Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Appendix C International Experiences Limiting Resident Duty Hours Altering the resident experience through reduced duty hours is an ongoing process in many countries. Some have implemented regulations to adjust resident schedules, while others have adopted more guideline-based meth- ods. These diverse approaches accommodate very different healthcare sys- tems, medical education programs, and cultures than in the United States. For most of the countries examined by the committee, the primary reason presented for modification of total resident hours is worker safety. Each country has experienced difficulties implementing its intended reductions due to workforce shortages and some have faced strains on educational training, including reduced clinical contact and procedural experience. The committee reviewed the strategies used by these countries to overcome the challenges encountered from reducing resident duty hours and identified educational redesign, scheduling flexibility, and a period of phase-in to adjust to changes as relevant lessons for the U.S. graduate medical educa- tion system. The desire to identify optimal resident duty hours is not unique to the United States. Duty hour regulation has been in place for more than 20 years in some countries (NZRDA, 2007). The statement of task for this Institute of Medicine (IOM) report specified an examination of the experi- ences of Europe (primarily the United Kingdom), New Zealand, Australia, and Canada to gain insights on additional strategies, practices, interven- tions, and tools employed by these countries in their efforts to adapt to changes in resident schedules. International efforts to modify the duty hours and work environment of medical residents provide useful perspectives on alternative systems. It is difficult to draw direct solutions from these 339
340 RESIDENT DUTY HOURS experiences given that medical education, medical training, and the overall infrastructure of health systems vary substantially between other countries and the United States. However, lessons learned from abroad might help residency programs in the United States anticipate potential challenges of implementing additional changes to resident duty hours and new scheduling practices, if they were to occur. Therefore, this review first provides a snapshot of current duty hour regulations and available patient safety data in the aforementioned coun- tries. Duty hour regulations are then discussed in more detail, along with comments on compliance with and enforcement of those regulations. Con- sequences of reduced duty hours on the resident workforce labor supply, and the impacts of hour reductions on resident education and training are also examined, with a final look at efforts to overcome these challenges through program redesign. General Overview of Country Experiences The rationale for regulations to reduce duty hours in New Zealand and Europe and for duty hour reform in Australia has been primarily from the perspective of worker safety, more so than patient safety, as it has been in the United States or Canada. Yet a combination of these concerns has caused each of the countries examined to make various changes in its medi- cal resident training, and each has used different approaches to implement them. It is important to note that many of those changes are taking place within an extended training system compared to U.S. training programs (Foundation Programme, 2007; Jarvis, 2002). The average duration of medical school in most European countries is 6 years, similar to New Zealand and Australia, with some schools including a year or two of intern- ship training (World Health Organization, 2007). The duration of medical school in Canada and the United States is normally 4 years. Upon being awarded a medical degree, graduates in the United States and Canada spend at least 3 years of residency training for general practice and additional years of training thereafter (usually 3 or more) depending on the specialty pursued (Medical Council of Canada, 2008; National Recruitment Office for General Practice Training, 2008; NHS, 2008a; Royal Australasian College of Physicians, 2007). One example that highlights how different training in other countries can be is the New Zealand experience. After 6 years of medical school, residents there typically spend another 4 years in basic training for general practice before qualifying to undertake specialty training (e.g., surgery, pediatrics, pathology). New Zealand residents who then train in those specialties (known there as âregistrarsâ) âcan continue providing services to patients for more than 10 years while they complete their trainingâ (NZRDA, 2008a, p. 1). In contrast, most specialty residency
APPENDIX C 341 programs in the United States typically last 3-5 years, similar to programs in Canada or the United Kingdom. The maximum weekly duty hours permitted by various recommenda- tions, regulations, and legislation also vary considerably across the coun- tries examined: from 37 hours per week to an unlimited number of hours per week that residents may work (Table C-1). Much of the available infor- mation on duty hours comes from national websites of medical professional organizations, resident organizations, or government agencies. TABLE C-1â Resident Duty Hour Regulations in Various Countries, 2008 Maximum Maximum Minimum How Hours Hours Consecutive Hours of Are Averaged Hours per Rest Between Country Regulated per Week Shift Shifts Europea European 48 13 11 Comission and collective agreements â Franceb Government 52.5 10 â â United Kingdomc Ministerial 56-64 14-24 8-12 agreements â Denmarkd Legislation and 37 13-16 11 collective agreements New Zealande Collective 72 16 8 agreements (not averaged) Australiaf Collective Unrestricted NA NA agreements Canadag Provincial Unrestricted Varies by Varies by collective nationally province province agreements â Manitoba â 89 24 + 2 NA â British Columbia â NA 24 8 â Ontario â 60 (ICU, ED) 24 + 1 (Anes, NA OB/GYN, ICU, CCU) NOTE: Anes = anesthesiology; CCU = cardiac care unit; ED = emergency department; ICU = intensive care unit; NA = Not applicable; OB/GYN = obstetrics-gynecology. aEuropean Trade Union Confederation, 2006. bCode du Travail, 2006; Woodrow et al., 2006. cNHS Employers, 2008b. dDanish Medical Association, 2008; Ministry of Science Technology and Innovation, 2008. eNZRDA, 2007. fAustralian Medical Association, 2005. gPAIRO-CAHO, 2005; PAR-BC, 2008; PARIM, 2008.
342 RESIDENT DUTY HOURS The committee performed a qualitative analysis of the evidence found for the countries investigated in this report that related to residentsâ duty hours and effects on resident training, education outcomes, and patient outcomes. The committee found limited documentation of the impact of reduced resident duty hours on patient outcomes, but a substantial litera- ture on resource management and medical training outcomes linked to duty hour reductions. For example, despite differences in the duty hour regula- tions of each country, a common implementation problem that all have faced is labor supply shortages. Providing residents with sufficient exposure to learning opportunities within the guidelines has also been difficult, but some countries are working on ways to address the issues. Discussion of these challenges follows the country descriptions of current regulations and rates of compliance. Patient Safety Events in the United States and Abroad Regarding the international evidence on patient safety, there have been a few large epidemiologic studies describing adverse events (AEs) and prevent- able adverse events (PAEs) in several countries. Each study has sought to characterize the nature and causes of these events, although none measured the contribution of resident fatigue or duration of work hours. Table C-2 represents the results of major studies conducted in six different countries. Among these studies, only three countries identified admissions from teaching hospitals in their sample and stratified their data by hospital type. AE rates of teaching hospitals compared to non-teaching hospitals were available for Canada, Australia, and the United States (those of the United States are only for the states of Utah and Colorado), and only the Canadian study provided PAE rates as well (Baker et al., 2004; Thomas et al., 2000b). The AE rates of major teaching hospitals in Australia did not differ from those in non-teaching or private hospitals (10.8 percent vs. 10.7 percent) (Thomas et al., 2000b). However, the difference in AE rates of major teaching hospitals in the United States compared to non-teaching hospitals was greater (4.0 percent vs. 2.5 percent) (Thomas et al., 2000b), as it was for Canadian teaching hospitals (10.3 percent vs. 5.2-6.0 percent) (Baker et al., 2004). Baker and colleagues were the only authors that com- mented on the difference in rates between hospital types, highlighting the following explanations: (1) differences in acuity of patient populations â IOM staff searched Medline and Embase databases using a combination of the follow- ing terms: junior doctors, doctors in training, residency, resident work, workload, patient outcomes, adverse medical events, mortality, morbidity, medical education, medical training, Europe, United Kingdom, France, Denmark, Germany, Australia, New Zealand, and Canada. Websites of national medical and resident organizations were also searched.
