the official management and practice journal

Volume 6 - Issue 1 - Spring 2006

ISS

N=

13

77-

75

64

The Official Management and Practice Journal

icu-management.org 1

Safety must be a key priority for any service

driven industry where errors have the potential to

cause harm to people (Dorman 2005). All humans

make errors, and doctors are no exception. Each

year the media highlight cases where the wrong

limb has been amputated or the wrong organ

removed, where chemotherapy has been given

intrathecally instead of intravenously, where

excess doses of radiotherapy have been given, …

the list goes on. Medical errors are very much a

fact of life; doctors, even intensivists (!), are not

infallible. One recent study (Rothschild et al.

2005) reported that 223 serious medical errors

occurred in one ICU over a 6-month period,

equating to 149.7 errors per 1000 patient days, or

1.5 errors per day for a 10-bed ICU; 11% of these

errors were potentially life-threatening

(Rothschild et al. 2005). Another study focusing

on medication errors found one preventable error

for every five doses of medication administered

(Kopp et al. 2006). These statistics will be of no

great surprise to any of us, but what can and

should be done to reduce the incidence of such

errors, to improve safety in our ICUs?

Over the years, healthcare services have devel-

oped a ‘cover-up’ culture where mistakes have

been hidden, or even denied, and reporting dis-

couraged. However, this traditional attitude is

beginning to change as we learn from other

industries where great harm is possible,

Jean-Louis VincentHead, Department of

Intensive Care, Erasme

Hospital Free University of

Brussels, Belgium

Safety in Care

e.g. aviation and nuclear power, which approach

the concept of safety with a no-fault or limited

fault approach (Pronovost et al. 2005). Systems

are being designed to make it harder for people

to make errors and easier for them to report

errors when they do occur without fear of unwar-

ranted personal reprisal. Medical errors are rarely

the fault of one individual, but more likely the

result of multiple failings in the system behind

the individual. Changing the system is more likely

to prevent a similar error occurring in the future

than chastising one person. Those involved need

to look beyond the immediate event to the

assumptions and conditions that gave rise to it,

and then introduce global, as well as local,

reforms to prevent it happening again (Reason et

al. 2001).

In this issue, Drs Gaba, Manser, Pronovost, Flin,

Al-Ansari and colleagues present some of the

potential organizational and management mecha-

nisms that can be used to improve patient safety

on our ICUs. Improving safety is a key step in

improving quality of care (Institute of Medicine

2000), surely a primary aim for us all.

Ed

ito

rial


Volume 6 - Issue 1 - Spring 2006

ICU Management is

the Official Management

and Practice Journal of the

International Symposium on

Intensive Care and

Emergency Medicine and

was previously published

as Hospital Critical Care.

Business Publication Audit

(BPA) applied for.

Editor-in-chiefProf. Jean-Louis Vincent

Belgium

Editorial boardProf. Antonio Artigas

Spain

Dr. Richard Beale

United Kingdom

Dr. Todd Dorman

United States

Prof. Hans Kristian Flaatten

Norway

Prof. Luciano Gattinoni

Italy

Prof. Armand Girbes

Netherlands

Dr. Claude Martin

France

Prof. Konrad Reinhart

Germany

Prof. Jukka Takala

Switzerland

CorrespondentsProf. David Edbrooke

United Kingdom

Dr. Anders Larsson

Denmark

Prof. Esko Ruokonen

Finland

Prof. Reto Stocker

Switzerland

Dr. Patricia Wegermann

Germany

Table of Contents

EDITORIAL

Safety in Care J.-L. Vincent 1

NEWS AND INNOVATIONS 4

COVER STORY: SAFETY IN CARE

What can critical care learn from HRO T. Manser, D. M. Gaba, G. K. Lighthall 8

A novel collaborative model to improve ICU care in Michigan P. J. Pronovost , C. A. Goeschel 10

Team skills for safety R. Flin,T. Reader 12

Training residents for safety M. A. Al-Ansari 16

MATRIX FEATURES

Nephrology R. Venkataraman, J. A. Kellum 20

Anaesthesiology & Cardiology F. Spöhr, B. W. Böttiger 22

Pharmacy R. S. Bourne 23

PRODUCT COMPARISON: CARDIAC OUTPUT MONITORS

Cardiac output monitoring with intravascular catheters M. Cecconi , E. McKinney, A. Rhodes 25

Cardiac output monitors ECRI Europe 28

MANAGEMENT

Communication with the Administration R. Pauldine,T. Dorman 34

A new risk-adjustment system for IC R. P. Moreno, P. G.H. Metnitz 36

Objective triage for the elderly: ELDICUS update C. L. Sprung, D. Edbrooke 37

Measuring patient satisfaction T. Heidegger, D. Saal,Y. Husemann, M. Nuebling 38

IC beds cf. other healthcare provision in four European countries D. L. Edbrooke, B. Guidet, A. Csomos,T. Prien, G. H. Mills 40

Australian and New Zealand IC databases C. George, L. Norton,T. Higlett 42

VIEWS & INTERVIEWS

Therapeutic hypothermia for all cardiac arrest patients M. Holzer 44

An interview with Professor Antonio Artigas K. Edwards 46

EU FOCUS: INSTITUTION SERIES S. Planitzer, R. Watson 48

COUNTRY FOCUS: US

Healthcare in the US S. S. Rajaram, C. Bekes 52

Progress in Safety and Quality in ICU Care in the US? C. G. Durbin 54

Current State of critical care medicine in the US G. A. Maccioli 56

Remote ICU care in the US M. J. Breslow 58

Billing and documentation: a primer T. Dorman, R. Pauldine 60

CONGRESS PREVIEWS J. M. Hunter 62

AGENDA 64

LETTERS TO THE EDITOR & REQUESTS FOR REFERENCES CITED IN ICU MANAGEMENT

[email protected]

Choscal Hospital (St. Catherine’s Hospital) andthe Chapi Health Centre in the heart of CitéSoleil, where volunteer staff performed nearly12,000 medical consultations and 800 emer-gency interventions during the first threemonths. At both sites, MSF collaborates with thestaff of the Ministry of Health. During the lastmonths of 2005, MSF has faced a huge increasein casualties of violence due to the intense fight-ing between local armed groups and UN troopsof MINUSTAH (United Nations StabilizationsMission in Haiti). In Choscal Hospital, MSF volun-teers treated 34 victims of gunshot in November2005, 80 in December and 103 in January 2006.Half of the wounded were women, children orelderly.

Médecins Sans Frontières is a NonGovernmental Organization and is looking foranaesthetists with knowledge of French to workat least one month on its project in Cité Soleil.Visit our website for more information:www.msf.be

UK: Critical Care Contingency PlanningBruce Taylor, Chair CCCP Group

The UK Intensive Care Society (ICS) and theDepartment of Health has established a workinggroup to plan for expansion of ICU capacity in theevent of an infectious pandemic. The multidisci-plinary Critical Care Contingency Planning(CCCP) group has received valuable advice fromcolleagues in Hong Kong and Canada, who havelearned from experience with the SARS out-breaks. Potential options for expanding capacityby using other clinical areas such as theatrerecovery and high-dependency areas as surro-gate ICUs have been outlined, and an informationgathering questionnaire will assess existingpotential for ICU expansion. A draft document onphased responses / triaging has been produced.Outline plans have been developed for expandingcritical care nursing assistance, identifying oper-ating theatre staff as being the most appropriate-ly skilled for this purpose, and guidelines on corecompetencies for support staff training are beingdeveloped. The ICS is also working to produce aUK-specific core training programme in criticalcare for medical staff.

US: Family Support® in NICUswww.marchofdimes.com

The March of Dimes NICU Family Support®project provides information and comfort tofamilies of premature and other critically ill new-borns in hospital-based neonatal intensive careunits (NICUs) throughout hospitalization, for thetransition home, and in the event of a newborndeath. Currently, NICU Family Support® isbeing implemented in 23 hospitals in the US,and is planned for 38 hospitals by the end of

analysis of recent events and future policies.WHO’s work in this field has intensified recentlyand the Health Action in Crisis (HAC) departmenthas been reorganized into three main pro-grammes: Emergency Response Operations,Recovery and Transition, and EmergencyPreparedness & Capacity Building. The consulta-tion in Geneva acted as a kick-off meeting for arenewed Emergency Preparedness & CapacityBuilding Programme. The main objective of theprogramme is to increase WHO’s support to helpcountries set up their management capabilitiesfor cases of emergency and disaster. The meet-ing concluded a set of recommendations forglobal strategy to scale up emergency nationalpreparedness and mitigation activities. Furtherinformation about HAC and the recommenda-tions can be found online at www.who.int.

LETTERS TO THE EDITOR: [email protected]

4 ICU Management 1-2006

Haïti: Medicin Sans Frontières www.msf.be

On August 2005Medicine Sans Frontières(MSF) started an inter-vention in Cité Soleil, oneof the most neglected

and dangerous slums in Port au Prince, Haiti’scapital city, where 250,000 people live in poverty,violence and social instability. MSF re-opened

UN Pandemic Planning and Preparednesswww.undg.org

To counter the direct impact that an influenzapandemic may have at a global level, the UnitedNations System Influenza Coordination (UNSIC)has developed “UN System Pandemic Planningand Preparedness Guidelines”. These includechecklists, templates and toolkits showing bestpractice. According to the guidelines, each coun-try has to formulate not only a preparednessstrategy but also a contingency or response planfor in the event of a pandemic. Plans for nationsworldwide are accessible on line atwww.undg.org. UNSIC is responsible for themonitoring and coordination of their implementa-tion in cooperation with the World Bank, UnitedNations Development Group and United NationsDevelopment Programme.

WHO: Emergency Preparedness & CapacityBuildingwww.who.int

The World Health Organization (WHO) held anExpert Consultation, “Emergency Preparednessfor the Health Sector Communities: Challengesand Way Forward” from 15th to 17th February inGeneva, Switzerland. The main topics of discus-sion were preparedness for crisis situations, and

NewsWorldw ide

NewsNat iona l

Dear Editor,

W ith reference to Dr Cochard’s article on evaluation of pressure sensors inthe autumn issue of ICU Management, quality of a pressure measure-ment is more dependent on the quality of the fluid-filled tubing, thecatheter and the monitor filter than on the transducer. We recommend thefrequency response test to obtain the Bode diagram of the system: ampli-tude and phase shift vs. frequency (Billiet and Colardyn 1998). The resultneeds to be interpreted taking into account the frequency bandwidth inwhich the pressure signal needs to be measured, which is 12 Hz for theright heart and the radial artery.

Our investigation has shown that all commercial disposable transducershave a 1/1 amplitude ratio far beyond the bandwidth of 12 Hz and aretherefore of good quality.

To obtain a usable technique when evaluating commercial products, wedeveloped the Gabarith test to select different configurations with a pre-set “maximum distortion“ of 2%, 5% or 10% (i.e. with accuracy of 98%,95% and 90%).

Yours sincerely,Lic. Ing. E. Billiet, UZ Gent Biomedisch Technische Dienst,

integral backup battery system maintains patientventilation in the event of an interruption inpower, also allowing intra-hospital patient trans-port from one care area to another with uninter-rupted ventilatory care.

ZOLL: ResQPOD® Circulatory Enhancerwww.zoll.nl

Impedance thre-shold devices(ITD) have beengiven a Class IIarecommendationby the AmericanHeart Association(AHA) in its newguidelines for

CPR and emergency cardiac care. The AHAreport that the use of the ITD has been shown toimprove circulation during CPR and increase thereturn of spontaneous circulation (ROSC) in car-diac arrest patients (AHA Guidelines for CPR andemergency cardiac care 2005).

Advanced Circulatory System's ResQPOD®Circulatory Enhancer – sold in Europe by ZollMedical Corporation – helps increase bloodflow to the heart and brain during assisted ven-tilation. It works in conjunction with all standardresuscitation techniques and equipment. TheResQPOD uses proprietary technology thatincreases circulation by regulating airflow intothe lungs during the chest wall recoil (or decom-pression) phase of CPR. In multiple preclinicalinvestigations and seven different publishedclinical studies with patients in cardiac arrest,the ResQPOD has been shown to increaseblood return to the heart and blood flow to thevital organs during CPR (Aufderheide et al.2005; Pirrallo et al. 2005; Thayne et al. 2005 (inpress); Wolcke et al. 2003; Plaisance et al. 20002004 & 2005).

MAQUET: MRI-compatible SERVO-iwww.maquet.com

MAQUET haslaunched itsMRI compati-ble SERVO-i,which enablesthe same venti-lator system tobe used in theintensive careunit and in the

magnetic resonance imaging (MRI) facilities, forthe treatment of all categories of patients. Thismakes it possible for continuous care, with thepatient connected to the same ventilator fromthe ICU bed to the MRI examination and back.SERVO-i can now be used everywhere in thehospital and also in airborne or road transport.These new capabilities of the product will be offi-cially introduced at the International Symposiumon Intensive Care and Emergency Medicine(ISICEM) in Brussels, March 21-24, 2006, whereMaquet will also present a non-magnetic cart andother non-magnetic mounting solutions, whichcan be used in MRI rooms.

Respironics® Launches BiPAP® Focus™Noninvasive [email protected]

Respironics havereleased the BiPAP®Focus™ NoninvasiveVentilator, a basicBiPAP delivery sys-tem designed specif-ically for the institu-tional setting.Ventilatory assistance is provided to stable, lessacute patients with respiratory insufficiency orfailure. Features of the BiPAP Focus Systeminclude:

M Digital Auto-Trak SensitivityM Large easy-to-use user interfaceM Display of patient parametersM Pressure bar graphM Easy to use alarmsM Backup battery

The BiPAP Focus includes Respironics’ propri-etary Digital Auto-Trak™ Sensitivity that assuresoptimum triggering and cycling sensitivitythroughout changing breathing patterns andleaks. This eliminates the need for a perfect sealof the patient interface and constant adjustmentwhile promoting patient-ventilator synchrony. Thepressure bar graph indicates patient breaths andpressure ranges. Pressure alarms are automati-cally set to 5 cm/H2O above IPAP and belowEPAP settings, minimizing nuisance alarms. An

HAMILTON MEDICAL: RAPHAEL XTCwww.hamilton-medical.com

In March 2006 atISICEM in Brussels,Hamilton Medicalwill launch theRaphael XTC, a sin-gle ventilation solu-tion for both nonin-vasive positivepressure ventilation(NPPV) and invasive

ventilation. Combining the RAPHAEL platformwith a single-limb breathing circuit and a largemonitor, the RAPHAEL XTC is suitable for step-down or -subacute care units, long-term carecentres, ICUs, recovery rooms or transit.To adaptto frequently changing leakage conditions,HAMILTON MEDICAL has developed triggertechnology that automatically and continuouslyadjusts the trigger threshold to the leakage.

The RAPHAEL XTC is an Intelligent Ventilationsolution that features Adaptive SupportVentilation (ASV), a closed-loop mode of ventila-tion for the respiratory management of patientsfrom intubation to weaning. ASV requires nomode changes during ventilation. It employslung-protective rules and adjusts the respiratorypattern based on the patient’s pulmonarymechanics and spontaneous respiratory activity.Studies show that ASV facilitates shorter timeson the ventilator, while at the same time lessuser interaction is required and fewer alarmsoccur (Sulzer et al. 2001; Cassina et al. 2003).


ESICM Research AwardsJean Daniel ChicheHead of Research Committeewww.esicm.org

The European Society of Intensive CareMedicine (ESICM) has launched a programme tofund European research initiatives led by younginvestigators. Thanks to a genuine partnershipwith industrial leaders, 6 Industry ResearchAwards ranging from 5,000 to 80,000 € are avail-able:

NewsIndustry

NewsSoc iet ies

2006. The service will be offered in at least 50NICUs by 2007. The project includes educationand activities for siblings, parent-to-parent sup-port, photography, professional developmentopportunities for NICU staff, and bereavementprograms. The March of Dimes is a national not-for-profit health agency whose mission is toimprove the health of babies by preventing birthdefects, premature birth and infant mortality.NICU Family Support® is part of the March ofDimes Prematurity Campaign, which addressesthe growing problem of premature births, theleading cause of newborn death in the US.According to the latest government figures,more than 500,000 US babies -- or 1 in 8 -- areborn prematurely each year. The most recentcollaborating site is Memorial Health UniversityMedical Centre in Georgia. In 2003 alone, 17,762babies were born prematurely in Georgia.

icu-management.org

Research: SCCM Awards

The 35th Critical Care Congress of the Society ofCritical Care Medicine was held in San Francisco7th to 11th January 2006. A selection of theSCCM awards is reported here and a full list isavailable on the SCCM website: www.sccm.org/membership/awards/recipients/index.asp

The winner of the 2006 SCCM Vision Grant,sponsored through contributions to the SCCM’sCritical Care Education and Research Foundation,was Nasia Safdar (Clinical Instructor at theUniversity of Wisconsin-Madison MedicalSchool) for her research, “Enhancing patient safe-ty by preventing nosocomial infections: a ran-domized controlled trial of preemptive precau-tions for prevention of infection by multi resistantpathogens”. The 2006 Norma J. ShoemakerCritical Care Nursing Research Grant, sponsoredby KCI USA Inc., was awarded to Sandra K.Hanneman. Dr Hanneman is Professor of nursingresearch, Associate Dean for research andDirector, Centre for Nursing Research, Universityof Texas Houston. She won the award for herresearch “Multi-site randomized clinical trial ofhorizontal positioning to prevent and treat pul-monary complications in mechanically ventilatedcritically ill patients; a pilot study.”

Sandra Swoboda (John Hopkins, Baltimore) wonan educational scholarship for her research“Does isolation status impact the frequency ofadverse events in a surgical intermediate careunit”. From this prospective observational studyof all patients admitted to a surgical intermediatecare unit with a length of stay over 2 days, DrSwoboda and her colleagues concluded thatadverse events reported underestimated actual

events for the isolated and non-isolated patientgroups, and that isolated patients were morelikely to experience an adverse event comparedwith the non-isolated patient group.

In addition to the ten educational scholarships, 10specialty awards were also presented. Theseincluded the Clinical Pharmacy and Pharma-cology Award, which was won by LorianneWright, (Vanderbilt Children’s Hospital Nashville),for her research “Impact of patient-specific deci-sion support on medication errors in critically illchildren”. The Healthcare Industry Award waswon by James Shaffer (Health First, Inc.) for“Remote ICU Management improves outcomesin patients with cardiopulmonary arrest.” Co-researchers on this project included MikeBreslow from VISICU Baltimore (see DrBreslow’s article on page 58 in this issue of ICUManagement). The Paediatrics Award was pre-sented to Kshitij Mistry (Paediatric Critical CareMedicine, University of North Carolina), for“Communication error during post-operativepatient hand off in the paediatric intensive careunit.” This research study defined miscommuni-cation as absence or inaccuracy in 18 categoriesof information necessary to provide adequatepost-operative care. Recordings of the verbalsign-out process and medical records were usedto evaluate 134 admissions. The researchers con-cluded from their results that miscommunica-tions commonly occurred during post operativepatient hand over and that most involved multipleevents. See also the article, “Teamwork forSafety” on page 12 in this issue of ICUManagement.

Two nursing section research awards were alsopresented, and there were five research citationwinners, including one for “Blood Volume meas-urements: Impact on fluid management”.Elisabeth Biuk-Aghai and colleagues from theSurgical Critical Care Unit at the University ofHawaii, Honolulu, hypothesised that blood vol-ume measurement may alter treatment deci-sions regarding fluid and blood product manage-ment. Following a prospective observationalstudy of blood volume measurements followedby a retrospective chart review of 42 surgicalintensive care unit patients, the researchers con-cluded that traditional means to evaluate the fluidstatus of critically ill patients may not be accu-rate. Blood volume measurements resulted intreatment changes in 40% of the cases in thisstudy.

Full abstracts from the Congress are published inthe “Society of Critical Care Medicine’s 35thCritical Care Congress Abstracts” Supplement toCritical Care Medicine Volume 33 Number 12December 2005.

7

NewsResearch

The ”Alain Harf Award on Applied RespiratoryPhysiology”, supported by Hamilton Medical,aims to fund research related to respiratory phys-iology and its applications to the field of mechan-ical ventilation.

The “Eli Lilly – ESICM Sepsis Elite Award” is a 2-year award to fund a PhD thesis or a post doctor-al fellowship on sepsis.

The ”Spacelabs Intelligent Monitoring Award”aims to fund research focused on patient moni-toring in intensive care and emergency medicine.

The ”iMDSoft Patient Safety Award” is a 10,000 €

award to fund a research project focused onpatient safety.

The “Edwards MinimallyInvasive HaemodynamicMonitoring Award” andthe “Edwards NursingScience Award” are sup-ported by EdwardsLifesciences to advanceknowledge respectively

related to minimally invasive haemodynamicsand nursing skills and procedures.

In addition, investigators can also apply to one ofthe prestigious 20,000 € ECCRN awards fundedby the ESICM. Details (including eligibility criteriasuch as ESICM membership) and applicationmaterials for the ESICM Young InvestigatorAward, the ECCRN Clinical Research Award, theECCRN Basic Science Award or each of the 6Industry Research Awards are available on theESICM website www.esicm.org.

SCCM release iCriticalCare Podcastswww.sccm.org/podcast

The Society of Critical Care Medicine (SCCM)became the first medical and only critical caresociety to provide podcasts, with the release ofits iCriticalCare Podcasts in July 2005. Sincethen, the breadth and depth of the podcasts con-tinues to grow, as was evident at the Society’sCritical Care Congress in January 2006. A pod-cast delivers information in audio format, which

can be downloaded from the Internet to a com-puter, an iPod, or another MP3 player. SCCMpodcasts are available at http://podcasts.yahoo.com and www.sccm.org/ podcast, free ofcharge. The iCritical Care Podcasts include newsreleases, interviews, and article summaries fromthe SCCM publications Critical Care Medicine,Paediatric Critical Care Medicine and CriticalConnections. SCCM have also released a vod-cast (video podcast), with content on SCCM’sRight Care, Right Now™ philosophy. Accordingto SCCM’s President, Charles G. Durbin, “thislaunch was a major achievement for SCCM, as itwas timely, strategic and ahead of the curve.”See also Professor Durbin’s article on page 54 ofthis issue of ICU Management.


Drs Manser, Gaba and Lighthall review how high reliability organization theory can beapplied to improve patient safety and quality of care.

C O V E R S T O RY : S A F E T Y I N C A R E

What can critical care learn fromHigh Reliability Organizations

IntroductionThe public expects that healthcare should be a high reli-ability undertaking with little risk of preventable harm.Other industries, such as commercial and military avia-tion or nuclear power production have come to gripswith the routine conduct of activities that bear intrinsichazard. At their best, organizations in these domainshave been described as High Reliability Organizations(HRO).

Of course, healthcare is not the same as generatingelectricity, flying airliners, or building a space station. Forone thing we do not design or manufacture humanbeings, nor do we receive an official instruction manual.Also, unlike some other industries in which the activitiesare relatively "elective" (the airliner doesn’t have to flyfrom New York to Chicago tonight) in healthcare thereare situations in which procedures must be performedeven if the hazards are particularly high. Nonetheless,healthcare still has much to accomplish to qualify as ahigh reliability undertaking, and we have much to learnfrom HRO theory (Gaba 2001).

What lessons can we learn from HRO theory?HRO theory (HROT) is now a complex amalgam ofapproaches and viewpoints (Roberts 1990; Weick andSutcliffe 2001). In general terms, HROT states thatappropriate organizational control can yield nearly failurefree results despite high hazard and high tempo, if theorganization (and overall industry) embodies characteris-tics listed in Table 1.

