adopting newprocesses and not the other way around.

Most quality improvement methods are based onthe application of continuous quality improvementtheory developed by the manufacturing industry.The principle underpinning quality improvementwas that quality was not something controlled atthe end of the line, but rather throughout theentire work process. This topic explains some ofthe underlying theory.

Traditional attempts to persuade and influenceclinicians to change behaviour, such ascompliance with a protocol or vigilance in regardto drug interactions in the interest of improving thequality of patient care, have by and large failed.There have been thousands of recommendationsby hundreds of committees and peer groups forimproving the safety and quality of patient careover decades, yet there is little evidence thatclinicians have changed their practice because ofthem. The publication of evidence in peerreviewed journals, does not necessarily lead in ofitself , to clinicians' changing their practice. Manyarticles outline best practice and make urgentrecommendations for changes as a result of newinformation [3].

Quality improvement methods have successfullyaddressed this gap and provide clinicians with thetools to: (i) identify a problem; (ii) measure theproblem; (iii) develop a range of interventionsdesigned to fix the problem; and (iv) test whetherthe interventions worked.

Tom Nolan, Brent James, Don Berwick and othershave applied quality improvement principles in

1

Topic 7: Introduction to quality improvement methods

166


developing quality improvement methods forhealth clinicians and managers. The identificationand examination of each step in the process ofhealth-care delivery is the bedrock of thismethodology. When students examine each stepin the process of care they begin to see how thepieces of care are connected and measurable.Measurement is critical for safety improvement.

A range of quality improvement methods havebeen designed. Below are some more commonexamples:• clinical practice improvement (CPI);• root cause analysis to retrospectively examine

what went wrong; • failure modes and effects analysis to

prospectively consider what might go wrong.

Keywords

Quality improvement methods, PDSA cycle,change concepts, continuous improvementmethods, variation, CPI, root cause analysis,flowcharts, Cause and effect diagrams(Ishikawa/fishbone), Pareto charts, histograms,run charts.

Learning objective

The objectives of this topic are to describe theprinciples of quality improvement and to introducestudents to the basic methods and tools forimproving the quality of health care.

Learning outcomes: knowledge andperformanceThis topic is an important one for students tounderstand because improvement will only beachieved and sustained through continuousmeasurement. However, it will also be one of themost difficult because many hospitals and clinicsdo not measure the processes of their care. Aneffective way for students to understand thebenefits of using quality improvement methods is

to apply the principles and use the tools toundertake their own improvement project.

What students need to know (knowledgerequirements):• the science of improvement;• the quality improvement model;• change concepts;• two examples of continuous improvement

methods;• methods for providing information on

clinical care.

What students need to do (performancerequirement): • know how to perform a range of

improvement activities and tools.

WHAT STUDENTS NEED TO KNOW(KNOWLEDGE REQUIREMENTS)

The science of improvement W Edwards Deming, the father ofimprovement theory, described the following fourcomponents of knowledge that underpinimprovement: [4]• appreciation of a system;• understanding of variation; • theory of knowledge; • psychology.

Deming stated that we do not need to understandthese components in depth to apply theknowledge. An analogy used by improvementleaders (e.g. Langley; Nolan [4,5]) is that we candrive a car without understanding how it works.Students beginning their medical careers onlyneed a basic understanding of the science ofimprovement. It is more important to be familiarwith the methods used to improve the processesof care.

5

4

3

2

167

Appreciation of a systemIn applying Deming’s concepts to the health care,we need to remember that most patient careoutcomes or services result from a complexsystem of interaction between health-careprofessionals, treatment procedures and medicalequipment. Therefore, it is important that medicalstudents understand the interdependencies andrelationships among all of these components(doctors, nurses, patients, treatments, equipment,procedures, theatres and so on) therebyincreasing the accuracy of predictions about anyimpact that changes may have on the system.

Understanding of variationVariation is the differences between two or moresimilar things such as different rates of successfor appendectomies in two different parts of thecountry. There is extensive variation in health careand patient outcomes can differ from one ward toanother, from one hospital to another and oneregion to another. Variation, though, is a feature ofmost systems. Shortages of personnel, drugs orbeds can lead to variations of care. Deming urgespeople to ask questions about variation. Studentscan get into the habit of asking their clinicalsupervisors what their outcomes are for aparticular treatment or procedure. Do the threepatients returned to theatres after surgery indicatea problem with surgery? Did the extra nurse onduty make a difference with patient care or was ita coincidence? The ability to answer suchquestions and others like them is part of thereason for undertaking improvement activities.

