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.
2 Current Duty Hours and Monitoring Adherence In 2003, the Accreditation Council for Graduate Medical Education (ACGME) established a maximum but not required 80-hour workweek for residents, averaged over 4 weeks. The best available national data show that first-year residents across various specialties reported working 66.6 hours a week on average during 2003-2004. Hours of work tend to be higher for residents in their first year of training, during rotations with overnight call responsibilities, and for certain specialty programs (e.g., general surgery). Lack of adherence to the ACGME limits remains an is- sue in some programs, particularly with respect to the limitations on the number of consecutive hours a resident can work and requirements for adequate time off for recovery sleep and personal activities. As a result, residents remain susceptible to acute and chronic sleep deprivation, despite the intent of the 2003 duty hour limits to prevent fatigue. The commit- tee found the need to enhance monitoring of and adherence to duty hour rules. These changes should include (1) unannounced audits of duty hours by ACGME and strengthened whistle-blower protections at the local and national levels for better detection and resolution of violations; (2) over- sight by the Centers for Medicare and Medicaid Services (CMS) and the Joint Commission of duty hours in relation to patient safety and quality improvement; and (3) evaluation of the hours worked and the frequency of violations, including documentation of their causes (such as patient needs or unnecessary workload) by specialty. Residency is a unique career stage, a time to focus exclusively on train- ing and professional development. Residents experience a sense of growing every day as they become more competent while performing important, 47
48 RESIDENT DUTY HOURS meaningful work. The design of this training ideally maximizes rich edu- cational experiences and has traditionally meant working long hours. In settings where camaraderie with faculty and fellow residents is strong, trainees experience a healthy learning environment and exponential growth as physicians, not just the fatigue associated with working long hours. However, concerns have been raised about the quality of care delivered by fatigued residents and the humane treatment of residents themselves. These concerns have led to duty hour reforms. The committeeâs task has been to consider whether the current Ac- creditation Council for Graduate Medical Education (ACGME) duty hour limits are optimal for resident safety, patient safety, and education. Before determining this, it was important to understand how the current limits came to be, whether implementation of any of the limits has been problem- atic, how the limits have been monitored, and whether data on adherence to these limits are reliable. Additionally, the committee reviewed duty hour limits in safety-sensitive transportation industries, including the processes for establishing regulations in these industries. Based on this appraisal, the committee makes recommendations in this chapter for future duty hour monitoring practices at the local and national levels. SETTING DUTY HOUR LIMITS Resident work schedules first received widespread public attention in 1984 after the death of Libby Zion, an 18-year-old woman treated in the emergency department of a New York hospital for fever and earache. Her family charged that she died due to the poor care by overworked and un- dersupervised medical residents (no attending physician saw her, although one was consulted by phone). Upon investigation of her death, a grand jury in 1986 exonerated the doctors involved in the case (Ludmerer, 1999) and, instead, faulted the broader system of graduate medical education (New York Supreme Court, 1986): the underlying causes of the medical deficient care and treatment in this case might be prevalent in other Level One hospitals .ââ.ââ. the most serious deficiencies can be traced to the practice of permitting inexperienced physi- cians to staff emergency rooms and allowing interns and junior residents to practice medicine without supervision.ââ.ââ.ââ.ââMoreover, those patients who are admitted into these hospitals for treatment are often cared for by interns and residents who are not required to have contemporaneous, in person consultations with senior physicians before they initiate a course of treatment. As a consequence, the most seriously ill patients may be cared for by the most inexperienced physicians. The grand jury concluded that the long duty hours of residents at that time were counterproductive to patient care and to resident learning, and
current DUTY HOURS AND MONITORING ADHERENCE 49 it recommended limitations on resident duty hours and enhanced super- vision (Bell, 2003). During further court proceedings in 1995 related to Libby Zionâs death, questions continued to be raised about whether duty hours and supervision were the only contributing factors (Andrews, 1995; Douglas, 1995). Regardless of the cause, concerns about the circumstances surrounding her death led to changes in resident scheduling in New York State and ultimately throughout the country. Table 2-1 outlines changes in duty hour limits over the past 25 years. The New York State Commissioner of Health appointed the New York State Ad Hoc Advisory Committee on Emergency Services, commonly re- TABLE 2-1â Evolution of Duty Hour Limits 1981 ACGME begins to require âtime for restâ 1984 Patient (Libby Zion) dies in a New York teaching hospital, her father claiming she died due to poor care delivered by overworked, fatigued, and inadequately supervised residents 1986 Responding to Zionâs death, a grand jury recommends patient care improvements including limits on resident duty hours in New York State 1987 The Bell Commission recommends specific limits on resident duty hours and increased supervision of their work 1989 New York State sets duty hour requirementsa 1989-1990 ACGME sets an 80-hour limit in several specialties (internal medicine, dermatology, ophthalmology, and preventive medicine) and limits in-house call to every third night with 1 day off in 7 in all specialties, on average 1998 Surprise inspections in 12 New York hospitals find extensive violations of New York duty hour limits 2001 Public Citizen, Committee of Interns and Residents of the SEIU, and American Medical Student Association petition OSHA; Representative Conyers introduces the Patient and Physician Safety and Protection Act of 2001 (not enacted) 2001-2002 ACGME Work Group on Resident Duty Hours and the Learning Environment develops common duty hour limits for all specialties 2003 ACGME requires current common duty hour limits; Representative Conyers and Senator Corzine introduce the Patient and Physician Safety and Protection Act of 2003 (not enacted) 2005 Representative Conyers and Senator Corzine introduce the Patient and Physician Safety and Protection Act of 2005 (not enacted) NOTE: OSHA = Occupational Safety and Health Administration; SEIU = Service Employees International Union. aNew York State Laws and Regulations: Title 10 NYCRR, Â§ 405.4 (1998). SOURCES: ACGME, 2007b; GovTrack.us, 2005a,b; IPRO, 2007a; Steinbrook, 2002.