The emphasis on systems does not mean that peopleand their skills are unimportant. Systems are made up ofpeople working in organizations. The dynamism, com-plexity, and risk in industries of high intrinsic hazardrequires special attention to decision making processesof individual personnel and the teams they work in. Infact, it appears that well-functioning clinical “microsys-tems” within healthcare institutions may be the bestcurrent examples of HROs in healthcare (Mohr andBatalden 2002; Nelson et al. 2002). Such HRO-likemicrosystems are still found only sporadically.

Creating high reliability characteristics in ICOVERVIEWUsing the framework above and what we know aboutcare delivery in intensive care units, it is possible to mapout a few categories of activity that would be seen if crit-ical care were a high reliability organization. These areasof activity could be termed overall quality improvement,

practice improvement, and performance analysis. It isimmediately obvious that these categories are inter-twined, but will be separated here for purposes of clear-er discussion.

QUALITY IMPROVEMENT ACTIVITIESIn medicine there is often a huge gulf between credibleevidence of measures that can improve patient safetyand survival, and actual practice change. Examples per-tinent to critical care include programmes for reducingcatheter related bloodstream infections, tight glycaemiccontrol, protective lung ventilation, perioperative betablockade, and specific early haemodynamic manage-ment of severe sepsis. Evaluation of clinical evidenceand generation of relevant local data to analyze for appli-cability should be an ongoing effort of patient safety andquality improvement. Having the personnel and data col-lection procedures in place to make these assessmentsin a timely manner is the type of proactive self aware-ness that characterizes an HRO.

PRACTICE IMPROVEMENTMedical training is based on the intensive learning ofbasic science concepts and is followed by a variablystructured apprenticeship. In the ICU and in otherenvironments, the practice of medicine is based onteamwork and cooperation between multiple disci-plines. Across the board, medicine offers incrediblylittle exposure to team management, communica-tion, and group processes. HROT highlights thatteam skills are not acquired unless specificallytaught. HROs recognize that intensive training andperformance assessment in both routine work and insimulations and drills pays off. HROs ensure thatteams and work units hone their skills during routineoperations. They debrief themselves routinely andkeep track of individual and team performance. Forexample, every Naval aviator – regardless of experi-ence level or seniority - is graded on every carrierlanding. HROs also use simulation and drills exten-sively to ensure maximum readiness for critical butuncommon situations and to optimize team perform-ance. Training is built into the work environment – itis not an add-on for the individual. The emphasis is ontraining the system, not just individuals. Moreover,training continues for the entirety of one's career andis not limited to those learning the job. In healthcarethe nursing profession adheres to these principlesmore avidly than does medicine. Physicians rely on afairly weak and haphazard system of "continuingmedical education" to maintain individual abilities.

Tanja Manser, PhDCentre for Organizational

and Occupational Sciences

ETH Zurich

[email protected]

David M. Gaba, MDStanford University

Patient Simulation Centre of

Innovation, VA Palo Alto

Health Care System

Geoffrey K. Lighthall

PhD, MDAnaesthesiology and Critical

Care, Stanford University

http://anesthesia.stanford.edu/VASimulator/simulator.htm

This system is largely at the discretion of theindividual, their time and expense. There is lit-tle systematic training of teams.

From the earliest training on, doctors, nurses andother healthcare workers should not only betaught basic knowledge and skills that traditional-ly define medical competence but also the con-cepts that are essential to teamwork: effective commu-nication and coordination, establishing and maintainingshared situation awareness, joint decision making andproblem solving, and conflict resolution. To be mosteffective, the training of team skills should – like all train-ing – be recurrent rather than a one-time event. Teamtraining needs to be integrated at different levels of train-ing and should continue throughout one's career. Basedon the concept of crew resource management from avi-ation, Howard et al. (1992) introduced a similar trainingparadigm to medicine with anaesthesia crisis resourcemanagement (ACRM).The latter is a comprehensive cur-riculum that examines the sources of errors as well astraining in communication, leadership, dynamic planningand decision making, workload management, and team-work as tools for effective crisis management (Gaba etal. 1994). The teaching points are practiced during simu-lations and discussed during video-assisted debriefingsessions.

The ACRM concept has since been adopted by many dif-ferent medical disciplines including critical care. Courseshave to be tailored to the specific challenges to team-work in critical care by involving different medical andallied health professions in training activities.

In bridging the gaps between the current system ofmedical practice and that of an HRO, teamwork in highfrequency (septic shock, myocardial infarction), high risk(massive haemorrhage), and even rare but catastrophicconditions (cardiac tamponade, anaphylaxis), would pro-vide an unmistakably high yield to all involved. The useof human patient simulation with videotaped debriefinghas become the state of the art method for immersiveand simulation learning for crisis management in the ICU(Lighthall 2004).

PERFORMANCE ASSESSMENTAlthough assessment of team performance in health-care is considered important, tested and validatedmethodologies for evaluating not only clinical or techni-cal performance, but also skills that are needed for effec-tive crisis resource management and good teamwork,are scarce. Review of critical events immediately afterthe fact is considered a valuable experience from manypoints of view including error reduction, performancemodification and improvement, and processing of a

stressful event. Review of near misses, even if anony-mously submitted, opens a constructive dialog thatencourages error reporting and ensures that futurepatients derive the maximum benefit from knownerrors. Cultural changes that encourage such reportingrather than generating shame are of obvious value.

The experience in simulation training has led many tobelieve that the educational value is bidirectional.Simulations often reveal shortcomings in how traineesare prepared to manage certain situations, and short-comings of the system at large (Hammond 2004;Helmreich 2000; Lighthall 2004). Our experience sug-gests that to be maximally effective, it is important thatsimulation activities bring together the multiple profes-sions present in the critical care unit.

SUMMARYHROs have employed anumber of commonsense practices as partof their everyday opera-tion without asking forevidence to justify theiruse. While there isgreat evidence thatmedical errors are com-mon, we don’t incorpo-rate findings from welldesigned studies intoclinical practice as effi-ciently as possible, andwe don’t train practitioners to act as teams despite thefact that this is how many conditions are managed.Because many ICUs function as a closed shop and areattended to by subcultures of larger departments andprofessions, and because the patients are at a higheroverall risk for adverse events, there may be greaterinterest and potential to apply HRO concepts to thisdomain of healthcare. In this article we emphasized therole of self-examination and multidisciplinary, multipro-fessional team training as the centrepieces of qualityimprovement, practice improvement, and performanceassessment. Improved cooperation amongst all stake-holders in a critical care unit will create a greater senseof shared purpose and move towards achieving higherreliability in the realm of patient safety.


Figure 1.

Training with simulation

© 2005 Patient Simulation

Centre of Innovation

Table 1. Key Elements of a High Reliability OrganizationA culture of safety permeates the organizationSystems, structures, and procedures conducive to safety and reliability are in placeIntensive training of personnel and teams takes place during routine operations, drills and simulationsSafety and reliability are examined prospectively for all the organization’s activities; organizationallearning by retrospective analysis of accidents and incidents is aggressively pursued

A novel collaborative model to improve ICU care was implemented in 127 ICUs in thestate of Michigan. Interventions were used to improve culture of the working environ-ment, and eliminate blood stream infections in >80% of ICUs.

Peter J. Pronovost

MD, PhDDept. of Anaesthesiology &

Critical Care Medicine

Johns Hopkins University

Baltimore, USA

[email protected]

Christine A. Goeschel

RN, MPA, MPSKeystone Centre for Patient

Safety and Quality

Michigan Health and

Hospital Association

Lansing, USA

www.mhakeystonecenter.orgwww.safetyresearch.org

AcknowledgementFunding provided by the

Agency for Healthcare

Research and Quality

(AHRQ), grant no.

1UC1HS14246



A novel collaborative model toimprove ICU care in Michigan

The need for dramatic improvements in healthcare qual-ity and patient safety is widely recognized. Global health-care stakeholders including academic researchers, regu-lators, purchasers, payers, hospitals, health systems andconsumers have taken the message posed by theInstitute of Medicine reports To Err is Human (1999) andCrossing the Quality Chasm (Institute of Medicine 2001)seriously. This message made the 98,000 lives lostannually in US hospitals from preventable errors a publicissue. A new challenge has risen in the wake of thesereports: how can we assess whether, using valid meas-ures, our efforts are actually improving quality and safe-ty? (Pronovost et al. 2006)

Quality and safety researchers at the Johns HopkinsUniversity, the MHA-Keystone Centre for Patient Safetyand Quality and 127 Michigan intensive care units havetaken on this challenge and made wide-scale improve-ments using a novel collaborative model (Pronovost,Goeschel 2005). In this project, participants were giveninterventions to improve safety and a safety scorecard totrack their progress. Participating ICUs can now state thatthey are safer than they were when the project startedtwo years ago.To state with confidence that they are safernow, teams worked to answer the following questions:

M How often do we harm patients?M How often do we do what we should?M How often do we learn from our defects?M How well do we improve culture?

Answering these questions created a safety scorecardthat addresses two related components: (1) genericmeasures that are not rates (qualitative), apply through-out a health system, and focus on learning from mis-takes and improving culture (Pronovost et al. 2005a); and(2) discipline specific measures that are rate based(quantitative) and focus on ensuring we do what weshould do and eliminate harm (Pronovost et al. 2006).

Process outcomesTo support teams we provided evidence-based meas-ures, standardized data collection tools and performancereports, dedicated project web space, weekly recordedconference calls, consistent encouragement and individ-ual coaching upon request.

Beyond long distance sharing, biannual workshopsallowed teams and Hopkins faculty to work side-by-sideas colleagues. As project leaders, we periodically sentletters to hospital CEOs providing them with progressreports and asking for specific executive assistance.

Ninety percent of teams routinely participated in confer-ence calls. Nearly 100% of the 127 teams attended thebiannual workshops and participant evaluations aver-aged 4.5 on a 5.0 scale.Teams are increasingly cognizantof their power to change the way care is delivered, toimprove patient outcomes, and to enhance unit culture.Administrative leaders have started speaking firsthandabout infection rates, safety issues, and the renewedsense of satisfaction that is permeating their ICUs.Variation in achievement spans the demographic mix ofparticipants, and sharing what works and does not workis universally valued.

Clinical outcomesDuring the first nine months of data collection, the entirestate moved from the 50th percentile in the country incatheter related blood stream infections (CR-BSI) andventilator associated pneumonia (VAP) to the 10th.Indeed, over 80% of participating teams eliminated CR-BSI; VAP data is still being analyzed. In addition, cultureimproved throughout Michigan ICUs.

Lessons learnt and next stepsThe clinical and cultural improvements achieved byMichigan teams are important and we believe replicable.Importantly, this study provides significant new insightsregarding how to measure and improve patient safetyand how to run a large scale improvement collaborative.However, the resources needed, and clinical, measure-ment and management expertise required for this col-laborative exceed the capacity of any single organization.

We learned several lessons; first, package educationalmaterials as specific interventions or behaviours. Forexample, we converted the 100-page guideline for pre-venting CR-BSI into 5 interventions; wash hands, usebarrier precautions, clean skin with chlorhexidine,remove unnecessary lines and avoid femoral site(Berenholtz et al. 2004). Second, efforts to improve cul-ture should accompany efforts to improve specific out-comes (Pronovost et al. 2005b). Third, valid measuresand data management, like any clinical research project,are imperative. Such measures are uncommon in quali-ty improvement projects. Beyond the immediate results,this project has been important in creating a virtual learn-ing community across participating ICUs, the hospitalassociation and the research team. The partnership inMichigan continues and lessons learnt have beenapplied to implement an improved program in NewJersey and Rhode Island.



IntroductionThe intensive care unit (ICU) is a high-risk environmentthat requires multidisciplinary teams to cope with rapid-ly evolving challenges, heavy workloads and intensetime pressure. The complexity of the ICU environmentrenders it particularly prone to errors resulting from fail-ures in teamwork and communication.

This is exemplified in a root-cause analysis of an adverseevent where a patient being treated for heart and renalfailures suffered an air embolism (resulting in consider-able neurological damage) after a large central venouscatheter was removed whilst the patient was sitting up(Pronovost et al. 2004).The analysis of the causal factorsunderlying the incident revealed that the attending doc-tor, a first-year renal fellow, did not know the propertechnique for removing central lines, yet was allowed toperform the procedure unsupervised and without train-ing, with the institution having no system to ensurecompetency for the procedure, and with the fellow feel-ing unable to admit his lack of knowledge. Furthermore,a nurse observed the fellow conduct the procedureincorrectly, and could have expressed concern, yet wasreluctant to speak up or correct the physician.

In high-risk industries that share similar characteristics tothe ICU (e.g. aviation and nuclear power), team workingskills are considered especially important for protectingagainst errors and in enhancing performance, with con-siderable investment being made in understandingteams and their training requirements (Burke et al.2004). This has resulted in training programmes beingdeveloped to meet the challenge of reducing errorthrough making better use of human resources(Helmreich et al. 1999). Team researchers have demon-strated the importance of team cognition (e.g. sharedmental models), as well as interpersonal behaviours andtask coordination (Salas et al. 1997). If properlydesigned, the programmes are based on a trainingneeds analysis that defines the team skills required for aparticular environment, as well as the present levels ofaccomplishment.

In light of evidence demonstrating that a high proportionof medical errors result from failures in team skills suchas communication, an Institute of Medicine report

encouraged healthcare providers to emulate high-riskindustries in the application of human factors research toenhance safety (Kohn et al. 2000). In particular, as theICU is an environment that requires a high level of coop-eration between multidisciplinary team members, itwould appear likely that good team working skills areimportant for reducing error, and thus enhancing patientsafety.

Teamwork and safety in the ICUA number of studies have published data demonstrat-ing the association between good teamwork andpatient safety in the ICU. An examination of ICU criticalincident studies, where the causal factors underlyingincidents that either harmed or could have harmedpatients are identified, reveals that poor team workingis frequently associated with occurrences of criticalincidents. Team working factors such as poor or inade-quate communication, poor supervision, a lack of open-ness between team members, and not ensuring otherteam members follow protocol are all associated withoccurrences of errors and critical incidents (Beckmannet al. 1996; Beckmann et al. 2003; Buckley et al. 1997;Wright et al. 1991).

However critical incident studies can lack a fine-grainedanalysis of the precise contributory teamwork factorsbehind incidents, as often the actual role of a particularcausal factor in the incident is not described (Fletcher etal. 2002). Observational simulator studies, where theperformance of ICU teams is video recorded and laterreanalysed, can overcome this. Simulator researchbased in the ICU has shown that failures in team work-ing, such as a lack of leadership, failures to communi-cate priorities, not sharing patient status information,and overloading nurses with requests, have a negativeimpact upon performance and result in the occurrenceof errors (Lighthall et al. 2003). Additionally, real-life ICUobservational studies have shown communicationbetween nurses and doctors to be associated with ahigh proportion of errors. Verbal communicationsbetween nurses and doctors were observed during rela-tively few activities and were associated with a high pro-portion of errors, leading the researchers to concludethat nurses and doctors were not combining their know-ledge and skills efficiently.

Team skills for safety

Effective teamwork is crucial for ensuring patient safety within the ICU. In other high-risk industries (see also Manser, Gaba and Lighthall in this issue ofICU Management) where teamwork has also been found crucial for safety, specialcourses have been developed to train the team skills required for maintaining safety.To develop similar training programmes within intensive care, an analysis of the skillsrequired for effective teamwork in the ICU is necessary.

Rhona Flin, PhDProfessor of Applied

Psychology

Thomas Reader, MAPostgraduate Research

Student

Industrial Psychology

Research Centre, University

of Aberdeen

[email protected]/iprc



Finally, research using performance measures and self-reported ratings of teamwork has revealed that higherratings by ICU staff of interdisciplinary collaborationskills (e.g. communication) are associated with lowerrisk-adjusted lengths of stay for patients, lower rates ofstaff turnover and lower rates of patient mortalities(Baggs et al. 1999).

Training team working skills In other high-risk industries where failures in teamworkhave also been shown to be a major source of humanerror, Crew Resource Management (CRM) training pro-grammes have been developed to meet the challenge ofreducing error. These programmes were initially desig-ned by psychologists and pilots to increase the effec-tiveness of teams through enhancing skills such as com-munication, leadership and cooperation. CRM coursesgenerally consist of lectures, role plays, discussions,accident analyses, case studies, and video re-enact-ments of accident scenarios. The team working skillstaught on CRM courses can include: assertiveness andspeaking up; asking questions; listening; providingappropriate feedback; attending to and observing non-verbal signals; maintaining team focus; supporting otherteam members; team decision making; resolving con-flicts; considering others; and sharing mental models(Flin et al. 2002). In Europe, CRM training is mandatedfor airline pilots and in the UK their CRM skills are re-evaluated during licence checks.

The skills taught on any CRM programme should bebased on a framework developed from an empiricalanalysis of the skill requirements for a specific domain.It is not sufficient to simply take the training materialsused in aviation, replace the word ‘pilot’ with ‘doctor’,and then apply the materials in intensive care. A CRMcourse developed for any setting should reflect the spe-cific challenges and needs of that setting (Helmreich &Merritt, 1998). An analysis of skill requirements involvesinvestigating the root-causes of accidents and nearmisses, analysing performances during normal or emer-gency operations (e.g. from voice recorders), andresearching organizational climate factors that mayaffect performance. This provides diagnostic data thatenables the identification of the key cognitive and inter-personal skills that are relevant for the particular job,thus providing an empirical basis for the training content.To evaluate the subsequent use of the skills, and there-fore the effectiveness of a CRM training programme,behavioural marker systems are developed (Flin et al.2002). These are empirically derived taxonomies ofobservable requisite non-technical skills for a particulardomain. Behavioural marker systems allow trainedraters to assess team skills through rating of behavioursknown to be indicative of good cognition and team work-ing. Through the use of behavioural marker systems,CRM training in aviation has been shown to result in sig-nificant increases in the target skills by flight crewsalongside improved attitudes towards the importance ofteamwork skills. Behavioural marker systems based onempirically derived taxonomies of non-technical skillshave been developed for several high-risk industries, aswell as medical domains such as anaesthesia (see table1) and surgery (Patey et al. In press; Yule et al. 2005).They have also been used to demonstrate that simulatortraining can enhance anaesthetists’ non-technical skills(Yee et al. 2005).

Training team skills for the ICUAs the ICU is an environment in which good team work-ing skills are essential for safety and performance, thepotential benefits of team training are apparent.However, the ICU is a unique and complex domain thatmakes specific demands from those who work within it,and thus the precise team skills required for effectiveteam working within the ICU need to be identified. Anumber of techniques can be employed to identify theseskills, including attitudinal surveys, structured inter-views, observations of behaviours in real-time and insimulation, and studies of cognition. This process facili-tates the development of a skills taxonomy for guidingand structuring the training of those skills. Furthermore,it allows the development of a behavioural marker sys-tem which enables the evaluation of team working skillsthrough an observable set of exemplar skills.

Table 1. Teamwork skills outlined in the Anaesthetists Non-technical skills (ANTS) taxonomyFor more details please see the ANTS website (www.abdn.ac.uk/iprc/ants), and Reader et al.(BJA, in press) for relevance to the ICU.

Team Working

Skill elementsCo-ordinating activities with teammembers

Exchanging information

Using authority and assertiveness

Assessing capabilities

Supporting others

Skills for working in a group context, in any role,to ensure effective joint task completion andteam member satisfactionDefinitionsWorking together with others to carry out tasks, for bothphysical and cognitive activities; understanding the rolesand responsibilities of different team members, andensuring that a collaborative approach is employedGiving and receiving the knowledge and data necessaryfor team coordination and task completionLeading the team and/or task (as required), accepting anon-leading role when appropriate; adopting a suitablyforceful manner to make a point, and adapting this for theteam and/or situationJudging different team members’ skills, and their abilityto deal with a situation; being alert to factors that maylimit these and their capacity to perform effectively (e.g.level of expertise, fatigue)Providing physical, cognitive or emotional help to othermembers of the team

© 2004 University of

Aberdeen.


Residents and patient safetyPatient safety has emerged as a compelling issue inhealthcare. Intensive care is one of the largest, mostexpensive, and complex components of healthcare,accounting for approximately 30% of acute carecosts (Pronovost 2002). Complex systems – of whichICUs are an example – are breeding grounds forerrors, because interdependent components interactin unexpected ways. Errors resulting in adverseevents are common in ICUs (Pronovost et al. 2004)and adverse events related to resident errors arewell-documented (Wu et al. 1991). Focusing theattention of physicians-in-training towards patientsafety represents a major opportunity for improvingthe safety of care in the ICU.

Little discussion has emerged, however, regardinghow physicians-in-training can integrate their workand learning into these new patient safety initiatives.This observation has stimulated educators to call fordramatic changes in undergraduate medical educa-tion to incorporate elements of patient safety(Rosebraugh et al. 2002). James Reason (1995)defined seven broad categories of conditions thatpromote medical errors: high workload, inadequateknowledge, ability or experience, poor interfacedesign, inadequate supervision or instruction, stress-ful environment, mental state (e.g. fatigue, boredom)and change. Residents who enter an ICU rotationface most of these error-promoting conditions. ICU isan inherently hazardous environment in which humanerror is inescapable.

Goals in training should therefore focus on how toimprove systems of care and how to manage theconsequences of errors if they occur. John Heffnerand colleagues have written extensively on the issueof residents in training in the ICU (2005), and I sum-marize here the key educational issues they raise forpromoting the contribution of residents to patientsafety.

New teaching methodologiesTraditional didactic courses and methods of teachingare limited in their effectiveness compared withnewer approaches, such as problem-based learning,interactive forms of education, small group discus-

sion and simulation with videotape feedback duringresidency training. The best predictor of perform-ance is practice. This underpins the new vision ofmedical education with the adoption of simulation.Simulation allows trainees to learn all the skillsrequired in a risk-free environment. A broad array ofsimulation models should be incorporated into theresident patient safety curriculum to ensure compe-tency in a variety of cognitive and procedural sub-jects. These new techniques promote greater under-standing of patient safety and successful change inclinical performance (see also Gaba et al. in thisissue of ICU Management).

Resident orientation to the ICUIt is no longer acceptable to have an increase inadverse events with the induction of new residentsinto an ICU. More errors occur in new work environ-ment, which implies that ICU orientation is an essen-tial element of patient safety. Detailed orientationsmust include training in the processes and proce-dures of the ICU, an introduction to key staff, a copyof relevant protocols and guidelines, and a discussionof the duties of the rotation and degree of responsi-bility. Clear instructions are essential on who to call ifin doubt, especially for support outside normal work-ing hours.

Resident supervisionDuring the training, a call for assistance is often inter-preted as a sign of weakness. This promotes reluc-tance of residents to involve senior staff in care deci-sions after hours. An appropriate safeguard to over-come this problem is to encourage nurses and othercaregivers to activate their chain of command whenthey note resident deviation from standard practices.Additionally, the Accreditation Council of GraduateMedical Education (ACGME) now recommends thatthe responsibilities of the faculty be extended tomonitor residents for signs of fatigue or stress.Supervision is an essential requirement for promot-ing patient safety in the ICU and resident well-being.Levels of supervision can be graduated appropriatelywith trainees’ progress.

Training residents for safety

Medical errors related to residents are well-documented. Focusing the attention ofphysicians-in-training towards patient safety represents an opportunity for improvingsafety in the ICU. In this article, Dr Mariam Al-Ansari summarises the key educationalissues for promoting the contribution of residents to patient safety drawn from thework of John Heffner and colleagues.