Theory of knowledgeDeming says that the theory of knowledgerequires us to make predictions that any changeswe make will lead to an improvement. Predictingthe results of a change is a necessary step toenable a plan to be made even though the futureis certain. Many students will have experience ofsuch predictions, having written study plans

predicting what is necessary to pass exams.Those with specific experiences may be better atfocused predictions. For example, healthprofessionals who work in particular health-caresettings such as a rural clinic may be better atpredicting the results of a change in thisenvironment. Because they have more knowledgeabout these clinics and the way they function orshould function and the way the change willimpact on the patients and their families. Whenhealth professionals have experience andknowledge of the area they wish to change it ismore likely that the change will result in animprovement. Comparing the results with thepredictions is important learning. Deming saysthat building knowledge by making changes andmeasuring the results or observing the differencesis the foundation of the science of improvement.

PsychologyThe last component is the importance ofunderstanding the psychology of how peopleinteract with each other and the system. Making achange whether it is small or large will have animpact and knowledge of psychology helps tounderstand how people might react, and why theymight resist change. A medical ward, for example,includes a number of people who will varyenormously in their reactions to a similar eventsuch as introducing an incident monitoringsystem. The potential different reactions must befactored in when making an improvement change.

Deming stresses that successful improvementscan only be achieved when all four componentsare addressed; he calls this the system ofknowledge underpinning improvement. Demingsays it is impossible for improvement to occurwithout the following action: developing, testingand implementing changes.


168


The role of measurement in improvement

Quality improvement activities require healthprofessionals to collect and analyse datagenerated by the processes of health care. Forexample, a student cannot study the change in hisstudy habits without obtaining some informationabout his current study habits and the environmentin which he lives and studies. He first needs thedata to see if he has a problem with study habitsand, second, he needs to decide what informationhe requires to measure whether he has made anyimprovements.

In this analogy, the aim of the improvement projectis to make a change in the study habits of thestudent leading to improved success in leading toimproved success his examinations, rather thansimply identifying students with poor study habits.

Measurement is an essential component of qualityimprovement because it forces people to look atwhat they do and how they do it. Most activities inhealth care can be measured, yet currently theyare not. There is strong evidence to show thatwhen people use the appropriate measures tomeasure change, significant improvements can bemade. All quality improvement methods rely onmeasurement. The medical student will only knowhe has improved his study habits by measuringthe before and after situation.

Medical students will be familiar withmeasurement in the basic sciences; the measuresin quality improvement are different from thoseused in formal medical research. The IHI hasproduced the following chart (see Table 15) todistinguish between the two measures. 6

Measurement for research Measurement for learning and processimprovement

Purpose To discover new knowledge To bring new knowledge into daily practice

Tests One large “blind” test Many sequential, observable tests

Biases Control for as many biases as possible Stabilize the biases from test to test

Data Gather as much data as possible, “just in case” Gather “just enough” data to learn and completeanother cycle

Duration Can take long periods of time to obtain results “Small tests of significant changes” acceleratesthe rate of improvement

Table 15. Institute for Healthcare Improvement: different measures

169


Three main types of measures

Outcome measures Examples of outcome measures include patientsatisfaction surveys and other processes thatcapture the patients’ and their families’ viewsabout their health care. This includes surveys andother methods such as interviews that seek toascertain peoples’ perceptions or attitudes to theservice and their level of satisfaction with thehospital or clinic.

Some specific examples include:• access: time waiting for surgery• critical care: number of deaths in the

emergency department;• medication systems: the number of

medication dosing or administration errors.

Process measures Process measures refer to measurements aboutthe workings of the system. These measures areusually used when a clinician or manager wants tofind out how well a part or aspect of a healthservice or system is working or being performed.

Some specific examples:• access: number of days the ICU is full and

has no spare beds;• surgical care: number of times swab count

completed.

Balancing measures This measure is used to ensure that any changedoes not create additional problems. It seeks toexamine the service or organization from adifferent perspective. If a medical student makes achange to his study habits that leaves no time forhim to see his friends that may have an impact onhis well-being. A specific example is: • reducing the length of stay in hospital: ensure

readmission rates are not increasing.

The quality improvement modelThe quality improvement model is acombination of building and applying knowledgeto make an improvement by asking threequestions and using the PDSA (plan, do,study,act) cycle developed by Deming.

The questions are:1. What are we trying to accomplish?2. How will we know whether a change is an

improvement?3. What changes can we make that will result

in an improvement?

Deming stresses that the questions can be askedand answered in any order. This model takes thesimple concept “trial and error” and transforms itinto the PDSA model that can be used to makeimprovements for all sorts of problems, both bigand small.

What are we trying to accomplish?The idea behind this question is to guide and focusthe efforts of the health-care team doing theimprovement. It is important that the team agreesthat a problem exists and that it is worthwhile fixing.

Does everyone agree that the infection rate inpatients who have had a knee operation is too high?Confirmation that there is a problem requires thatsupporting evidence (qualitative or quantitive) existsindicating the extent of the problem.

Do we have the figures indicating the high infectionrate?It is not a good idea to put a lot of effort intosomething that only one person thinks is a problem.