50 RESIDENT DUTY HOURS ferred to as the Bell Commission, to investigate the conditions of residency. The Bell Commission, named after its chair Bertrand Bell, recommended changes in graduate medical education, including limiting duty hours and improving supervision of residents. In 1989, as a result of the recommenda- tions, New York State implemented rules limiting total resident duty hours per week and the length of extended duty periods, as well as providing for time off from work to address resident fatigue (Table 2-2). Table 2-2 compares New York State rules with subsequent proposals (Public Citizen petition and legislative proposals) and the 2003 ACGME limits established for all residencies nationwide. Attempts to limit resident duty hours through regulatory or legisla- tive bodies separate from the medical establishment have repeatedly been stopped both in the U.S. Congress and in state legislatures other than New York and Puerto Rico (IPRO, 2007a). The ACGME acknowledges that its âinitiative to institute common minimum standards for duty hours unfolded against a political backdrop in which groups pursued federal intervention to regulate resident hoursâ (ACGME, 2003, p. 1; Steinbrook, 2002). Previ- ously in 2001, Public Citizen petitioned the U.S. Occupational Safety and Health Administration (OSHA) to establish national duty hour limits for residents by arguing that long duty hours are physically and mentally harm- ful to medical residents (Table 2-2) (Public Citizen, 2001). OSHA turned down the groupâs petition in 2002 saying, âBecause the issues involved with medical resident hours go well beyond job safety and affect hospital patient safety, because other knowledgeable groups are taking action to work on this problem .ââ.ââ.ââ, the Agency has decided to deny your petitionâ (Depart- ment of Labor, 2002). Several bills to legislate duty hour limits have come before the U.S. Congress since 2001âall have been referred to committee, and none has ever been called up for a vote. The most recent bills, introduced by Rep- resentative Conyers and Senator Corzine, are the Patient and Physician Safety and Protection Act of 2005 (H.R. 1228 and S. 1297, respectively). This legislation would have amended title XVIII (Medicare) of the Social Security Act and authorized the U.S. Secretary of Health and Human Services to establish regulations on resident duty hours, supervision, and whistle-blower protections (Table 2-2). The bills would have authorized funding for training facilities to help meet regulations and required fines for nonadherence (GovTrack.us, 2005a,b). â New York State Laws and Regulations: Title 10 NYCRR, Â§ 405.4 (1998).
current DUTY HOURS AND MONITORING ADHERENCE 51 ACGME 2003 Duty Hour Rules Beginning July 1, 2003, the ACGME required that duty hours for resi- dents âmustâ meet the following provisions: â¢ An 80-hour workweek averaged over a period of 4 weeks, includ- ing all in-house calls; â¢ 1 day in 7 off without any educational or clinical duties or call, averaged over 4 weeks; â¢ In-house overnight call frequency of no more than every third night, averaged over 4 weeks; â¢ A maximum onsite duty period of 24 hours with up to 6 additional hours available for didactic education as well as transfer of patients (residents may not take any new patient after 24 hours on duty); and â¢ Although at-home or pager calls do not count toward the every third night or 24 + 6 hour limit, they âmustâ not be utilized so frequently that the resident is unable to rest or to have a reasonable amount of personal time (ACGME, 2007b). Additionally, residents âshouldâ have 10 hours off between shifts for ad- equate rest (ACGME, 2003). The ACGME used âmustâ to designate man- datory requirements (ACGME, 2008a). These duty hour limits apply to trainees in their fellowship years as well. The 80-hour workweek specified in the New York and ACGME rules was not empirically determined. The Bell Commission put forth the 80-hour week using the following heuristic: âThere are 168 hours in a week. It is rea- sonable for residents to work a 10-hour day for 5 days a week. It is humane for people to work every fourth night. If you subtract the 50-hour week .ââ.ââ. from 168 hours, you end up with 118 hours. If you then divide 118 by 4 (every fourth night), it equals 30. If you then add 50 to 30, that equals an 80-hour weekâ (Bell, 2003, p. 40). Similarly, Dr. Paul Friedmann, co-chair of the later ACGMEâs duty hours working group said that 80 hours is âa num- ber with some general acceptance, without much scientific underpinningâ (Steinbrook, 2002, p. 1298). The ACGME adopted the 80-hour limit to help protect against the sleep loss associated with working long hours (ACGME, 2003). ACGME indicated that the +6 hours was added to the extended duty period to prevent residents from driving home at their circadian nadir and to provide time for learning activities and handover of patient information (ACGME, 2003).