Mariam A. Al-Ansari

MD, FRCSIConsultant Intensivist

Salmaniya Medical Complex

Kingdom of Bahrain

[email protected]



Team buildingTeamwork is especially important in the ICU. Thehierarchical model of patient care wherein the physi-cian prefers to be preeminent on the healthcareteam, does not promote optimal patient care andsafety. Team building is assisted by interdisciplinarytraining, which rarely occurs in resident level educa-tion, despite having been shown to improve patientcare. An ICU safety curriculum should provide knowl-edge on how to work effectively as part of a team inthe dynamic environment ofthe ICU, where the commongoal in the team is patient safe-ty and optimal outcome.

Information technologyResidents face multiple distrac-tions that interfere with theirprovision of quality ICU care. Innoncomputerized ICUs, forexample, they often divert timefrom patient care to find charts and information.Information technology in the ICU has a proven effecton the physician performance, with reduction in thefrequency of errors and associated adverse events,and a decrease in patient mortality (Dimick 2005).Physicians in training should have a thorough knowl-edge of applications and the value of new informa-tion technologies towards improving patient safety.

Error and adverse event reportingRecognizing and reporting medical errors andadverse events represents a critical opportunity forresidents’ learning. This positive response to medicalerrors is likely to promote residents’ willingness toreport errors when they occur. Confidential inter-views can facilitate the reporting of errors by resi-dents who are rotating on clinical services.

Medico-legal education and disclosureResidents demonstrate little understanding of how todisclose adverse events and errors (Lester and Tritter2001; Pilpel et al. 1998). 76% of residents reportedthat they had never disclosed a serious error to apatient (Wu et al. 1991). A patient safety curriculumshould teach the resident the importance of fullacceptance of responsibility for errors. Open disclo-sure by communicating adverse events in an honest,open, and empathetic manner is likely to decreasemedico legal risk (Mazor et al. 2004).

Residents should be taught to provide patients andfamilies with statements explaining that an error oradverse event has occurred, a description of thenature of the event, why it occurred, how recur-rences will be prevented and an apology.

Resident support after errorsPhysicians are often the secondvictim of medical errors.Contributing to adverse patientevents triggers anger and selfblame and sometimes the tempta-tion for intellectual dishonesty bynot disclosing errors. Blame hasbeen the traditional response to

errors in recent years and this philosophy needs tobe eliminated, a point to be emphasized in thepatient safety curriculum. As a core component, cur-ricula need to support the emotional needs of theresident, and educate on the real sources of errors,which are usually to be found in faulty systems.

Clinical auditingAnalysis of adverse events is an essential componentof care improvement. In these analyses, the role ofthe individual and underlying system defects shouldbe correctly emphasized to improve care, and oppor-tunities used for teaching residents. Such analysesprovide valuable opportunities for residents’ educa-tion, for gaining clinical skills and promoting individ-ual and organizational performance improvement.Trainees in the ICU should therefore participate inclinical audits and mishap analyses.

Quality improvement methodologyCurricula should increase the awareness of physi-cians in training about quality improvement method-ologies. Participation in at least one ICU qualityimprovement project should be a requisite of trainingfor critical care fellows. A patient safety curriculumteaches the residents and fellows in the critical train-ing program the nature of healthcare systems andgrounds them in quality improvement and total qual-ity management methodologies.

SEND LETTERS TO THE EDITOR & REQUESTS FOR REFERENCES CITED IN ICUMANAGEMENT TO: [email protected]

adverse events related

to resident errors are

well-documented


ContentArticles may focus on any management or practice issuein intensive care related to economics, quality of care orpatient outcome. We only accept scientific papers with aclear connection to management and practice issues. Wealso invite Viewpoints for publication, which are personalopinions of the author, and Letters to the Editor, which arepublished at the discretion of the Editors.Submissions may not have been published previously orbe currently submitted for publication elsewhere. Articlesmust be written by independent authorities and any spon-sors for research named. If manufacturers are named inan article, the text must present an unbiased view, notsupporting any particular company.

Submission guidelinesAuthors are responsible for all statements made in theirwork, including changes made by the editor and author-ized by the submitting author.The text should be provided as a word document via e-mail to [email protected]. Please provide acontact e-mail address for correspondence.Following review, a revised version, which includes theeditors’ comments and recommendations, is returned tothe author (at the contact e-mail address) for autho-rization.

LengthArticles: maximum 700 words (less if figures or tables areincluded)Viewpoints: maximum 700 wordsLetters to the editor: maximum 175 wordsPlease note that contributions longer than the specifiednumber of words may not be accepted.

StructureArticle texts must contain:• a title;• names of authors with abbreviations for the highest aca-

demic degree; • affiliation: department and institution, city an country;• main authors are requested to supply a portrait photo

(see specifications below);• a summary of one or two sentences (no more than 30

words) describing the content;• one contact name for correspondence and an e-mail

address which may be published with the article;• website, if appropriate;• acknowledgements of any connections with a com pany

or financial sponsor;• an introduction, main text and summary/conclusion,

with subheadings as appropriate;• authors are encouraged to include checklists and/or

guidelines which summarise findings or recommenda-tions;

• references or sources, if appropriate, as specifiedbelow.

Writing styleArticles must be written in English (with English spelling),with short sentences, a clear structure (see above) andwithout bias. Full stops in numbers may only be used toindicate a decimal place; otherwise use commas as sep-arators.

ImagesMain authors are invited to supply a portrait photo for pub-lication with their article. This and any other relevantimages for publication with an article should be sent by e-mail as separate files (only high resolution images with300dpi) and their order of placement in the article mustbe clearly indicated. Only the electronic formats “.tif” or“.jpeg” can be used for images, i.e. not Microsoft Word orPowerPoint. Images must be no smaller than 9cm x 9cmat 100% scale. Only images meeting these specificationscan be published. If an image has been published before,permission to reproduce the material must be obtainedby the author from the copyright holder and the originalsource acknowledged in the text, e.g. “© 2004 KirstieEdwards”.

Format for referencesPlease use the Harvard reference system. Citations with-in the text for a single author reference should include theauthor surname and year of publication; for a citation withtwo authors include both author surnames and year ofpublication; for more than two authors, include the firstauthor surname followed by "et al." and the year of publi-cation. Multiple citations should be separated by a semi-colon, and listed in alphabetical order.Example of within text citation: (Edwards 2004; Edwardsand Miller 2002; Miller et al. 2003).The format for listing references in submitted articlesshould follow the Harvard reference system.Example of standard journal reference: Sydow Campbell,K. (1999) "Collecting information; qualitative researchmethods for solving workplace problems", Technicalcommunication, 46 (4) 532-544.Readers will be provided with an e-mail contact for refer-ences, which will be kept on file and supplied onrequest.Authors are responsible for the accuracy of the refer-ences they cite.

AcceptanceIt is always at the discretion of our editorial board to acceptor refuse submissions.We will respond to submissions with-in 8 weeks of receipt. We reserve the right to revise the arti-cle or request the author to edit the contents, and to publishall texts in any Euromedical Communications journal, on theInternet and to list them in on line literature databases.

Thank youThe ICU Management Editorial [email protected]

Author guidelines for ICU Management

The appropriate dose of renal replace-ment therapy (RRT) of patients withacute renal failure (ARF) is a matter ofconsiderable debate. The dose of inter-mittent dialysis is quantified using theunitless index, Kt/V, where K representsurea clearance, t is the time of dialysisand V is the volume of distribution ofurea. Existing literature supports a single-pool Kt/V urea of at least 1.2 per treat-ment, at least 3 times weekly, as the min-imum dose in patients with end-stagerenal disease (ESRD) (National Kidney

Foundation 1997). Arguably, most patients with ARF aremore ill, malnourished and catabolic than ESRD patientsand hence warrant “more” RRT. While supported bysmall, retrospective and non-randomized studies(Paganini 1996; Schiffle 1997), until recently this hypoth-esis had not been tested by randomized trials.Although an optimal dialysis dose has not been estab-lished in patients with ARF, it is generally accepted thatthe delivered dose of dialysis should be at least as greatas that recommended for ESRD. Despite this, a recentprospective study of 40 patients (136 dialysis treat-ments) with ARF treated with intermittent haemodialy-sis (IHD), reported that prescribed Kt/V was less than 1.2in 49% of treatments, and more importantly deliveredKt/V was less than 1.2 in nearly 70% of treatments(Evanson et al. 1998). In a recent study, Schiffl et al.assigned 160 critically ill, but haemodynamically stable,patients with ARF to daily or every other day haemodial-ysis, in alternating order (Schiffl et al. 2002). The twostudy groups were similar at baseline. Mortality was28% in patients assigned to daily IHD as compared to46% with alternate-day dialysis (P=0.01). Dailyhaemodialysis also resulted in faster resolution of ARF(mean [± SD], 9 ± 2 days vs. 16 ± 6 days; P=0.001), bet-ter control of uraemia and fewer hypotensive episodesduring haemodialysis than conventional haemodialysis.Although this study is supportive of a more intensivedialysis prescription in ARF, it has several important limi-tations. First, the exclusion of haemodynamical-ly unstable patients eliminated the sickestpatients (these patients were treated with con-tinuous RRT instead) and diminished generaliz-ability. Second, the non-random assignment ofpatients to groups may have introduced bias,although the reported baseline characteristics ofthe two groups appear similar. Finally, the deliv-ered dose of therapy in the alternate-day group

was substantially lower than accepted as “adequate”haemodialysis as described above.In continuous haemofiltration, dose of therapy correlateswith effluent flow rate (Clark et al. 1992; Clark et al.2003). Using effluent flow as an index of dose of thera-py, a recent single centre RCT demonstrated that higherhaemofiltration doses improved patient survival in ARFcompared to conventional doses, while further increas-es in dose were not helpful (see table 1: Ronco et al.2000). Of note, more than 90% of patients in this studyreceived the prescribed dialysis dose. However, a sec-ond recent smaller RCT did not show similar results. Inthis study, 106 patients with oliguric ARF (defined asrefractory to furosemide) were randomized to three(almost equal) groups for early high- and low-volume andlate low-volume haemofiltration (see table 1: Bouman etal. 2002). No differences in 28-day survival or duration ofARF were found between these groups. This study washowever limited in that it was underpowered and proba-bly enrolled less sick patients as suggested by very highsurvival rates.Similar to IHD in ARF, CRRT delivery may not reach thelevels prescribed. In a single centre retrospective reviewof CRRT dosing patterns, we found that the mean CRRTdose prescribed for patients with ARF was only 24.46 ±6.73 ml/Kg/h, and that the mean dose delivered wasmerely 16.55 ± 5.41 ml/Kg/h (68% of the prescribeddose, p<0.000001) (Venkataraman et al. 2002). Whilethere was high concordance between prescribed anddelivered effluent flow rates in this study, treatmenttime was reduced (16.1 ± 3.53 hours/day) due to inter-ruptions in therapy.

Although these clinical studies suggest that more inten-sive renal support may improve survival, they have sig-nificant limitations and hence are not widely acceptedinto clinical practice. An ongoing multicentre trial(Palevsky et al. 2005) is now comparing intensive renalsupport to conventional management of renal replace-ment therapy in critically ill patients with acute renal fail-ure to provide a more definitive answer to this question.


In the winter issue of ICU Management 2005, Drs Kellum and Venkataramanreviewed evidence on timing of initiation of renal replacement therapy in Acute RenalFailure (ARF). In this article, they review the limitations of research to date on theappropriate dose of renal replacement therapy of patients with ARF.

M AT R I X F E AT U R E S

Dialysis Dosing in Acute Renal Failure

RameshVenkataraman

MD

John. A. Kellum, MDThe CRISMA Laboratory

(Clinical Research,

Investigation, and Systems

Modelling of Acute Illness)

Department of Critical Care

Medicine

University of Pittsburgh

USA

[email protected]

ICU StakeholderAnaesthes io logy

Cardiology

Pharmacy

In terna l medic ine

M icrob io logy

Nephrology

Respira tory

. . .

Table 1. Results of two effluent flow studies Study Dose SurvivalRonco et al. 2000 20ml/kg/hr 41 %

35ml/kg/hr 57 %45ml/kg/hr 58 %

Bouman et al. 2002 Early 72-96 L per 24 hrs 75 %Early 24-36 L per 24 hrs 68.8 %Late 24-36 L per 24 hrs 74.3 %

ICU Management 1-200622

Out-of-hospital cardiac arrest isassociated with a very poor prog-nosis. Only 5-14% of all patientssuffering cardiac arrest are expect-ed to be discharged from hospital(Böttiger et al. 1999; Newman et al.2000). Although several drug thera-pies during cardiopulmonary resus-citation (CPR) have been proposed,none has proved to improve long-term outcome in these patients

(Brain resuscitation clinical trial I study group 1986 & trialII study group 1991; Kudenchuk et al. 1999). The twomajor underlying diseases leading to sudden cardiacarrest in more than 70% of cases are acute myocardialinfarction (MI) or ischemia-related arrhythmia and mas-sive pulmonary embolism (PE) (Silfvast 1991; Spauldinget al. 1997; Zipes and Wellens 1998). Systemic thrombol-ysis is an effective therapy for acute MI or PE withhaemodynamic instability (Arcasoy and Kreit 1999).Thrombolytic therapy during CPR causes direct throm-bolysis at the site of coronary or pulmonary artery occlu-sion. In addition, experimental and clinical studies haveshown a marked activation of coagulation during CPRwhich is not counterbalanced by an adequate activationof fibrinolysis (Böttiger et al. 1995; Gando et al. 1997).This generalized hypercoagulable state leads to forma-tion of microcirculatory thrombi which may severelyimpair organ function even after restoration of sponta-neous circulation, especially in the brain (Fischer et al.1996).

The fear of causing life-threatening bleeding complica-tions which may be increased by mechanical CPR, how-ever, has been a major drawback for using thrombolyticdrugs during CPR. Consequently, thrombolytic agentshave historically been withheld in the setting of cardiacarrest. However, in patients failing to achieve restorationof spontaneous circulation despite immediate advancedcardiac life support after in-hospital cardiac arrest, throm-bolysis during CPR has been used as a last resort thera-py. As a result, case reports and small case series ofthrombolysis published over the last 30 years have sug-gested an improved long-term outcome of neurological-ly intact survivors without causing critical bleeding com-plications (Padosch et al. 2002).

In addition, several clinical studies on thrombolysis duringout-of-hospital cardiac arrest have shown an improvedshort-term outcome. The Heidelberg thrombolysis trial(Böttiger et al. 2001) was the first prospective, controlledstudy that compared thrombolytic treatment during CPRwith standard treatment in patients who had failed toachieve spontaneous circulation after more than 15 min-utes of conventional resuscitation. Patients treated withthe thrombolytic drug alteplase were admitted to hospitalsignificantly more frequently as compared to the controlgroup (58% vs. 30%). In addition, there was a trendtowards improved survival at hospital discharge in thethrombolysis group (15% vs. 8%). These results wereconfirmed by a retrospective study with 108 out of hospi-tal patients who received alteplase during CPR (Ledereret al. 2001). Compared to 216 conventionally resuscitatedpatients both short and long term survival were improvedin the thrombolysis group. Significantly more patientswho were treated with alteplase survived to discharge(25.0%, as compared to 15.3% in the control group). Therecent guidelines for cardiopulmonary resuscitation rec-ommend considering thrombolytic therapy in patients suf-fering cardiac arrest in whom an acute thrombotic etiolo-gy for the arrest is suspected (Nolan et al. 2005).

In order to assess the efficacy and safety of a generalizeduse of thrombolytic drugs during CPR in patients sufferingcardiac arrest of presumed cardiac origin (i.e. acute MI orPE), a large randomised study is currently under way inEurope. The Thrombolysis in Cardiac Arrest (TROICA) trialis a double-blind, placebo-controlled multicentre trial onthrombolysis during CPR after out-of-hospital cardiacarrest, designed to enrol more than 1000 patients in tenEuropean countries. Adult patients who suffer witnessedout-of-hospital cardiac arrest of presumed cardiac origincan be randomised, if basic or advanced life support isstarted within 10 minutes of onset. Patients presentingwith asystole, however, are not included. Patients are ran-domised to receive a weight-adjusted dose oftenecteplase or placebo. Primary endpoint will be survivalat 30 days and at hospital admission; secondary end-points include neurological performance of survivingpatients and survival after 24 hours. Safety endpoints willassess the incidence of symptomatic intracranial haemor-rhages and major bleeding complications. Enrolment isexpected to be completed by summer of 2006.

Risk of life-threatening bleeding complications has prevented the use ofthrombolytic drugs during CPR, although recent cases suggest improved short andlong-term outcomes in certain patient groups. The Thrombolysis in Cardiac Arrest(TROICA) trial is assessing the efficacy and safety of a generalized use of throm-bolytic drugs during CPR in patients suffering cardiac arrest.

Thrombolytic therapy during cardio-pulmonary resuscitation: the TROICA trial

M AT R I X F E AT U R E S


Cardiology

Pharmacy


M icrob io logy

Nephrology

Respira tory

. . .

Fabian Spöhr, MD

Bernd W. Böttiger

MD, DEAA

Dept. of Anesthesiology

Ruprecht-Karls-University

Heidelberg, Germany

http://erc.edu/index.php/newsitem/en/nid=146

[email protected]


Consultant pharmacist posts have been estab-lished and will be expected to provide signifi-cant leadership, research and educational roles,in addition to expert clinical practice.Importantly, these standards have been pro-duced in conjunction with a new pay schemeaffecting pharmacists, in which they mustdemonstrate the necessary competencies andexperience (i) to take up a post and (ii) toprogress along the pay scale within that post.In spring 2006, accredited pharmacists will beable to prescribe medication independently forthe first time. This legislation provides a huge opportu-nity for clinical pharmacists to take even greater respon-sibility for individual patient care within the multidiscipli-nary team. It will allow pharmacists, for example toundertake TDM, dose alterations in multiorgan failure,parenteral nutrition and optimise antimicrobial therapy.

Further workNationally, the current pharmacy workforce is lacking thenumbers of pharmacists with appropriate critical careexperience, as demonstrated by particular difficulties inrecruitment to these positions. It is therefore paramountthat larger critical care units and pharmacy departmentshave recognised rotations for clinical pharmacists, there-by increasing their skills and fostering an interest in acritical care career. It is only in this way that we can hopeto meet the needs of the 24/7 activity of critical care(Department of Health 2005b). While some centres haveachieved positive steps to address these needs, this isnot the norm. Clearly, the future workforce developmentplan needs a national accreditation system that incorpo-rates these requirements.Clinical pharmacy services are not alone in experiencingthe challenges of funding their activity. It is hoped thatnational acceptance of staffing levels in addition toagreed staff skills and experience will ease the incorpo-ration of clinical pharmacy costs in any critical careexpansions.

ConclusionClinical pharmacy services are accepted as integral tonational recommendations for critical care. The chal-lenges ahead include reducing national variations anddeveloping a skilled pharmacy workforce not only capa-ble of meeting the service needs of this specialty, butalso to contribute to the clinical, audit and researchagendas to improve the future care of the critically illpatient.

Richard S. Bourne

MSc, MRPharmSLead Critical Care

Pharmacist

Sheffield Teaching

Hospitals, UK

[email protected]

Development of critical care clinicalpharmacy services in the UK

New legislation and national standards in the UK promote the role of the critical care clini-cal pharmacist towards integration in the critical care team and improving patient care.

Traditionally, critical care clinical pharmacy services haveoften been developed in response to requirements toimprove cost effective use of medicines and/or reducemedicine risk. While these aspects of medicine manage-ment remain paramount, there is increasing recognitionof the skills clinical pharmacists contribute to the care ofthe critically ill patient (Intensive Care Society 1997).Critical care pharmacists reduce adverse drug reactions,optimise medicine utilisation, improve fluid manage-ment, and reduce patient morbidity and mortality as wellas having a positive pharmacoeconomic impact (Kane2003; Papadopoulos et al. 2002). These results areachieved by integration within the critical care multidisci-plinary team. Pharmacists attend medical ward rounds,review individual patient's medication, develop drugguidelines, educate staff, and manage therapeutic drugmonitoring (TDM), medication risk, and medicine auditand research.Appropriate use of pharmacy technicians for roles suchas medicines management, drug audit and drug expen-diture reporting enable clinical pharmacists to spendmore time on clinical care as recommended in ASpoonful of Sugar (Audit Commission 2001).A major difficulty for clinical pharmacy services for criti-cal care in the UK has been the lack of national stan-dards, demonstrated in the variation in practice andservice provision (Timmins 2000). Some critical careareas had no pharmacist input; others had newly quali-fied, inexperienced pharmacists clearly attempting topractice beyond their current capability, while others hadteams of experienced critical care pharmacists(Department of Health 2004). Predictably the perceivedvalue of critical care clinical pharmacy services fluctu-ates widely from hospital to hospital accordingly. Therecent development of national recommendations forboth critical care pharmacist staffing levels and experi-ence required, has gone some way to correcting thisvariation (Department of Health 2003).

Developments in Clinical Pharmacy ServicesAs the role of the critical care pharmacist gains furthernational recognition, it underlines the need to make fur-ther improvements in the standardisation of serviceswhich units receive. In the UK, we have made some sig-nificant developments in this area and now have nation-al recommendations for core knowledge standards forspecialist critical care pharmacists. A career frameworkfor clinical pharmacists within critical care has beenestablished including the competencies required forthese positions (Department of Health 2005a).


Cardiology

Pharmacy


M icrob io logy

Nephrology

Respira tory

. . .


IntroductionHaemodynamic monitoring is part of routine clinicalpractice in the intensive care scenario and its applica-tion is now spreading further to other critical careareas such as the emergency department. The firstdevice able to give some estimate of the cardiac out-put at the bedside was the pulmonary artery catheter(PAC). It has been used now for nearly 30 years andeven if the PAC is still considered the gold standardfor measurement of cardiac output, several concernshave been raised regarding its invasiveness, useful-ness and associated complications (Boyd et al. 1983).Indeed despite the growing evidence that monitoringcardiac output can guide fluid and drug managementin the haemodynamically unstable patient, there isstill a lack of evidence on which form, if any, is bestfor monitoring cardiac output. The use of the PACopened the way to the use of dilutional techniquesfor the determination of cardiac output. In this paperwe will briefly describe devices using intravascularcatheters for the determination of cardiac output inthe intensive care unit.

Thermal dilution was the first dilution technique usedin clinical practice. Nowadays two kinds of thermaldilution are possible with PAC: intermittent thermaldilution and semi-continuous inverted thermal dilu-tion (Boldt et al. 1994). Another device developedmore recently is the PiCCOplus by Pulsion. It usesthermal dilution to measure the cardiac output and toderive volumetric indices. This is then used to cali-brate a continuous measure of stroke volume (SV),taken from the arterial waveform (Goedje et al. 1999).As well as thermal dilution, dye dilution techniqueshave been used in clinical practice. Currently the onlydevice available on the market that uses a dye dilu-tion technique to measure cardiac output is LiDCO,by LiDCO, UK. These measurements can be used tocalibrate an algorithm for the continuous determina-tion of cardiac output (Jonas and Tanser 2002).

Is a dilution technique fundamental to the determina-tion of cardiac output? Recently two new deviceshave been developed that analyse the arterial pres-sure waveform. PRAM, FIAB, Italy and VigileoMonitor with Flotrac Sensor, Edwards Lifescience,CA, USA, can track stroke volume and cardiac outputfrom the analysis of any arterial pressure waveform(Manecke et al. 2004; Romano and Pistolesi 2002).Nowadays some devices can give an estimate of the

cardiac output without the need for calibration, forinstance through thoracic bioimpedance and oeso-phageal doppler monitoring; neither of these twotechniques require the placement of an intravascularcatheter.

