Review Article Article de revue

Defining medical

Ethan D. Grober, MD, MEd;* John M.A. Bohnen, MD†

Medical represent a serious public health problem and pose a threat to safety. As institutions establish “error” as a clinical and research priority, the answer to perhaps the most fun- damental question remains elusive: What is a medical error? To reduce medical error, accurate measure- ments of its incidence, based on clear and consistent definitions, are essential prerequisites for effective action. Despite a growing body of literature and research on error in , few studies have defined or measured “medical error” directly. Instead, researchers have adopted surrogate measures of error that largely depend on adverse patient outcomes or (i.e., are outcome-dependent). A lack of standard- ized nomenclature and the use of multiple and overlapping definitions of medical error have hindered data synthesis, analysis, collaborative work and evaluation of the impact of changes in health care deliv- ery. The primary objective of this review is to highlight the need for a clear, comprehensive and univer- sally accepted definition of medical error that explicitly includes the key domains of error causation and captures the faulty processes that cause errors, irrespective of outcome.

Les erreurs médicales constituent un grave problème de santé publique et menacent la sécurité des pa- tients. Alors même que les établissements de santé accordent la priorité dans les interventions cliniques et en recherche à l’«erreur», la réponse à la question peut-être la plus fondamentale nous échappe tou- jours : qu’est-ce qu’une erreur médicale? Afin de réduire les erreurs médicales et d’intervenir efficace- ment, il est essentiel de commencer par en mesurer précisément les incidences en fonction de définitions claires et uniformes. En dépit d’une masse croissante de documents et de recherches sur l’erreur en mé- decine, peu d’études ont défini «l’erreur médicale» ou l’ont mesurée directement. Les chercheurs ont plutôt adopté des substituts de mesures de l’erreur qui reposent en grande partie sur les résultats indési- rables pour les ou les traumatismes (c.-à-d. liés aux résultats). Le manque de nomenclature nor- malisée et les multiples définitions de l’erreur médicale qui se chevauchent ont nui à la synthèse des données, à l’analyse, à la collaboration et à l’évaluation de l’incidence du changement sur la prestation des soins de santé. Cette étude vise principalement à mettre en évidence le besoin d’une définition claire, complète et universelle de l’erreur médicale incluant explicitement les domaines clés des causes d’erreur et saisissant les processus défectueux à l’origine des erreurs, sans égard aux résultats.

he issue of plays a ity to error is becoming apparent. US$17-billion and US$29-billion per T prominent role in health care. Its Medical errors are a leading cause of year in lost income, lost household prominence is fueled by an expanding death in North America;7 between production, disability and additional body of literature that shows a high 44 000 and 98 000 patients are esti- health care costs.9 incidence of error in medicine,1–5 mated to die each year in the USA as To reduce the incidence of errors, coupled with well-publicized medical a result of medical errors.7 Using health care providers must identify error cases that have raised public conservative estimates, deaths due to their causes, devise solutions and concern about the safety of modern medical errors exceed the number at- measure the success of improvement health care delivery.6 tributable to the 8th leading cause of efforts. Moreover, accurate measure- As empirical literature on medical death in North America.8 Medical er- ments of the incidence of error, based error expands, medicine’s vulnerabil- rors are estimated to cost between on clear and consistent definitions,

From the Divisions of *Urology and †General , Department of Surgery, and Centre for Research in Education at the University Health Network, , Toronto, Ont.

Accepted for publication Jan. 26, 2004

Correspondence to: Dr. John M. A. Bohnen, Department of Surgery, University of Toronto, 100 College St., Toronto ON M5G 1L5; fax 416 978-3928; [email protected]

© 2005 Canadian Medical Association Can J Surg, Vol. 48, No. 1, February 2005 39 Grober and Bohnen