52 RESIDENT DUTY HOURS TABLE 2-2â Comparison of Duty Hour Provisions New York Code 405 Rules Duty Hour Provisions (1989, updated in 1998)a Maximum hours per week 80 hours, averaged over 4 weeks Maximum shift length 24 hours + 3 hours for transitional activities Maximum in-hospital on-call Every third night, with averaging frequency Minimum time off between 8 hours scheduled shifts Mandatory time off duty 24 hours off per week, no averaging Emergency room limits 12-hour limits in hospitals with more than 15,000 unscheduled visits Whistle-blower protections Enforcement Civil penalties issued by the state; originally $2,000 per violation; in 2000 raised to $6,000 per item, plan of correction within 30 days; $25,000 penalty for noncompliance with correction plan; additional $50,000 penalty for subsequent noncompliance with correction plane Funding (for additional staff Initially, yes; to make up for loss of resident also funding for compliance monitoring duty time) a SOURCES:â New York State Laws and Regulations: Title 10 NYCRR, Â§ 405.4 (1998). bPublic Citizen, 2001. cACGME, 2003. dGovTrack.us, 2005a,b. eNew York State Department of Health, 2002.
current DUTY HOURS AND MONITORING ADHERENCE 53 Public Citizen Petition to OSHA ACGME Accreditation H.R. 1228 and S. 1297 (2001)b Standards (2003)c (2005)d 80 hours, no 80 hours, averaged over 4 80 hours, no averaging averaging weeks; 88 hours for select programs for a sound educational rationale 24 hours 24 hours + 6 hours for 24 hours + 3 hours for transitional activities transitional activities (exception for patient emergencies) Every third night, Every third night, with Every third night, no averaging averaging no averaging 10 hours 10 hours 10 hours 24 hours off per 24 hours off per week, averaged 24 hours off per week, one full week, no averaging over 4 weeks weekend off per month; no averaging 12-hour shift limit, at least 12 hours an equivalent period of time off between shifts; 60-hour workweek with additional 12 hours for education Yes Some confidentiality protection Yes in complaint procedure Civil penalties Potential loss of accreditation; Civil penalties, not to exceed sufficiently large plan of correction $100,000 per training to deter violations; program in a hospital, with unannounced corrective action plans to the inspections Secretary of Health and Human Services; public disclosure on a hospital and residency training program-specific basis No No Yes, to cover hospital incremental costs to comply with regulations
54 RESIDENT DUTY HOURS Definition of Terms ACGMEâs definition of graduate medical trainee duty hours includes all time spent in âclinical and academic activities related to the program, that is: patient care (both inpatient and outpatient), administrative duties relative to patient care, provision for transfer of patient care, time spent in-house during call activities, and scheduled activities, such as conferences. Duty hours do not include reading and preparation time spent away from the duty siteâ (ACGME, 2008a). The term extended duty period (also known as âlong callâ) is used in this report to refer to the 30-hour (24 + 6) maximum continuous duty period allowed under the 2003 limits (ACGME, 2008a). The term shift is applied to any other scheduled period of work, whether during the day, evening, or night. Residents may or may not have any time to sleep during extended duty periods, depending on how busy their service is and the presence or absence of mechanisms for distributing responsibilities to other residents in a ânight floatâ system or to a hospital- ist service who will admit new patients or respond to the needs of patients already in the hospital. On a day or night float schedule, residents are not assigned to a single service but float across services or teams to help with admissions and follow-up (PAIRO, 2008). Cross-coverage means being available to care for patients admitted by other residents when the resident who has had primary care responsibility for these patients is not at the hospital or is otherwise unavailable. Other definitions of terms are available in a glossary in Appendix D. Suggested Refinements to Duty Hours The committee heard testimony from organizations representing the graduate medical education community, which favored continuation of the current rules over any further reductions in duty hours (AAMC, 2008; ABNS, 2008; ACGME, 2007b, 2008c; ACS, 2008; AMA Resident/Fellow Section, 2007). Other speakers, however, pointed out that resident train- ing in different countries used much shorter workweeks and suggested that the committee consider shortening the total number of hours allowed (Landrigan, 2007; Public Citizen, 2007). Speakers from the surgical com- munity described implementation problems encountered in the United Kingdom as it has reduced hours substantially. Their perception was that fewer hours did not ensure sufficient operative experience to attain the level of competence required for independent practice (ABMS, 2008; ACS, 2008; Royal College of Surgeons of England, 2008). Appendix C contains an expanded discussion of the lessons from the international experience. The main objection to the 2003 rules raised in testimony concerned the extended duty period of 24 + 6 hours. Public Citizen, the Committee
current DUTY HOURS AND MONITORING ADHERENCE 55 of Interns and Residents (CIR) of Service Employees International Union (SEIU) Healthcare, and Drs. Landrigan and Czeisler advocated that resi- dents work no more than 16 hours straight, including time for transfer of patient care and resident education (American Medical Student Association, 2007; CIR/SEIU, 2007; Czeisler, 2007; Landrigan, 2007). The CIR/SEIU Healthcare also recommended that averaging of duty hours no longer be allowed in order to prevent large variations in the number of duty hours from week to week (CIR/SEIU, 2007). ADAPTING TO 2003 DUTY HOURs Have duty hours changed in response to 2003 duty hour limits? The best available evidence shows that mean hours have been reduced over time (Baldwin et al., 2003; Landrigan et al., 2006). Despite progress in reducing overall hours, residents and their residency programs do not always adhere to every aspect of the 2003 ACGME limits (ACGME, 2004; IPRO, 2007b; Landrigan et al., 2006). Certain elements of the rules are more problematic than