To the Editor: Haynes et al. (Jan. 29 issue)1 report on a surgicalsafety checklist to reduce morbidity and mortality in a globalpopulation. Transferring the concept of checklists from aviationto surgery sounds intuitively sensible. However, to claim thatthe use of checklists can reduce the perioperative rate of deathby more than 30%, based on extrapolation across a mixture ofhospitals in developed and developing countries, may be misleadingand counterproductive. In any case, all except one of the participatinghospitals in developed countries had a preintervention rateof death that exceeded the published normal range of 0.4% to0.8%. Indeed, the rate of death in the only hospital at theextreme end of the normal range increased from 0.8% to 1.4%after the intervention.
In the United Kingdom, the National Patient Safety Agency hasresponded rapidly to this study by issuing a safety alert toall National Health Service hospitals; this alert requires thehospitals to use a modified 26-point checklist by 2010.2 Althoughwe support this initiative because it is likely to promote greaterteam cooperation, we are concerned that the implied reductionin the perioperative rate of death is unlikely to be realizedin the United Kingdom, and ultimately, this may adversely affectcredibility and compliance with this potentially valuable adjunctto safety measures for surgical patients.
Ian C. Martin, F.D.S.R.C.S., F.R.C.S. Marisa Mason, Ph.D. George Findlay, F.R.C.A. National Confidential Enquiry into Patient Outcome and Death London W1T 5HD, United Kingdom imartin{at}ncepod.org.uk
References
Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med 2009;360:491-499. [Free Full Text]
To the Editor: Haynes and colleagues show the benefit of thesurgical safety checklist for improving perioperative care.Although the authors state that the intervention was neither"costly nor lengthy," information to support this statementwas not presented. The length of the checklist must be considered,especially for the anesthetized patient who is prone to physiologicaldisturbance, including hypothermia. Thus, it is unclear whya "time out" should be conducted after the induction of anesthesia,rather than immediately before the induction of anesthesia (thisis particularly important in lower-income environments withless availability of physiological support). The significantbenefits of education, checklist awareness, and increased resourceutilization must be realized. The adoption of increased useof antibiotics and pulse oximetry may have accounted for thesurvival advantage in lower-income sites. It is possible thatimplementation of other monitoring devices, such as temperaturemonitoring, would further improve perioperative care. However,although all sites in the current study had access to pulseoximetry and prophylactic antibiotics, the additional costsassociated with these resources were not disclosed. Many lower-incomesites may lack funding for these resources.
Robert D. Sanders, M.B., B.S. Imperial College London London SW10 9NH, United Kingdom robert.sanders{at}ic.ac.uk
Simon S. Jameson, M.B., B.S. Stirling Royal Infirmary Stirling FK8 2AU, United Kingdom
To the Editor: As reported by Haynes et al., the World HealthOrganization (WHO) Safe Surgery Saves Lives study shows strikinglylarge and important effects of checklists on the rates of inpatientcomplications, including death. The use of checklists will bepromoted globally.1 Unfortunately, in addition to the weak preintervention–postinterventiondesign, the inference of effectiveness is not convincing forother reasons. The Hawthorne effect is discussed as a possiblemechanism of effect, rather than as a study limitation. Apartfrom consideration of the possible effects of direct observationin the operating room, other artifacts of the research processare not considered. Clinical teams were fully aware that theywere participants in a study of their own behavior. Surgicalpractice may well have been altered by this research contextrather than by the checklist. Blinding is recommended to reducesuch performance biases.2,3 Clinicians could have been keptunaware of the study, and the checklist could have been introducedas a matter of hospital policy in precisely the ways intendedfor subsequent routine use. The opportunity to reliably estimatethe size of the effects of checklist introduction in a trialhas been missed. Dedicated, sophisticated study of the Hawthorneeffect is long overdue.4
Jim McCambridge, Ph.D. London School of Hygiene and Tropical Medicine London WC1E 7HT, United Kingdom jim.mccambridge{at}lshtm.ac.uk
Kypros Kypri, Ph.D. University of Newcastle Newcastle, NSW 2300, Australia
Diana R. Elbourne, Ph.D. London School of Hygiene and Tropical Medicine London WC1E 7HT, United Kingdom
Boutron I, Guittet L, Estellat C, Moher D, Hróbjartsson A, Ravaud P. Reporting methods of blinding in randomized trials assessing nonpharmacological treatments. PLoS Med 2007;4:e61-e61. [CrossRef][Medline]
Higgins JPT, Altman DG, eds. Assessing risk of bias in included studies. In: Higgins JPT, Green S, eds. Cochrane handbook for systematic reviews of interventions, version 5.0.0. (updated February 2008). Oxford, England: Cochrane Collaboration, 2008. (Accessed May 7, 2009, at http://www.cochrane-handbook.org/.)