are essential prerequisites for effective health care professionals; and pro- Noxious episode: all untoward events, action. Unfortunately and under- pose a new definition of medical er- complications, and mishaps that resulted standably, what is considered a med- ror and justify its use in clinical prac- from acceptable diagnostic or therapeu- ical error (if the term is used at all) tice and research. tic measures deliberately instituted in the hospital (Schimmel, 1964).18 has been influenced by differing con- texts and purposes, such as research, Outcome- versus process- Reflecting the growing presence quality control, ethics, insurance, dependent definitions in the 1970s of third-party insurance legislation, legal action and statutory companies in the economics of health regulation.1,3,5,10 As a result, a lack of Historically, patient safety researchers care, The California Medical Insur- standardized nomenclature and the investigating the impact of error in ance Feasibility Study10 adopted the use of multiple and overlapping defi- medicine have adopted outcome- term potentially compensatable event nitions of medical error has hindered dependant definitions of medical er- to reflect errors that could potentially data synthesis and analysis, collabora- ror and its surrogate terms, and have lead to malpractice claims. tion and evaluation of the impact of limited their focus to patients experi- Potentially compensatable event: an changes on health care delivery.11–15 encing adverse outcomes or injury as event due to medical management that Furthermore, few published studies a consequence of medical care.1–5 Per- resulted in disability, which led to or have measured medical error direct- haps this tendency stems from a guid- prolonged a hospitalization (The Cal- ly.16,17 Instead, researchers have adop- ing principle of medical practice cred- ifornia Medical Insurance Feasibility 10 ted surrogate measures of error such ited to Hippocrates, prium no nocere, Study, 1977). as noxious episodes,18 iatrogenic ill- which translates to “First, do no In the 1990s, the publication of ness,19 critical incidents,20 potentially harm.”21,25,26 Moreover, the manner in the 3 most extensive investigations on compensatable events,10 negligence,1 which patient safety has been defined medical error — the Harvard Medical preventable adverse events,3 slips,21 promotes an outcome-dependant ap- Practice Study,1,2 the Quality in Aus- mistakes21 and violations.14 Recogniz- proach to defining medical error. tralian Health Study,3,4 and the Utah ing the limitations in defining med- Patient safety: the avoidance, preven- and Colorado Medical Practice Stu- ical error, leading medical error com- tion and amelioration of adverse out- dy5 — gave prominence to the term mentators Eric Thomas and Troyen comes or stemming from the adverse event. Brennan22 have warned “Reader, be- process of health care (US National Pa- Adverse event: unintended injury to ware” when making comparisons of tient Safety Foundation, 1999).27 Free- patients caused by medical management error rates in the published literature. dom from accidental injury (Institute of (rather than the underlying condition of 11 As health care institutions establish Medicine, 2000). the patient) that results in measurable error as a research priority, the answer In the earliest studies on patient disability, prolonged hospitalization or to perhaps the most fundamental safety in the 1950s, medical errors both (the Harvard Medical Practice Study,1,2 1991, and the Utah and Colo- question remains elusive: What is a were largely considered “ of rado Medical Practice Study,5 1999). medical error? A theme that reson- medical progress” 28 and dismissed as Unintended injury or that ates throughout the current research “the price we pay for modern diag- results in disability, death, or prolonged 29 literature is the need for a clear, com- nosis and .” These reports hospital stay and is caused (including prehensive and universally accepted tended to be limited to unusual pa- acts of omission and acts of commis- definition.11,16,17,21,23,24 tient reactions or those of magnitude sion) by health care management rather and consequence.28,29 than the patients (Quality in Objectives In “Hazards of hospitalization,” a Australian Health Study, 1995).3,4 pioneering investigation on error in Although adverse events typically The objective of this paper is to medicine, Schimmel18 maintained that result from medical intervention, not highlight the need for a clear, com- “assessment of all untoward reactions, all adverse patient outcomes are the prehensive and universally accepted regardless of severity, is essential to de- result of error. Reflecting this fact, definition of medical error and pro- termine their total incidence and to in- many investigators suggest that only pose a definition that fulfills those dicate the cumulative risk assumed by preventable adverse events be attribu- needs. As part of this proposition, we the patient exposed to the many drugs ted to medical error.1,3,11,21,30 Patient- will review critically how the term and procedures used in his care.” safety experts have considered an ad- medical error has been defined in the With this imperative, he adopted the verse event to be preventable when literature published over the past term noxious episode as a surrogate …there is a failure to follow accepted half-century; describe how other term for medical error, and studied practice (the current level of expected safety-critical industries define the prospectively the type and frequency performance for the average practitioner term error; consider the impact of of such episodes in patients admitted or system that manages the condition in the term error on the psyche of to a university medical service. question) at an individual or system level