others, and certain specialties have more problems with adherence than others. Change in Mean Duty Hours Mean hours are useful information but they alone do not capture the variation within and across institutions or specialties. Mean duty hours for first-year residents (interns) appear to have declined nationwide in response to duty hour reforms from approximately 83 hours per week in 1998-1999 to 66.6 hours after the 2003 limits. Table 2-3 shows the results from a national survey of a randomly drawn sample of residents (n = 3,493) in the 1998-1999 academic year, with interns from two-thirds of special- ties working more than 80 hours (83.3 hours) on average (Baldwin et al., 2003). Second-year residents averaged 76.2 hours per week, with one-third of specialties (all but one a surgical specialty) over 80 hours (Baldwin et al., 2003). Landrigan and colleagues (2006) looked at the duty hours of a national self-selected sample of first-year residents (n = 1,278) from at least 13 different specialties (e.g., internal medicine, pediatrics, psychiatry, general surgery, obstetrics-gynecology [OB/GYN]) using a monthly web- based survey to track duty hours. They found a decline from a mean of 70.7 hours before duty hour reform (2002-2003) to 66.6 hours in the first year of implementation (2003-2004) (Landrigan et al., 2006). The difference in pre-duty hour levels between Baldwinâs 83 and Landriganâs 70.7 hours may be due to many programs starting to transition to expected limits even before the ACGME limits became official, although it may also reflect
56 RESIDENT DUTY HOURS TABLE 2-3â Average Reported Weekly Work Hours and Percentage of PGY-1 and PGY-2 Residents Working Over Proposed 80-Hour Limit by Specialty, 1998-1999 National Survey PGY-1 PGY-2 Percent of Percent of Residents Residents Working Working Over 80- Over 80- Specialty Mean (SD) Hour Limit Mean (SD) Hour Limit Anesthesiology 78.1 (18.1) 44.1 77.7 (16.2) 42.3 Dermatology 59.9 (16.7) 6.9 Emergency medicine 80.1 (17.5) 41.2 71.0 (15.2) 14.3 Family practice 78.1 (16.1) 39.1 67.6 (17.1) 17.9 Internal medicine 83.7 (15.5) 52.4 77.1 (17.3) 35.6 Internal medicine/pediatrics 81.6 (13.9) 46.7 77.5 (16.5) 37.3 Neurological surgery 110.6 (14.4) 100.0 Neurology 82.4 (17.5) 41.0 Obstetrics/gynecology 90.5 (13.7) 69.5 90.8 (17.1) 71.1 Ophthalmology 72.4 (18.9) 23.1 Orthopedic surgery 94.5 (19.1) 75.9 93.8 (16.3) 70.7 Otolaryngology 88.6 (16.5) 57.7 Pathology 60.8 (16.2) 9.4 56.7 (11.2) 5.1 Pediatrics 81.3 (14.9) 43.6 78.1 (14.9) 36.2 Physical â 64.2 (18.6) 27.3 â medicine/rehabilitation Preventive medicine 58.7 (18.9) 0.0 Psychiatry 69.7 (16.5) 20.4 59.2 (14.9) 7.1 Radiation oncology 67.4 (10.8) 9.1 Radiology 66.5 (14.7) 20.4 Surgery (general) 102.0 (16.1) 89.0 105.7 (13.6) 93.3 Transitional year 80.1 (17.2) 38.2 Urology 98.5 (19.3) 66.7 Overall 83.0 (17.7) 49.7 76.2 (19.9) 35.1 NOTE: PGY-1 = postgraduate year 1; PGY-2 = postgraduate year 2. Reprinted with permission by Academic Medicine. Baldwin, D. C., Jr., S. R. Daugherty, R. Tsai, and M. J. Scotti, Jr. 2003. A national survey of residentsâ self-reported work hours: Thinking beyond specialty. Academic Medicine 78(11):1154-1163. methodological differences in sampling and recall period for hours worked. No other national study is available since the 2003 reforms on mean duty hours across such a spectrum of specialties; studies tend to be specialty or institution specific. Landrigan et al. (2006) also noted a significant decline in the mean length of extended duty periods from 32.1 to 29.9 hours for interns in multiple specialties. In a subsequent analysis of the change of pediatric residentsâ extended duty hours in three institutions, Landrigan et al. (2008)
current DUTY HOURS AND MONITORING ADHERENCE 57 found a significant decline in the mean from 29.3 Â± 3.2 hours before the 2003 rules were implemented to 28.5 Â± 2.4 hours afterward. Degree of Compliance with Current Limits The responsibility for ensuring that residents and institutions adhere to the 2003 duty hour standards falls to institutions themselves and the ACGME as part of its announced accreditation visits. The ACGME ac- creditation review occurs once every 1-5 years, or once every 3.7 years on average (ACGME, 2008b). After 10 years of experience with the 80-hour workweek, New York State mandated yearly, unannounced audits of its training institutions because surprise inspections in 1998 found widespread violations: 94 percent of residents in New York City and 37 percent of those throughout the rest of the state worked more than 85 hours per week, and 77 percent of surgical residents in New York City and 60 percent in the rest of the state worked more than 95 hours per week. Further 38 percent of all residents and 67 percent of surgical residents worked more than 24 consecutive hours. Emergency room residents were in compliance with their 12-consecutive-hour limits (DeBuono and Osten, 1998; Kennedy, 1998). To achieve annual review, the New York Department of Health has contracted with IPRO since 2001 to focus solely on monitoring duty hour compliance (IPRO, 2008). IPRO, an independent, not-for-profit healthcare and quality improvement organization, is the New York Medicare Quality Improvement Organization and its Medicaid Utilization Review/Quality Assurance Agent. All 124 New York teaching hospitals have monitoring by both IPRO and ACGME. Substantial Compliance Versus Counting Each Violation ACGME and IPRO both use what they term a âsubstantial compli- ance modelâ for monitoring duty hours. Using this model, a program will not receive a citation for single or isolated violations of duty hour rules. The violations need to be more systemic. ACGME examines programs more closely if 15 percent or more residents report violations on three or more standards through ACGMEâs yearly resident survey. IPRO also uses a threshold of 15 percent for resident nonadherence to a single duty hour rule before giving a citation. IPRO does not trigger a violation at precisely the hour limit. Instead 15 percent of residents would have to be over an 85- hour week average, for example, to trigger a violation or 15 percent would have to be over 28 hours on an extended duty period (New York limit is 24 + 3 hours). Various accreditation bodies (e.g., The Hague Accreditation â Personal communication, V. Wilbur, IPRO, June 9, 2008.