TABLE C-2â International Comparison of Adverse Events (AEs) and Preventable Adverse Events (PAEs) Country Number Hospital AE Rate from Portion of AE (year data of Medical Medical Rate That Was collected) Hospitals Records AE Definition Records (%) Preventable (%) United Kingdoma 2 â 1,014 Unintended injury caused by medical management 10.8 47.0 (1999-2000) rather than by disease process Denmarkb 17 â 1,097 Unintended injury caused by medical management â 9.0 40.4 (1998) that resulted in disability, death, or prolonged hospital stay New Zealandc 13 â 6,579 Unintended injury or complication that resulted in 12.9 37.1 (1998) disability, death, or prolonged hospital stay and was caused by healthcare management rather than by the underlying disease Australiad,g 28 14,179 Same as New Zealand Adverse Events Study 10.6 NA (1992) Canadae (2000) 20 â 3,745 Same as New Zealand Adverse Events Study â 7.5 36.9 United States â New Yorkf (1984) 51 30,195 Unintended injury caused by medical management â 3.7 that resulted in disability 27.6 â Utah and 28 14,700 Injury caused by medical management that resulted â 3.2 NA â Coloradod,g in prolonged hospital stay or disability at discharge â (1992) aVincent et al., 2001. bSchiÃ¸ler et al., 2001. cDavis et al., 2003. dThomas et al., 2000b. eBaker et al., 2004. fBrennan et al., 1991. gThe data in these two rows come from the Thomas et al. (2000b) study referenced above, which attempted to harmonize the variance in study methods between two earlier and separate 1992 studies: one on AEs in Australia (Wilson et al., 1995) and the other on AEs in Utah and Colorado 343 in the United States (Thomas et al., 2000a). The AE and PAE rates reported in those earlier studies are AE 16.6%, PAE 51% (Wilson et al., 1995); AE 2.9% (Utah, Colorado), PAE 32.6% (Utah), 27.4% (Colorado) (Thomas et al., 2000a).
344 RESIDENT DUTY HOURS between settings are difficult to capture with precise accuracy, (2) teaching hospitals may receive patients at different points in their care that place them at higher risk for AEs than at other hospitals, and (3) care delivered by multiple health professionals in teaching hospitals may increase risk of AEs due to miscommunication or lack of coordination of care (Baker et al., 2004). These same conclusions could be drawn to explain the differences in AE rates among U.S. hospital types, although insufficient data exist to do so. Insufficient data also prevented the committee from being able to conclude whether a correlation exists between duty hours and AEs. The rates of AEs and PAEs found in each of the jurisdictions studied represent a distillation of local practices, documentation, and culture. They also demonstrate methodological differences between studies, temporal changes in care, and the changing nature of defining AEs. In trying to deter- mine a relationship between duty hours and AE rates, inspection of Tables C-1 and C-2 reveals that patients in countries with lower duty hours were not necessarily at less risk of AEs or PAEs than patients in countries with longer duty hours. The AE rates for the United States, represented by Utah and Colorado, are lower than those of all the other countries by more than 50 percent. More recent data on amenable mortality rates for international populations under the age of 75 (Nolte and Mckee, 2008) also show no ap- parent association between preventable events and duty hours across coun- tries. Although these studies do not establish whether a correlation exists between AEs and duty hours, the results appear inconsistent with general expectations that reduced duty hours improve patient safety. Furthermore, they underscore the variability of health systems and patient safety across countries and the complexity of achieving improved patient outcomes. Current Resident Duty Hour Regulations Europe The main objective of the European Working Time Directive (EWTD), issued by the European Council, âis to promote health and safety at work, given the clear evidence that people who work long hours run higher risks of illness and accidentsâ (European Trade Union Confederation, 2006). The EWTD was first established in 1993 to place limits on all workersâ hours throughout Europe. That directive included physicians but excluded âdoc- tors in training.â In 2000, a new directive passed to include the âjunior doctorâ constituency after European resident groups lobbied for the change (Woodrow et al., 2006). The amended article established that medical residents are subject to all laws of the EWTD, accompanied by a require- ment that by 2009 all health systems in the European Union limit resident work to a maximum of 48 hours averaged per week (European Trade Union Confederation, 2006). Regulated transition periods allow countries
APPENDIX C 345 to reduce duty hour limits gradually until they reach the 2009 goal (NHS, 2007). For example, the United Kingdom is currently in a transitional phase consisting of 56-hour workweek schedules (NHS, 2007), while Denmark already meets the 48-hour workweek goal (Ministry of Science Technology and Innovation, 2008). Some countries, whose limits were much higher in the past and are working toward achieving the directive targets for 2009, have found it more difficult to restructure resident schedules to fit the new requirements while maintaining service to patients and educational op- portunities for residents. Some reasons for disparities in transitions include workforce supply issues (e.g., physician-to-patient ratios and resident-to- physician ratios of each country), the general health status of a countryâs population and subsequent effects on physician workload, and the different organizational structures of entire health systems. The EWTD that will apply to European residents by 2009 also includes the following (Council Directive No. 93/104/EC, 1993): â¢ A minimum daily rest period of 11 consecutive hours, â¢ A minimum rest period of 1 day (24 hours) per week, â¢ A maximum of 8 hours of night work on average per 24 hours, and â¢ A right to 4 weeks of paid annual leave. Prior to the establishment of the EWTD, each European country had different duty hour regulations for its medical residents, ranging from a 65- hour-per-week maximum in Ireland, to 56 hours averaged over 24 weeks in Germany, to no hour restrictions at all in Denmark and France (Australian Medical Association, 1998). New Zealand New Zealand has been enforcing duty hour regulations for many years in attempting to address the issue of overworked medical residents. Since 1985, the maximum permissible duty for New Zealand residents has been 72 hours a week, with a set limit of 16 consecutive hours a day. They are also not to work more than 12 consecutive days without a 48-hour break (NZRDA, 2007). However, the regulations allow residents to be scheduled to work seven consecutive night shifts of 10 hours or more roughly once a month, which has been identified as being counter productive to resident well-being and performance (Dula et al., 2001; Powell, 2004). Australia Australia reacted very differently to the issue of duty hour restrictions for its medical residents. In 1996, the Australian Medical Association ad-