PACThe PAC provides measurement of cardiac output,(through intermittent thermodilution), of pulmonaryartery pressures and of pulmonary artery occlusionpressure (PAOP). It has been modified in recentyears by adding a thermal coil which semi continu-ously produces thermal waves. This allows a semi-continuous measurement of the cardiac output. Themost recent development in PAC technology is thevolumetric PAC. This is equipped with a rapidresponse sensor and integrates the signal from theelectrocardiogram. In this way it calculates the rightventricular ejection fraction and then, combining thisdata with the SV, calculates the right ventricular enddiastolic volume (RVEDV: Cheatham and Right 2000).Intermittent measurement of cardiac output usingthe PAC is still considered by many to be the goldstandard of clinical cardiac output measurement.Semi-continuous cardiac output has been validatedagainst intermittent thermodilution and offers thegreat advantage of not requiring any operator todirect its measurement. The accurate placement ofthe catheter is the most important requirement forreliable measurement. Although cardiac output cannow be measured by other less-invasive devices, thePAC is the only device that allows continuous meas-urement of pulmonary artery pressure and determi-nation of the pulmonary artery occlusion pressure(PAOP). Many PAC’s also continuously monitor themixed venous oxygen saturation (SvO2). Despiteconcerns regarding the usefulness of both PAOP andRVEDV to predict preload status, there are somepatient groups, for instance acute right ventriculardysfunction, where the continuous monitoring of pul-monary artery pressures remain both important anduseful.

Despite much criticism of the PAC in recent years,especially regarding the lack of evidence forimproved outcome in critically ill patients, there areseveral studies of haemodynamic optimisation usingPAC to monitor cardiac function and to direct therapywhich have shown some improvement in outcome(Boyd et al. 1993).

P R O D U C T C O M PA R I S O N : C A R D I A C O U T P U T

Cardiac output monitoring withintravascular cathetersIn this paper we present the main characteristics of devices that can provide cardiacoutput measurement and which require the placement of an intravascular catheter.

M CecconiDept. of Anaesthesia and

Intensive Care

University of Udine, Italy

E McKinney St George's Hospital,

London

A RhodesSt George's Hospital,

London

[email protected]

lithium dilution and continuous cardiac output ofLIDCO have been validated in several conditions(Hamilton et al. 2002; Tsutsui et al. 2004). A recentstudy utilising the LiDCOTMplus to target-directedtherapy versus standard management in high-risksurgical patients showed an improved outcome inthe treatment group (Pearse et al. 2005). This sys-tem, like the PiCCO can also provide the functionalparameters that can predict fluid responsiveness.

Vigileo monitor and PRAMThe Vigileo system and PRAM are the most recenthaemodynamic monitors that can track changes instroke volume from the analysis of the arterial wavetrace. The most interesting characteristic of thesemonitors is that they require no calibration and onlyan arterial line for their use. Flotrac is the sensorused by Vigileo to transduce the arterial line trace tothe monitor. The algorithm, based on the standarddeviation of arterial pressure waveforms, uses age,weight and sex of the patient and incorporates theinformation obtained from the analysis of the arterialtrace to calculate stroke volume and cardiac output.Vigileo has been validated so far in only two studies,showing reasonable accuracy and precision in com-parison with the PAC (Manecke et al. 2004; McGee etal. 2005). PRAM uses an analysis of the arterial pres-sure waveform which is based on the physics of per-turbation. It analyses the points of “perturbance onthe wave trace” and from this analysis it calculatesthe impedance of the system, bypassing the problemof calibration. PRAM has already been validatedagainst the PAC in cardiac surgery patients, showinggood accuracy and precision (Giomarelli et al. 2004).Whilst preliminary studies performed with thesetechnologies appear promising, further investigationsare required to establish if they can replace moreinvasive monitors.

ConclusionsThe intensivist can now choose from several devicesfor monitoring cardiac output in critically ill patients.Degree of invasiveness is very different dependingon which device is chosen while some monitors pro-

vide parameters that others cannot. It willbe important in coming years not only tosee if the new devices that are replacingthe PAC prove to be as accurate, but alsoto determine what clinical impact treat-ments guided by their use can provide.We will hopefully have an answer soon.So far it is difficult to proclaim which, ifany, is best.

PiCCOplus by PulsionThis monitor is less invasive than PAC, requiring onlya central line and a specialised arterial catheter(femoral or radial) for set-up. Like PAC it uses ther-modilution as the technique for measurement of car-diac output. A bolus of cold saline is injected throughthe central line and transpulmonary thermodilution isregistered and computed via the specialised arterialcatheter. The value of the cardiac output is then usedto calibrate the algorithm of the pulse contour analy-sis available on PiCCOplus. In this way the monitortracks the SV beat to beat giving a continuous esti-mate of cardiac output. Both transpulmonary ther-modilution and continuous cardiac output monitoringhave been validated in clinical practice in several sit-uations (Della Rocca et al. 2003; Rodig et al. 1999).

In addition to the cardiac output, the PiCCOplus,through the analysis of the thermodilution curve,gives “volumetric parameters” such as global enddiastolic volume (GEDV), intrathoracic blood volume(ITBV) and extravascular lung water (EVLW). Theseparameters have proven to be better indices of pre-load than PAOP and CVP. EVLW is an interestingparameter that represents the amount of water in theextravascular thoracic space. High values of EVLVcorrelate well with pulmonary oedema and havebeen associated with poor outcome in the intensivecare population (Marting et al. 2005). In addition,PiCCO is able to monitor the functional haemody-namic parameters of stroke volume variation (SVV)and pulse pressure variation (PPV). These have beenshown to reliably reflect fluid responsiveness inpatients who are intubated and sedated (Michard etal. 2000).

LiDCOLiDCOTMplus is a new cardiac output monitor. Itmeasures cardiac output using transpulmonary dilu-tion of lithium. This technique requires the injectionof 0.3 mmols of lithium through a central or peripher-al line and the dilution curve is detected via a sensorthat can be attached to an arterial line already in-situ.Technically LiDCOTMplus requires only a peripheralline and an arterial line. Through the transpulmonarydilution of lithium the system calibrates an algorithmfor continuous cardiac output monitoring derivedfrom the arterial pressure wave analysis. Both the


P R O D U C T C O M PA R I S O N : C A R D I A C O U T P U T M O N I T O R S

LETTERS TO THE EDITOR &REQUESTS FOR REFERENCESCITED IN ICU MANAGEMENT:[email protected]

All of ECRI’s products and services are availablethrough the European Office, addressing the specialrequirements of Europe and the UK. Utilising someof the world’s largest health related databases, help,support and guidance can be given to our Europeanclients at a local level.

Amongst its many products and services ECRI ispleased to provide readers of ICU Managementwith sample information on products for cardiac out-put measurements using thermal dilution, designedfor use in critical care from its Healthcare ProductComparison System (HPCS), which contains over 280reports. The HPCS reports contain extensive informa-tion about the technology, its purpose, its principlesof operation, stage of development and reportedproblems.

This extract from our database contains model bymodel specifications for easy assessment andreview and also includes ECRI’s ‘Recommended

specifications’ (generic tem-plates) which can be usedfor comparison and tender-ing purposes.

The data are extracted fromECRI’s 2003 database andhave additionally beenreviewed and updated by therespective manufacturers.Publication of all submitteddata is not possible: for fur-ther information please con-tact ECRI or [email protected].

ECRI is a totally independent non profit researchagency designated as a Collaborating Centre of theWorld Health Organization (WHO). Such organiza-tions are appointed to contribute to WHO’s publichealth mission by providing specialized knowledge,expertise, and support in the health field to the WHOand its member nations. ECRI is widely recognizedas one of the world’s leading independent organiza-tions committed to advancing the quality of health-care with over 240 employees globally.

ECRI’s focus is medical device technology, health-care risk and quality management, and health tech-nology assessment. It provides information servicesand technical assistance to more than 5,000 hospi-tals, healthcare organizations, ministries of health,government and planning agencies, voluntary sectororganizations and accrediting agencies worldwide.Its databases (over 30), publications, informationservices and technical assistance services set thestandard for the healthcare community.


P R O D U C T C O M PA R I S O N : C A R D I A C O U T P U T M O N I T O R S

ECRI EuropeWeltech Centre Ridgeway

Welwyn Garden City

Herts AL7 2AA, United

Kingdom

Tel. +44 (0)1707 871511Fax. +44 (0)1707 393138

[email protected]

Cardiac output monitors

Footnotes used in pages 29 and 30

ECRI recommendations1 These recommendations are the opinions of ECRI's technology experts.ECRI assumes no liability for decisions made based on this data.

EDWARDS LIFESCIENCESE1 (9.5 x 11.5 x 8.2)

PHILIPSPh 1 The M1012 is just the CO measurement module. The M3012A measurementextension provides invasive pressures and temperature with CO as an option.They can be used with a large number of different patient monitors from Philips,from small transport monitors that can run on batteries for several hours to largehigh end systems with multiple high resolution touch displays. Obviously, weightand dimensions of the complete monitoring devices are larger. http://www.medical.philips.com/main/products/patient_monitoring/products/intellivue_mp70_mp60/

PULSIONPu 1 ContinuousPu 2 Not currently available in the USA, FDA clearance pending


Healthcare Product Comparison System

ADtext

titelMODELWHERE MARKETEDFDA CLEARANCECE MARK (MDD)CONFIGURATION

CO RANGE, L/minTHERMAL INJECTATE

Injection vol, mL

Temp range, °CThermistor for injectatetemp

DYE DILUTIONDATA ENTERED

Method

REMOTE START SWITCHOUTPUT COMPUTATION

DATA DISPLAYED

NO. STORED CURVES/CALCULATIONS

NO. CURVES AVERAGEDSELF-TEST CURVECIRCUITS SELF-TESTEXTERNAL DISPLAY DEVICES

CATHETERSCompatible brands

THERMISTOR CIRCUIT FAULTINDICATOR

H x W x D, cm (in)WEIGHT, kg (lb)

POWER SOURCE, VAC (Hz)

Battery typeRechargeableOpen time, hrChargerLow-battery indic

PURCHASE INFORMATIONPrice, unit 7 Fr, balloon tipWarrantyDelivery time, AROYear first soldFiscal year

OTHER SPECIFICATIONS

ECRI RECOMMENDEDSPECIFICATIONS1

BASIC

Stand-alone or modular

0-12

3, 5, 10

0 to 25Any

OptionalCatheter resistance and sensitivity,injectate volume and temperature

User selected

OptionalCO, blood and injectate tempera-tureThermal dilution curve, blood andinjectate temps, CO consumption,trace size, position controls, haemo-dynamic calculations

6

5OptionalYesOptional

Visual

When a correct calculation cannotbe provided, a message shoulddescribe the problem and give anexplanation.

HEMO 2:4: HEMONED PODSWorldwideYesYesPods for SC7000, SC8000,SC9000XI, Delta, Delta XL, Kappaand Kappa XLT. Vista XL andGamma X XL (without CO) supportthe HemoMed Pod0.5 to 20

3, 5, 10

-5 to +30T-piece

NoPatient height and weight, injectatevolume, catheter size and type

Start key on pods, menu on themonitorRemote KeypadIntegration cuts off at 50% of peak;exponential extrapolationCO, blood temperature, injectatetemperature; optional with calcula-tions package: BSA, CI, CCI, SV,SVI, SVR, SVRI, TVR, PVR, PVRI,TPR, LVSW, LVSWI, RVSW, RVSWI,LHCPP, RPP50

5NoYesScope, recorder, printer atMultiview Workstation

Arrow, Baxter, Ohmeda, Edwards,Abbott/MedexMessage on monitor, MultiviewWorkstation14 x 20.5 x 6 (5.5 x 8.1 x 2.3)Hemo4 0.9 (1.9) Hemo 2 andHemoMed 0.7 (1.6)110/220 (50/60) from monitor

NoneNANANANA

Not specified1 yearNot specifiedNot specifiedNot specifiedAll data stored in monitor/ multiviewworkstation.

VIGILANCE IIWorldwideYesYesStand-alone

1-20

3, 5, 10

0 to 30Edwards Immersion or flow throughprobeNoCatheter size, injectate volume,computation constant automaticallyset when most EdwardsLifesciences catheters are selected;manually set for othersNavigation knob

NoIntegration cuts off at 30% of peak;exponential extrapolationCCO, CCI, CO, CI, SvO2,ScvO2,EDV, EDVI, SVR, SVRI, RVEF, SV,SVI (additional measured or calculat-ed data available in the smallparameter frame windows)

6

6OptionalAutomatic2 serial ports, 1 analog output port

Edwards Swan Ganz catheters, PreSep oximetry cathetersVisual message, audible alarm

24.1 x 29.2 x 20.8 E1

3.43 (7.5)

100-240 (50/60) power consump-tion: 40 watts maximumNoneNANANANA

Not specified1 yearNot specifiedNot specifiedJanuary to DecemberTrending; CEDV and SvO measure-ment capabilities optional. UL certi-fied.

M1012AWorldwideYesYesModule for the V24, CMS, and Intelli-vue patient monitoring systems. TheM3012A measurement extensionprovides invasive pressures & tem-perature with CO as an option0.1 to 20

As specified with computation con-stant0 to 27 Bath or in-line probe

NoComputation constant based oninjectate temperature, injectate vol-ume, and catheter characteristics

Softkeys

Optional handswitchCO, CI, blood and injectate temper-aturePA, injectate temperatures; bloodtemperature; computation constant;CO; CI; BSA; average CO, curve

6

Selectable up to 6YesYesPrinter, optional recorder

Any with standard connector

Message on display

9.9 x 3.6 x 9.7 (3.9 x 1.4 x 3.8)0.23 (0.5)

NA (internal)

NoneNANANANA

Not Specified1 year (module)Not Specified1989January to DecemberUp to 6 curves displayed for review& selection for use in determinationof avg CO value; directly links toCMS/ Intellivue haemodynamic cal-culations; measurement reliabilityby multiplecurve alert message;when abnormal thermodilutioncurve is detected, alerts includenoisy baseline, thermal drift, slope,multiple peaks, delayed injection,and injectate temp.Ph1


ADICU design series

Does it matter?

Ergonomics

Guidelines

IC in the Battlefield

Improvising with

architecture

Healthcare Product Comparison System

MODELWHERE MARKETED

FDA CLEARANCECE MARK (MDD)CONFIGURATION

CO RANGE, L/minTHERMAL INJECTATE

Injection vol, mLTemp range, °C

Thermistor for injectatetemp

DYE DILUTIONDATA ENTERED

MethodREMOTE START SWITCHOUTPUT COMPUTATION

DATA DISPLAYED

NO. STORED CURVES/CALCULATIONS

NO. CURVES AVERAGEDSELF-TEST CURVECIRCUITS SELF-TESTEXTERNAL DISPLAY DEVICES

CATHETERSCompatible brands

THERMISTOR CIRCUIT FAULTINDICATOR

H x W x D, cm (in)

WEIGHT, kg (lb)POWER SOURCE, VAC (Hz)Battery type

RechargeableOpen time, hrChargerLow-battery indic

PURCHASE INFORMATIONPrice, unit 7 Fr, balloon tipWarrantyDelivery time, AROYear first soldFiscal year

OTHER SPECIFICATIONS

DASH 3000 - 4000 - 5000Worldwide

YesYesConfigured

0.2 to 15

3, 5, 100 to 30

Depending on catheter brand

NoPatient height and weight, catheterbrand, injectate temp sensing type(in-line or bath), catheter size, injec-tate volume; optional: computationconstantTrim knobAuto mode or Remote controlBSA, HR, MAP, CVP, PAM, PAW,CO, CI, SC, SV, SVR, SVRI, PVR,PVRI, LVSWI, RVSWIThermal dilution curve, CO, blood &injectate temps, computation con-stant, time stamp

No curves, all calculations withinthe last 24 hours4NoYesPrinter, central station (CIC)

Edwards (Baxter), Abbott, B&D(Ohmeda, Spectramed), Arrow,OtherVisual message on display

Dash 3000: 26 x 28 x 20Dash 4000: 27 x 29 x 24Dash 5000: 29 x 31 x 245.1 to 6.4 kgDepending on regionLithium ion; 2x hot-swappableYes4 to 5 hrYes: integrated and optional externalYes

Not specified1 year parts & labor4 weeks1999Not specifiedComprehensive messages relatedto technical/ physiological problems,like "Instable Blood Temperature"

SOLAR/TRAMWorldwide

YesYesModular

0.2 to 15

3, 5,100 to 30

Depending on catheter brand

NoPatient height and weight, catheterbrand, injectate temp sensing type(in-line or bath), catheter size, injec-tate volume; optional: computationconstantTouchscreen or trim knobAuto mode or Remote controlBSA, HR, MAP, CVP, PAM, PAW,CO, CI, SC, SV, SVR, SVRI, PVR,PVRI, LVSWI, RVSWIThermal dilution curve, CO, blood &injectate temps, computation con-stant, time stamp

No curves, all calculations withinthe last 24 hours4NoYesPrinter, central station (CIC)

Edwards (Baxter), Abbott, B&D(Ohmeda, Spectramed), Arrow,OtherVisual message on display

Not applicable (modular system)

Not applicable (modular system)Depending on regionNo batteryNot applicableNot applicableNot applicableNot applicable

Not specified1 year parts & labor4 weeks1990Not specifiedComprehensive messages relatedto technical/ physiological problems,like "Instable Blood Temperature"

PICCO PLUSEU, USA, Canada, Russia,Asia/Pacific (incl. Japan, China,India), South America, Middle East.YesYesStand-alone, integrated in COmodule for Philips and DrägerInfinity SmartPod

0.25 to 25

2-200 to 24

PULSION injectate temperaturesensorNoPatient height and weight, injectatetype, injectate volume, CVP, auto-matic catheter detection

Membrane keypadNo; acoustic signal "ready"Stewart Hamilton formula

Cardiac output: COtherm, PCCOPu1,SVPu1, Preload: GEDV, ITBV;Afterload: SVRPu1 Contractility:GEF, CFI, dPmaxPu1,2 Lung water:EVLWPu2, PVPIPu2 Volume respon-siveness: PPVPu1,2, SVVPu1

Others: MAPPu1, APsysPu1,APdiaPu1, HRPu1, R-L-shunt Pu2

5 curves, 50 thermo dilution results,7 day trendUp to 5NoYesBlood pressure data transfer to anybedside monitor; Interface solutionfor bedside monitors and patient datamanagement systems via data port

PULSIOCATH

Message on display

15.8 x 26 x 25 (6.2 x 10.2 x 9.8)

4.8 (10.6)115/230 (50/60)Sealed lead D-cells Yes0.5 minimumInternalSymbol on screen

Not specified2 yearsdevices on stock1997calendar yearMinimally invasive; transpulmonarythermodilution measurement fordetermination of cardiac output,preload volume, and extravascularlung water; Continuous cardiac out-put and stroke volume by pulsecontour analysis; continuous vol-ume responsiveness R-L-shuntdetection and quantification; Verbalerror messaging on screen.

91496/90496Worldwide

YesYesModular; compatible with anySpacelabs Ultraview or PCMS moni-tors2. With appropriate softwarelevels.0.1 to 18 ±10%

5, 10 0-27.5°C injectate and 17.2-43°Cblood.In-line or bath probe

NoPatient height and weight, HR, BSA,MPAP, MAP, CVP, PCWP, computa-tional constant

User selectedNoStewart Hamilton formula

CO, CI, PVR, SVR, SVRI, PVRI,LVSW, SV, SVI, RVSW, LVSWI,RVSWI, blood temperature, injec-tate temperature

5/30 measurements

Up to 5NoYesMonitors, bedside and centralrecorders/printers

All major

Visual

11.3 x 5.7 x 18 (4.5 x 2.2 x 7.1)

0.8 (1.8)From monitor; <5 WNoneNANANANA

Not specified1 yearNot specified2004 (91496); 1998 (90496)July to JuneEdit; auto average; user prompts;self-tests; haemodynamic calcula-tions; 24 hr trends.


M A N A G E M E N T

Clear and effec-tive dialogue isneeded betweenthe intensive careteam (physicians,nurses and sup-port staff) and thea d m i n i s t r a t i v emanagers of the

hospital and healthcare system. Such dialogue is nec-essary to ensure the efficient function of the health-care organization. From the perspective of the inten-sivist, administrative understanding and support isparamount in maintaining the financial and physicalresources necessary to provide excellent and safepatient care. From the perspective of the administra-tion, intensive care units represent a significant finan-cial outlay, comprising as much as 30% of the totalhospital budget (Bekes et al. 2004). In order to appre-ciate potential barriers to effective communicationbetween these entities, it is important to understandthe culture and general priorities of each and todesign strategies and systems to overcome theseobstacles when they are present.

Barriers to Communication with AdministratorsA number of potential barriers to effective communi-cation exist between clinical providers and adminis-trators. In many situations, the clinicians have noformal business training and may look upon adminis-trative tasks as an unwelcome intrusion into theirbusy practices. Even when enthusiastic about admin-istrative responsibility, lack of management trainingand true differences in the language of clinical medi-cine and business administration may impair mean-ingful discourse (Atun 2003). A clear understandingof business terms is required for the intensivist teamto communicate their needs in a means that allowsadministration to understand the need, and demon-strates to administration that the intensivist and theirteam are dedicated to institutional goals. Thus a thor-ough understanding of fixed and variable costs isrequired. Furthermore, an appreciation for the con-text in which these costs exist is also required. Forexample, the intensivist must understand whetherpayment is based upon a per-diem basis or bundledinto diagnosis-related groups (DRG). Reducing length-of stay under a per diem payment system may bebeneficial in the long term but acutely may lessencash flow and thus hurt the institution, whereas

reductions in length-of-stay under a DRG-like systemwill immediately contribute positive margin.

The goals of clinicians and administrators are notoften aligned. Intensive care clinicians frequentlyvalue autonomy of practice, commitment to theirpatients and specialty while gaining a source of iden-tity from their profession. Administrators are morelikely to value a team and committee approach withloyalty to the organization, and tend to identify withtheir position in the organizational hierarchy (Nowickiand Summers 2002). External pressures includingthe cost of information technology, new patient tech-nology, and expensive therapies that may not neces-sarily demonstrate a proven cost savings benefit mayfurther polarize physicians and administrators.Increasing consumerism and access to informationon the worldwide web has led to greater expecta-tions from patients and their families. Additionally,escalating malpractice costs, decreasing reimburse-ment for services, government budget cuts, an agingpopulation, pressure to increase clinical workloadand an ever more restrictive policy environment allserve to widen the schism between the perceivedgoals of practitioners and administrators (Rundall etal. 2004). Tension over resource allocation can lead toan environment where administrators are viewed asonly being concerned about the bottom line, and cli-nicians are seen as wasteful of valuable resources.Distrust, alienation and power struggles may result.It is clear that intensive care unit directors are facingan increasing demand for services in an environmentof rapidly increasing costs and diminishingresources. In order to create and sustain a positiveenvironment of care and practice for patients and theclinical staff, while contributing to the financial healthof the organization as a whole, intensivists andadministrators need to remove the barriers to effec-tive communication, seek common ground and worktogether interdependently.

Improving Communication with AdministratorsStrategies to improve communication between clini-cians and administrators begin with appreciating thedifferences in focus and values outlined above.Formal or informal training in principles of businessadministration is beneficial for intensivists in bridgingthe cultural and language gap with managers.Learning to speak in the language of the administra-tion cannot be stressed enough. Understanding the

In previous articles in this series, Drs Pauldine and Dorman have discussed communica-tion skills in conflict resolution, and strategies for effective communication in intensivecare practice. In this article, the authors focus on how to communicate effectively withadministrators.