40 J can chir, Vol. 48, No 1, février 2005 Defining medical error

(Quality in Australian Health Study, Outcome-dependant definitions of lead to both near misses and adverse 1995); 3 …it is widely established that a medical error have provided valuable events are identical, identification and high incidence of this type of complica- insight into the costs, morbidity and analysis of the processes that poten- tion reflects low standards of care or tech- magnitude of harm resulting from tially can (near misses) and actually nical expertise (Lucian Leape, 1994).21 such events. Nonetheless, quality im- do (adverse events) lead to adverse The relationship between negli- provement initiatives require under- patient outcomes is critical.11,16,34 Ex- gence and preventable adverse events standing of the processes that lead to amples of process-dependant defin- was characterized in the publications such errors.22 Building a safer health itions of medical error in the pub- of the Harvard1,2 and Utah and Col- care system will depend on our suc- lished literature include 5 orado Medical Practice studies. Re- cess at designing processes of care Medical error: the failure of a planned duction of malpractice claims and lit- that ensure patients are protected action to be completed as intended (an igation against health care providers from the threat of injury.11–13,16,21,24,32 error of execution) or the use of a wrong were established as primary objec- Therefore, a definition of medical er- plan to achieve an aim (an error of plan- tives in both of these investigations. ror should capture process or system ning) (Reason, 1990).35 An unintended Negligent adverse events represent a failures that cause errors, irrespective act (either of omission or commission) subset of preventable adverse events of outcome (a process-dependant ap- or one that does not achieve its intended 21 that satisfy the legal criteria used in proach). Ideally, process-dependant outcome (Leape, 1994). Deviations determining negligence. definitions of medical error should from the process of care, which may or capture the full spectrum of medical may not cause harm to the patient (Rea- Negligence: failure to meet the stan- son, 2001).14 errors, namely, errors that result in dard of care reasonably expected of an 35 average qualified to take care adverse patient outcomes as well as Reason’s definition distinguishes of the patient in question (Brennan et al, those that expose patients to risk but between errors of execution and er- 1991).1 Care that fell below the stan- do not result in injury or harm.11,16,33,34 rors in planning, acknowledging that dard expected of in their com- Errors that do not result in injury are mental/judgmental and physical/ munity (Thomas et al, 1995).5 often referred to as near misses, close technical failures both contribute to calls, potential adverse events or war- errors. However, his definition neg- Negligent adverse event: injury caused ning events.34 lects errors of omission: What if there by substandard medical management (Leape, 1991).2 Near miss: any event that could have was no plan, or no action? Leape’s had an adverse patient consequence but definition recognizes that both ac- Adverse patient outcomes represent did not, and was indistinguishable from tions (acts of commission) and inac- a limited subset of medical errors. The a full-fledged adverse event in all but tion (acts of omission) contribute to vast majority of errors do not result in outcome (Barach and Small, 2000).34 medical errors,21 but omits intended injury to patients because the error Given that many of the factors that acts that are based on wrong plans was identified in time and mitigated; because the patient was resilient; or because of simple good luck.11,14 Layers of defence Medical James Reason’s “Swiss cheese” model error of error causation31 illustrates how this notion applies to health care (Fig. 1). According to Reason, most complex Holes in defence systems and work environments (such as hospitals) have several layers of de- fence that offer protection against the adverse consequences of error (signi- fied by several slices of Swiss cheese). In spite of such safeguards, several holes or flaws exist within each indi- vidual layer of defence (the holes in each individual slice). Injury to pa- Adverse event tients occurs only when circumstances arise that cause the flaws in each indi- vidual layer of protection (or holes in FIG. 1. James Reason’s “Swiss cheese” model of error causation. In complex orga- the cheese slices) to align in a way that nizations (hospitals, clinics, biomedical institutes), medical errors cause adverse allows an error to penetrate their de- patient outcomes only when they penetrate through the holes or flaws in the multi- fences and reach the patient.31 ple layers of defence (slices of Swiss cheese).