346 RESIDENT DUTY HOURS opted federal policy for safer working environments for medical residents and, in 1999, further adopted a National Code of Practice for them to follow (Scallan, 2003). Its purpose is to provide âpractical guidance on how to eliminate or minimise risks arising from the hazards associated with shiftwork and extended working hours,â which include guidelines on performance-based scheduling, incident reporting, and education on fatigue mitigation techniques (Australian Medical Association, 2005, p. 4). This code is not a legal regulation nor is it mandated by any organization within or outside the medical community. The code is simply a set of guidelines that hospitals and doctors are strongly encouraged to follow according to their best judgment and acknowledge the responsibilities of employers and employees under Australiaâs Occupational Health and Safety legislation (Australian Medical Association, 2005). The committee that composed the code based it on available empirical evidence on sleep and fatigue and their impact on work performance. This evidence showed that extended hours of work had negative effects on medical training for three main reasons: âlack of time for formal and independent study, lack of motivation due to fatigue, and work patterns that failed to provide necessary supervision and feedback for effective learningâ (Scallan, 2003, p. 910). Canada In Canada, individual provinces establish duty hour regulations through collective agreements negotiated by resident associations with hospitals, re- sulting in variations in these regulations across the country. Currently, the only province that has an explicit hour limit is Manitoba, which enforces a maximum of 89 hours per week averaged over 4 weeks (Fok et al., 2007; PARIM, 2008). In the provinces of Alberta, Quebec, and the Maritimes, collective agreements limit resident duty to 12 hours per routine day (not averaged over a week). Ontario and Saskatchewan have no overall hour limit, but the Professional Association of Internes of Ontario does have limitations of 60 hours per week for residentsâ performing in-hospital shift work, such as in intensive care units (ICUs) and emergency departments (EDs) (Fok et al., 2007). Shifts in EDs cannot exceed 12 hours, and in ICUs they cannot exceed 24 hours (plus handover); in other hospital areas, after 24 hours on call residents cannot admit new patients and have to be at home by noon (PAIRO-CAHO, 2005). Enforcement of DUTY Hour Regulations To ensure compliance with the regulations outlined in the previous sec- tion, each country has developed its own method for enforcing them; none seem to constitute an ACGME equivalent.
APPENDIX C 347 Europe In the United Kingdom, National Health Service (NHS) Employers en- force compliance with hour regulations. They are under contractual obliga- tion to monitor compliance by collecting and reporting duty hour patterns of their residents on a semiannual basis, which the NHS Executive and the British Medical Association review jointly. Recent compliance reports show that most health authorities perform at near-perfect compliance (NHS Employers, 2008a). However, a 2004 independent study on resident com- pliance in a large urban U.K. hospital showed residents working beyond the duty hour regulations (whose weekly maximum was 56 hours at the time): the average number of hours worked by physicians was 63.6 hours per week and by surgeons 70.2 hours per week (Jagsi and Surender, 2004). Such mixed results imply that monitoring efforts may not be completely rigorous or accurate, perhaps a common problem shared with the United States. Data from a 2002 survey gathered by the Royal College of Physicians indicated that residents and physicians in most European countries worked total hours or continuous duty periods in excess of those established in the EWTD (many worked 30 or more continuous hours) despite claims that these countries were compliant with the directive (Royal College of Physicians, 2002). Such excesses are consistent with the previously stated findings in the United Kingdom. The committee did not find more current compliance rates for other European countries. The committee also had difficulty uncovering a clear picture of enforc- ing bodies for EWTD regulations in the remaining European countries (Mayor et al., 2004). However, an article reporting, in part, on France, whose national government is the regulatory force for duty hour limits, mentioned that it has not assigned an organization the responsibility for enforcing the established limits, nor does it currently monitor compliance rates itself (Woodrow et al., 2006). In Germany, since residents are fully licensed physicians and regular employees of the hospitals, their duty hours are enforced by the same state and local institutions that are responsible for enforcing the duty hours of all employees (trade supervisory board). These few examples indicate that monitoring and enforcement methods likely vary significantly across European countries. New Zealand The New Zealand Resident Doctorsâ Association (NZRDA) negotiates resident hours for its members through collective employer agreements and regulates the provisions therein (NZRDA, 2008b). In 2006, the average duty hours of New Zealand doctors was 45.8, and the average worked by doctors aged 24 years or younger was 57.7 hours per week (Medical
348 RESIDENT DUTY HOURS Council of New Zealand, 2006). Both numbers fall well below the national maximum of 72 hours. A nationwide survey of residents (63 percent re- sponse rate: 1,366 responses) conducted by Gander and colleagues in 2003 showed that 57 percent of residents worked between 50 and 70 hours a week and that 13 percent worked more than 70 hours, substantially more than the average hours reported in 2006 to the Medical Council (Gander et al., 2007). Australia Australiaâs approach to resident work scheduling is unique and may reflect a culture with expectations that resident hours should be similar to those of other workers in the population. Resident duty hour guidelines are advisory, as opposed to being binding rules, and there is no designated enforcement body. The guidelines of the code imply that working more than 50 hours per week puts a resident at âsignificant riskâ of fatigue and associated negative consequences, while working 70 hours or more is considered to put residents at âhigher riskâ (Australian Medical Associa- tion, 2005; Scallan, 2003). The Australian Medical Association takes it upon itself to conduct national surveys of physician duty hours as a way to gauge current practice. Results of the 2001 national survey showed that 70 percent of Australian medical residents worked an average of more than 50 hours a week (Scallan, 2003). The independent study by Gander et al. (2007) had similar results. In May 2006, the Australian Medical Associa- tion conducted another national survey representing all doctors (not just junior doctors), and results indicated that 62 percent of hospital doctors are in the âsignificantâ and âhigher riskâ categories for doctor safety based on hours of work, with 85 percent of surgical doctors falling in those catego- ries. Although the later survey noted some relative improvements (e.g., the longest continuous period of work was 39 hours, down from 63 hours in 2001), it is inaccurate to compare the results of the two surveys since the earlier one surveyed only residents, whereas the latter surveyed all physi- cians. Also, there was no mention in either survey of any fatigue-mitigating methods used by doctors during extended shifts on duty, such as periods for sleep. The authors of the later report concluded that 39 hours is still too many to work consecutively and that hospitals need to continue tak- ing steps to mitigate any safety risks to residents and patients (Australian Medical Association, 2006). Canada The Royal College of Physicians and Surgeons of Canada (RCPSC) and the College of Family Physicians of Canada (CFPC) are the regula-