58 RESIDENT DUTY HOURS and Approval Standards) use such substantial compliance models (Council on Accreditation, 2008). Both the ACGME and IPRO monitor residents and fellows in all years of graduate medical training and annually report duty hour violations based on substantial compliance. Annually, ACGME surveys residents across the country by asking about their compliance with duty hours within the past week. When on site, ACGME looks back at a longer period, and IPRO reviews the previous 3 months of records. Their reported rates are not directly comparable because ACGME reports violations by specialty program and by residents while IPRO reports by facility. In academic year 2006-2007, 8.8 percent of the residency programs reviewed by ACGME received one or more citations for a violation of âany ruleâ related to duty hour limits (Table 2-4A) (ACGME, 2007a). In its 2006-2007 contract year, IPRO found that 16 percent of 124 New York teaching facilities had violated âany ruleâ (IPRO, 2007b). Since initiating duty hour monitoring, 46 percent of sponsoring facilities have received a duty hour citation from ACGME for one or more of their programs (ACGME, 2008c). Most research studies on duty hours include counts of every violation reported by residents and do not use a substantial compliance threshold. Thus, these studies can be expected to report higher levels of duty hour violations than the two monitoring organizations. One study found that for the first year of duty hour rule implementation (academic year [AY] 2003-2004), 83.6 percent of interns, 85.4 percent of residency programs, and 90.8 percent of teaching facilities had a violation of âany ruleâ during at least 1 month of the year (Landrigan et al., 2006). While it is clear from this work that duty hour violations are common, these especially high rates of nonadherence might be accounted for in several ways: respondents were first-year residents (interns typically work more hours than residents in other years of training), the data are from the first year that duty hour rules were implemented, the data were collected monthly covering 11 months rather than a more limited period examined during an accreditation or audit visit, and no threshold is applied. Violations of Specific Duty Hour Rules Certain duty hour rules have been more difficult to adhere to than others, chiefly those limiting duty periods to 30 hours and requiring 1 day off in 7. Tables 2-4A and 2-4B present data on violation rates for breaking âany ruleâ as well as each of the components. These data are most useful in determining which duty hour limits are most difficult to adhere to across multiple monitoring systems rather than determining which monitoring sys- tem discovers the most violations. The basis for reporting differs by source, with violation rates reported by âfacilityâ (IPRO, 2007b; Landrigan et al.,
current DUTY HOURS AND MONITORING ADHERENCE 59 TABLE 2-4Aâ Comparison of Reported Duty Hour Violation Rates by Facility and Program Pre- 2003- 2004- 2005- 2006- 2003a 2004 2005 2006 2007 Rule and Source of Data (%) (%) (%) (%) (%) Rates Reported for Facilities (% of facilities) Any rule violation â IPRO 46, 42 21.0 13.0 17.0 16.0 â Landrigan et al. (2006) 90.8 80-hour rule violation â IPRO 28, 10 â 0.0 â 2.0â 0.0 â 0.0 â Landrigan et al. (2006) 81.8 Extended continuous hour shift â IPRO 45, 32 15.0 â 5.0 18.0 12.0 â Landrigan et al. (2006) 79.8 Average 1 day off in 7 â IPRO 14, 5 â 3.0 â 4.0 â 8.0 10.0 â Landrigan et al. (2006) 63.6 Separation between shifts â IPRO 18, 13 â 4.0 â 2.0 â 1.0 â 1.0 Rates Reported for Programs (% of programs) Any rule violation â ACGMEb â 5.0 â 7.3 â 7.9 â 8.8 â Landrigan et al. (2006) 85.4 80-hour rule violation â ACGME (15% of residents) â 2.0 â ACGME (any resident) â 9.8 â Landrigan et al. (2006) 69.7 Extended continuous hour shift â ACGME (15% of residents) â 9.5 â ACGME (any resident) 20.9 â Landrigan et al. (2006) 70.2 Average 1 day off in 7 â ACGME (15% of residents) 33.4 â ACGME (any resident) 45.9 â Landrigan et al. (2006) 50.9 Separation between shifts â ACGME (15% of residents) 12.0 â ACGME (any resident) 27.5 Average call no more than every third ââ night â ACGME (15% of residents) 10.5 â ACGME (any resident) 25.0 aIPRO percentages in this column represent values for the first 2 years of its contract with New York State. bPercentage of ACGME site-reviewed programs; other ACGME data come from resident surveys. SOURCES: ACGME, 2008f; IPRO, 2007b; Landrigan et al., 2006.