Communications seriesConflict and communication - autumn 2005

Effective communication - winter 2005

Communicating with administration - spring 2006

Communication with the Administration

Ronald Pauldine

MD, Fellow

Todd Dorman

MD, FCCM

Dept. of Anaesthesiology

and Critical Care Medicine


School of Medicine

Baltimore, US

[email protected]

35

overall goals or mission of the healthcare organiza-tion is useful in ensuring that the short-term andlong-term plans for intensive care resources arealigned with the vision of senior leadership (Burmahl2003). The ability to demonstrate and clearly articu-late the benefits and return on investment of a well-run and well-supported intensive care unit can makea strong argument for the provision of responsibleleadership and active support of this expensiveresource.

Plans to be presented to administration should beresearched and prioritized in advance. Recall that com-munication can be accomplished through a variety ofmedia and formats. Communication can take the formof face-to-face meetings, group encounters, telephoneconversations, written correspondence, email, or otherprinted media such as newsletters. An appreciation andin-depth understanding of your administrator’s leader-ship style and preferred learning style can be extreme-ly helpful. Is live presentation preferred over writtencommunication? If so, does the administration preferinformal presentations or formal briefings with audiovi-sual aids? Should there be a written follow-up? Isemail effectively utilized or is a formal document pre-ferred? Do scheduled meetings at regular intervalsmeet the need or should meetings be scheduled on an“as needed” basis? Who should request the meeting?Are your communication needs met by existing stand-ing committees? Obviously, the answers may vary withthe issues at hand and the culture of the institution.However, the point of tuning in to the right method forthe people involved and the question presented is animportant one. Finally, these alternate means can beextremely useful, but the value of face-to-face meet-ings must not be underestimated.

If lines of communication are closed or difficult toaccess, methods such as structured dialogue can beconsidered to direct the formulation and presentationof specific plans, by soliciting meaningful input fromthe clinical staff and administration. Structured dia-logue has been recommended as a method for admin-istrators to partner with physicians in a systematicscheduled exchange of information that allows clini-cians to be part of the decision-making process, there-by increasing their influence and stake in their profes-sional and economic affairs (Cohn et al. 2005). A seriesof scheduled meetings with the input of a committeeof clinician leaders has been suggested as a format. Inorder to be successful, the environment must fostercooperation between all of the stakeholders with acommitment to implement the recommendations ofthe panel. Milestones to measure progress should be

implemented. In building trust and strengthening rela-tionships between the intensivist staff and administra-tion, communication techniques such as appreciativeinquiry utilize principles that recognize achievementand success, as opposed to dwelling on shortcomingsand problems. Such techniques are based on the con-cepts that people respond to positive reinforcementthat strengthens self esteem, shared vision is theengine that drives lasting change, and affirmation andenvisioning of goals increases the likelihood that thegoals will transform into reality. While it is importantfor the intensivist to understand administrative styles,goals, and values, communication is clearly a two waystreet. Administrators send very potent messagesthrough the way they listen, seek to build partnershipswith clinical staff, and exhibit willingness to share riskand reward.

Another important area for communication withadministrators is in creating an institutional climate ofsafety. The use of executive walk rounds has beendemonstrated to be beneficial in improving nursingattitudes regarding the climate of safety on inpatientunits where leadership participated in regular discus-sions with the providers on the unit (Pronovost et al.2004; Thomas et al. 2005).

The goal of communication up the administrative chainis for problem solving or resource allocation as itrelates to the delivery of services in a particular health-care organization. However, the importance of com-munication systems is magnified in times of crisis.Effective and efficient communication between thecritical care group and hospital administrators wasdemonstrated to be of extreme importance during theSARS outbreak in Toronto in 2003. In an assessmentof lessons learned from the situation, a recommenda-tion was made for critical care communities to consid-er existing systems for communication in advance ofcrisis (Booth and Stewart 2003). This interplay stress-es the importance of the relationship between clini-cians, their hospital administration and the regionalcritical care community as a whole.

Intensive care professionals can improve communi-cation with administrators by becoming familiar withtheir culture, language, values and specific manage-ment styles, while using techniques to align goalsand build teams. The long-term result of establishingimproved communication between clinicians andadministrators is in the development of shared per-spectives that result in an increase in mutual respectand trust, leading to the most effective and efficientuse of intensive care resources.

icu-management.org

Risk-adjustment, i.e. the “normalization” of raw mortali-ty rates according to the severity of illness for eachpatient is a method that has been used for a long timein epidemiology. In intensive care medicine, risk adjust-ment became popular in the early nineties, shortly afterpublication of the first systems by Knaus et al. 1981 &1985, Le Gall et al. 1983 & 1993 and Lemeshow et al.1993.

Today, severity-of-illness scores have a fixed place inintensive care research: the adjustment for the risk todie within the hospital that a certain patient carries at thetime of ICU admission or within the first hours in theICU. At the end of the nineties, however, some studiesunveiled a lack of prognostic performance of the currentsystems: in most cases an underestimation of mortalityin low-risk patients and an overestimation in high-riskpatients. This pattern was observed for all the publishedmodels (Apolone et al. 1996; Moreno and Morais 1997;Moreno et al. 1983; Metnitz et al. 1999). For this reason,several researchers tried to improve the prognostic per-formance of various systems through recalibration (alsocalled “customization”), which generally did not solvethe various problems inherent to the models.

These problems can be classified as either user-, patient,or model-dependent. User-dependent problems includedifferences between, or ambiguous definitions, and theirnon-uniform application. Patient-dependent problemsare mainly related to shifts in the baseline characteristicsof the populations and the therapeutic options overtime: besides differences in the age distribution, severalnew techniques have been developed (such as advancesin minimal invasive surgery or liver replacement therapyetc.). Model-dependent problems often also include thelack of important prognostic variables, such as diagnos-tic information or the presence of infection. For this rea-son a group of intensivists (SAPS 3 Outcome ResearchGroup, SORG) have developed a new risk-adjustmentsystem for inten-sive care. Afterseveral years ofpreparation, theSAPS 3 study wascompleted at theend of 2002. Morethan 340 ICUsworldwide docu-mented all admit-ted patients over aperiod of two

months. The subsequent data analysis then took morethan one and a half years.

The main model – the SAPS 3 Admission Score - whichhas been published recently in Intensive Care Medicine(Metnitz et al. 2005; Moreno et al. 2005) uses data fromwithin one hour prior and after ICU admission. The scoreconsists of 20 variables, which can be ordered in threelogical boxes (see table 1). The score can thus be usedon a variety of levels. First, patients can be comparedselectively on one or more levels, such as status beforeICU admission (Box I) or the physiological derangement(Box III). Moreover, the sum score (Box I + Box II + BoxIII) can be used as a measure of the overall severity of ill-ness. In addition, the sum score can be entered into alogistic regression equation and a predicted hospitalmortality can be calculated. This “expected” mortalitycan then be compared with the observed mortality,which gives the so called O/E ratio. The O/E ratio canthen be used to statistically compare mortality ratesfrom different groups of patients (or ICUs).

Intensive care medicine is, however, a very heteroge-neously plasticized medicine. For this reason, seven dif-ferent regions were defined and separate mortality pre-diction equations developed to allow comparisons atboth general and regional levels. Additionally, a tool forcomparisons by major patient typologies is under devel-opment. Comprehensive supporting information for cal-culations has been published with the study reports(www.springer.com) and additional information is avail-able at www.saps3.org.

SAPS 3 has successfully addressed some of the prob-lems of currently used prognostic models, providing themost modern and accurate risk-adjustment system avail-able. Whether the SAPS 3 is adopted in ICUs dependson the willingness and motivation of intensivists to inte-grate a new system into their environment.


M A N A G E M E N T

A new risk-adjustment system for ICProblems with existing risk assessment models for assessing the likelihood of a patientdying following admission to intensive care have recently been addressed by the SAPS3 study and development of a new model, the SAPS 3 Admission Score.

Rui P. Moreno

MD, PhD, Prof.Unidade de Cuidados

Intensivos Polivalente

Hospital de St. António dos

Capuchos

Lisboa, Portugal

[email protected]

Philipp G.H. Metnitz

MD, PhD, Prof., DEAADept. of Anaesthesiology

and General Intensive Care

University Hospital of

Vienna, Austria

On behalf of the SAPS 3

Investigators

Table 1. SAPS 3 Admission Score Variable CategoriesCategory Type of variablesBox 1 Information on patients before ICU admission: age, previous health status,

co-morbidities, location before ICU admission, length of stay in the hospital before ICUadmission, and use of major therapeutic options before ICU admission

Box 2 Information on circumstances of ICU admission: reason(s) for ICU admission, surgicalstatus and anatomic site of surgery (if applicable), planned or unplanned ICU admission,

, and infection at ICU admissionBox 3 Information on the presence and degree of physiologic derangement at ICU admission,

within 1 hour before or after admission

On 29th and 30th January 2006, 34 experts cametogether in Lisbon for a consensus meeting followingpreliminary research results of the ELDICUS project.This project aims to make triage decisions for the elder-ly in Europe more transparent, fair and cost-effective,and to deliver recommendations for public policytowards harmonising European standards. Literatureresearch, data on more than 8000 triages across Europe(the largest trial of its kind) and the expertise of theinvestigators have contributed to this second stage of aconsensus on 60 statements, designed to promote anobjective triage decision making process. In addition todirectors and professors of intensive care, the consen-sus group comprised Jane Barratt, Secretary General,and Yitzhcak Brick, President, of the InternationalFederation for Aging, Avi Israeli, Director General of theIsraeli Ministry of Health, and specialists from externalhealthcare consultancies, anaesthesiology, gerontology,statistics, economics and biophilosophy.

The ELDICUS project has comprised a number of challeng-ing aims, including the design of a standardized triage scorefor ICU, the comparison of ICU costs across France, Spain,Italy, Denmark, the UK, the Netherlands and Israel, andsocial linkage analysis to help identify effective triage deci-sion making solutions. Patient data before and after triagewere collected, including demographic and physiologicdata, and mortalities. Participating ICUs completed activityand cost forms based on the cost block methodology pre-viously used in the UK (Edbrooke et al. 2004). Additionally,a social linkage analysis was completed by Professor ofApplied Ethics, Guido Van Steendam, from K.U. Leuven.Based on the results of the study, and in part on earlierguidelines published by the Society of Critical CareMedicine (Society of Critical Care Medicine Ethics Com-mittee 1994), 60 statements and guidelines for decisionmaking in triage were drawn up and circulated in December2005 for consideration by all the investigators and expertsattending the consensus meeting in January 2006.

On the first day of the consensus meeting, the prelimi-nary results of the ELDICUS project were presented. Inaddition, investigators and experts discussed results fromthe first evaluation stage for each of the 60 statements indepth, including recommendations for amendment, rejec-tion or deletion of statements, and voted on each point.With consensus defined as agreement of over 80%, 42 ofthe 60 statements were accepted or amended as appro-priate and agreed on during this second stage. All state-

ments were reconsidered in a third stage on the secondday of the meeting, with emphasis on discussing thosewithout consensus. A further eight statements wereagreed on and detailed comments on the remaining state-ments recorded to guide a subgroup of experts in a finalround of consideration and amendment.

In the ELDICUS multi-centred trial, 50% of the triagedpatients were over 65 years, with significant variation inrefusal rates for the elderly between the different coun-tries. Further in depth qualitative analysis was proposed atthe consensus meeting to research what elements of thedecision making process may lead to lower refusal ratesof the elderly. Particularly focusing on the role of age intriage decision making, statement 31, which acquired100% consensus in the third round, was concluded as“Age should never be the sole determining factor in triagedecisions.” Social linkage analysis, however, identifies thatone selection criterion can be used to hide another morecontroversial criterion. To further protect the elderly sothat no decisions based on age can be hidden, and thosewho can benefit will benefit, the group agreed unani-mously on a further statement related to physiologicalstatus and age: “Physiological status is more importantthan chronological age in triage decisions”.

Perhaps the most significant statement agreed on in thisthird round of consideration, was “An objective triagescore should be used by physicians to help make triagedecisions for individual patients”. Social linkage analysissuggests that selection criteria cannot be truly objective,because many criteria such as “quality” and “benefit”require subjective interpretation. The group debated,however, that objectivity and subjectivity are not adichotomy, but a continuum, so that aiming for theobjective end of the continuum remains a realistic goal.Additional solutions to not achieving the ideal of anentirely objective tool are recognition of the fact, andkeeping the door open for continuous improvement.

Final consensus on all the statements is yet to be con-cluded following further statistical analysis and interpreta-tion of the data collected in the ELDICUS study, furtherconsideration of all the statements by the subcommittee,and final voting by all the experts and investigators.

ELDICUS is an EU funded project and closes at the endof April 2006, although analyses, meetings and publica-tions are expected to continue long after.

Charles L. SprungDirector General Intensive

Care Unit

Department of

Anaesthesiology and Critical

Care Medicine

Hadassah University

Hospital, Jerusalem, Israel

David EdbrookeProfessor of Intensive Care

Medicine, Royal Hallamshire

Hospital, Sheffield, UK

On Behalf of the ELDICUS

investigators

AcknowledgementThe ELDICUS project is

funded as part of the

European RTD project

ELDICUS – “Triage decision

making for the Elderly in

European Intensive Care

Units” funded by the

European Commission

(Contract No. QLK2-CT-2002-

00251) and in part by the

European Society of

Intensive Care Medicine.

Objective triage for the elderly:ELDICUS update

37icu-management.org

ELDICUS investigators move another step towards identifying the elements for an objectivetriage decision making tool in a consensus meeting of 34 experts. ELDICUS is the projectacronym for “Triage decision making for the ELDerly in European Intensive Care UnitS.

M A N A G E M E N T

Outcomes as an indicator of quality of care have become increasingly important in thepast decade. Assessment of patient satisfaction reflects care from the patient’s viewpoint.The development of valid and reliable instruments to measure patient satisfaction is the first step in continuously improving patient care.


Satisfaction is a part of outcomequality, in addition to clinically ori-entated 'traditional' outcomes(e.g. mortality), economic meas-urements (costs) and health-related quality of life, and hasbecome an important endpoint inoutcomes research (Cleary et al.1988; Orkin 1999). The concept

of satisfaction is complicated, and influenced by cultural,socio-demographic, cognitive and affective components(Aharony et al. 1993). Many theories include patients'expectations as the basic concept of satisfaction (Calman1988; Thompson et al. 1995; Wu et al. 2001). A traditionaldefinition of satisfaction is therefore the degree of congru-ence between expectation and accomplishment (Pascoe1983). Consequently, the involvement of patients in thedevelopment of an instrument to measure satisfaction isvery important and must be an integral part of develop-ment. Unfortunately, most instruments have not consid-ered this aspect and are therefore of questionable value(Le May et al. 2001). This has contributed to the poor rep-utation of patient satisfaction as an indicator of the qualityof healthcare services (Westbrook 1993).From surveys of the US and Europe that consciouslyadopted the patient's perspective, we know that patientsatisfaction is primarily determined by aspects such as'respect for patients’ values', 'information', ‘coordinationand continuity of care', 'physical comfort', 'emotional sup-port', and 'involvement of family' (Allshouse 1993;Delbanco 1992). This has also been shown when measur-ing patient satisfaction with anaesthesia care in Europeancountries (Auquier et al. 2005; Heidegger et al. 2002).Measuring patient satisfaction requires the application of avalid and reliable method of measurement. Only a highquality psychometric instrument will be able to generatehigh quality data (Avis 1997; Roberts et al. 1987). Mostinstruments used are questionnaires that are completedby the patients themselves. This technique allows surveyswith (relatively) higher numbers and a lower budget thanface-to-face or other personal interview methods. For thisquantitative research, usually highly standardised instru-ments are applied. Qualitative interviews are of greatimportance in the phase of generating instruments inorder to evaluate all relevant aspects. This approach is,however, (usually) too expensive for broad-based data col-lection.The most important points during the construction ofquestionnaires are content validity, criterion validity, con-struct validity, reliability, and practicability (DeVellis 1991;Hall et al. 1988; Streiner et al. 1998).

Content validity: All relevant aspects of satisfaction need tobe included in the questionnaire, integrating patient andexpert views, and evaluation of the state of the art for sim-ilar constructs. Focus groups with patients who havealready gained experience with healthcare, help to collectitems, assure content validity and avoid relevant parts ofpatient perception of care from being omitted. An evalua-tion of the ‘state of the art’ considers and incorporatesaspects from other studies measuring similar constructs,if appropriate.

Criterion-related validity: Aspects, which are related statis-tically to central outcome parameters such as overall satis-faction show criterion-related validity - in a causal interpre-tation also called predictive validity.Thus, items and scalesbelieved to assess an important aspect of patient satisfac-tion must demonstrate such a relationship in terms of acorrelation to a central outcome parameter.

Construct validity: Construct validity is the extent to whicha measure 'behaves' in the same way as the construct itrepresents (DeVellis 1991). An important point is whetherthe relevant aspects are translated in a comprehensiveway into questions which truly measure them. Questionssimply relating to overall satisfaction are inadequate;patient satisfaction should be measured multi dimension-ally using a multi-item technique for each aspect.

Reliability: Besides test-retest reliability, scale reliability(internal consistency) is of great importance. This is basedon correlation and determines to what extent the incorpo-rated items (or questions) are measuring the same under-lying construct (latent variable), for example “information“.

Practicability: The instrument should be as economicalas possible, and include everything that is necessary forthe measurement of patient satisfaction (content validi-ty), but no more. To achieve high response rates(> 60%), questionnaires should be concise and sentwithin five weeks after discharge (Saal et al. 2005), withone reminder, if possible.

Patient satisfaction as part of outcome quality has gainedgreat importance in the past decade. The development ofhighly standardised, valid and reliable instruments is a pre-requisite to gain plausible data. The patient's involvementshould be an integral part of this process. Results of singleitem ratings of overall satisfaction are over-optimistic anddo not represent the true indication of care. Conclusionsshould only be drawn from results of well-designed, psy-chometrically developed instruments.

Measuring patient satisfaction

Thomas Heidegger

MD, DEAA

Quality Manager in Healthcare

Consultant Anaesthetist

Cantonal Hospital St. Gallen

Switzerland

[email protected]

Detlef Saal

MD, DEAAConsultant Anaesthetist

Cantonal Hospital St. Gallen

Switzerland

Yvonne Husemann

Lic. Phil.Healthcare Consultant

Switzerland

Matthias Nuebling

PhDGEB: Empirical Consulting

Denzlingen, Germany

Research to practice seriesPerception and practice, autumn 2005

Research from nursing practice, winter 2005

Appraising the evidence, winter 2005

Outcome measures

Evidence to practice

This small study investigated Intensive Care provi-sion in four countries, the United Kingdom (UK),France, Hungary and Germany, from data collected in2000. There has been no published data on the provi-sion of Intensive Care beds for about 7 years andthus this research provides a much needed updatefor this decade. Whilst there is little information tocompare the differences internationally, the few avail-able studies have reported the number of beds per100,000 population (Burchardi et al. 1994; Rapoportet al. 1995; Miranda 1986).

It’s difficult to assess the number of Intensive CareUnit (ICU) beds per 100,000 population due mainly todifficulties regarding ICU definition. In France, ICUsand intermediate care units are merged in the sameunit, whilst in the UK, high dependency units (HDUs)are often organisationally separate from ICUs (conse-quently in table 1, HDU provision is included withinthe ICU totals in the UK figures). In addition mostEuropean ICUs admit a larger number of electivepost-operative patients than the UK.

For France and Germany the number of ICUs avail-able from national data was utilised and the mediannumber of beds for each country was applied to thisfrom the sample available. Population estimateswere derived from a standard source (www.popula-

tionwold. com) (see table 1). This data shows that theUK provides fewer ICU beds than the other threecountries.

Table 2 shows other data on healthcare within thesame four countries for comparison. GDP per capitais defined as a measure of the total goods and serv-ices produced within a specific territory and within adefined time period per person. When convertedusing cross country comparators, such as purchasingpower parities or exchange rates, it can be used tocompare simplistically the wealth of a country. TheGDPs in France, Germany and the UK are similar, butthe UK seems to expend approximately 30% less percapita on healthcare. As a percentage of GDP,Hungary and the UK spend considerably less thanFrance and Germany.

Is this spending of the four countries reflected inmarkers of healthcare provision? Available markersinclude the number of hospital beds and physiciansavailable per 100,000 population and if France andGermany were used as a standard, one would expectthe number of hospital beds per 100,000 populationin the UK to be 604. In fact, it is much lower at 421.The predicted number of physicians in the UK usingthe same model would be 259 whereas the actualprovision is 212. Surprisingly, Hungary provides as

40

M A N A G E M E N T

IC beds cf. other healthcare provision infour European countries

This paper describes differences in ICU provision (reported as beds per 100,000population) in France, Germany, Hungary and the UK.These data are then compared towider healthcare indices including markers of expenditure and healthcare provision.The large differences highlight the necessity to collect new international data to explainthese differences.

ICU Management 1-2006

Table 1. Provision and Costs of ICUs per 100,000 populationCountry Population Total No. ICU beds Source & Date of ICU bed data

ICU beds per 100,000population

UK 59,040,300 3213 5.44 Department of Health 2005:this figure includes highdependency beds

France 59,303,800 5310 9.8 Garrouste-Orgeas et al. 2005:Predictors of intensive careunit refusal

Germany 81,904,100 22,887 27.94 www.gbe-bund.de 2003: this figure includes intermediate care.

Hungary 10,164,100 1153 11.41 National Registry, HungarianSociety of Anaesthesiology andIntensive Care(www.anesz.info) 2004

David L. Edbrooke

FRCAProfessor Intensive Care

Royal Hallamshire Hospital

Sheffield, UK

dedbrooke @aol.com

Bertrand Guidet, MDProfessor, Intensive Care

Hospital St. Antoine

Paris, France

Akos Csomos

MD, DEAAConsultant, Intensive Care

Markhot Teaching Hospital

Eger, Hungary

Thomas Prien, MScProfessor, Intensive Care

University Hospital

Munster, Germany

Gary H Mills

PhD, FRCAConsultant Intensive Care

Royal Hallamshire Hospital

Sheffield, UK

many hospital beds and physicians as France andGermany, despite the GDP per capita of Hungarybeing lower. Hungary provides nearly twice as manyhospital beds and 50% more physicians per 100,000population than the UK.

Returning to ICU beds, if the number of ICU beds per100,000 patients was also utilised as an indicator ofhealthcare provision, it is clear that the UK’s health-care expenditure per person is not reflected in thenumber of ICU beds provided. There is no gold stan-dard as to how many beds should be provided per100,000 population, but the frequent transfers thatoccur between ICUs and the cancellation of electivesurgery in the UK, may well indicate an under provi-sion of intensive care.

In conclusion, whilst the authors accept that the fig-ures calculated for each country are estimates andmay not be representative of the whole country andthat it is difficult to calculate the number of ICU bedswith the lack of an internationally accepted definition,we contend that the reporting of these figures repre-sent a starting point for expanding and refining datacollection for ICU at an international level, which is anessential step to explain the large differencesobserved.