APPENDIX C 349 tory bodies for duty hours in Canada. âHowever, the RCPSC has neither developed nor endorsed specific policies regarding duty hours. As a result, there has been no national drive to implement duty hour restrictions, but most provinces have nonetheless adopted them independently of each other within a relatively short time frameâ (Woodrow et al., 2006, p. 1047). Just as duty hour regulations vary according to province, so does enforcement of those regulations. Data on actual hours worked in Canada are very limited, but a 2007 survey of British Columbia first-year residents is helpful in describing some general patterns that are occurring at least in that part of Canada. The survey reported that these residents work an average of 65.4 hours a week (including on-call work) and sleep an average of 41.9 hours a week (Fok et al., 2007). However, when looked at separately, surgical residents reported working many more hours than their non-surgical colleagues. On aver- age, surgical residents worked 80.4 hours per week compared to 57 hours per week for non-surgical residents. Of the surgical residents, orthopedic surgeons and general surgeons work the most hours, 102 hours and 88.9 hours a week, respectively (Fok et al., 2007). Overall Although governing bodies try to enforce compliance with duty hour limits, there is some indication that the number of hours reported officially by residents through their institutions are not necessarily accurate. Gener- ally, the residents in separate surveys seem to report working more than the guidelines suggest. Therefore, the committee cannot conclude that these countries uniformly enforce their stated hour limits or that the reported compliance rates are accurate. The following section discusses some reasons why countries may encounter difficulties achieving compliance with their own regulations along with other challenges that medical training has faced because of reducing duty hours. Consequences and Barriers Reducing resident duty hours abroad has led to several changes in health systems that have pushed countries to create new ways to provide continuous services to patients and maintain educational opportunities for residents. The most significant organizational change among hospitals to achieve this has been the trend to replace traditional 24-hour (or longer) shifts with shorter shifts of 12 or 16 hours in accordance with local require- ments or preferences (BBC, 2004). Concerns about shift-based schedules re- garding continuity of care and resident training have been articulated (Carr, 2003; Jagsi and Surender, 2004; Powell, 2004; Royal College of Surgeons
350 RESIDENT DUTY HOURS of England, 2008), as in the United States, and countries are taking steps to address related changes (e.g., improving handover protocol). However, another persistent problem caused by this change in work system is that of labor supply shortage. Labor Supply Shortages Employing shift systems, with explicit or implicit reductions in total duty hours, requires more doctors to provide 24-hour coverage. The imple- mentation of more shift-based schedules to comply with the duty hour regulations and recommendations in Europe, New Zealand, and Australia has thus created or exacerbated workforce shortages and strained financial resources. This has been identified as one of the most difficult challenges that countries face as a result of restricting work hours (Child and Old, 2004; Powell, 2004; Sheldon, 2004). New Zealand hospitals faced a sudden shortage of residents after the new regulations passed in 1985; they were unable to ensure resident staffing to care for patients at all times. Hospitals decided to employ more residents because increasing resident staffing was the most immediate solution avail- able to provide sufficient coverage of care while being asked to comply quickly with reduced hour restrictions (Child and Old, 2004). However, as the demand for more residents began to outstrip the supply of medical graduates, it became necessary to employ more senior doctors and nurses to cover gaps in working hours (Powell, 2004). Residents in New Zealand are currently in short supply, not only because of increased demand from hour restrictions but also because of increased service demands on the healthcare system by a rapidly growing elderly population (a trend that countries worldwide are facing). In Europe, recruiting more residents to fill the needed positions requires additional funding to support their salaries, but hiring higher-level health professionals (with accompanying higher salaries) to fill the same gaps significantly increases the costs associated with supplying enough workers. In an attempt to comply with the mandated changes of the EWTD, âmany countries maintain that a significant increase in physicians and residents will be needed to adequately cover service requirements. Figures ranging from 6,000 to 27,000 have been cited at a cost of up to EUR 1.75 billionâ (Sheldon, 2004; Woodrow et al., 2006, p. 1048). The United Kingdom alone estimated the need for an additional 15,000 doctors to staff the NHS in order to meet the requirements of the Working Time Directive (Child and Old, 2004). Australiaâs health workforce is also under pressure as doctors work fewer hours and show preferences for work-life balance, such as having more time to spend with family (Productivity Commission, 2005). âThe
APPENDIX C 351 medical workforce shortage [in Australia] is both absolute and relative: we do not train enough doctors, and once trained, doctors are not working in areas of greatest need. Chronic vacancies exist across the health care systemâ (Skinner, 2006, p. 35). In February 2004, the state of New South Wales reported more than 900 vacancies at the resident and registrar levels (Greater Metropolitan Clinical Taskforce Metropolitan Hospitals Locum Issues Group, 2005). Exacerbating the problem is the decision of many junior doctors to opt out of the public hospital workforce to work locum positions instead (part-time fill-in work), which pay nearly three times more than full-time salaried public positions. This, in turn, significantly increases hospital budgets and reduces the number of doctors training in needed specialties (Chisholm, 2008) since locum certification does not require specialty training and many are general practitioner positions. However, a recent recruitment assessment in Australia indicates that limiting shifts to 12 hours is a positive and effective solution to recruitment and retention challenges among critical care nurses that may hold promising solutions for residents as well (Dwyer et al., 2007). Impact on Training and Education Limiting duty time has the potential to reduce the educational oppor- tunities residents have (Carr, 2003). There is speculation from various U.K. sources that âshift working has had a detrimental effect on learning for doc- tors in trainingâ and that new rotations âhave led to an erosion of the team system and a loss of mentorship in trainingâ (Carr, 2003, p. 623). Residents also perceive adverse effects on continuity of care and on resident-patient relationships and resident-supervisor relationships (Ardagh, 2003; Jagsi and Surender, 2004). Supervision seems to be equally important for resident learning abroad as it is here (Scallan, 2003), and its importance receives attention in Chapter 4 of this report. Surgeons particularly express significant concern about the potential effects of hour limits on resident performance because of reduced operative experience (Royal College of Surgeons of England, 2007b, 2008; Thorne et al., 2006) or reduced follow-through with postoperative patient care (Mestres et al., 2006). A large British survey in 2002 showed that a third of orthopedic surgical residents were not taught in the operating theater or in clinic (British Orthopaedic Association, 2002), and a senior resident in the British training system commented that âmany senior house officers arrive at posts halfway through their rotations without any real competence in operative skills as basic as suturing and tying knotsâ (Chikwe et al., 2004, pp. 418-419). Another large survey of cardiothoracic residents throughout the United Kingdom reported a unanimously perceived negative impact on their training since EWTD implementation (West et al., 2007). Less