Can J Surg, Vol. 48, No. 1, February 2005 41 Grober and Bohnen

except for when the actions based on injury or in which the aircraft receives sion of such outcomes obscures the those plans lead to unintended out- substantial damage (Federal Aviation goal of preventing and managing its 37 comes. Regulations, 2001). causes and effects. Human-factor sci- Reason’s and Leape’s definitions The nuclear power industry has entists have acknowledged the im- have further limitations. Although adopted a more inclusive (process- portance of recognizing error (with action plans may not “be completed and outcome-dependant) definition appropriate, timely feedback) as a as intended” or “achieve their inten- of accident that accounts for both the powerful tool for learning, shaping ded outcomes,” errors do not neces- actual and potential consequences of behaviour, and achieving goals.11,14,40,41 sarily account for all of those failures. the event. Reason35 has pointed out that Often, circumstances beyond the commonly, errors occur from the Accident: any unintended event, in- physician’s control influence patient cluding operating error, equipment fail- convergence of multiple and complex 36 outcomes. For example, consider a ures or other mishaps, the consequences contributing factors. Public and leg- patient with no known history of al- or potential consequences of which are islative intolerance for medical errors lergies who experiences an allergic not negligible from the point of view of illustrate a lack of understanding of drug reaction upon starting a new protection or safety (the International Reason’s observations of complex . The outcome is unin- Atomic Energy Agency, 2000).38 human systems. The human factor tended, yet not convincingly attribu- will always be a problem, and ack- table to medical error. To suggest Should the term error be nowledging such factors does permit that all unintended outcomes can be used at all, in medicine? improvement strategies to be under- attributed to medical error is not jus- taken that promote both system tifiable. With growing public concern over changes and education. However, Reason’s definition14 is appropri- the impact of medical errors on pa- blaming or punishing individuals for ately both process-dependant and tient populations, we must not ignore errors related to underlying systemic outcome-independent. Unfortunate- the profound psychological effect causes will not change or address ly, it is in our opinion too general, that errors can have on the health those causes, nor prevent repetition by simply referring to “the process care professionals who make them.11,39 of the errors. of care” rather than stating those The term error carries with it a stigma For this reason, patient-safety ex- processes explicitly. that can evoke feelings of guilt, anger, perts are focusing not on the perpe- inadequacy and depression.14 The trators of individual errors or getting How do other safety-critical threat of legal action compounds such rid of “bad apples,” but on building industries define error? feelings. Some authors have main- safer health care systems to reduce tained that the term error is exces- the probability of errors and miti- Medicine is beginning to benefit sively negative and antagonistic, and gate their effects on patients, em- from interdisciplinary and collabora- perpetuates a culture of blame.23,24,33 ployees and society when they do tive efforts between specialists in A physician or nurse whose confi- occur.11,35,36,42 Errors represent oppor- health care and other high-reliability dence and morale has been shattered tunities for education and construc- and safety-critical industries with a as a result of an error may work less tive changes in health care delivery. long history of recording and ana- effectively and efficiently, and may Ultimately, we may regard them as lyzing errors.11,14,36 Such experiences even consider abandoning a career in “beacons of safety.” 24 reveal that the distinction between medicine.14 This raises an important process and outcome-dependant ap- question: Should the term error be An outcome- and process- proaches to defining error is not used at all? dependent definition of unique to medicine. We acknowledge that it may be medical error The aviation industry uses terms prudent to limit the use of the term like incidents and accidents as substi- error when seeking causes in specific, We propose the following definition: tutes for medicine’s near misses and identifiable cases or patient popula- adverse events, respectively. tions, especially if the documentation Medical error: an act of omission or may become public record. How- commission in planning or execution Incident: an occurrence other than that contributes or could contribute to ever, apprehension over the use of an accident associated with the opera- an unintended result. tion of an aircraft, which affects or could the term error should not lead to its affect the safety of operations (Federal complete removal from work to im- This definition of medical error in- Aviation Regulations, 2001).37 prove patient safety and redesign cludes explicitly the key domains of er- Accident: an occurrence associated health care systems. Adverse patient ror causation (omission and commis- with the operation of an aircraft…in outcomes do occur because of errors; sion, planning and execution), and which any person suffers death or serious to delete the term error from discus- captures faulty processes that can and