Table 2. WHO healthcare data (World Health Report 2004 reporting for 2002)

France Germany Hungary UK

GDP per capita 26,809 26,205 13,473 26,273

Healthcare Expenditure per person (US $) 2,567 2,820 914 1,989

Health Expenditure as % of GDP 9.7 10.9 7.8 7.7

Hospital Beds per 100,000 population 780.11 892.69 783.53 421.82

Physicians per 100,000 population 334.86 336.93 324.57 212.61

icu-management.org

The Medical Economics and Research Centre

(MERCS) in Sheffield has launched an initiative with

ICU Management to collect data on how many

adult general ICU beds are available in different

countries. If you can help, please contact

[email protected]

Adult Patient Database (APD)The APD was established in 1992 (Stow et al.2006) and currently holds over 560,000records. Local data collection software is pro-

vided free by the Society to all Australian and NewZealand sites. Units submit electronic de-identified dataquarterly and receive periodic reports comparing their per-formance against similar units. Annual reports are alsoperiodically generated and published in the public domain(Bristow 2003; Bristow & George 2002; Victorian DataReview Committee 2005).The minimum dataset (ANZICSAdult Patient Database 2003) contains the required vari-ables needed to use the APACHE II, SAPS II and APACHEIII-J prognostic models. Recalibration of the APACHE IIalgorithm has been undertaken on the 2000 – 2003 cen-tral repository data (Bishop et al. 2004). Whilst better per-formance was achieved for APACHE II, APACHE III-J wasfound to be better calibrated with good discrimination andis now the main prognostic model used to benchmarkperformance (Bristow et al. 2004).

Paediatric Intensive Care Registry (ANZPIC)ANZPIC was established by the ANZICS Paediatric StudyGroup (PSG) in 1997 with three aims:

M To describe paediatric intensive care practices and out-comes in Australia and New Zealand;

M To provide contributing units with efficacy and efficiencyreports comparing performance against national andinternational standards;

M To facilitate research in paediatric intensive care.Participating hospitals (All 8 PICUs and 8 adult ICUs)maintain unit specific databases and submit electroniccopies of de-identified data to the ANZPIC registry eachquarter. Annual reports are generated for each participat-ing unit and compare performance to that of other par-ticipating units. An aggregate annual report is publishedby ANZICS each year and is available at: http://sas.anz-ics.com.au/Portal. This report publishes data on thedemographics of critically ill paediatric patients in

Australia and New Zealand, and includes admission char-acteristics, length of stay and mortality.

Research Centre for Critical Care Resources(ARCCCR)Established in 1993, ARCCCR catalogues the critical careresources and infrastructure in Australian and New Zea-land. It aims to provide reliable information upon whichhealthcare providers, policy makers and government canbase decisions to improve healthcare services. Theresearch focus is quality-oriented and directed towardintensive care infrastructure, processes of care, criticalcare workforce (medical and nursing), demographics andinternational comparators. Over 95% of the 197 units inAustralia and New Zealand submit data each year, and theannual surveys completed by ICU staff assist in monitor-ing trends in intensive care service delivery. Reports areavailable at http://sas.anzics. com.au/Portal.

ChallengesThe geographical spread of the Australian and NewZealand populations has provided unique challenges tothe development of ANZICS intensive care registries.Australia, with a population of approximately 20 million,covers an area of some 7.7 million square km, thirty twotimes greater than the UK (Australian Bureau ofStatistics 2005). Lying over 2000 km to the east, NewZealand, with a population estimate of 4 million, occu-pies 270,500 sq km, similar in size to the UK (StatisticsNew Zealand 2000). See figure 1. The vast distances inAustralia and across to New Zealand has challenged theprovision of data collection training to ICUs, and the sub-mission and reporting processes. Participation in educa-tional workshops in regional centres and informationsessions held at scientific meetings continues to be dif-ficult for geographically isolated units. However, rigorousverification and validation are centrally undertaken at thetime of data submission, and extensive data validationand training are conducted annually by independentauditors for the paediatric registry. In 2006, a pilot studywill evaluate the feasibility of extending this to the 189adult ICUs in Australia and New Zealand.

Web-based reporting software has recently been imple-mented for direct query and reporting access to previ-ously submitted data to the ANZICS projects. Improvedweb access to the central data repository for authorisedresearchers and regional audit committees will facilitateepidemiological research and clinical audits. A key areaof focus for the future is to integrate the informationheld by the adult and paediatric registries, together withthat of the research centre.

M A N A G E M E N T

Australian and New Zealand IC database

This article describes the three intensive care resource and quality assurance projectsmanaged by the Australian and New Zealand Intensive Care Society (ANZICS).


Figure 1.

Distribution of ICUs in

Australia and New Zealand.

Carol George

MBus (IT)

Lynda Norton, MPH

Tracey Higlett, MPHAustralian and New Zealand

Intensive Care Society

(ANZICS)

[email protected]://sas.anzics.com.au/Portal

www.anzics.com.au

AcknowledgementThe ANZICS database proj-

ects are funded at a region-

al level by grants from the

Ministry of Health (New

Zealand) and State and

Territory Health

Departments through the

Australian Health Ministers

Advisory Council.

Database series NICE, NL

ICNARC, UK

ANZICS

s

ICU Management subscription form

Ways to subscribe:

• Log on to www.icu-management.org and complete the form under ’Subscription’;

• Send an E-mail with name and address to [email protected];

• Complete the form below and post it toICU Management - 28, rue de la Loi - B-1040 Brussels - Belgium;

• Complete the form below and fax it to +32 2 286 8508.

Name:

Institution:

Address:

Postcode & Town:

Country:

Telephone:

E-mail:

o 2 year subscription o 1 year subscription o Participant ISICEM 2005

Subscription ratesOne year: Europe 50 € Overseas 65 €

Two years: Europe 85 € Overseas 100 €


About 17 million people worldwide die from cardiovascu-lar diseases each year (World Health Organization 2002),and many of these deaths are due to sudden cardiacarrest. The incidence of out-of-hospital sudden cardiacarrest in industrial countries lies between 36 and 128 per100,000 inhabitants per year. Unfortunately, full cerebralrecovery after cardiac arrest is still a rare event and goodneurological recovery of patients admitted to a hospitalcan only be achieved in 11% to 48% of the cases. Therest of the patients die during the hospital stay or remainin vegetative state (Vreede-Swagemakers et al. 1997).

Therapeutic hypothermia is a new therapy that acts in amultifactorial way on the mitigation of post-resuscitationdisease. This includes the slowing of destructive enzy-matic processes, the protection of lipid membranes flu-idity and the reduction of oxygen needs without impair-ing microvasculatory blood flow in low-flow regionsduring reperfusion after ischemia. Additionally, therapeu-tic hypothermia also acts on lipid peroxidation, brainoedema, intracellular acidosis and apoptotic neuronalcell death (Chopp et al. 1989; Lei et al. 1994; Sterz et al.1992; Zhu et al. 2004). Therapeutic hypothermia not onlyprotects neurons, but also has beneficial effects onwhite matter injury and astroglial cell proliferation(Hachimi-Idrissi et al. 2004; Roelfsema et al. 2004).

Recently two randomized trials documented that reduc-ing the body temperature to 32-34°C after successfulrestoration of spontaneous circulation could substantial-ly improve neurological recovery (Bernard et al. 2002;The Hypothermia After Cardiac Arrest (HACA) studygroup 2002). Based on these trials, recent publishedguidelines of resuscitation recommend cooling coma-tose patients after cardiac arrest due to ventricular fibril-lation. The guidelines state that this might also be bene-ficial in patients with a cardiac arrest due to a non shock-able rhythm or in-hospital arrest (European Resus-citation Council 2005).

Experimental evidence shows that therapeutichypothermia is also beneficial in other cerebral ischemicstates (e.g. stroke). Studies analysing the protectivemechanisms have shown that the effect of therapeutichypothermia is largely independent of the underlyingcause of ischemia. Further, a small randomized trialincluding only patients resuscitated after asystole orpulseless electrical activity found a non significantincrease in survival and improvement of neurologicalrecovery (Hachimi-Idrissi et al. 2001). A recent meta-analysis of all three randomized hypothermia trials

Dr Michael Holzer presents the rationale for cooling all comatose patients after cardiacarrest.

V I E W S A N D I N T E R V I E W S

Therapeutic hypothermia for allcardiac arrest patients

Michael Holzer

MDDepartment of Emergency

Medicine, Medical

University of Vienna

[email protected]

showed a significant improvement of survival and short-term neurologic recovery (Holzer et al. 2004). It couldtherefore be concluded that as long as the patient hasrestoration of spontaneous circulation, there is somebrain tissue left which could be rescued by therapeutichypothermia, independent of the underlying cause, loca-tion or rhythm of cardiac arrest.

There are also side effects of using therapeutic hypother-mia and in each case whether the risk outweighs thebenefit needs to be assessed. Negative experienceswith lower temperatures (26-32°C, moderate hypother-mia) at the beginning of the therapeutic hypothermia eraraised concerns that the developing complications out-weighed any favourable effects of hypothermia.

More recent experiences with mild therapeutichypothermia (32-34 °C), however, show that the benefi-cial effect exceeds the complications by far. The follow-ing complications can occur and should be consideredwhen treating patients with therapeutic hypothermia. Asignificantly higher rate of pneumonia has been report-ed, although only in one study (Yanagawa et al. 1998), inwhich it was also reported that none of the cases ofpneumonia was a direct cause of death. A higher inci-dence of complications has been found in other studieswith arrhythmias, haemodynamic instability, bleedings,thrombocytopenia, pneumonias, sepsis and convulsions(Bernard et al. 1997; Bernard et al. 2002; TheHypothermia After Cardiac Arrest (HACA) study group2002). The total complication rate was not significantlyhigher in the cooled group in any of the reported studies.However, the number of patients studied was small.Treatment with therapeutic hypothermia requiresdefined protocols, to identify possible arising complica-tions promptly.

Induced mild hypothermia after resuscitation from car-diac arrest improves neurological outcome. Extendingfurther the existing resuscitation guidelines, all uncon-scious adult patients with spontaneous circulation afterout-of-hospital cardiac arrest should be cooled to 32 to34 °C for 24 hours as long as there is no contraindication(ongoing haemorrhage, pregnancy, terminal disease).Although the optimal target temperature, duration andmode of re-warming are still the subject of investigation,it is clear that therapeutic hypothermia should be start-ed as soon as possible to yield the maximum benefit. Avery effective method of inducing therapeutic hypother-mia is the rapid infusion of 30 ml/kg cold (4°C) lactatedRinger’s solution.

the part of the staff, although often they findfamilies keen to participate in care of thepatients. The centre is currently researching thefeelings of families in ICU as a special qualitycontrol analysis. The atmosphere in the unit ispleasant, with relaxing, grey and pink colourschemes. Televisions are available for entertain-ment or informative programs on the patient’sillness or the ICU.The ICU design, built 14 yearsago, lends itself to the maintenance of a cleanICU, with cleaning stations and the laboratoryconveniently located on the same floor (see fig-ure 1), and stratified levels of care managed inseparate ICUs.

Highly invasive treatments for the more severe-ly ill patients are managed in the ICU and noninvasive treatments and monitoring in the inter-

mediate ICU (see figure 1). We are also responsible forpre-hospital care of critically ill patients for a populationof 600,000. We’re called out by phone and the ambu-lance which goes to the patient has a physician, nurseand driver/technician. Patients are given early treatmentin the ambulance and come straight to the ICU. There’sa flow of 7% patients who worsen and move to the ICUfrom the intermediate ICU, and 40% of patients fromthe ICU move to the intermediate ICU for non invasivetreatment and monitoring. It’s probably due to this strat-ified system that the readmission rate is only 1.29% andthe occult mortality in the ward is only 2.25%.

The Critical Care Centre handles medical and surgicalpatients. The same staff serves pre-hospital care and inboth intensive care units. Staff comprise intensivists,one anaesthesiologist for 24 hour cover, who links withthe operating room, and consultants from various spe-cialties such as cardiology, neurology etc. who dedicate25% of their time to critical care. The critical care teamcan also be consulted for any patient with vital risk crite-ria in the hospital. Physicians and nurses from the criti-cal care centre go to the ward to check the patient. Asenior is responsible for triage and optimising bed use inthe critical care centre, taking the surgery schedule intoaccount. For severely ill patients who stay for longerthan 2 weeks, e.g. sepsis patients, ARDS patients etc.the physician also follows up for the first three or fourdays after the patient has been transferred to the ward,to monitor for complications and prevent respiratory orcardiac arrest.

Professor A. Artigas has organized 20 scientific meet-ings, published 15 books, more than 120 papers in inter-nationals scientific journals, and is a board member andpeer reviewer for numerous critical care journals.Research activities have included Eurosepsis, RCT newpharmacological treatments, APC, TFPI, AT III, the devel-opment of the new severity scores SAPS II, SAPS III andMPM II, ALI ARDS consensus conferences for NorthAmerica and Europe, and epidemiologic studies on inci-dence, outcome and risk factors.

Since 1988, Professor Artigas has been managing theCritical Care Centre at Sabadell hospital in Sabadell,Barcelona. The hospital has 660 acute care beds. TheICU part of the Critical Care Centre has 16 beds and theintermediate ICU 10 beds, with 12 medical staff, 6 resi-dents and 40 nurses. Managing 2500 patients per year,the nurse to patient ratio is around 1:3 (1:2.3 in the ICUand 1:1.5 in the intermediate ICU). ICU beds cost threetimes the cost of other hospital beds and the critical carecentre accounts for 15% of the hospital budget.Mortality rates are 16.5% for the ICU and 1.1% for theintermediate ICU.

Describe the organization of your unitAll rooms are fully equipped with mechanical ventilatorsand monitoring equipment, and the unit is a closed ICUwith open visiting hours. Single visiting family memberscan stay for as long as they wish between 13:00 and22:00 hours. A second visitor or friend may stay for halfan hour at scheduled times. This requires flexibility on


V I E W S A N D I N T E R V I E W S

An interview withProfessor Antonio ArtigasProfessor Antonio Artigas describes the design and operation of the Critical CareCentre at Sabadell Hospital, his management experiences and strategic aims forimproving intensive care.

Antonio Artigas Critical Care Centre

Sabadell Hospital

University Institute Parc

Tauli, Sabadell

Barcelona, Spain

[email protected]

Figure 1. Organization of the

Critical Care Centre

icu-management.org

What are the main directives of your role?I would estimate that my time is divided between 30%teaching, 40% research and 30% clinical care. We arealso heavily involved in designing quality control strate-gies and promoting discussion for improvement. Wemonitor and analyse internal data annually. Every 5years, we present the results and propose improvementplans to the hospital management. We are also respon-sible for the critical care department donor organs, andpacemaker and cardiac arrest programmes.

The centre has different working groups for infections,severity scores, management of acute respiratory fail-ure, renal failure, trauma and sedation etc. Each grouphas a member of staff as leader and two nurses fromthe critical care centre, and consultants or physiciansfrom other departments e.g. for infections we have amicrobiologist and someone from infectious diseases.We meet daily to learn about new cultures, for exampleThe idea is to bring people from the critical care centretogether with experts from all these specialty areas. Theworking groups are responsible to develop guidelines,treatment and diagnostics, to provide annual reports onquality control, and to design the teaching activities andresearch programmes in their specialty areas. The work-ing groups meet monthly and provide a multidisciplinaryplatform for critical care, presenting their work annuallyto the centre for discussion.

What skills are essential in ICU Management?I believe experience is an essential component to train-ing in management. In addition to this, I attended anMSc programme with the University of Jerusalem. Ihad to study week-ends for two years to achieve this.The course was designed to train people in how to pro-mote improvements in hospitals and act as facilitators.Our Professor flew from Israel every week-end and weworked together for one and half days. We also had tocomplete homework in groups. These were projects toimprove care in the hospital or healthcare system,using the same methodologies as commercial compa-nies. For example, one problem to be addressed wasthe relationship between primary care and hospitals.We presented our solution as a contract to the hospitalmanagement for approval, thus treating the hospital asa company. I apply many of the things I learnt on thiscourse in my practice. An example in intensive care isthe admission of stroke patients, for which there is acompetence mismatch between the neurologist andintensivist. Working together with the objective todetect and start treatment within 3 hours, we definedthe needs and a protocol, according to the availableresources. The result is that neurosurgeons and radiol-ogists are now happy to work with us. In this way, wedetect problems, consider the opinions of everyoneand encourage people to find solutions.

Describe your best and toughest experiencesIt’s been especially wonderful to have the opportunity ofworking with our new organizational design. A very spe-cial experience, though, was the treatment of a younglady who arrived in the ICU with severe ARDS manyyears back. She was treated with mechanical ventilationfor a month, and survived. I followed her progress for ayear afterwards and was eventually invited to her wed-ding. I have a fantastic relationship with this family andI’m now Godfather of her 16 year old child. This familyhave shown enormous appreciation. Human relation-ships have no price.I’ve also had two very sad experiences. My sister suf-fered a car accident on holiday a few years ago and diedwithin two days.The resources were simply not in place tosupport her care in the ICU she was taken to; I couldn’thelp her. Also in our own critical care centre, a youngmale nurse died of cancer and this was a very difficultyear for all of us.On a day to day basis in intensive care, it’s sometimesvery difficult to explain to administration the special dif-ficulties of our work; it takes a long time to gain the sup-port we need.

How would you improve intensive care?We’re addressing several areas to improve intensivecare more generally:

M we’re using new technologies to explore the poten-tial of monitoring and advising in smaller hospitalsusing telemedicine. We can also manage patients inambulances in this way. This may avoid unnecessarytransfers in the next few years;

M e-consultation, improved numbers of ICU beds, andgood coordination of existing facilities are all poten-tial solutions to improving intensive care in regionssuch as Catalonia, for example;

M we’re also working towards earlier care, pre-hospitaland in the emergency area, and towards supportingthe crucial follow-up after discharge;

M we believe in the value of stratifying care accordingto severity e.g. with the intensive care unit and inter-mediate intensive care unit in our centre;

M another important area is prevention of adverseevents to improve short and long term outcomes.

What are your personnel management strategies?We practice a human management style, consideringthe expectations of colleagues, talking with them andprioritising what they want to do, to keep them motivat-ed. We’ve had the same staff for 14 years, except forone colleague, who recently moved to a more seniorposition in another hospital. The head of nursing staffhas changed more regularly (four times in six years), buton the whole, we’re more like a family.

Thank you, Professor Antonio Artigas.

47

The key-role of the Council of the EUThe Council of the European Union was set up by thefounding Treaties in the 1950s, similar to the EuropeanParliament. The Council of the EU is the main decision-making body. It represents the member states, and itsmeetings are attended by one minister from each of theEU´s national governments.

The Council of the EU has the main role of agreeing leg-islation, although in recent years this has been sharedmore and more with the Parliament under the codeci-sion procedure. The European Commission makes pro-posals, for which in principle, the Council acts as legisla-tor. The Council may modify the proposals before adopt-ing them.

The Council consists of one government minister fromeach member state. Although there is just one Council,different groups of ministers meet depending on whattopic is being discussed at the weekly meeting. Eachminister is empowered to commit his or her governmentand is accountable to their own national parliaments forthe decisions in the Council.

Council configurationsDepending on the matters under discussion, the Councilmeets in nine different configurations within which eachcountry is represented by the minister responsible forthat subject. If the Council is to discuss environmentalissues, for example, the meeting will be attended by theEnvironment Minister from each country and is calledthe ”Environment Council”. The nine Council configura-tions are:

M General Affairs and External RelationsM Economic and Financial Affairs (ECOFIN)M Justice and Home Affairs (JHA)M Employment, Social Policy, Health and

Consumer AffairsM Competitiveness (Internal Market, Industry

and Research)M Transport, Telecommunications and EnergyM Agriculture and FisheriesM EnvironmentM Education, Youth and Culture

Each minister in the Council is empowered to commithis or her government. This means that the minister’ssignature represents the signature of the whole govern-ment. Moreover, each minister in the Council is answer-able to his or her national parliament and to the citizenswho the parliament represents, which ensures the dem-ocratic legitimacy of the Council’s decisions.

Responsibilities of the Council of the EUThe Council has the following six key responsibilities

M To pass European laws. As mentioned above, theCouncil shares legislation jointly with the EuropeanParliament in many fields.

M To co-ordinate the broad economic policies of themember states. This co-ordination is carried out bythe economics and finance ministers, who collective-ly form the ECOFIN Council.

M To conclude international agreements between theEU and one or more states or international organiza-tions.

M To approve the EU`s budget, jointly with theEuropean Parliament.

M To develop the EU`s Common Foreign and SecurityPolicy (CFSP).

M To co-ordinate cooperation between the nationalcourts and police forces in criminal matters.

Most of these responsibilities relate to the”Community” domain, for example, areas of action forwhich the member states have decided to pool theirsovereignty and delegate decision-making powers to theEU institutions. This domain is the “first pillar” of theEuropean Union. However, the last two responsibilities

European Institutions Series This is the third part in a series which covers the structure and operations ofthe EU institutions. In the first of the series (autumn 2005) Helicia Hermanintroduced the European Commission (EC).The second part (winter 2005)described the composition, functioning and main role of the EuropeanParliament.

In this third instalment, the structure, role and operation of the Council ofthe European Union are explained by our EU Editor for European Affairs,Sonja Planitzer. Representations on the Council, and the decision makingprocesses are presented. The European Council, and the Councils of Europe,and the European Union are carefully distinguished. Evolutions in the EUdecision making processes are presented, and finally our EU Correspondent,Rory Watson describes some of the current decision making activities relat-ed to healthcare.

The final instalment in the series will be published in the summer issue2006, and will cover the Court of Justice.


Sonja PlanitzerEditor European Affairs

E U F O C U S

Council of the European Union:the main decision making body of the EU


relate largely to areas in which the member states havenot delegated their powers, but are simply workingtogether. This is called ”intergovernmental coopera-tion” and covers the second and third “pillars” of theEuropean Union.

European CouncilArticle 4 of the Treaty on European Union says: “TheEuropean Council shall provide the Union with neces-sary impetus for its development and shall define thegeneral political guidelines thereof.”

The European Council brings together the heads of stateor government of the European Union and the presidentof the Commission. It defines the general political guide-lines of the European Union.The European Council meetsat least twice yearly (in practice, four times yearly, andsometimes, if necessary more often), usually in Brussels.

The European Council provides the impetus for themajor political issues relating to European integration:amendments to the Treaties and changes to the institu-tions, declarations on external relations in the context ofthe common foreign and security etc; its guidelines anddeclarations, however, are not legally binding. To be putinto effect, such political guidelines must undergo theroutine procedures by the European Parliament and theCouncil of the European Union, followed where neces-sary by implementation at national level.

Procedures of the Council of the EUCOREPERIn Brussels, each EU member state has a permanentrepresentation to the European Community, represent-ing and defending its national interest at EU level. Thehead of each representation is, in effect, his or her coun-try’s ambassador to the EU.

These ambassadors (also known as “permanent repre-sentatives”) meet weekly within the PermanentRepresentatives Committees – the “COREPER”. Therole of this committee is to prepare the work of theCouncil, with the exception of most agricultural issues,which are handled by the Special Committee onAgriculture. “COREPER” is assisted by a number ofworking groups, made up of officials from the nationaladministrations.

PRESIDENCY OF THE COUNCILThe ”Presidency of the Council“ rotates every sixmonths. In other words, each EU country in turn takescharge of the Council agenda and chairs all the meetingsfor a six month period, promoting legislative and politicaldecisions and brokering compromises between the

member states. Currently, Austria chairs the EU (formore information see page 51); in July Finland will takeover the EU Presidency until December 2006.

GENERAL SECRETARIATThe Presidency is assisted by the General Secretariat,which prepares and ensures the smooth functioning ofthe Council’s work at all levels.