352 RESIDENT DUTY HOURS hands-on training and an increase in shift-based duty has led to reports of residentsâ dissatisfaction with their degree of learning or the teaching style provided (Rawnsley et al., 2004; West et al., 2007). The committee listened to testimony by Dr. Bernard Ribeiro, president of the Royal Col- lege of Surgeons of England, that reiterated many of these same sentiments and presented additional data on reduction in the number of operations performed across surgical specialties since the 2004 EWTD, showing some residents performing as much as 25 percent fewer procedures (Royal Col- lege of Surgeons of England, 2008). These findings imply that decreasing residentsâ time to perform surgical procedures may decrease their general level of competence in these tasks due to the demands and intricacies of the work (Chikwe et al., 2004). The committee found no other objective measures of such outcomes, and as noted below, evaluation of these impacts on training has yet to be published. To compensate for reductions in training time for U.K. surgical resi- dents, redesign efforts have focused on providing more time concentrated on procedural activities than other activities. For example, the study by Lim and colleagues (2006) suggests that reorganizing institutional structure can maintain the competency levels of cardiac surgical trainees in these activities despite the reduction in work hours. Their institution adopted a team-based model to achieve the desired results, ensuring adequate time with trainers to maximize learning (Lim and Tsui, 2006). However, others in the field have noted the difficulty that less robust programs may have in implementing such changes and the likelihood that many programs would not be able to do so, given their lack of resources (Mestres et al., 2006; West, 2007). In response to these issues, the Royal College of Surgeons has developed its own set of rotation guidelines in compliance with the EWTD for surgeons in training to follow, hoping to maximize patient safety and resident surgeon learning (Royal College of Surgeons of England, 2007a). This issue has yet to be resolved in the United Kingdom, and no surgeons have yet emerged from the shortened training system that would allow the NHS to evaluate its overall impact and costs. Observations of resident training in Australia suggest that much resi- dent time is spent on administrative tasks and providing service demands for the health system (Gleason et al., 2007). A consequence of this has been variable quality in resident education because of insufficient time to attend some of their didactic sessions (Gleason et al., 2007). Australian medical education leaders have observed that âthe existing systems for delivery of education and training are inefficient, under-resourced and under pressure, and they will not be sustainable into the futureâ (McGrath et al., 2006, p. 348). This may not necessarily be a result of reduced hours, but reducing hours could further affect such limitations.
APPENDIX C 353 The committee concludes that reports from other countries indicate that reducing total duty hours places strain on the healthcare workforce and creates challenges to maintaining the quality of resident learning en- vironments. Several countries have experienced financial and manpower strains trying to staff hospitals sufficiently to provide adequate 24-hour care, and many have reported insufficient time to teach residents or to have them practice procedures. Noneducational resident workload has also been found to be an impediment to the resident learning experience. Applicable Lessons for Residency Programs In the United States Retaining Flexibility in Work Schedules A key lesson from these countries is the need for flexibility in how programs arrange schedules within the confines of established regulations. In the United Kingdom, since the introduction of the Working Time Regu- lations, the NHS has moved toward achieving the goal of safer work con- ditions through reduced hours for medical residents (NHS, 2007). While considering how best to implement the newest directive with further re- duced hours, the NHS has taken into account an array of rotation sched- ules with various combinations of night shifts, weekend shifts, day shifts, and rest hours that its hospitals could adopt (Royal College of Physicians, 2006). In each instance, individual hospitals need to determine the most effective combination of workforce (e.g., numbers of residents and other staff on duty in a given shift) and required hours to achieve safe patient care and quality resident education. Similar to the medical education system in the United Kingdom, different types of residency programs in the United States have national requirements to fulfill regarding educational content within the duty hour restrictions. Therefore, the different organizational methods across programs require flexibility in the way work schedules are arranged. The committee concludes that, given the differences among healthcare institutions, different specialties, and the needs they must satisfy, maintain- ing flexibility in the way work hours are scheduled is necessary. This echoes the findings in Chapter 3 of the experiences of programs in the United States as they adjusted to the 2003 duty hour limits. It appears that other coun- tries have experienced problems with adherence to established duty hour limits, as have programs in the United States. (See Chapter 2 on adherence to duty hours and its enforcement in the United States.)
354 RESIDENT DUTY HOURS Time for Phase-in The international experience indicates that a well-planned period of transition would be necessary as part of any major changes in total work hour limits. In Europe, 4 years after residents became subject to the EWTD laws, as programs were still transitioning to current requirements, multiple countries in the European Union (EU) were protesting the mandated sched- ule claiming that they lacked the resources, in either workforce, finances, or both, to successfully comply with the regulations by the given deadline (Sheldon, 2004). Based on the international experience, a phase-in of committee- r Â ecommended changes would be beneficial. Changes specifically related to resident duty hours and schedules might require some time for planning to accommodate constraints in workforce and other resources. Chapter 7 contains the committeeâs recommendations for adjusting duty hours, and Chapter 9 includes macro-level estimates of the potential costs, as well as the type and number of clinical personnel that would be needed to replace existing residents under further adjusted duty hours. Redesign of Resident Education and Training Systems The introduction of duty hour restrictions for residents in other coun- tries has created an impetus for changing the fundamentals of medical education and training programs. Complying with new regulations caused hospital providers to alter their work practices, staffing, and delivery meth- ods in order to ensure a degree of quality care in light of these significant changes. For example, in Australia the National Code clearly states that it was created as âone part of a broader education and awareness program to change the current individual and organisational beliefs and culture that support working hours and patterns that would be considered unaccept- able in most other industry sectorsâ (Australian Medical Association, 2005, p. 4). Culture change is at its core. To achieve this goal, the Australian Medical Association developed a risk assessment strategy to evaluate the extent of hazard caused by residentsâ working extended hours (working more than 50 hours per week is considered âsignificant risk,â working 70 hours or more is âhigher riskâ) and recommends principles that should be at the foundation of work schedule design to minimize risk to patient and resident safety. Aside from these assessments, high service demands and inadequate funding of education and skills training have also been identi- fied as risks to resident training (Gleason et al., 2007). As institutions try to increase the number of residents to satisfy service demands, the authors of the study note that merely increasing the numbers of residents will not