42 J can chir, Vol. 48, No 1, février 2005 Defining medical error do lead to errors, whether adverse out- universally accepted definition of Zbar BI, Howard KM, Williams EJ, et al. Costs of medical injuries in Utah and Col- comes occur or not. The inclusivity medical error hinder data collection orado. Inquiry 1999;36:255-64. and explicitness of the definition and collaborative work to improve should make it useful for research into health care systems. If health care 10. Mills DH, Boyden JS, Rubamen DS. Re- the etiology of errors from the per- providers and researchers are to im- port on The Medical Insurance Feasibility spective of the provider: given this def- prove patient safety, we must all Study. San Francisco (CA): Sutter Publica- tions; 1977. inition, a health care worker has a clear speak the same language. The defini- roadmap with which to designate a tion of medical error we have pro- 11. Kohn LT, Corrigan JM, Donaldson MS. process as error-prone or error-laden. posed explicitly addresses the key do- To err is human: building a safer health By including potential adverse out- mains of error causation and includes system. Washington: National Academy comes, the definition includes the process faults that have the potential Press; 2000. “silent majority” of errors that do not for, but do not necessarily lead to, 12. Hebert PC, Levin AV, Robertson G. Bio- cause harm but reflect faulty processes. adverse patient outcomes. ethics for clinicians 23: disclosure of medi- At the same time, it ignores trivial mis- cal error. CMAJ 2001;164(4):509-13. takes (for example, taking the wrong Competing interests: None declared. route to visit a patient) that have no 13. Hatch D. Incidence and acceptance of er- References rors in medicine. Schweiz Arzte, Bull Med potential for adverse outcome. Swiss 2001;82:1339-43. To stimulate collaborative efforts 1. Brennan TA, Leape LL, Laird NM, He- and facilitate data collection, syn- bert L, Localio AR, Lawthers AG, et al. 14. Reason JT. Understanding adverse events: thesis and analysis, we have defined Incidence of adverse events and negli- the human factor. In: Vincent C, editor. medical error explicitly and directly, gence in hospitalized patients: results of Clinical risk management: enhancing pa- tient safety. London: BMJ Publishing avoiding the use of surrogate terms. the Harvard Medical Practice Study I. N Engl J Med 1991;324:370-6. Group; 2001. p. 9-30. We believe that this will help focus error measurement for research, qua- 2. Leape LL, Brennan TA, Laird N, Law- 15. Senders JW. Medical devices, medical er- lity control and legislative purposes, thers AG, Localio AR, Barnes BA, et al. rors and medical accidents. In: Bogner but allow the use of surrogate terms The nature of adverse events in hospital- MS, editor. Human error in medicine. for regulatory, legal and insurance ized patients: results of the Harvard Medi- Hillsdale (NJ): Lawrence Erlbaum Associ- cal Practice Study II. N Engl J Med 1991; ates; 1994. purposes. 324:377-84. Medical, legal and governmental 16. Senders JW. On errors, incidents and acci- institutions must work collaboratively 3. Wilson RM, Runciman WB, Gibberd RW, dents. Human Factors Science. Available: to break down the culture of blame Harrison BT, Newby L, Hamilton JD. www.visualexpert.com/Resources/roadac while retaining methods for account- The Quality in Australian Health Care cidents.html (accessed 2005 Jan 14). Study. Med J Aust 1995;163:458-71. ability. When this challenge has been 17. Hofer TP, Hayward RA. What is an error? met, health care institutions will not 4. Wilson RM, Harrison GB, Gibberd RW, Eff Clin Pract 2000;3(6):261-9. be constrained from measuring the Hamilton JD. An analysis of the causes of most useful target for process im- adverse events from the Quality in Aus- 18. Schimmel EM. The hazards of hospitaliza- provement: all errors, whether or not tralian Health Care Study. Med J Aust tion. Ann Intern Med 1964;60:100-9. 1999;170:411-5. they lead to adverse outcomes. When 19. Steel K, Gertman PM, Crescenzi C, An- that happens, health care can join the 5. Thomas EJ, Studdert DM, Burstin HR, derson J. Iatrogenic illness on a general ranks of similar high-reliability and Orav EJ, Zeena T, Williams EJ, et al. Inci- medical service at a university hospital. N safety-conscious industries such as dence and types of adverse events and Engl J Med 1981;304:638-42. aviation. negligent care in Utah and Colorado. Medical Care 1999;38(3):261-71. 20. Cooper JB, Newbower RS, Long CD, McPeek B. Preventable anesthesia mishaps: Conclusions 6. Cook R, Woods D, Miller C. A tale of two a study of human factors. stories: contrasting views of patient safety. 1978;49:399-406. Medical errors represent an impor- Chicago: National Patient Safety Founda- tant public health problem and pose tion; 1998. 21. Leape L. Error in medicine. JAMA 1994; 272:1851-7. a serious threat to patient safety. The 7. American Hospital Association. Hospital growing awareness of the frequency, statistics. Chicago: the Association; 1999. 22. Thomas EJ, Brennan TA. Errors and ad- causes and consequences of error in verse events in medicine: an overview. In: medicine reinforces an imperative to 8. Center for Disease Control and Preven- Vincent C, editor. Clinical risk manage- improve our understanding of the tion (National Center for Health Statis- ment: enhancing patient safety. London: tics). Deaths: final data for 1997. National BMJ Publishing Group; 2001. p. 31-43. problem and to devise workable sol- Vital Statistics Reports. 1999;47(19):27. utions and prevention strategies. 23. Wears RL, Janiak B, Moorehead JC, Variations in nomenclature without a 9. Thomas EJ, Studdert DM, Newhouse JP, Kellermann AL, Yeh CS, Rice MM, et al.