The person who said the infection rate was highhad only worked one shift in three and had a patientwho had an infection-is this a significant problem? Many countries will have national and internationaldatabases for specific disease indicators that are

87

170


useful, particularly for benchmarking. These dataare important because it allows the team to focuson the right area. In some cases, there might not bemuch available to help answer the question;however, irrespective of the extent of information,the simple rule is to keep the answer short andconcise.

How will we know that a change is animprovement?An improvement can only be confirmed when themeasures show things were improved over time.

What changes can we make that will result in animprovement?This last question involves the team testing thedifferent interventions used to make theimprovements. PDSA is a method designed toassist testing a range of ways to see if anintervention worked.

Using the improvement model developed byLangley, Nolan and others, the IHI has created aPDSA template to assist health-care practitionersimplement PDSA cycles to improve health-careservices or processes (see Figure 4). The modelfor improvement, promoted by the IHI wasdeveloped by Associates in ProcessImprovement. Their version of qualityimprovement is different from other changemodels in that it seeks to accelerate improvement.Hundreds of health-care organizations havesuccessfully used the model to improve health-care processes and outcomes.

Plan-do-study-act cycle

Figure 4 Model for improvement

IHI has summarized a range of qualityimprovement methods, which can be accessedon their web site athttp://www.ihi.org/IHI/Topics/Improvement/ImprovementMethods/Tools/ accessed May 2008.

One of the rules of quality improvement is regulartesting of any changes introduced becauseunexpected things may happen. The cycle beginswith a plan and ends with an action. The studysection is designed to build new knowledge. Thisis an important step in improvement sciencebecause the new knowledge allows betterpredictions about the impact of changes. Theapplication of the model can be simple orcomplex, formal or informal. It can be used toimprove waiting times in the clinic or decreasesurgical infection rates in theatres. A formalimprovement activity may require detailed

ACT PLAN

STUDY DO

The PDSA CYCLE

Langley, Nolan, Nolan & Provost 1999

Determines what changes are to be made

Change or test

Summarizes whatwas learned Carry out the plan

ACT PLAN

STUDY DO

What are we trying to accomplish

How we will know that a change is an improvement

What change can we make that will result in an improvement?

The Model for ImprovementLangley, Nolan, Nolan & Provost 1999

9 10

171


documentation, more complex tools for dataanalysis or more time for discussion and teammeetings. The PDSA model depends on a formatthat repeats steps over and over until animprovement has been effected and sustained.

The IHI model has two parts:• three fundamental questions, which can be

addressed in any order (as set out byDeming);

• the PDSA cycle to test and implementchanges in real work settings—the PDSAcycle guides the test of a change todetermine if the change is an improvement.

Forming the teamIncluding the right people on a processimprovement team is critical to a successfulimprovement effort. Teams vary in size andcomposition. Each organization builds teams tosuit its own needs. For example, if theimprovement project is to improve dischargeplanning than the team should have people whoknow about discharge-nurses, doctors, patients,primary care physicians and nurses.

Setting aims and objectivesImprovement requires setting aims andobjectives. Objectives should be time-specific andmeasurable and should also define the specificpopulation of patients that will be affected. Thishelps keep the team focused on the project.

Establishing measuresTeams use quantitative measures to determine if aspecific change actually leads to an improvement.

Selecting changesAll improvement requires making changes, but notall changes result in improvement. Organizations,therefore, must identify the changes that are mostlikely to result in improvement.

Testing changesThe PDSA cycle is shorthand for testing a changein the real work setting—by planning it, trying it,observing the results and acting on what is learnt.This is the scientific method used for action-oriented learning.

Implementing changesAfter testing a change on a small scale, learningfrom each test and refining the change throughseveral PDSA cycles, the team can implement thechange on a broader scale—for example, for anentire pilot population or on an entire unit.

Spreading changesSuccessful implementation of a change orpackage of changes for a pilot population or anentire unit can permit the team or managers tospread the changes to other parts of theorganization or in other organizations.

Change conceptsMany people intuitively use change conceptsin their daily lives such as asking what changescan be made to improve a particular situation—improved study habits, tension with a familymember, a teacher or difficulties at work. Theyask: “What can I do to make the situation better—to make an improvement?” A change concept inquality improvement is a general notion (a goodidea, an approach) that has been found useful indeveloping specific ideas for change that willresult in improvement.

Nolan and Schall [5] defined a change concept asa general idea, with proven merit and soundscientific or logical foundation, that can stimulatespecific ideas for changes that lead toimprovement.

They identify a number of sources for thinkingabout possible changes: critical thinking about thecurrent system, creative thinking, observing the

11

172


process, a hunch, an idea from the literature, apatient suggestion or an insight gained from acompletely different area or situation. A health-careteam that wants to improve patient care takes aconcept and moulds it to fit their localenvironment, situation or the task they are trying toimprove. This is an important step because itengages the local team in the process. Teammembers will be more committed to theimprovement project and it caters for the particularvariations that can occur in different settings.Langley and his colleagues have developed 70change concepts that have been grouped into thefollowing nine general categories listed in their1996 landmark book on improvement, Theimprovement guide: a practical approach toenhancing organizational performance.