APPENDIX C 355 alleviate their work intensity, nor will it solve training deficiencies unless adequate resources are provided. They conclude that a strong emphasis on training, supported by sufficient resources, is necessary to deal with these changes and reference the U.K. efforts to redesign its educational program (examined below). Similarly, although restrictions on hours vary by prov- ince in Canada, many medical training programs throughout the coun- try have been focused on improving patient outcomes through improved resident training strategies for several years. Strategies have consisted of increasing the educational value and success of residency programs rather than using them to produce residents as a cheaper form of healthcare work- force (Landau, 2007). Competency-Based Training Perhaps the foremost issue posed by reduced duty hours for residents is how to ensure competence during and at the end of training. To address this issue and create long-term improvements, changes need to occur at the educational level. This has much to do with the fact that the duty hour restrictions provide less time to train residents. As a result, both the United Kingdom and New Zealand have been significantly reshaping their resi- dency programs toward competency-based or performance-based training, which determines a traineeâs level of expertise by their ability to demon- strate specified required skills at a given point in training, as opposed to a more time-based educational model, which focuses on completing physi- cian training in a certain number of years. Their programs now center on providing an experience with more educational value for residents and have been fundamentally restructured to ensure the desired results (NHS, 2008b; Workforce Taskforce, 2007). One such example from the United Kingdom is the new Foundation Programme, which constitutes the first 2 years of residency training upon graduating medical school. Intended to âbridge the gap between under- graduate school and specialist/medical trainingâ (Foundation Programme, 2007, p. 5), the Foundation Programme focuses on patient safety and improving the quality of care by having residents demonstrate competence in communication and consultation skills, patient safety, and teamwork, in addition to more typical clinical skills (Major shake-up in medical training, 2005). This training replaces the previous introductory 2-year program and is a departure from that program, which focused nearly exclusively on demonstrating competence in clinical skills (Foundation Programme, 2007). No assessments have been made to date as to whether the quality of resi- dent work has been affected by the new program, although one study has reported concerns on behalf of both the trainees and the educators about the generic nature of some of the skills trainees were expected to acquire
356 RESIDENT DUTY HOURS (OâBrien et al., 2006). Still in its nascent years, the Foundation Programme has opportunities to modify its curricula. Monitoring Quality Measures and New Models of Care The NHS has also developed several pilot programs that not only alter resident rotations and schedules, but also incorporate new quality improve- ment practices such as regularly assessing patient and staff satisfaction and reporting patient safety measures, as well as measuring the compliance of residents with the time limits. In some instances the pilot programs show that they achieved reductions in patient wait times for treatment or surgery, reductions in the time between prescribing and administering certain medi- cations, shorter lengths of stay in hospital wards, and improved discharging processes, even with reduced resident hours (NHS, 2004). One pilot created the Hospital at Night program, a new model of care that reduces the risk of adverse events by improving the quality of care tak- ing place during nighttime hours in hospitals, while also aiming to enhance resident learning during daytime hours (Hospital at Night and The NHS Information Centre, 2008; Institute for Healthcare Improvement, 2007; NHS, 2008c). With the completion of more pilots, the NHS anticipates providing additional solutions to the EWTD that are transferable across residency programs and improve resident learning and the quality of patient care across the entire health system. Based on the above review of educational changes, it is evident that some countries have begun to redesign their educational system, focusing changes in both curriculum and competency-based aspects, and that these changes have been in response to decreased duty hours and some negative impacts on resident training. Conclusion When considering the experiences of the countries examined in this chapter with duty hour regulations, it is important to remember that most of these countries have lower duty hour limits than the United States (e.g., Europe, 48; New Zealand, 72). Even Australia, where no national maxi- mum limit exists, tries to have residents work no more than 70 hours per week. With these lower limits, key stakeholders in those countries report considerable challenges as they adjust to the reduced availability of resi- dents to provide 24-hour care. The concerns about workforce shortages and financial constraints because of these changes are particularly noteworthy, as are the frequently voiced concerns that the educational experience of residency is deteriorating. However, some countries have produced innova- tive designs and promoted culture change.
APPENDIX C 357 Yet based on the evidence examined from other countries, it is not possible to determine an ideal number of hours that residents should be required to work. There are limited or no data to show that decreasing duty hours improves patient safety, and many redesign efforts have yet to be fully developed or evaluated for effect on resident learning or patient safety. Furthermore, the international data on the quality of patient safety indicate that duty hour regulations may not be the greatest factor affecting quality of care and that other facets of delivery systems or educational programs likely deserve more attention as areas for improvement. Despite the limited evidence from other countries, the committee was able to draw lessons on retaining scheduling flexibility, allowing phase-in time, and redesigning educational programs for residency training. How- ever, the committee concluded that no single system is directly applicable to that of the United States given the different construct and culture of our healthcare system. Although the committee lacks systematic multinational evidence on resident education and patient outcomes as a result of imple- menting duty hour regulations, from the evidence gathered it appears that any changes in duty hour limits would necessitate modifications to redesign and enhance the medical training system, to have an available workforce to substitute for hours that residents are not available to staff, and to create substantial financial resources. A similar call to redesign medical education and training programs exists in the United States (Meyers et al., 2007). It may be important to maintain some degree of flexibility as residency training programs determine which innovative educational approaches and new scheduling designs to adopt for their healthcare delivery and resident learning needs. REFERENCES Ardagh, M. 2003. Beyond Ashburton: Junior hospital doctor employment in New Zealand. The New Zealand Medical Journal 116(1168):1-2. Australian Medical Association. 1998. Overseas experience in regulating hours of work of doctors in training. Kingston ACT, Australia: Australian Medical Association. âââ. 2005. National code of practiceâHours of work, shiftwork and rostering for hospital doctors. Kingston ACT, Australia: Australian Medical Association Ltd. âââ. 2006. Safe hours = safe patients: AMA safe hours audit 2006. Kingston, Australia: Australian Medical Association. Baker, G. R., P. G. Norton, V. Flintoft, R. Blais, A. Brown, J. Cox, E. Etchells, W. A. Ghali, P. Hebert, S. R. Majumdar, M. OâBeirne, L. Palacios-Derflingher, R. J. Reid, S. Sheps, and R. Tamblyn. 2004. The Canadian Adverse Events Study: The incidence of adverse events among hospital patients in Canada. CMAJ: Canadian Medical Association Journal 170(11):1678-1686. BBC (British Broadcasting Corporation). 2004. Q&A: Junior doctorsâ hours. BBC News. http://news.bbc.co.uk/1/hi/health/3941073.stm (accessed January 23, 2008).