In 2004, Minister Javier Solana was re-appointedSecretary General of the Council. He is also HighRepresentative for the Common Foreign and SecurityPolicy (CFSP), and in this capacity he helps coordinatethe EU’s action on the world stage. Under the new con-stitutional treaty, the High Representative would bereplaced by an EU Foreign Affairs Minister. TheSecretary General is assisted by a Deputy Secretary-General, in charge of managing the GeneralSecretariat.

The power of legislationThe power to legislate is shared between the Counciland the European Parliament. In most situations,European laws are made by a co-decision procedure.This means that the Council and the Parliament jointlyadopt proposals for legislation originating from theEuropean Commission. The Council and the Parliamentmay make amendments to the legislation under this pro-cedure. However, there are certain important areas, forexample, tax legislation, where the Parliament may onlygive an opinion on whether a proposed piece of legisla-tion can become law.

The Council only acts, as a rule, on a proposal from theCommission, and the Commission normally has respon-sibility for ensuring that EU legislation, once adopted, iscorrectly applied.

Not to be mixed up!

We have now the ”European Council”, the ”Council of the European Union” and last but notleast, the ”Council of Europe”. These three different institutions should not be confused:

The European Council comprises, as described above left, the heads of state or govern-ment of the European Union, and the president of the Commission. The role of the EuropeanCouncil is crucial, but differs to that of the Council of the European Union.

Members of the Council of the European Union are Ministers from the Member States.The Council of the European Union exercises the power conferred on it by the Treaty subjectto review by the European Court of Justice, and adopts Community legal instruments.

The Council of Europe, currently with 46 members, is another institution distinct from theEuropean Council and the Council of the European Union.This is an international organisationoutside the European Union, which deals with education, culture and above all the protectionof human rights.

Minister Javier Solana

© Council of the European

Union, 2000 - 2005

EU decision making proceduresIn general it is the European Commission that proposesnew legislation, but it is the Council and Parliament thatpass the laws. Other institutions and bodies also haveroles to play.

The rules and procedures for EU decision making are laiddown in the Treaties. Every proposal for a new Europeanlaw is based on a specific treaty article, referred to as the“legal basis” of the proposal. This determines which leg-islative procedure must be followed. The three main pro-cedures are “consultations”, “assent” and “co-decision”.

CONSULTATIONUnder this procedure, the Council consults Parliamentas well as the European Economic and SocialCommittee (EESC) and the Committee of the Regions(CoR). The Parliament has three options: 1) to approvethe Commission proposal 2) to reject it or 3) to ask foramendments. If the Parliament asks for amendments,the Commission will consider all the changes Parliamentsuggests. If it accepts any of these suggestions it sendsthe Council an amended proposal.The Council examinesthe amended proposal and either adopts it or amends itfurther. In this procedure, as in all others, if the Councilamends a Commission proposal, it must do so unani-mously.

ASSENTThis procedure means that the Council has to obtain theEuropean Parliament`s assent before certain very impor-tant decisions are taken. In this case the Parliament can-not amend a proposal – it must either accept or reject it.Acceptance (“assent”) requires an absolute majority ofthe vote cast.

CO-DECISIONThis is a procedure now used for most EU law making.In the codecision procedure, Parliament does not

merely give its opinion: itshares legislative powerequally with the Council.If Council and Parliamentcannot agree on a pieceof proposed legislation,it is put before a con-ciliation committee, com-posed of equal numbers

of Council and Parliament representatives. Once thiscommittee has reached an agreement, the text is sentonce again to Parliament and the Council, so that theycan finally adopt it as law.

Council decision making proceduresThere are different ways that the Council makes its deci-sions. A unanimous decision is required in importantareas, such as common foreign and security policies, andtaxation. Each member state has a vote in these areas.

In other fields, the Council makes its decisions byQualified Majority Voting. Each member state has aspecific number of votes (see below), which is relat-ed to the size of its population. A qualified majoritywill be reached, if a majority of member statesapprove, and if a minimum of 72.3 % of the votes arecast in favour.

Since 1st November 2004 the total number of votes hasbeen 321. The numbers of votes each country can castare shown in table 1.

Modernising the system with the ConstitutionThe EU is growing bigger and bigger. Although the deci-sion making system has evolved over half a century, itwas originally designed for a community of just sixnations. The EU now has 25 member states and itsmembership will increase further in the years ahead.Thedecision making system, therefore, needs simplifyingand streamlining. To avoid paralysis, most decisions willhave to be taken by “qualified majority voting” ratherthan requiring every single country to agree.

The proposed Constitution agreed by the EuropeanCouncil in 2004 tackles these questions head on. Itspells out much more clearly than in previous treatieswhat the European Union is and where it is going. It laysdown the new rules for more streamlined decision mak-ing. It is due to come into force in 2006, but first it hasto be approved by all 25 member countries – in somecases by referendum. Meanwhile the situation is at a”standstill”, with some member states having approvedthe Constitution – with referendums in some countries,such as in France and the Netherlands – and some mem-ber states returning a “negative” response.There is nowwhat some politicians are describing as a period ofreflection.

E U F O C U S

Table 1. Number of Council votes per countryGermany, France, Italy and the UK: 29Spain and Poland 27Netherlands 13Belgium, Czech Republic, Greece, Hungary and Portugal 12Austria and Sweden 10Denmark, Ireland, Lithuania, Slovakia and Finland 7Cyprus, Estonia, Latvia, Luxembourg and Slovenia 4Malta 3



The Austrian and Finnish governments have alreadyestablished the priorities for their two presidencies asthey steer European Union business throughout theyear. Whether they manage to achieve their objectiveswill depend not just on their diplomatic skills, but also onthe willingness of the European Parliament and EU gov-ernments to strike compromises, particularly on legisla-tive proposals.

Vienna will have its work cut out if it is to reach agree-ment on the provisions of an updated working timedirective. As negotiations between employment minis-ters in Brussels shortly before Christmas demonstrated,there is a huge gulf between those countries whichwant to retain the opt out from the 48-hour week andthose that wish to phase it out eventually.

Surprisingly, a large part of the complex negotiationswere filmed and broadcast live to media and the public,sitting elsewhere in the building, and provided a fascinat-ing insight into the way deals are normally put togetherbehind closed doors. But, even with the evident goodwill that existed on all sides, the gap proved too wide tobridge. If that remains the case, then behaviour in thisarea, particularly on on-call time, will be determinedmore by rulings from the European Court of Justice, asin the past, than by legislation agreed by Europe’s politi-cians.

The two governments may have greater success on theservices directive – an ambitious piece of legislation thataims to liberalise the cross-border market in this area.The probable outcome should become clearer in mid-February when the European Parliament will vote on thedraft text. At stake, from the medical point of view, iswhether health services should be excluded from thescope of the legislation. The Parliament is split over theissue. The Left basically supports exclusion. The Rightaccepts such a solution for public health services, butbelieves that private services should be covered.

The result, either way, will have implications for thehealth sector. The European Commission, which hasdrawn up its own public health priorities programme for2006, will have to take this into account as it finalises awide-ranging strategy paper. Due to be completed bythe end of the year, this aims to set the framework and

provide a more coherent approach to EU public healthactivities.

While fully accepting that the provision and manage-ment of healthcare remains a national responsibility, thiswill emphasise where Union activity can bring addedvalue. This is notably the case in developing the EU’scapacity to respond to health emergencies. Here, therecently established European Centre for DiseasePrevention and Control based in Stockholm, which ishelping to put in place a structure for handling pan-demics, will have a key role to play.

The strategy paper will also examine how to tackleinequalities in health treatment and how to strengthenthe Union’s role in international health organisations andits relations with national health systems.

Further measures to highlight the dangers of tobaccoloom large on the Commission’s agenda. It will launch anew awareness programme aimed at the young,deglamourising the practice of smoking and is consider-ing setting up a European Youth Parliament to discusstobacco control.

Member States which have failed to fully implement theEU legislation banning tobacco advertising that cameinto effect last August face legal action. The main culpritis Germany. Berlin tried unsuccessfully to persuade theEuropean Court of Justice to declare the legislation ille-gal and has still not transposed the EU directive intonational law. But some countries – Italy, Spain andHungary – are believed to be flouting the new rules byallowing advertising at Formula One racing events.Others, such as the Czech Republic and Portugal, havestill not notified the Commission of the measures theyhave taken to implement the legislation.

Organ transplants is another area where theCommission is exploring the possibility of further EUaction. Union rules already cover blood, human tissueand cells. The Commission is now examining issuessuch as the donation and trafficking of organs andintends to table a legislative proposal later this year toguarantee their quality and safety.

ICU Management EU Correspondent, Rory Watson, describes current EU decisionmaking activities concerning the working time directive, services directive, theEuropean Commission’s health priorities programme 2006, and a new organtransplants legislative proposal.

The priorities of the Austrian andFinnish presidencies

Rory WatsonEU Correspondent

[email protected]


HealthcareThe US has a mixed system of public and private insur-ance. Most working-age Americans receive privatehealth insurance through their employers. Privatehealth insurance covers about 70% of the population,but it accounts for only 35% of the healthcare spend-ing (Levit et al. 2004). Over 40 million Americans donot have health insurance and about half of bankrupt-cies in the US involve a medical reason or large med-ical debt. Private and government programmes forhealthcare exist and are explained below.

Private programmes1. Health Maintenance Organizations (HMOs): An

HMO is a prepaid “managed” health plan deliveringcomprehensive care to members through designat-ed providers, having a fixed periodic payment forhealth services.

2. Preferred Provider Organizations (PPO): A PPO hasarrangements with doctors, hospitals and otherproviders who have agreed to accept the plan’sallowable charges for covered medical services thatare similar to a fee-for-service plan. This givespatients a choice of using doctors and hospitals in anetwork with a co-payment and outside networkwith an annual deductible and a percent of the bill.More Americans with job-provided insurance areenrolled in PPOs (41%) than in HMOs (29%)(Oberlander 2002).

Government programmes1. Medicare: A federal program provides health insur-

ance to all Americans over 65 years of age, personswith disabilities and end-stage renal disease.

2. Medicaid: This health insurance program providesfor certain low-income families with children; aged,blind, or disabled people on supplemental securityincome, certain low-income pregnant women andchildren, and people who have very high medicalbills. Medicaid is funded and administered througha state-federal partnership. Although there arebroad federal requirements for Medicaid, stateshave a wide degree of flexibility to design their pro-gram. However all states must cover basic servic-es: inpatient and out patient hospital services,skilled nursing and home health services, familyplanning, and periodic health check ups. Medicaidreaches about 40% of Americans at the 100%poverty level (defined as an annual income of$9,570 for a family size of one person; Dept of

Health and Human Services 2005; US CensusBureau 2006).

3. State Children’s Health Insurance Program (SCHIP):This provides health benefits coverage to childrenliving in families whose income exceeds the eligibil-ity limits for Medicaid with incomes at or below200% of the federal poverty level (annual income of$32,180 for a family size of 3).

4. There is also a military plan for active and retiredservicemen and women.

Healthcare statistics and costsUS life expectancy was 77.6 years in 2003 (74.8 formen and 80.1 for women). Deaths from heart disease,cancer and stroke continue to drop (National Centrefor Health Statistics 2005). Heart diseases are thenumber one cause of death followed by malignantneoplasm and cerebrovascular diseases (ibid). Infantmortality has dropped to 6.9 deaths per 1,000 livebirths.

As a percentage of GDP, healthcare spending reached15.4% in 2004 (Centres for Medicare and MedicaidServices 2005). In 2005 it is estimated that the totalNational Health expenditure was $1,921 billion.National healthcare expenditures are projected toreach $3.6 trillion (18.7% of GDP) in 2014, growing atan average annual rate of 7.1% per year from 2003 to2014 (Centres for Medicare and Medicaid Services2004 & 2005). Intensive care units spend 10-30% of ahospital budget which accounts for to 0.5-1% of theGDP (Polderman and Metnitz 2005).

The US has the highest per capita health expenditureof any nation (Anderson et al. 2003). It spent $5267per person for healthcare in 2002, compared to thesecond most expensive system in Switzerland ($3445per capita; Bodenhemier 2005). Ten percent of thepopulation accounts for 70% of the cost(Bodenhemier and Fernandez 2005). Figure 1 presentsrelative healthcare expenditures (Bodenhemier 2005;Levit et al. 2004). Prescription drugs have been thefastest growing expenditure, increasing at a rate of11% over the last 3 years.

The US had fewer physicians and hospital admissionsper 1,000 population, physician visits per capita, acutecare beds and acute care days per capita than themedian of industrialized countries (Anderson et al.2003). The medical school enrolment has been con-

This article describes the healthcare system in the US, current issues,cost and management.

C O U N T RY F O C U S : U S

Healthcare in the United States

S. Sujanthy Rajaram

MDAssistant Professor of

Medicine

Cooper University Hospital

[email protected]

Carolyn Bekes

MD, MHAProfessor of Medicine

Cooper University Hospital


stant since 1980 in the US, and the increase in num-ber of physicians has mostly come from physicianswho immigrated to the US following medical educa-tion in other countries (Anderson et al. 2003). In 2000,per 1000 population there were 8.3 nurses, 2.8 physi-cians, 3 acute care beds, and 118 admissions. Therewere 5.8 physician visits per capita (ibid).

US healthcare system: an analysisA rapidly emerging trend in every American metropol-itan area is the formation of health networks made upof hospitals, physicians and insurance underwriters.Managed care, an organized way to manage the cost,use and quality of the healthcare system has had aprofound impact on the delivery of medical services,transforming traditional insurance arrangements(Oberlander 2002). Most studies have found little dif-ference in quality of care between traditional insurersand managed care plans, though there is evidence ofworse outcomes for chronically ill seniors in HMOs(Miller and Luft 1997). The functional status of the eld-erly has improved recently and there is a decreaseddeath rate. Recent advances are cost effective at gen-erally accepted values of an added year of life (Cutlerand McClellan 2001).

While rising costs may not create major problems forthe economy as a whole, they negatively affectemployers, employees, government and patients. Theaging population is not an adequate explanation for theincreased cost since it is too gradual a process to rankas a major cost driver in healthcare (Reinhardt 2003).The lack of well developed competitive markets inhealthcare may be partially responsible for the higherexpenditure. Technologies such as magnetic reso-nance imaging, computed tomography, coronaryartery bypass graft, angioplasty, intensive care units,positron emission tomography and radiation oncologyfacilities are associated with higher costs and areused extensively in the US (Bodenheimer 2005).However, Japan’s healthcare system has the highestusage of CT and MRI scanners (84.4 and 23.2 com-pared to the USA’s 13.6 and 8.1 per million populationin 2000) and a relatively high use of dialysis, with theleast expensive health system among developedcountries (Anderson et al. 2003).

The US also has the highest cost per unit of care,physician fees, payment per hospital day and pharma-ceutical prices. Even though physician visits and hos-pital days per capita have been lower in the US thanmany other developed nations, use of expensive tech-nologies, market power of hospitals and physicians,who are able to garner high prices for services, morerapid diffusion of innovative technologies, and a high-

Figure 1. Relative healthcare expenditures. Adapted with permis-

sion from Bodenheimer, T. (2005) High and rising healthcare

costs. Part 1: seeking an explanation. Annals of Internal

Medicine, 142 (10), pp. 847-854

er cost for administering the healthcare system hasdriven the overall healthcare cost to be high(Bodenheimer 2005). One proposed driver of health-care spending growth is the medical malpractice sys-tem, which encourages physicians to practice “defen-sive medicine” by ordering unnecessary diagnostictests or treatments to avoid malpractice litigation(Anderson 1999). Defensive medicine may account for5-9% of health expenditure (Hessler and McClellan1996).

Approximately 63% of growth in healthcare spendingis the result of an increased prevalence of obesity,stress, ozone, changing treatment threshold for hyper-tension, diabetes, hyperlipidaemia and osteoporosisand new innovations like statins, antidepressants, andother medications (Thorpe et al. 2005). Treatment oflow-birth weight babies and heart attacks has alsoaccounted for 37% of growth in healthcare spending(ibid).

One strategy for reducing the growth in healthcarecosts is to focus on slowing or reversing prevalence ofobesity including school based interventions to reducechildhood obesity (15% of school-age children wereobese in 2000), changing certain behaviours like smok-ing and driving while intoxicated, work place healthpromotion programmes, and cost effective use of highcost, low benefit medical technologies.

IntroductionEvaluation of and increasing the quality of healthcareis an international issue. Patients, physicians, insur-ance entities and governments share a concern forimproving quality and reducing cost of care. Eachinterested community has their own agenda but thecommon understanding is that the current system ofcare in the United States is neither consistently safenor universally effective. The intensive care unit is ofparticular interest as the cost of intensive care is veryhigh and the risk to life and limb from errors is greatin the vulnerable patients cared for in ICUs.

Multi-professional teams improve careThe Society of Critical Care Medicine was foundedover thirty years ago with the belief that having amulti-professional team of experts overseeing care atthe critically ill patient’s bedside is the optimal way todeliver safe and effective care. Scientific studies haveoverwhelmingly supported this hypothesis. Moststrongly supported is the improvement in patient sur-vival of having a board-certified intensivist involvedwith patient care and as leader of the multi profes-sional ICU team.

In a systematic metaanalysis, Pronovost defined a“high-intensity unit model” as one in which therewas a mandatory consultation with an intensivist onall ICU patients or all patients were managed by thecritical care team, and a “low-intensity unit model” inwhich there was no required or only elective consul-tation by an intensivist (Pronovost et al. 2002). 16 of17 published studies demonstrated lower hospitalmortality with a high-intensity model compared to alow-intensity model. This difference was highly signif-icant. The high-intensity model was also associatedwith shorter length of stays. No study demonstratedincreases in mortality with a high-intensity model ofstaffing in place. This metaanalysis included allpapers between 1979 and 2000, and the selected 27studies comprised over 27,000 critically ill adults andchildren.

Critical care nurses’ team contributionOther studies have documented the value of havingother expert members of the ICU team at the bed-side. These individuals include critical care pharma-cists, specially trained ICU nurses, and respiratory

therapists. Each profession can bring state-of-the-artpractice to care.

Nursing qualifications and workload have been associ-ated with patient outcome. Nursing-to-patient ratiosdirectly influence the incidence of preventable adverseevents in ICU patients. In a paediatric ICU, factors thatpredicted unplanned extubations included patient agi-tation and a nurse-to-patient ratio of less than 1:1.Having a nurse-to-patient ratio of less than 1:2 in caringfor patients following repair of an abdominal aorticaneurysm increased patient complications and lengthof stay in a study of Maryland hospitals (Pronovost etal. 2001). Using non-ICU trained nurses to care forpatients with central venous lines inserted for hyperal-imentation was associated with a higher line infectionrate than when patients were cared for by ICU special-ist nurses (Alonso-Echanove et al. 2003). Providingadditional nursing hours reduced errors even when theextra staff members were not ICU trained (Binnekadeet al. 2003). This was due to allowing ICU trained nurs-es to concentrate on the higher level, more risk-proneactivities. Nursing inexperience contributed to at leasthalf of the more than 1400 adverse events reported inthe Australian survey of ICU outcomes study (Morrisonet al. 2001).

Respiratory therapists for quality careCritical care respiratory therapists have contributed inimportant ways to ICU care. The use of patient-drivenprotocols for increasing and withdrawal of respiratorysupport has shortened duration of mechanical ventila-tion and reduced the incidence of ventilator associatedpneumonia (VAP) in critically ill patients, by promotingconsistency in ventilation (Ely et al. 1999; Restepo etal. 2004). These protocols are developed locally, arecollaborative and supported by the team. They includestate-of-the-art ventilatory care recommendations,which are continually updated as knowledge increases.Identification and reduction of risk factors for the devel-opment of VAP is an area of respiratory care in whichlarge gains have been made. Use of inline suctiondevices and reducing the frequency of ventilator circuitchanges has saved money and reduced VAP (MacIntyre2005; Stamm 1998). Instituting therapist-initiated dailyspontaneous breathing trials has the potential for fur-ther shortening the duration of mechanical ventilationand reducing morbidity (Ely et al. 1996).

Scientific studies support delivery of ICU care by a multi-professional team of expertsstationed at the patient’s bedside. A looming shortage of healthcare providers in theUnited States will threaten ICU safety and quality.

Charles G. Durbin,

Jr. MD, FCCMPresident of the Society of

Critical Care Medicine

[email protected]



Progress in Safety and Quality in ICUCare in the US?


Pharmacists’ contributionsMany of the errors occurring in the ICU are fromdrugs. Preventable adverse drug events (ADEs) aremore frequent in the ICU than on acute patientwards. ADEs occur as high as 19 per 1000 patient-days in the ICU (Cullen et al. 1997). Adding a seniorpharmacist to the ICU rounding team reduces thisproblem. In one unit, the ADE rate fell 66% followingaddition of a pharmacist to ICU rounds (Leape et al.1999). See also Dr Bourne’s article on the developingrole of critical care pharmacists in the UK in thisissue of ICU Management.

Team interaction impacts qualityWhile most benefits of care by a multiple profession-al team have been shown with individual members,the question remains whether the team actuallymakes a difference. In reporting the actual to predict-ed mortality in the 13 units used for the originalAPACHE data set, one unit was found to perform sig-nificantly better (only 41 deaths when 69 were pre-dicted) and one unit was significantly worse thanaverage (58% more deaths than predicted; Knaus etal. 1986). Comparison of the organization of care inthese units demonstrated increased coordination ofcare at the better performing unit. Physicians andnurses worked better together in the best unit. Whencaregivers work as a team, patient outcomeimproves.

Zimmerman and colleagues studied structural andorganizational characteristics of nine ICUs selectedfor their different performance in patient mortalityand efficiency (Zimmerman et al. 1993). Superiorunits demonstrated patient-centred culture, strongmedical and nursing leadership, effective communi-cation, coordination of care, and an open, collabora-tive approach to problem solving and conflict resolu-tion. The best units experienced more than the aver-age amount of conflict, but they had effective andopen means for achieving resolution. No particularstructure in this observational study was predictive ofsuperior performance, but the factors in table 1 wereassociated with unit performance.

Caregiver shortages threaten quality Despite the accumulation of supportive data, littleprogress has been made towards increasing thenumber of ICUs with an intensivist led team of dedi-cated experts. A recent survey by the AmericanCollege of Critical Care Medicine noted that only25% of general medical/surgical ICUs had a full-time,board-certified intensivist as a medical director. This

is little changed from a survey in 1991 (Groeger et al.1992). Less than 50% of physicians staffing the5,800 US ICUs had any specialized training in criticalcare medicine and less than 25% of ICU nurses wereICU certified.

Increased demand for critical care beds has resultedin caring for a larger percentage of these patientsoutside traditional ICUs with resultant prolongedstays in post operative care units (PACU) and theemergency department. The number of ICU beds hasexpanded significantly since the survey of 1991;there are about 30% more ICU beds in the US now.This expansion may account in part for the failure ofa larger percentage of beds to be staffed by certifiedpractitioners.

The major threat to sustaining ICU quality is the dimin-ishing workforce in the United States (Buerhaus et al.2000). The average age of bedside caregivers contin-ues to rise and exceeds 45 years for nurses.Retirement and burn-out are further affecting the quan-tity of ICU nurses (Chen and McMurray 2001; Odem2000). There is a projected short fall in trained and cer-tified intensivists and fewer individuals in the US areentering the medical workforce (Ewart et al. 2004). Thisis an ongoing and increasing issue for all concernedparties that will affect quality and safety of critical caretreatment in the future.