1. Eliminate wasteLook for ways of eliminating any activity orresource in the hospital or clinic that does not addvalue to patient care.

2. Improve workflowImproving the flow of work in processes is animportant way to improve the quality of patientcare delivered by those processes.

3.Optimize inventoryInventory of all types is a possible source of wastein organizations; understanding where inventory isstored in a system is the first step in findingopportunities for improvement.

4. Change the work environmentChanging the work environment itself can be ahigh-leverage opportunity for making all otherprocess changes more effective.

5. Enhance the health provider/patient relationshipTo benefit from improvements in quality and safetyof health care, the health-care professionals andpatients must recognize and appreciate the

improvements.

6. Manage timeAn organization can get more achieved byreducing the time to deliver health care, developnew ways of delivering health care, reducingwaiting times for services and cycle times for allservices and functions in the organization.

7. Manage variationReducing variation improves the predictability ofoutcomes and helps reduce the frequency ofadverse outcomes for patients.

8 Design systems to avoid mistakesOrganizations can reduce errors by redesigningthe system to ensure that there is redundancy i.e. multiple checks and balances to combathuman error.

9. Focus on the product or serviceAlthough many organizations focus on ways toimprove processes, it is also important to addressimprovement of products and services.

Example: change concept

A health-care team may want to adhere to theWHO protocol Clean hands are safer hands.Infection control is a good idea and the WHOguidelines are based on evidence, expertopinion and the literature. One could predictthat if the guidelines were implemented thenan improvement would be made, i.e. adecrease in the transmission of infection viahands. Implementing a guideline is anexample of an abstract concept.

The team is required to then make more specificstatements about implementing the guideline intheir workplace. This process will move theabstract change concept to a practical aim. If the

12

173


change concept is abstract as opposed topractical then it should be backed by literatureand evidence-based medicine.

As the concept becomes more local and practicalit should be increasingly concrete, logicallyconnected and sensitive to the local situation.One of the benefits of lists such as the 70 changeconcepts described by Langley et al. is that theycan speed up the process by not having toduplicate long searches for ideas to test using thePDSA cycle.

A number of catalogues have been publishedcovering topics such as medical errors, waitingtime, delays, intensive care and asthma.

Two continuous improvement methodsThere are a number of examples of qualityimprovement methods in health care but the twomost relevant to medical students are:• CPI (Clinical practice improvement)

methodology;• root cause analysis.

Clinical practice improvement Slides

CPI methodology is used by health-careprofessionals to improve the quality and safety ofhealth care. It does this through a detailedexamination of the processes and outcomes inclinical care. The success of a CPI projectdepends on the team covering each of thefollowing five phases. An example of a completedCPI project is provided in the second part of thistopic and in the Case Study Bank in Appendix 1.

Project phase: The team needs to askthemselves what it is they wish to fix or achieve.They do this by developing a mission statement orobjective that describes what it is they wish to doin a few sentences. This is the time to select theteam members who should be selected on thebasis of their knowledge about the problem.

Patients should always be considered asappropriate members of the team rather than asan after thought. At this stage the team shouldconsider the type of measures they may use.

Diagnostic phase: Some problems are annoyingyet may not be worth fixing because of the littlebenefit they add. Therefore, the team needs toask if the problem they have identified is worthsolving. The team should establish the full extentof the problem by gathering as much informationabout the problem as possible. A brainstormingexercise by the team will generate possiblechanges that could lead to an improvement. Adecision about how to measure the improvementneeds to be resolved during this phase.

The following activities will assist the team tocomplete the diagnostic phase.1. Team members collect and analyse quantitive

and qualitative data of the process beinginvestigated to establish causes of andpotential solutions.

2. Members discuss the different causesinteract to produce the problems.

3. Members identify solutions using the followingactivities.• process flowchart;• brainstorming;• consumer focus groups;• nominal group technique;• tally chart.

Members organize and prioritize information byusing the following tools.• cause and effect diagram;• affinity diagram;• Pareto chart. Members prepare graphs of current data-runchart, statistical process control chart.

Intervention phase: By now, the team will haveworked out what the problems are and their

13 14

174


possible solutions. Each of the solutions have tobe tested through a trial and error process byusing the PDSA cycles to test changes, observethem and keep the bits that work.

Impact and implementation phase: This is thetime to measure and record the results of the trialsof the interventions. Did they make anydifference?

All changes are required to be measured forimpact so that the change can be said to trulyhave made a difference rather than a coincidenceor a one-off effect. The goal is to introduce achange that has sustained improvement. The datashowing the evidence of the change are displayedusing run charts and statistical process control.Using the study habits of a student as anexample, we can say that the student hasimproved his study habits if he has maintained hisimproved study habits for a period of months andnot returned to the old habits.