60 RESIDENT DUTY HOURS TABLE 2-4Bâ Comparison of Reported Duty Hour Violation Rates by Residents Pre- 2003- 2004- 2005- 2003 2004 2005 2006 Rule and Source of Data (%) (%) (%) (%) Rates Reported for Residents (% of Residents) Violation of any rule by residents â ACGME â Landrigan et al. (2006) 83.6 Violation of the 80-hour rule â Baldwin et al. (2003), PGY-1s 49.7 â Baldwin et al. (2003), PGY-2s 35.1 â ACGME â 2.2 â Landrigan et al. (2006) 43.0 Extended continuous hour shift â ACGME â 7.6 â Landrigan et al. (2006) 67.4 Average 1 day off in 7 â ACGME 15.3 â Landrigan et al. (2006) 43.7 Separation between shifts â ACGME â 5.2 Average call no more than ââ every third night â ACGME â 5.7 SOURCES: ACGME, 2008f; Landrigan et al., 2006. 2006), âprogramâ (ACGME, 2004, 2006, 2007a; Landrigan et al., 2006), or âresidentâ (ACGME, 2004, 2006, 2007a; Baldwin et al., 2003; ÂLandrigan et al., 2006). In addition, ACGME released data for AY 2005-2006 on what the âprogramâ violation rate would be if it counted every resident-reported violation or applied a 15 percent threshold (Table 2-4A). Where data are available, Table 2-4A includes compliance rates for each component by âfacilityâ or âprogramâ and Table 2-4B by âresidents.â The 80-hour rule is more often adhered to than other limits, based on ACGME and IPRO reporting (Tables 2-4A and 2-4B). IPRO reports that it seldom finds excessive violations of the 80-hour rule any more in New York, although some flexibility is factored into its monitoring as noted ear- lier. Similarly, ACGME data show few program or resident violations of the 80-hour limit when the substantial compliance threshold of 15 percent is applied. Even when every resident is counted, the 9.8 percent program â Personal communication, V. Wilbur, IPRO, January 30, 2008.
current DUTY HOURS AND MONITORING ADHERENCE 61 violation rate is lower than the violation rates for other limits (Tables 2-4A and 2-4B). Only 2.2 percent of residents report violating the 80-hour rule in the past month on the ACGME resident survey, less than reported viola- tions for other limits (Table 2-4B). Landrigan et al. (2006) reported that 43 percent of first-year residents violated the 80-hour limit in the first year of implementation, and other institution-specific and specialty-specific reports, that also do not use a substantial compliance threshold, show variable rates of violation for the 80-hour week including some that were quite high (e.g., 16-94 percent of residents) (Carpenter et al., 2006; Jagsi et al., 2008; Lin et al., 2006; Reiter and Wong, 2005). Two studies that examined trends over time found decreasing levels of violations of the 80-hour week (Jagsi et al., 2008; Landrigan et al., 2006). The elements of the 2003 duty hour limits that provide opportunities for recovery from fatigue (days off per week, separation between shifts, limiting frequency of call) and limit consecutive hours on duty have had higher violation rates than the 80-hour limit by all measures (Tables 2-4A and 2-4B). In 2005-2006 according to ACGME data, 15.3 percent of resi- dents went without 4 days off in a month and 45.9 percent of programs had at least one resident without the required days off (Tables 2-4A and 2-4B). Even in closely monitored New York, 8-10 percent of facilities failed to always deliver the days off (Table 2-4A). Over time, the extended duty period has had the highest violation rate of any limit in New York facilities. ACGME found fewer, but still frequent, violations of the long duty period than of providing mandatory days off (Tables 2-4A and 2-4B). Other institution-specific or specialty-specific stud- ies show greater adherence problems relative to other limits than either the ACGME or the IPRO data indicate for the 30-hour extended duty period limit. For example, at one major training center in 2005, 85 percent of medical and general surgery residents reported violations of the 24 + 6 hour limit compared with 65 percent in violation of the 80-hour limit and 28 percent in violation of the 1 day off in 7 rule (Carpenter et al., 2006). When asked if they âalwaysâ comply with a rule, 50 percent of otolaryn- gology surgery residents surveyed across the country reported violating the 30-hour limit, 39.5 percent reported violating the 80-hour workweek (aver- aged over 4 weeks), 30 percent reported missing their 1 day off in 7, and 66 percent reported not having the proper separation between shifts (Reiter and Wong, 2005). Rates of violation of 30-hour extended duty periods were more likely on inpatient wards, intensive care rotations, and surgical rota- tions (Cull et al., 2006; IPRO, 2007b; Landrigan et al., 2006). Jagsi et al. (2008) reported improved compliance with the 30-hour limit after the 2003 limits in 76 programs at two institutions. Their analysis com- pared programs that made substantial reductions in their total workweek hours (reduced-hours programs) to those that did not (other programs).