358 RESIDENT DUTY HOURS Brennan, T. A., L. L. Leape, N. M. Laird, L. Hebert, A. R. Localio, A. G. Lawthers, J. P. N Â ewhouse, P. C. Weiler, and H. H. Hiatt. 1991. Incidence of adverse events and neg- ligence in hospitalized patients. Results of the Harvard Medical Practice Study I. New England Journal of Medicine 324(6):370-376. British Orthopaedic Association. 2002. Education and training for SHOs: A snap-shot of the moment and recommendations for the future. London: British Orthopaedic Association. Carr, S. 2003. Education of senior house officers: Current challenges. Postgraduate Medical Journal 79(937):622-626. Chikwe, J., A. C. de Souza, and J. R. Pepper. 2004. No time to train the surgeons. BMJ 328(7437):418-419. Child, S., and A. Old. 2004. Resident medical officer working conditions in New Zealand: Results of a recent survey. New Zealand Medical Journal 117(1204). Chisholm, D. 2008. Junior doctors opting out. http://www.stuff.co.nz/sundaystartimes/ 4464669a20455.html (accessed May 27, 2008). Code du Travail (Labor law in France). 2006. Paris, France: MinistÃ¨re du travail, des relations sociale, et de la famille et de la solidaritÃ©. Council Directive No. 93/104/EC. 1993. European working time directive. Council of the European Union, Brussels. Danish Medical Association. 2008. Dutiesâin brief. http://www.laeger.dk/portal/page/portal/ LAEGERDK/LAEGER_DK/ENGLISH/DUTIES_IN_BRIEF (accessed June 11, 2008). Davis, P., R. Lay-Yee, R. Briant, W. Ali, A. Scott, and S. Schug. 2003. Adverse events in New Zealand public hospitals II: Occurrence and impact. New Zealand Medical Journal 116(1183). Dula, D. J., N. L. Dula, C. Hamrick, and G. C. Wood. 2001. The effect of working serial night shifts on the cognitive functioning of emergency physicians. Annals of Emergency Medicine 38(2):152-155. Dwyer, T., L. Jamieson, L. Moxham, D. Austen, and K. Smith. 2007. Evaluation of the 12-hour shift trial in a regional intensive care unit. Journal of Nursing Management 15(7):711-720. European Trade Union Confederation. 2006. Working time directive. http://www.etuc.org/ a/504 (accessed November 2, 2007). Fok, M. C., A. Townson, B. Hughes, and W. C. Miller. 2007. Work hours, sleep depriva- tion, and fatigue: A British Columbia snapshot. British Columbia Medical Journal 49(7):387-392. Foundation Programme. 2007. Curriculum, Academy of Medical Royal Colleges, London. Gander, P., H. Purnell, A. Garden, and A. Woodward. 2007. Work patterns and fatigue- related risk among junior doctors. Occupational and Environmental Medicine 64(11): 733-738. Gleason, A. J., J. O. Daly, and R. E. Blackham. 2007. Prevocational medical training and the Australian curriculum framework for junior doctors: A junior doctor perspective. Medi- cal Journal of Australia 186(3):114-116. Greater Metropolitan Clinical Taskforce (GMCT) Metropolitan Hospitals Locum Issues Group. 2005. Meeting the medical locum challengeâOptions for action within the NSW public hospital system. Sydney, Australia: GMCT. Hospital at Night and the NHS Information Centre. 2008. The case for hospital at nightâThe search for evidence. Manchester, England: NHS. Institute for Healthcare Improvement. 2007. The hospital at night project: Reducing risks at our most vulnerable time of day. Boston, MA: Institute for Healthcare Improvement.
APPENDIX C 359 Jagsi, R., and R. Surender. 2004. Regulation of junior doctorsâ work hours: An analysis of British and American doctorsâ experiences and attitudes. Social Science & Medicine 58(11):2181-2191. Jarvis, J. 2002. Response from an RMO. New Zealand Medical Journal 155(1166). Landau, L. I. 2007. Quality junior doctor training, improved workforce outcomes and patient safety. Australian Health Review 31(Suppl 1):S106-S108. Lim, E., and S. Tsui, on behalf of the Registrars and Consultant Cardiac Surgeons of Papworth Hospital. 2006. Impact of the European Working Time Directive on exposure to operative cardiac surgical training. European Journal of Cardiothoracic Surgery 30(4):574-577. Major shake-up in medical training, UK. 2005 Medical News Today. http://www.medicalnews today.com/articles/29074.php (accessed November 12, 2007). Mayor, S., J. Burgermeister, K. Kosner, T. Villanueva, A. Tuffs, B. Spurgeon, F. Turone, M. Houston, and T. Sheldon. 2004. Over the limit? Student BMJ 12(September). McGrath, B. P., I. S. Graham, B. J. Crotty, and B. C. Jolly. 2006. Lack of integration of medical education in Australia: The need for change. Medical Journal of Australia 184(7):346-348. Medical Council of Canada. 2008. Medical training/licensure system in Canada. http://www. img-canada.ca/en/licensure_overview/licensure.html#undergrad (accessed February 19, 2008). Medical Council of New Zealand. 2006. Annual report 2006: Medical workforce survey. http://www.mcnz.org.nz/portals/0/publications/2006.pdf (accessed May 29, 2008). Mestres, C.-A., J.-M. Revuelta, and A. C. Yankah. 2006. The European working time di- rective: Quo vadis? A well-planned and organized assassination of surgery. European Journal of Cardiothoracic Surgery 30(4):571-573. Meyers, F. J., S. E. Weinberger, J. P. Fitzgibbons, J. Glassroth, F. D. Duffy, and C. P. Clayton. 2007. Redesigning residency training in internal medicine: The consensus report of the Alliance for Academic Internal Medicine Education Redesign Task Force. Academic Medicine 82(12):1211-1219. Ministry of Science Technology and Innovation. 2008. Working hours. http://www.workin denmark.dk/Working_hours (accessed June 10, 2008). National Recruitment Office for General Practice Training. 2008. GP careers: The training program. http://www.gprecruitment.org.uk/gpcareers/programme.htm (accessed Sep- tember 18, 2008). NHS (U.K. National Health Service). 2004. Working time directive: Pilots programme re- port. http://www.healthcareworkforce.nhs.uk/pilotprojects.html (accessed November 14, 2007). âââ. 2007. European working time directive (EWTD). Coventry, UK: National Library for Health. âââ. 2008a. Modernising medical careers: Training in 2008. http://www.mmc.nhs.uk/ default.aspx?page=318 (accessed September 18, 2008). âââ. 2008b. The pastâThe principles of MMC. http://www.mmc.nhs.uk/default.aspx? page=310 (accessed June 2, 2008). NHS Employers. 2008a. Junior doctorsâ terms and conditions of service and associated docu- ments. http://www.nhsemployers.org/pay-conditions/pay-conditions-467.cfm (accessed May 30, 2008). âââ. 2008b. National Health Service hospital medical and dental staff and doctors in public health medicine and the community health service (England and Wales): Terms and conditions of service. NHS. âââ. 2008c. Hospital at night. http://www.healthcareworkforce.nhs.uk/hospitalatnight. html (accessed June 2, 2008).