Sustaining and Improvement phase: The finalphase requires the team to develop and agreeupon a monitoring process and plans forcontinuous improvement. Improvements madenow will become failures in the future if there areno plans to sustain the improvements.

This may involve:• standardization of existing processes and

systems for undertaking work activities;• documentation of relevant policies

procedures protocols and guidelines;• measurement and review to enable the

change to become routine;• training and education of staff.

Root cause analysis Many hospitals and health services are nowusing a process called root cause analysis todetermine the underlying causes of adverseevents or incidents. A root cause analysis is usedafter an incident has occurred to uncover theprimary possible causes. As such, it focuses onthe particular incident and the circumstancessurrounding it. However, there are many lessonsto be gained from this retrospective process thatmay prevent similar incidents in the future.

A root cause analysis is a defined process thatseeks to explore all of the possible factorsassociated with an incident by asking whathappened, why it occurred and what can be doneto prevent it from happening again.

Health-care workers require training in thismethod as they do in CPI methodology. Manycountries have introduced training programmesfor staff to develop skills in conducting root causeanalyses. The Veteran Administration in the UnitedStates has adapted root cause analysis toinvestigate adverse events; their model hasbecome a prototype for health-care organizationsthe worldwide.

It will be very difficult for a team of health-careprofessionals to conduct a root cause analysiswithout the support of the organization becausethe process requires resources to be effective—people, time, support from the managers andclinicians and the chief executive. Yet, the benefitsare real and can lead to improvements of thesystem as a whole. An effective root causeanalysis requires the following components.• multidisciplinary team:

- of no more than six people; - including lay people, particularly those

that may bring a patient’s perspective; - in which no one on the team has had an

actual involvement in the event under review;

19

15 16 17 18

21

20

175

- composed of people who can add valuebecause of their knowledge, position inthe organization or unique perspectivethey bring;

- made up of some members who havebeen trained in root cause analysis, whocan guide the rest of the team with just-in-time training about the root cause analysisprocess, wider system issues and factorsthat may be associated with the event;

- who will be committed to meeting weeklyfor two to four hours at a time over aperiod of five to six weeks;

• root cause analysis effort is directed towardsfinding out what happened:

- documentation and review (medicalrecords, incident forms, hospitalsguidelines, literature review, letters fromthe patient or their family or carer);

- site visit—the team will benefit fromvisiting the environment in which theevent occurred to examine theequipment, the surroundings and observethe relationships of the relevant staff;

• event flowchart is a key part of theinvestigation as it:

- helps to form a common understandingof what happened;

- allows the team to develop problemstatements to enable a cause and effectdiagram to be developed;

- outlines the story and defines whathappened chronologically;

• the team develops a problem statement that:- clearly states the problem to be

addressed;- focuses deeply about the problem(s) and

not the solutions;• establishing the contributing factors or root

causes are accomplished through:- a brainstorming process of all possible

factors; - the development of an event flowchart of

the events including documenting theprocess of questions about each eventand expanding the chart on the basis ofthe information:

• environmental factors: e.g. the workenvironment and its attention tosafety; the type of culture in the unitor locality; medico-legal issues;

• organizational factors: e.g. staffinglevels; quality and attention topolicies; morale of staff; workload andfatigue; access to essentialequipment; administrative support,attitudes to patients and their families;

• team staff factors: e.g. supervision ofjunior staff; quality and effectivenessof communication betweenprofessional groups; availability ofsenior doctors;

• individual staff factors: e.g. level ofknowledge or experience; fatigue andstress; expectations of staff;

• task factors: e.g. existence of clearprotocols and guidelines; system inplace for obtaining test results;definition and description of tasks;

• patient factors: e.g. distressedpatients; communication and culturalbarriers between patients and staff;multiple co-morbidities.

The VA root cause analysis process hasdeveloped a guide to staff about the possibleareas and questions they might ask to uncoverthe possible factors involved in an incident. • communication: Was the patient correctly

identified? Was information from patientassessments shared by members of thetreatment team on a timely basis?

• environment: Was the work environmentdesigned for its function? Had there been anenvironmental risk assessment?

• equipment: Was equipment designed for its


176

intended purpose? Had a documented safetyreview been performed on the equipment?

• barriers: What barriers and controls wereinvolved in this? Were they designed toprotect patients, staff, equipment orenvironment?

• rules, policies and procedures: Was there anoverall management plan for addressing riskand assigning responsibility for risk? Had aprevious audit been done for a similar event,were the causes identified and were effectiveinterventions developed and implemented ona timely basis?

• fatigue/scheduling: Were the levels ofvibration, noise and other environmentalconditions appropriate? Did personnel haveadequate sleep?