McCambridge J, Day M. Randomized controlled trial of the effects of completing the Alcohol Use Disorders Identification Test questionnaire on self-reported hazardous drinking. Addiction 2008;103:241-248. [CrossRef][Web of Science][Medline]
To the Editor: Haynes et al. discuss the beneficial effectsof using checklists for surgical procedures. Many years ago,before attending medical school, I was a fighter pilot flyingF-86 Sabrejets in the Air Force. I and most of my flying colleaguesalways used checklists that were strapped to our thighs whilewe were sitting in the cockpit. Every one of the myriad switches,gauges, dials, handles, and circuit breakers had to be properlyset or checked. Procedures had to be followed assiduously, especiallyduring an emergency. Checklists helped us do that. Each of usknew that a careless mistake could lead to our death. By contrast,if physicians or nurses make a careless mistake, someone elsesuffers or dies. Many of us evince too cavalier an attitudein working with patients. If all of us in medicine thought ourown lives were at risk, you can bet a lot fewer mistakes wouldbe made. Requiring the use of checklists is an excellent wayto reduce errors and keep our patients safer.
David C. Levin, M.D. Thomas Jefferson University Hospital Philadelphia, PA 19107 david.levin{at}jeffersonhospital.org
The authors reply: Martin et al. express concern that the ratesof death in our study were not "normal" and that the beneficialresults observed may therefore not be generalizable. We wellrecognize that an eight-hospital study cannot provide a preciseestimate of the magnitude of reduction in harm that is possiblefrom broad implementation of the checklist. However, the largercriticism is flawed. The rates of death that we reported werefor procedures conducted in the study operating rooms, not forhospital-wide procedures; the case mix varied widely among hospitals,and the hospitals themselves had enormous diversity. In addition,the normal rate of postoperative death is unknown for the mixof cases observed in this international group of hospitals;a comparison with data from limited studies in developed countrieswould be invalid.
Sanders and Jameson correctly point out that the cost and timeinvolved in such an intervention are important considerations.The checklist was designed to be brief; in testing, we aimedfor a total duration of less than 2 minutes in routine situations,as shown in a training video provided to the study sites.1 The"time out" occurs after induction of anesthesia because it isnot always practical to have the surgeon present before induction.In addition, errors may be introduced in the period betweenthe induction of anesthesia and skin incision, including incorrectdraping and delay of antibiotic administration. The cost ofproviding pulse oximetry and prophylactic antibiotics is animportant concern. However, the WHO recommends that these resourcesbe used as minimum standards for safe surgery and that the valueof elective surgery in their absence be critically examined.2In clinical settings without oximetry, as many as 1 in 150 patientshas been reported to have died from anesthesia-related causes,3and a surgical infection rate of more than 20% has been reportedin settings without appropriate prophylactic antibiotics.4
McCambridge et al. are legitimately concerned about the roleobservation may have played in the results. The precise cause-and-effectrelationship between the checklist program and the observedreduction in complications is unclear. Observation could haveproduced a Hawthorne effect by three possible mechanisms. First,the presence of an observer may have affected outcomes, but,as noted, we found no such effect. Second, teams may have usedthe checklist more assiduously because of the ongoing study,but this would not weaken any checklist effect. Finally, thereis the possibility that the performance of the operative teamsimproved because of their awareness of being studied. However,we would recommend that any attempt to implement the checklistinclude monitoring of basic surgical outcomes, resulting ina similar scrutiny of results.
Levin astutely points out that lessons from aviation can beapplied to improve safety in health care. The design of theWHO checklist was informed by experience from aviation and otherindustries. The use of checklists enhances both patient safetyand clinical professionalism.
Alex B. Haynes, M.D., M.P.H. Atul A. Gawande, M.D., M.P.H. Harvard School of Public Health Boston, MA 02115 safesurgery{at}hsph.harvard.edu
World Alliance for Patient Safety. WHO guidelines for safe surgery. Geneva: World Health Organization, 2008.
Ouro-Bang'na Maman AF, Tomta K, Ahouangbévi S, Chobli M. Deaths associated with anaesthesia in Togo, West Africa. Trop Doct 2005;35:220-222. [Free Full Text]
Fehr J, Hatz C, Soka I, et al. Antimicrobial prophylaxis to prevent surgical site infections in a rural sub-Saharan hospital. Clin Microbiol Infect 2006;12:1224-1227. [CrossRef][Web of Science][Medline]
Previous