360 RESIDENT DUTY HOURS Nolte, E., and C. M. McKee. 2008. Measuring the health of nations: Updating an earlier analysis. Health Affairs 27(1):58-71. NZRDA (New Zealand Resident Doctorsâ Association). 2007. Collective agreement. ÂAuckland, New Zealand: New Zealand Resident Doctorsâ Association. âââ. 2008a. Resident doctorsâ association fact sheet. Auckland, New Zealand: New Zealand Resident Doctorsâ Association. âââ. 2008b. What is NZRDA: About us. http://www.nzrda.org.nz/Site/About_Us/What_ is_NZRDA.ashx (accessed June 5, 2008). OâBrien, M., J. Brown, I. Ryland, N. Shaw, T. Chapman, R. Gillies, and D. Graham. 2006. Exploring the views of second-year Foundation Programme doctors and their educational supervisors during a deanery-wide pilot Foundation Programme. Postgraduate Medical Journal 82(974):813-816. PAIRO-CAHO. 2005. 2005-2008 PAIRO-CAHO agreement, edited by the Professional As- sociation of Internes and Residents of Ontario (PAIRO) and the Council of Academic Hospitals of Ontario (CAHO). Ontario, Canada. PAR-BC (Professional Association of Residents of British Columbia). 2008. Collective agree- ment & constitution: Scheduling. http://www.par-bc.org/collectiveagreement.php/20 (ac- cessed May 28, 2008). PARIM (Professional Association of Residents and Interns of Manitoba). 2008. Collective agreement: Article 14âduty hours. http://www.parim.org/50 (accessed February 22, 2008). Powell, D. 2004. Key issues facing Resident Medical Officers (RMOs). New Zealand Medical Journal 117(1204). Productivity Commission. 2005. Australiaâs health workforce: Research report. Canberra, Australia: Commonwealth of Australia. Rawnsley, A., K. Hurst, and M. Robinson. 2004. Clinical and education implications of shift work. Medical Teacher 26(1):71-73. Royal Australasian College of Physicians. 2007. Requirements for physician training: New Zea- land 2008 handbook. Wellington, NZ: The Royal Australasian College of Physicians. Royal College of Physicians. 2002. European working time directive: European dimensions. http://www.rcplondon.ac.uk/news/statements/doc_ewtd_european.asp (accessed June 10, 2008). âââ. 2006. Designing safer rotas for junior doctors in the 48-hour week. Prepared for on behalf of a multidisciplinary working group by N. Horrocks and R. Pounder. London, England: Royal College of Physicians of London. Royal College of Surgeons of England. 2007a. Rota planning: Guidance from the working time directive working party. London, England: Professional Standards and Regulation, Royal College of Surgeons of England. âââ. 2007b. Safe shift working for surgeons in training: Revised policy statement from the working time directive working party. London, England: Professional Standards and Regulation, Royal College of Surgeons of England. âââ. 2008. Lessons from surgery. Presentation by Bernard Ribeiro to Committee on Op- timizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Scallan, S. 2003. Education and the working patterns of junior doctors in the UK: A review of the literature. Medical Education 37(10):907-912. SchiÃ¸ler, T., H. Lipczak, B. L. Pedersen, T. S. Mogensen, K. B. Bech, A. Stockmarr, A. R. Svenning, and A. FrÃ¸lich. 2001. Incidence of adverse events in hospitalized patients. The Danish Adverse Event Study (DAES). Ugeskrift for Laeger 163(39):5378. Sheldon, T. 2004. Pressure mounts over European working time directive. BMJ (Clinical Research Ed.) 328(7445):911.
APPENDIX C 361 Skinner, C. A. 2006. Re-inventing medical work and training: A view from generation x. Medical Journal of Australia 185(1):35-36. Thomas, E. J., D. M. Studdert, H. R. Burstin, E. J. Orav, T. Zeena, E. J. Williams, K. M. Howard, P. C. Weiler, and T. A. Brennan. 2000a. Incidence and types of adverse events and negligent care in Utah and Colorado. Medical Care 38(3):261-271. Thomas, E. J., D. M. Studdert, W. B. Runciman, R. K. Webb, E. J. Sexton, R. M. Wilson, R. W. Gibberd, B. T. Harrison, and T. A. Brennan. 2000b. A comparison of iatrogenic injury studies in Australia and the USA I: Context, methods, casemix, population, patient and hospital characteristics. International Journal for Quality in Health Care 12(5):371-378. Thorne, L., S. Burn, S. Shaw, B. Arvin, and R. Bradford. 2006. Neurosurgical trainees opera- tive experience before and after introduction of the new deal for junior doctors. British Journal of Neurosurgery 20(1):31-35. Vincent, C., G. Neale, and M. Woloshynowych. 2001. Adverse events in British hospitals: Preliminary retrospective record review. BMJ 322(7285):517-519. West, D. 2007. European working time directive implementation and cardiothoracic training: Larger centers may optimise training [comment]. European Journal of Cardio-thoracic Surgery 31(5):958. West, D., M. Codispoti, and T. Graham. 2007. The European Working Time Directive and training in cardiothoracic surgery in the United Kingdom: A report of the specialty ad- visory board in cardiothoracic surgery of the Royal College of Surgeons of Edinburgh. The Surgeon 5(2):81-85. Wilson, R. M., W. B. Runciman, R. W. Gibberd, B. T. Harrison, L. Newby, and J. D. H Â amilton. 1995. The quality in Australian health care study. Medical Journal of ÂAustralia 163(9):458-471. Woodrow, S. I., C. Segouin, J. Armbruster, S. J. Hamstra, and B. Hodges. 2006. Duty hours reforms in the United States, France, and Canada: Is it time to refocus our attention on education? Academic Medicine 81(12):1045-1051. Workforce Taskforce. 2007. Reshaping medical education and training to meet the challenges of the 21st century: A report to the Ministers of Health and for tertiary education from the Workforce Taskforce. Wellington, New Zealand: Ministry of Health. World Health Organization. 2007. World directory of medical schools: Final updates through December 2007. http://www.who.int/hrh/wdms/en/ (accessed September 19, 2008).