Other activities involved in the root cause analysisprocess include:• cause and effect diagrams:

- a cause and effect diagram helps theteam to stay focused on all of thepossible causes rather then fixate on theone cause. The cause and effect diagramor fishbone diagram begins with a fewproblem statements and shows howthese may have been caused by a fewactions and many latent (underlying)conditions;

• root cause statements:- root cause statements should only be

made at the end of the process—the VAprocess provides the following guidanceto staff in writing root cause statements;

• the cause and effect relationship must beexplicit and:

- avoid negative value statements;- identify a preceding cause in each human

error;- each procedural deviation must have a

preceding cause;- failure to act is only causal when there

was a pre-existing duty to act;• preventing a reoccurrence:

- many root cause analyses fail at thisstage because they have not paidsufficient attention to the feasibility andpracticality of the recommendations—ifthere is no senior engagement andcommitment to implementing therecommendations then a root causeanalysis process can wither before itblooms;

- once the root cause analysis iscompleted the team should come upwith recommendations that aim to eithereliminate it (requires action), control it(requires action) or accept. They shouldbe very focused and not too general andcertainly not only about more resources.

All recommendations should be realistic and:• address the root cause of a problem;• be specific and concrete;• be easily understood;• be possible to implement;• define roles and responsibilities for

implementation;• define a timeframe for

implementation.

WHAT STUDENTS NEED TO DO(PERFORMANCE REQUIREMENTS) Know how to use a range of improvementactivities and tools.

Quality improvement tools The following tools are commonly used inquality improvement efforts to improve health care.They are simple to use and can be used byeveryone in the organization. Most hospitals andclinics routinely collect and use data about the healthservices being delivered and many will statisticallyanalyse the data to report to the health authorities orhead of the health service. The following tools arecommonly known as the seven tools of TQM (total

22


177


quality management). Four of these are listed below.• flowcharts;• cause and effect diagrams

(Ishikawa/fishbone);• Pareto charts;• run charts.Experience from other industries and qualityimprovement experts such as Ishikawademonstrate that 9 out of 10 operationalproblems may be solved by one of the toolsdescribed in this section. In health care thatmeans a health service problem can usually besolved by applying some or all of these tools.

The following case example of a CPI project willassist students to understand the tool and how itis used in the improvement process. The caseused below is a real case and was undertakenduring the CPI programme conducted by theNorthern Centre for Healthcare Improvement(Sydney, New South Wales, Australia). The nameof the hospital and the participants are de-identified, the title of the project is acceleratedrecovery colectomy surgery (ARCS).

The first thing that needs to be done is to identifyexactly what it is that needs fixing. Is the length ofstay for patients having colectomy surgery toolong and outside best practice? A missionstatement is developed that captures the aim ofthe project. Participants are encouraged to aimhigh, to develop stretch goals. The followingmission statement was agreed upon.

The mission statement:To reduce the length of stay for patientshaving colectomy surgery from 13 days to 4days within 6 months at the base hospital.The next thing to do is to make sure that the rightteam has been selected to undertake this project.The team members must have the fundamentalknowledge.

Guidance team members:• Health service manager;• Executive officer for the hospital; • Director of nursing for the hospital;• Area clinical nurse consultant pain

management;• Visiting medical officer surgeon.

Project team members:• Area clinical nurse consultant pain

management (team leader);• General surgeon;• Anaesthetist;• Acute pain nurse; • Peri-operative clinic;• Surgical ward registered nurses;• Physiotherapist;• Dietician;• Patient.

Is this a problem?This group of health-care workers decided thatthere was a problem with the length of stay andwanted to reduce the time patients were inhospital

FlowchartsThe next step is to understand the stepsinvolved for patients having colectomy surgery.A flowchart is a pictorial method for showing allthe steps or parts of a process that makes up thesystem. Health care is so complex and before wecan fix a problem we need to understand how theparts fit together and how they function. A rangeof people construct and contribute to a flowchart.It would be very difficult to only have a doctordraw a flowchart because he may not know thewide range of actions that occur in a particularsituation or have access to the documentation ofthe service. Flowcharts are good for setting outwhat people actually do at work rather than whatothers think they do.

24

23

178


Even though the actions described may differfrom the organization’s view, it is important todraw in the flowchart what actually happensbecause this provides a common reference pointand shared language that all members of theteam can share. Constructing a flowchart enablesa “real” and actual portrayal of the process.

The IHI has assembled a range of tools on theInstitute’s web page that is available to helppeople with improvement projects.

There are two levels of flowcharts:• high-level flowchart:

- there are only 6-12 steps described thatgives an overview of a process;

- these show any major blocks of activity,or the major system components, in aprocess;

- they are especially useful in the earlyphases of a project.

• detailed flowchart:- there are many steps described and is a

close-up view of the process;- it can identify loops and allows complex

causes of errors to be identified;- these are often shown using the cloud

symbol as shown in the slide below(flowchart of process);

- detailed flowcharts are useful after teamshave pinpointed issues or when they aremaking changes in the process.