Can J Surg, Vol. 48, No. 1, February 2005 43 Grober and Bohnen

Human error in medicine: promise and 31. Reason J. Managing the risks of organiza- /internet/fw_learn.htm (accessed 2005 pitfalls, part 2. Ann Emerg Med 2000;36 tional accidents. Aldershot: Ashgate; 1997. Jan 27). (2):142-4. 32. Liang BA, Storti K. Creating problems as 38. NUSAFE Nuclear Installation Safety Net. 24. Fish JM. Human error in medicine: pro- part of the “solution”: the JCAHO Sen- IAEA safety glossary. In: Regulatory con- mise and pitfalls, part 2 [letter]. Ann tinel Event Policy, legal issues, and patient trol of nuclear power plants, 1st ed. 2002. Emerg Med 2001;37(4):419-20. safety. J Health Law 2000;33(2):263-85. Vienna (Austria): International Atomic En- ergy Agency. 2002. Available: www-ns.iaea 25. Veatch RM. Cross cultural perspectives in 33. Holloway RG, Panzer RJ. Lawyers, litiga- .org/tutorials/regcontrol/intro/glossarya medical ethics readings. Boston: Jones and tion, and liability: Can they make patients .htm (accessed 2005 Jan 27). Bartlett Publishers; 1989. safer? Neurology 2001;56(8):24-5. 39. Kaufmann M. Medical error: the human 26. Nightingale F. Notes on hospitals. London: 34. Barach P, Small SD. Reporting and pre- perspective. Ont Med Rev 2002:50-1. Longman; 1863. venting medical mishaps: lessons form non-medical near miss reporting. BMJ 27. National Patient Safety Foundation. Agen- 2000;320:759-63. 40. Weinger MB, Paniskas C, Wilklund M, da for research and development in patient Carstensen P. Incorporating human fac- safety. Available: www.npsf.org (accessed 35. Reason J. Human error. Cambridge: tors into the design of medical devices. 2005 Jan 14). Cambridge University Press; 1990. JAMA 1998;280(17):1484.

28. Moser RH. Diseases of medical progress. 36. Cook RI, Woods DD. Operating at the 41. Andrews LB, Stocking C, Krizek T, Gott- N Engl J Med 1956;255:606-14. sharp end: the complexity of human error. lieb L, Krizek C, Vargish T, et al. An alter- In: Bogner MS, editor. Human errors in native strategy for studying adverse events 29. Barr DP. Hazards of modern diagnosis medicine. Hillsdale (NJ): Erlbaum; 1994. in medical care. Lancet 1997;349(9048): and therapy: the price we pay. JAMA p. 255-310. 309-13. 1955;159:1452-6. 37. Federal Aviation Administration. Aviation 42. Deming WE. Elementary principles of the 30. Nora PF. Improving safety for surgical pa- glossary, Federal Aviation Regulations. In: statistical control of quality: a series of lec- tients: suggested strategies. Bull Am Coll Aviation safety data [Web page]. Wash- tures. Tokyo: Nippon Kagaku Gijutsu Surg 2000;85(9):11-4. ington: FAA. Available: www.asy.faa.gov Remmei; 1952.

Change of address Changement d’adresse We require 6 to 8 weeks’ notice to ensure unin- Il nous faut de 6 à 8 semaines d’avis afin de vous terrupted service. Please send your current assurer une livraison ininterrompue. Veuillez faire mailing label, new address and the effective parvenir votre étiquette d’adresse actuelle, votre date of change to: nouvelle adresse et la date de la prise d’effet du changement, à l’attention du CMA Member Service Centre Centre des services aux membres de l’AMC 1867 Alta Vista Dr. 1867, prom. Alta Vista Ottawa ON K1G 3Y6 Ottawa ON K1G 3Y6 tél 888 855-2555 ou tel 888 855-2555 or 613 731-8610 x2307 613 731-8610 x2307 fax 613 236-8864 fax 613 236-8864 [email protected] [email protected]

44 J can chir, Vol. 48, No 1, février 2005