62 RESIDENT DUTY HOURS Nonadherence with the extended duty period of 30 hours was reduced from 40.8 percent to 11.4 percent in the reduced-hours programs and from 12.6 percent to 5.0 percent for other programs. As noted earlier, Landrigan and colleagues have found the mean length of the extended duty period to decrease since the 2003 rules (Landrigan et al., 2006, 2008). Provisions for time off between shifts and call frequency have similar violation rates according to ACGME data. IPRO reports fewer problems maintaining the proper separation between shifts, perhaps because New York requires that time off âmust beâ 8 hours long, whereas ACGME rules recommend that time off âshould beâ 10 hours. This is illustrated by a report from one surgical program in New York reporting 98 percent adherence to IPRO regulations but only 88 percent adherence to ACGME limits (Goldstein et al., 2005). The use of âshouldâ with respect to this provision of ACGME rules while its other duty hour rules use âmustâ has caused confusion in the extent to which it must be followed. ACGME is conducting pilot tests on whether to change this rest requirement to âmust be 8 hoursâ (ACGME, 2008f). Compliance by Year of Training and Specialty Year of training and type of specialty both influence duty hours worked and the potential for violation of duty hour limits. These observations raise questions about whether the same limits should apply across all specialties or years of training. Compliance Across Specialties Before implementation of the 2003 duty hour limits, there was great variability in total hours worked by different specialties, but even when mean duty hours were less than 80, on average, for a given specialty, a large percentage of its residents would still have been in violation of that limit at some point in the year (with the exception of those in pathology, dermatol- ogy, psychiatry, and preventive medicine) (Table 2-3) (Baldwin et al., 2003). In 2002, program directors anticipated greater relative difficulty for surgical programs in adhering to the duty hour changes, and since 2003, surgery programs have had to reduce the duty hours of residents by 20 percent or more (e.g., neurosurgery second-year residents averaged 110.6 hours per week before 2003) to meet the 80-hour limit (Baldwin et al., 2003; Brotherton et al., 2002; Lieberman et al., 2005). Only half (49 percent) of surgical residents compared with three-fourths of medical residents (73 percent) were expected to be in compliance with the proposed 2003 limits (Lieberman et al., 2005). Studies on adherence since initiation of the 2003 rules vary whether
current DUTY HOURS AND MONITORING ADHERENCE 63 TABLE 2-5â Duty Hour Violations in New York State by Specialty (2006-2007) Violation of Violation of Any Duty Extended Duty Violation of Hour Limit Period of 24 + 3 1 Day Off in (%) Hours (%) 7 (%) Statewide 16 12 10 Specialty â Anesthesiology â0 â1 â2 â Emergency department â0 â0 â1 â Family practice â5 â4 â4 â Internal medicine â7 37 33 â OB/GYN â0 â5 â9 â Pediatrics â7 10 10 â Surgery 13 44 42 SOURCE: IPRO, 2007b, Tables 36 and 40. they identify worse duty hour violations among surgery programs than other specialties. IPRO data show that surgery, pediatrics, internal medi- cine, and family practice programs in FY 2006-2007 still had violations of some aspect of duty hour rules even after years of intensive monitoring, with surgery having almost twice the rate of nonadherence to âany ruleâ as the others (Table 2-5) (IPRO, 2007b). Landrigan and colleagues (2006) re- ported that programs with interns in internal medicine, pediatric programs, emergency medicine, and even psychiatry were equally likely to violate some duty hour rule as surgical programs for at least 1 month in the year. In New York, IPRO has found that internal medicine and surgery programs are almost equally noncompliant with the 24 + 3 hour extended duty period and the day off per week, at rates three or more times greater than the other specialties (Table 2-5). ACGME data over time show a high portion of surgical programs cited for extended duty period violations, but frequent citations are also found in family medicine, internal medicine, pe- diatrics, and transitional year programs (ACGME, 2004, 2006, 2007a). ACGME grants exemptions to the 80-hour workweek rule that allow up to a maximum workweek of 88 hours, after determining a program has a âsound educational rationale.â The number of programs receiving exemptions decreased from 68 in 2004-2005 to 40 in 2007-2008: 34 of the 40 in 2007-2008 are in neurological surgery, 6 in thoracic surgery, and 1 in general surgery (ACGME, 2007a). Thus, approximately 40 percent of neurosurgery programs have an exemption from the 80-hour-a-week limit. Neurological surgery programs still have programs cited, although ACGME citation reports do not distinguish between programs that have the 88-hour limit and those that do not (ACGME, 2007a).