Using a flowchart has a variety of benefits: • it explains the processes involved in health-

care delivery; • it identifies the steps that do not add value to

the health-care service includingdelays; needless storage andtransportation; unnecessary work, duplicationand added expense; breakdowns incommunication;

• it helps health-care workers get a shared

understanding of the process and use thisknowledge to collect data, identify problems,focus discussions and identify resources;

• it serves as a basis for designing new ways todeliver health care;

• health-care workers who document theprocess also gain a better understanding ofeach other’s role and functions.

Not all flowcharts look the same. Slide 24 showsthe flowchart developed by the team who want toreduce colectomy patient length of stay from 13days to 4 days within 6 months.

The team also wants to understand theexpectations of the participants.

A cause and effect diagram is a tool for solvingproblems. This diagram is also called an Ishikawaor fishbone diagram. The diagram is used toexplore and display the possible causes of acertain effect. The content on each arm of thediagram is generated by members of the team ina brainstorm about possible causes. The fishbonediagram in slide 26 is the result of a brainstorm bya team of health-care professionals working onreducing length of stay post-colectomy.

A cause and effect diagram has a variety ofbenefits: • it identifies multiple causes that may

contribute to an effect; • it graphically displays the relationship of the

causes to the effect and to each other; • it focuses the team to the areas for

improvement.Continuing with the CPI project conducted bythe team at the base hospital who were tryingto reduce length of stay post colectomy, thePareto chart identifies the factors that they sawas contributing to the current time patientsstayed in hospital. 27

26

25

179


Pareto chartsIn the 1950s, Dr Joseph Juan used the words“Pareto principle” to describe a large proportion ofquality problems being caused by a small numberof causes. The principle that a few contributionsaccount for the majority of the effect is employed todetermine where to focus the effort in attempting tofix a problem. This is done by prioritizing problems,highlighting the fact that most problems areaffected by a few causes and indicating whichproblems to solve and in what order.

A Pareto diagram is a bar chart in which themultiple factors that contribute to the overall effectare arranged in descending order according to themagnitude of their effect. The ordering is animportant step because it helps the teamconcentrate its efforts on the factors that have thegreatest impact. It also assists them to explain therationale for concentrating on particular areas.

Slides 26 and 27 come from the IHI tool Paretodiagram. Slide 26 describes a sample data tablesetting out the types of errors discovered duringsurgical setup and Slide 27 is a bar chartdepicting in descending order the magnitude ofthe contributing factors.

Implementation of plan-do-study-act cycles A team can brainstorm a number of possibleinterventions.

Run chart example Slide 29 is a run chart produced by the basehospital team that tracks over time theimprovements. Run charts or time plots aregraphs of data over time. A run chart helps theteam know if a change is an improvement overtime or just a random fluctuation wronglyinterpreted as significant improvement. Run chartshelp identify if there is a trend. A trend is formedwhen a series of seven consecutive points

continually fall or rise.The benefits of using run charts include:• helping the team judge how a particular

process is performing;• helping the team to identify when a change is

truly an improvement by plotting a pattern ofdata that one can observe as the changesare made.

Strategies for sustaining improvementsMaking the improvement is not the end; theimprovement needs to be sustained over time.This means continuous measuring and makingadjustments through PDSA cycles. Slide 30describes the strategies for sustainingimprovement and this is where we leave our team.They have successfully reduced the length of stayfor patients having colectomy surgery at theirhospital. In doing so they have saved the hospitala significant amount of money as well asdecreased the chance of a patient receiving aninfection. Even so, they need to sustain theseimprovements. The above strategies wereidentified by the team to monitor and measure thelength of stay on a monthly basis.

SummaryThere is overwhelming evidence that patient careimproves and errors are minimized when cliniciansuse quality improvement methods and tools. Onlythen will the efforts of the team be rewarded byreal sustained improvements to health care. Thistopic set out the methods for quality improvementand described a range of tools that are used inquality improvement.

HOW TO TEACH THIS TOPIC

Teaching strategies/formats Teaching quality improvement methods tostudents can be challenging because it requiresclinicians who have had real experience with thetools and know the benefits. The best way to

30

29

28

180


teach this topic is to get the students to use thequality improvement tools and arrange forindividualized coaching in quality improvementmethods. Students should be encouraged to joinexisting projects to experience the teamapproaches and how patient outcomes aresignificantly improved with such methods.

This topic can be delivered in a number of ways.

An interactive/didactic lecture This topic contains a lot of underpinning

and applied knowledge that is suitable for aninteractive didactic lecture. Use the accompanyingslides as a guide, covering the whole topic. Theslides can be PowerPoint or converted tooverhead slides for a projector.

Panel discussionsInvite a panel of health professionals who havedone a root cause analysis or a CPI project to talkabout the process and whether the methods gavethem insights they would not have had except forthe root cause analysis or CPI.

Small group discussion session The class can be divided up into small

groups with three students in each group askedto lead a discussion about the benefits of qualityimprovement methods—root cause analysis andCPI and when they might be used.