Table 1. Factors associated with ICU performance (Zimmerman et al. 1993)

Operational and Organizational Characteristics of the Best Performing Units

Strongly shared beliefs and objectives about excellence in patient care

Team-satisfaction oriented culture

Nurse and therapist empowerment to take initiative

Strong educational programmes including mentorship and skills exercises

Nurse and therapist empowerment to challenge others decisions

Liberal use of rewards

Community celebration of successes

Highly developed sense of collegiality

Visible nursing leadership

Commitment of hospital administration to quality care

Characteristics Observed in the Worst Performing Units

Concern about personal job security

Focus on procedural issues

Concern with “rules”

Highly developed bureaucracy

Overly concerned with work hours, pay, and hiring

Little concern with patient care issues



The practice of critical care medicine (CCM) in theUnited States can be divided into the adult and pae-diatric patient populations. Paediatric CCM practice isa fully consolidated line of patient care. No general,non-specialist, paediatrician in community or aca-demic practice would attempt to manage a criticallyill child or adolescent. In contradistinction, practition-ers of adult CCM include physicians in the medicalspecialties of anaesthesiology, surgery, and internalmedicine. All offer post-residency fellowship trainingin CCM. The fragmentation of CCM may increase, asemergency medicine programmes may begin to offerCCM fellowship training. Currently, the majority ofpracticing American intensivists are internal medi-cine based physicians (dominated by pulmonary-CCM practitioners; Angus et al. 2000).

Despite a large body of literature demonstrating thata “closed” (specialist practitioner only) unit improvespatient outcomes (Brown and Sullivan 1989; Carsonet al. 1996; Ghorra et al. 1999; Li et al. 1984;Manthous et al. 1997; Multz et al. 1998; Pollack et al.1988; Pronovost et al. 1999; Pronovost et al. 2002;Reynolds et al. 1988; Rosenfeld et al. 2000) and opti-mizes resource utilization (Hanson et al. 1999), thevast majority of community (private hospital) inten-sive care units (ICUs) have “open” (any practitioner)admission and management policies. The traditional"open" model reduces friction between the medicalstaff and the intensivist in most instances, but doeslittle to improve the quality of care. The "closed"model has not yet taken hold due to issues ofresource allocation, control of patients, concerns bynon-intensivists over lost revenue, and fears ofrestricting non-specialist practice.

The majority of critical care consultants combineCCM practice with work in their parent specialty. Thismixed workload may be attributed to many factorsincluding but not limited to: economics, preservingskills and interest in their base specialty, and a moreflexible work pattern. The emotional and physicalchallenges, unpredictable work pattern, lower remu-neration and political factors may have limited manyphysicians from full time careers in CCM. Federalreimbursement for critical care services pays lessthan private insurance, and as Medicare (a federalhealth insurance program for people aged 65 andolder and for individuals with disabilities) covers a

large percentage of ICU patients, time spent in one'sbase specialty remains more economically attractive.This may be addressed by the recent increase in therelative value unit for critical care services negotiatedby the Critical Care Work Group, comprising sixnational societies related to intensive care.

In November 2000, the Leapfrog Group published astandard regarding Intensive Care Unit PhysicianStaffing (IPS; Birkmeyer et al. 2000). The LeapfrogGroup is a consortium of the largest US companies,and other large healthcare purchasers committed to acommon set of purchasing standards with full imple-mentation in 2003. As a result, the physician work-force projects an increased demand for intensivists.It has been estimated that 35,000 critical care physi-cians will be required to staff all adult American ICUs(Ewart et al. 2004). This demand outstrips supplywhich continues to hover around 9,500.

In an effort to provide the necessary patient cover-age, many institutions and group practices haveadded nonphysician providers to the clinical picture.Collaboration between advanced practice nurses andmedically directing physicians may be a clinically effi-cient response. The Acute Care Nurse Practitioner(ACNP) is a registered nurse with a graduate degreein nursing who is prepared for advanced practice inacute clinical care. Early studies suggest that thismodel can be clinically efficient and effective in somepatient populations (Jatremski 2001; Hoffman et al.2005). The addition of nurse practitioners to the criti-cal care team results in maintenance of thedecreased length of stay, and improved clinical out-comes seen in ‘closed’ units, with documentedevidence of improved family satisfaction and commu-nication (Schukman et al. 1995).

Over the coming years as the population ages and anincreased number of individuals survive with chronicdiseases, tertiary-care centred hospitals are likely toincrease the percentage of critical care and monitoredbeds to upwards of 50% of the total. The combinationof a sicker patient population, coupled with the payerdriven demand for quality care and the planned reim-bursement adjustments, may result in significantdemands for future critical care practitioners.

Current State of critical care medicinein the United States

Dr Maccioli discusses some of the issues which threaten to aggravate the predictedshortage of intensivists in the US, and some potential solutions.

Gerald A. Maccioli

MD, FCCMChairman, American Society

of Anaesthesiologists

Committee on Critical Care

Medicine

President-Elect

American Society of Critical

Care Anaesthesiologists

Director of Critical Care

Medicine, Critical Health

Systems of NC

Raleigh Practice Centre

[email protected]



IntroductionThere are insufficient numbers of intensivists, criticalcare nurses and other healthcare professionals toachieve consistent, high quality care for all high acuitypatients. Despite data demonstrating improved out-comes with the dedicated intensivist care model, < 15%of US ICUs have this model in place during daytimehours. Moreover, the simple presence of intensivistsdoes not translate into standardized care processes androutine use of best practices. Recognizing that signifi-cant restructuring of clinical care represents the onlyviable approach to achieving quality goals, many healthsystems across the US have implemented remote ICUcare programmes.This care model centralizes select ele-ments of ICU care to increase provider efficiency andeffectiveness.

StructureThirty-two health systems, representing more than 4000ICU beds, 300 ICUs and 150 hospitals, have elected toestablish remote ICU care programmes. The remote caremodel links ICU beds in multiple hospitals to a centralizedcare centre. The remote care centre is staffed with inten-sivists and critical care nurses from the health systemimplementing the program. The size of the remote teamvaries with the number of beds in the network. Most siteshave a single intensivist and a variable number of criticalcare nurses (~ 1 for every 30-35 patients). Hours of oper-ation range from 12 per day (nights) to 19-24 per day (ICUswithout dedicated intensivists). Most programs include aflagship centre (academic medical centre or major tertiarycare facility) and several smaller facilities. Some service asingle metropolitan area; others cover large geographiccatchment areas.

TechnologyThe technology infrastructure includes high-resolution,in-room cameras, speakers and microphones, remotebedside monitor viewers, an ICU clinical information sys-tem, an automated alerting system and local and widearea networks. This configuration allows the remoteteam to access all relevant clinical data and interact withon-site providers. It also provides them with tools tomanage the population of patients in the ICU network.

ServicesICU clinicians at the remote site work in concert with on-site providers to provide consistent, round-the-clock,quality care. The on-site clinicians are responsible forestablishing a comprehensive daily care plan for eachpatient.The remote team is responsible for ensuring thatall goals of the care plan are achieved. This entails fre-

quent review of clinical data (e.g. virtual rounding) andtitration of therapies, as needed. The remote team,through regular rounding and automated alerts, is alsocharged with identifying new problems promptly and ini-tiating timely countermeasures. Many sites have cen-tralized quality improvement activities as well (e.g. ven-tilator and sepsis bundles).

OperationsImplementation of a remote care program requires sig-nificant changes in how ICU care is organized and deliv-ered. In addition to individual practitioners masteringnew skills and new technologies, there is often a needfor major cultural change. Collaboration and standardiza-tion are central to the care model; unfortunately, theseconcepts are not universally embraced. Even in academ-ic medical centres there are often practice variationsamong the different sub-specialty ICUs, and cross-departmental collaboration is uncommon. Sites thathave successfully implemented remote care pro-grammes have recognized the scope of the clinicaltransformation and allocated the resources to managethe change process. This includes strong executive sup-port, enlistment of key thought leaders, extensiveadvance education and effective program management.

Outcomes dataThe first remote care program, implemented 5 yearsago, reported decreases in mortality, ICU LOS and hos-pital LOS of 27%, 17% and 13%, respectively (Breslowet al. 2004). The other programmes are < 3 years old,and thus there is a paucity of published data. A recentabstract reported a significant decrease in cardiopul-monary arrests and deaths from cardiopulmonaryarrest after program implementation (Shaffer et al.2005). Another observed a decrease in ventilator days,with the magnitude of the decrease correlated with thedegree of autonomy granted to the remote team(Cowboy et al. 2005). Sites that have centralized bestpractice oversight have reported significant increasesin compliance rates.

ConclusionRemote ICU care programmes are becoming more com-mon in the US, driven by both the shortage of inten-sivists and other healthcare professionals and the desireof health systems to improve the quality of ICU careacross all their facilities. The health systems implement-ing these programmes are changing how healthcare isdelivered and through their efforts are discovering howbest to implement these programmes and maximizetheir effectiveness.

Remote ICU care in the USRemote ICU care programmes have become more prevalent in the US over the past 3years as hospitals try to leverage clinician expertise and improve quality.

Michael J. Breslow

MD, FCCMExecutive Vice President

Research & Development

VISICU, Inc.

Baltimore, Maryland

[email protected]



Billing for critical care services in the United Statescan be quite confusing and time consuming. Everypatient represents a different billing scenario relatedto potentially a different payer or multiple payers.Each payer is permitted to utilize their own system ofcoding and to establish their own payment rules.Furthermore, each payer is permitted to changethese rules as they see fit and thus the combinationsof rules are endless and constantly changing. In addi-tion, few payers pay what is billed, with most negoti-ating a discounted rate such that the gross collectionratio is commonly in the 30-40% range.

Many payers base at least a component of their sys-tem on Medicare payment policy, which is estab-lished by the Centres of Medicaid and MedicareServices (CMS). Thus most physicians establish doc-umentation practices in line with Medicare rules andtypically bill according to Medicare principles. In thismanuscript we will briefly review the documentationcriteria for billing for critical care services for adults.

The guidelines for documentation are initially estab-lished as a component of the definition of criticalcare services, which is established by the AMACommon Procedural Terminology (CPT) Committee(AMA 2005). There are two CPT codes for criticalcare services. One for the first hour of critical care(99291) and one for each subsequent half hour ofcritical care (99292). It should be noted that CPTdefines the first hour of critical care as ranging from30 minutes to 74 minutes, thus establishing the prin-ciple that critical care services can never take lessthan 30 minutes. The time can accrue over multipleiterative sessions and can include the time to docu-ment the care. Furthermore, the time can includetime in family conferences if, and only if, the patientis unable to communicate or is deemed incompetentand the discussion is absolutely necessary for careprovided or withheld for that day.

In addition to time criteria, two additional criteriamust be met and documented; the patient mustmeet the CPT definition of critical illness and thephysician must provide direct critical care services.Critical illness is described as, “the critical illness orinjury [that] acutely impairs one or more vital organsystems such that the patient’s survival is jeopard-ized” (AMA 2005). Additionally, there is “a high prob-ability of sudden clinically significant or life threatening

deterioration in the patient that requires the highestlevel of physician preparedness to intervene urgent-ly” (AMA 2005).

Once the physician believes that the patient meetsthe definition of critical illness and that their time hasexceeded 30 minutes, the physician must documentthat they have provided critical care services to thepatient while on the unit or floor. There are threetreatment criteria which must be met and document-ed. The “critical care services require direct personalmanagement by the physician,” the “services are lifeand organ supporting interventions that require per-sonal assessment and manipulation by the physi-cian,” and “withdrawal of, or failure to initiate theseinterventions on an urgent basis, would likely resultin sudden clinically significant or life-threateningdeterioration of the patient’s condition” (AMA 2005).One should note that merely activating protocolsdoes not constitute critical care services.Furthermore, it should be stressed that all compo-nents (critical illness, critical care service and time)must be met and clearly documented or payment canbe withheld or claims of fraudulent billing practicecan be made against the physician.

Procedures done in addition to providing critical careservices can also be billed under certain conditions.First, the procedure must not be bundled into thedefinition of critical care. Procedures that are consid-ered bundled include: ventilator management, car-diac outputs, temporary pacing, arterial puncture,interpreting chest radiographs. If the procedure is notbundled then separate documentation is requiredand a bill can be submitted. Of course, the physicianmust ensure that the time to do the procedure is notcounted in the time accrued for billing the criticalcare services, as this would be considered doubledipping.

As one can see there are many nuances to docu-menting critical care services and subsequently tobilling and receiving payment for said services. Giventhe variability in the model of critical care physicianservices in the United States, it is unlikely that amore streamlined approach that principally servesthe full-time intensivist model will be established anytime soon.

Billing and documentation: a primer

Billing and documentation requirements are extremely complex for critical ill patientsand no relief is in sight. Drs Dorman and Pauldine review the present documentationrequirements for critically ill or injured adults.

Todd Dorman

MD, FCCM

Ronald Pauldine

MD, Fellow

Dept. of Anaesthesiology

and Critical Care Medicine


School of Medicine

Baltimore, US

[email protected]

led by Isabelle Murat (Paris), ably supported by D Patel(Manchester) and M Weiss (Zurich).

Perhaps the most interesting lecture for all delegateswill be the one opening the meeting on Saturday June3rd at 13.00 hr: Juan Navia Roque of Madrid will discussthe management of the casualties from the Madridbombings in “Mass Casualties: The Madrid Experience”.Emergency medicine in Europe will be discussed in asymposium chaired by J Andres (Cracow) and CDDeakin (Southampton), and German and Norwegianviews will be given in a workshop asking “Do we needdoctors to go out in Emergencies?”

A symposium on Incident Reporting will specifically dis-cuss this approach in intensive therapy, and workshopsmanaging a range of critical incidents will recruit audi-ence participation. A workshop on the influence ofhuman factors, such as stress and fatigue, on patientsafety will be of value to all practising clinicians. Otherworkshops include two onTEE, where the theory, anato-my and pathology as well as the applied physiology willbe discussed. Tips on how to pass the ECHO examina-tion will be given Italian style by Fabio Guarracino (Pisa),and assessment of mitral valve function using trans-oesphageal echocardiography (TOE) will be formally dis-cussed by the international expert Jan Poelaert (Gent).Monitoring of organ and cell function including for theliver and gut, will be addressed in a symposium organ-ised by Andreas Hoeft.

There is much for intensivists to enjoy at the ESAmeeting in June 2006. The increasing number of dele-gates – 5033 in Vienna in June 2005 – must indicatethat the programme for these events continues to berelevant and stimulating. Supported so ably by theSecretariat, and especially by Raf Kinnaer as theProgramme Administrator, the ESA is fully committedto the continued enhancement of this, the largest sci-entific anaesthetic meeting in Europe. The fusion of theEAA, CENSA and the “old” ESA at the beginning of2005 represented the beginning of a truly unitedapproach to the continued development of anaesthesi-ology, critical care and pain relief on this continent. Joinus, as we move forward in our continued attempts tounderstand the scientific basis of our ever-wideningclinical practice, and at the same time enjoy thedelights of Madrid and its environs.

C O N G R E S S E S


Euroanaesthesia 2006

The European Society of Anaesthesiology, amalgamated since 2005 with the EuropeanAcademy of Anaesthesiology (EAA), and the Confederation of European NationalSocieties of Anaesthesiologists, holds the largest scientific anaesthetic meeting inEurope, Euroanaesthesia. Jennifer Hunter, Chairman Elect of the Scientific ProgrammeCommittee, previews this year's meeting to be held in Madrid.

From June 3-6 2006, the European Society ofAnaesthesiology (ESA) will hold its annual meeting inMadrid, the beautiful capital of Spain. Our venue is thepurpose-built Juan Carlos Exhibition Centre in the NorthConvention Centre Building, conveniently accessiblefrom the airport and city centre by public transport. Alocal organizing committee led by Professor HectorLitvan has contributed significantly to the preparation ofthis meeting on behalf of the Spanish Society ofAnaesthesia, Reanimation and Pain Therapy.

The ESA Scientific Programme Committee (SPC) led byDr Gordon B Drummond (Edinburgh, UK) has put togeth-er an exciting four day scientific programme, coveringmany aspects of intensive care medicine. Its 17 subcom-mittees have planned the programme with such diversetopics as Monitoring, Patient Safety, Local and RegionalAnaesthesia, the Neurosciences, Respiration, andClinical and Experimental Circulation. Each subcommit-tee of seven members from throughout Europe has aChairman, nominated to the SPC chair by the subcom-mittee members, and approved for the appointmentthrough the Nominations Committee by the ESA Board.Each subcommittee recommends to the SPC (16months in advance of each scientific meeting) twoRefresher Courses, each given by recognised experts asdidactic lectures; two or three 90 minute symposia con-sisting of three 30 minute lectures, including discussion;and two 45 minute practical workshops.

The Chairman of the SPC subcommittee on IntensiveCare Medicine, Professor Gernot Marx (Jena) has beenaided by intensivists from throughout Europe to putthis part of the ESA programme together. DidacticRefresher Courses will cover the diagnosis and treat-ment of sepsis (F Stüber, Bonn), nutritional support inthe critically ill (RJ Beale, London), and global andregional oxygen transport in the critically ill (S Jakob,Bern). Symposia will cover blood transfusion and IV flu-ids in the critical care environment, antibiotic therapy,metabolism and acid-base, and an update on mechani-cal ventilation, and weaning from it. Management ofcritically ill children will be discussed, including the useof high frequency ventilation in this population. Thissymposium will be chaired by W Habre (Geneva) andincludes HFO in the child with ARDS, HFO in theneonate, and HFJV for laser airway surgery in children.A workshop on paediatric airway management will be

Jennifer M Hunter

MB ChB PhD FRCAChairman Elect

Scientific Programme

Committee, ESA

University Department of

Anaesthesia

University Clinical

Departments

Liverpool

[email protected]

www.euroanesthesia.org

Acknowledgement

We thank Dr Gordon

Drummond, who completes

his term of office as

Chairman of the Scientific

Programme Committee in

March 2006, for his flair,

energy and commitment to

the ESA and its annual sci-

entific meetings.

64 ICU Management 1-2006ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-2005ICU Management 4-200564ICU Management 4-200564ICU Management 4-2005

ICU MANAGEMENT is the Official Management and Practice Journal ofthe International Symposium on Intensive Care and EmergencyMedicine and was previously published as Hospital Critical Care.Business Publication Audit (BPA) applied for.

EDITOR-IN CHIEFJean-Louis Vincent, Head, Department of Intensive Care, ErasmeHospital, Free University of Brussels, Belgium [email protected]

EDITORIAL BOARDProf. Antonio Artigas (Spain) [email protected]. Richard Beale (United Kingdom)

[email protected]. Todd Dorman (United States) [email protected]. Hans Kristian Flaatten (Norway)

[email protected]. Luciano Gattinoni (Italy) [email protected]. Armand Girbes (Netherlands) [email protected]. Claude Martin (France) [email protected] Prof. Konrad Reinhart (Germany)

[email protected]. Jukka Takala (Switzerland) [email protected]

NATIONAL CORRESPONDENTSProf. David Edbrooke (United Kingdom) [email protected] Anders Larsson (Denmark) [email protected]. Esko Ruokonen (Finland) [email protected]. Reto Stocker (Switzerland) [email protected]. Patricia Wegermann (Germany)

[email protected]

MANAGING EDITORKirstie Edwards [email protected]

SCIENTIFIC EDITORDr. Sonya Miller [email protected]

EUROPEAN AFFAIRS EDITORSHelicia Herman [email protected] Planitzer [email protected]

EDITORIAL ASSISTANTVioleta Ribarska [email protected]

GUEST AUTHORSM. A. Al-Ansari, C. Bekes, B. W. Böttiger, R. S. Bourne, M. J. Breslow,M. Cecconi, J. D. Chiche, A. Csomos, T. Dorman, C. G. Durbin,D. Edbrooke, R. Flin, D. M. Gaba, C. George, C. A. Goeschel,B. Guidet, T. Heidegger, T. Higlett, M. Holzer, J. M. Hunter, Y. Husemann,J. A. Kellum, G. K. Lighthall, G. A. Maccioli, T. Manser, E. McKinney,P. G.H. Metnitz, G. H. Mills, R. P. Moreno, L. Norton, M. Nuebling,R. Pauldine, P. J. Pronovost, T. Prien, S. S. Rajaram, T. Reader,A. Rhodes, D. Saal, F. Spöhr, C. L. Sprung, B. Taylor, R. Venkataraman,J.-L. Vincent, R. Watson

PUBLISHING HOUSEEuromedical Communications s.a.28, rue de la Loi, B-1040 Brussels, BelgiumTel +32 2 286 8500Fax +32 2 286 8508E-mail [email protected] www.icu-management.org

PUBLISHERChristian Marolt

MEDIA CONTACT, MARKETING, ADVERTISINGLuk Haesebeyt [email protected]

SUBSCRIPTION RATESOne year Europe 50 Euros Overseas 65 EurosTwo years Europe 85 Euros Overseas 100 Euros

Note: Participants of the International Symposium on Intensive Care andEmergency Medicine receive a one year subscription as part of theirsymposium fee.

ART DIRECTORSarah Daeleman [email protected]

PRODUCTION AND PRINTINGImprimerie Centrale s.a., LuxembourgPrint run: 9,000 - ISSN = 1377-7564

© ICU Management is published quarterly. The publisher is to be notifiedof cancellations six weeks before the end of the subscription. The repro-duction of (parts of) articles without consent of the publisher is prohibit-ed. The publisher does not accept liability for unsolicited materials. Thepublisher retains the right to republish all contributions and submittedmaterial via the Internet and other media.

LEGAL DISCLAIMERThe Publishers, Editor-in-Chief, Editorial Board, Correspondents andEditors make every effort to see that no inaccurate or misleading data,opinion or statement appears in this publication. All data and opinionsappearing in the articles and advertisements herein are the sole respon-sibility of the contributor or advertiser concerned. Therefore the publish-ers, Editor-in-chief, Editorial Board, Correspondents, Editors and theirrespective employees accept no liability whatsoever for the conse-quences of any such inaccurate or misleading data, opinion or state-ment.

REFERENCESReferences cited in this journal are provided to EuromedicalCommunications by the authors and are available on request to [email protected].

Agenda

MARCH 2006

21-24 26th International Symposium on Intensive Care and

Emergency Medicine

Brussels Belgium

www.intensive.org

24-28 International Anaesthesia Research Society 80th Congress

San Francisco USA

www.iars.org

MAY 2006

10-13 8th Scientific Congress of European Resuscitation Council

Stavanger Norway

http://congress.erc.edu/

13-17 Australian and New Zealand College of Anaesthetists (ANZCA)

Annual Scientific Meeting

Adelaide, South Australia

www.sapmea.asn.au/conventions/anzca/index.htm

23-24 UK Intensive Care Society Annual Spring Meeting

Harrogate UK

www.ics.ac.uk

JUNE 2006

3-6 Euroanaesthesia

Madrid Spain

www.euroanesthesia.org

SEPTEMBER 2006

2-6 European Respiratory Society 2006 Annual Congress

Munich Germany

www.ersnet.org/ers/default.aspx?id=2078

24-27 19th Annual Congress of the European Society of Intensive

Care Medicine

Barcelona Spain

www.esicm.org

OCTOBER 2006

12-15 31st Australian & New Zealand Annual Scientific Meeting on

Intensive Care

Tasmania Australia

www.anzics.com.au

FEBRUARY 2007

18-21 36th Critical Care Congress of the SCCM

Orlando, USA

www.sccm.org

LETTERS TO THE EDITOR & REQUESTS FOR REFERENCESCITED IN ICU MANAGEMENT [email protected]

the official management and practice journal

Documents