Simulation exercises Different scenarios could be developed for

the students: practising the techniques ofbrainstorming; designing a run chart, cause andeffect diagram or histogram.

Teaching and learning activities This topic is best taught by getting the students topractise the tools and techniques used in qualityimprovement methods by undertaking their ownpersonal quality improvement project.

Examples of self-improvement projects: • develop better study habits;• spend more time with family;• give up smoking;• lose or put on weight;• perform more housework.

Students can implement the PDSA cycle to suittheir own personal circumstances and obtain abetter understanding of the process so they canapply it in their professional work as medicalstudents or members of a health-care team.Following the steps set out above in the caseexample, students can begin to experiment withthe tools and see how to use them and whetherthey helped them in their project.

Another activity, if available, is for students to asktheir clinical supervisors or other healthprofessionals if the hospital undertakes qualityimprovement. If so, they could ask to observe aroot cause analysis process of a CPI project.

After these activities students should be askedto meet in pairs or small groups and discusswith a tutor or clinician what they observed andwhether the features or techniques beingobserved were present or absent, and whetherthey were effective.

181

TOOLS AND RESOURCES

Web-based resources Root cause: Root cause analysis. USDepartment of Veteran Affairs National Center forPatient Safety, 2007(http://www.va.gov/NCPS/rca.html, accessedMay 2008).Flowchart: Flowchart. Institute for HealthcareImprovement Boston, 2004(http://www.ihi.org/NR/rdonlyres/9844A3FD-9F2F-44D7-A423-81F81891F19E/651/Flowcharts1.pdf, accessed May 2008).Improvement methods: Improvement Methods.Institute for Healthcare Improvement, Boston(http://www.ihi.org/IHI/Topics/Improvement/ImprovementMethods/Tools/, accessed May 2008).Root cause: Ask “why” five times to get to theroot cause. Institute for Healthcare Improvement,Boston(0htt10p://www.ihi.org/IHI/Topics/Improvement/ImprovementMethods/ImprovementStories/AskWhyFiveTimestoGettotheRootCause.htm, accessedMay 2008).Clinical improvement guide: Easy guide toclinical practice improvement: a guide for healthprofessionals. New South Wales HealthDepartment, 2002(http://www.health.nsw.gov.au/quality/pdf/cpi_easyguide.pdf, accessed May 2008).Health care improvement: Northern Centre forHealthcare Improvement(http://www.nchi.org.au/www/html/443-documentation.asp, accessed May 2008).

There are many good examples of completed CPIprojects.

Resources Langley GL et al. The improvement guide: a

practical approach to enhancing organizationalperformance. Institute for HealthcareImprovement. San Francisco, Jossey-Bass

Publishers, 1996.

Mozena JP, Anderson A. Quality improvementhandbook for healthcare professionals.Milwaukee, ASQC Quality Press,1993.

HOW TO ASSESS THIS TOPIC

A range of assessment methods are suitable forthis topic including:• complete and write up a self-improvement

project; • write a reflective statement about an

observation of the root cause analysis or CPI process;

• participate in a root cause analysis or CPIproject.

HOW TO EVALUATE THIS TOPIC

Evaluation is important in reviewing how ateaching session went and how improvementscan be made. See the Teacher’s Guide (Part A) fora summary of important evaluation principles.


182

References1. Berwick D. The science of improvement.

Journal of American Medical Association,2008, 299(10):1182–1184.

2. Davidoff F, Batalden P. Toward stringerevidence on quality improvement: draftpublication guidelines: the beginning of aconsensus project. Quality & Safety in HeathCare, 2005, 14:319–25.

3. Lundberg G, Wennberg J A. JAMA themeissue on quality in care: a new proposal and acall to action. Journal of the American MedicalAssociation, 1997, 278(19):1615–1618.

4. Langley GL et al. The improvement guide: apractical approach to enhancingorganizational performance. San Francisco,Jossey-Bass Publishers, 1996.

5. Nolan TW et al. Reducing delays and waitingtimes throughout the healthcare system, 1sted. Boston, Institutue for HealthcareImprovement, 1996.

SLIDES FOR TOPIC 7: INTRODUCTIONTO QUALITY IMPROVEMENTMETHODS

Didactic lectures are not usually the best way toteach students about patient safety. If a lecture isbeing considered, it is a good idea to plan forstudent interaction and discussion during thelecture. Using a case study is one way to generategroup discussion. Another way is to ask thestudents questions about different aspects of healthcare that will bring out the issues contained in thistopic such as the blame culture, nature of error andhow errors are managed in other industries.

The slides for topic 7 are designed to assist theteacher deliver the content of this topic. The slidescan be changed to fit the local environment andculture. Teachers do not have to use all of theslides and it is best to tailor the slides to the areasbeing covered in the teaching session.


topic 7: introduction to quality improvement · pdf file26.01.2009 · study...

Documents