64 RESIDENT DUTY HOURS Compliance by Year of Training First-year residents typically work longer hours than residents in other years, according to a 2007 ACGME resident survey and other studies (ACGME, 2008c; Baldwin et al., 2003; Carpenter et al., 2006). Interns have the most to learn and take longer to accomplish tasks while they are learning not only new medical information but how to work efficiently in the training environment. Learning how to manage oneâs time is an im- portant part of the first-year experience. Among interns, 9 percent are not âalways or usuallyâ in compliance with the 80-hour workweek, compared to 4.6 percent of residents in later years, and 11 percent of interns are not âalways or usuallyâ in compliance with the 24 + 6 shift length compared to 6 percent of residents in later years (ACGME, 2008f). Underreporting of Violations by Residents Residency programs monitor resident duty hours in a variety of ways, including by self-report on time sheets, telephone or computer log-in/log- outs in the hospital, badge readers at entries and exits of hospitals or park- ing garages, and personal monitoring by program directors to ensure that residents are not still on the floor outside of duty hour limits (Asad et al., 2006; Chao and Wallack, 2004; Goldstein et al., 2005; Landrigan et al., 2008). The degree of resident participation determines the quality of infor- mation from any system, and the ease of use and the degree of monitoring and enforcement activities all influence resident participation (Chang et al., 2006; Chao and Wallack, 2004). Testimony before the committee and other reports revealed that some residents have underreported the extent of their duty hours (American Medical Student Association, 2007; Arora et al., 2006; CIR/SEIU, 2007). Residents give multiple reasons for failing to report duty hours accurately. One rationale given for residents being disinclined to call ACGMEâs at- tention to duty hour violations is because this could lead to probationary accreditation for their program or loss of its accreditation altogether (AMA Resident/Fellow Section, 2007; CIR/SEIU, 2007). Residents perceive that graduation from a program that is on probation or without accreditation may hurt their chances when they seek employment. An American Medi- cal Association (AMA) survey of residents in 2005 revealed that 7 in 10 residents know how to report excessive duty hours but half would be un- comfortable actually reporting them (AMA Resident/Fellow Section, 2005). A subsequent AMA survey (2006) found that some residents experience in- timidation from attending physicians, senior residents, and fellows (AMA, 2006). These AMA survey data are not based on a representative response from residents and fellows. Such intimidation may be subtleâor at times
current DUTY HOURS AND MONITORING ADHERENCE 65 not so subtleâcultural expectations that long hours are necessary and should be borne without complaint (Arora et al., 2008; CIR/SEIU, 2007). Residents give other reasons for underreporting hours (Lamberg, 2002), such as wanting to experience more cases, not wanting to call at- tention to the fact that they do not work as fast as others, or knowing that everyone is overburdened by workload demands. Others attribute underreporting to the desire to be responsive to patient care needs and not wanting to be considered unprofessional if not following through in the care of a sick or unstable patient (Associated Press, 2003) or missing other patient-related activities such as meetings with families (Fletcher et al., 2008). A survey of internal medicine, pediatric, and general surgical residents at one major teaching center found that 85 percent exceeded duty hour limits at least once in the previous 3 months and 48 percent admit- ted underreporting their hours (Carpenter et al., 2006). Eighty percent of the residents noted that their concern for patient care was the greatest motivation in working the extra hours. Similar sentiments of not wanting to leave their patientsâ care to someone else were echoed in statements of residents interviewed in closed session by some members of the commit- tee (Resident Panel, 2008). Carpenter et al. (2006) expressed concern for the ethical dilemma facing residents: professionalism and care for their patients drive residents to exceed duty hour limits, but they then must act unprofessionally by falsely reporting their time to avoid negative conse- quences for their program. DUTY HOURS IN SELECTED INDUSTRIES This section presents a brief overview of the hours of service and the nature of rule making in certain safety-sensitive transportation industries, comparing the length of duty hours for residents and other workers. There is considerable variation among transport modes in federal work and rest requirements. A more extensive review of this topic can be found in an article by Rogers in the Institute of Medicine (IOM, 2003) report Keeping Patients Safe: Transforming the Work Environment of Nurses. Hours of Service per Week and per Shift The 80-hour workweek limit for residents, although long, is not sub- stantially different from those of some transportation industries that also have worker and public safety concerns. In examining the history of set- ting weekly limits on work time, one finds that the older the industry (e.g., trains and maritime vs. commercial trucking and aviation), the longer are its allowable hours in a workweek (Table 2-6). There is no limit on the total workweek for railroad conductors, and shipboard personnel on tank-
TABLE 2-6â Federally Mandated Work Hour Limitations for U.S. Transportation Modes and ACGME Duty Hours 66 Minimum Rest Weekly Limits of Single Minimum Rest Period Given Industry Limitations (h) Shift Duration (h) Between Shifts (h) Weekly (h) Regulatory Agency Enforcement Airline 30 h of flying 8 h of flight time 11 h of None Federal Aviation FAA pilotsa,b time in any 7 per 24 h continuous rest in Administration consecutive daysa; the 24 h prior to (FAA) 32 h of flying 9+ h of scheduled time in any 7 flight timec consecutive daysb Shipboard 84 h per week 15 h per every 24 None None personnel on h and 36 h per tankersd 72 h Railroad None 12 h 10 consecutive None Federal Currently hours are conductorse,f hours after a Railroad recorded by hand; 4 12 h shift and 8 Administration major railroads have consecutive hours (FRA) upgraded to electronic during the 24 h recordkeeping prior to any shift Long-haul 60 or 70 h duty 14 h on duty with 10 consecutive 34 h Federal Motor Drivers are required truck driversg time per 7- or 8- a maximum of 11 hours, drivers continuous rest Carrier Safety to record a log of day shift h spent driving with sleeper period prior to Administration hours for each 24 h berth must spend any 7- or 8-day (FMCSA) period, including a minimum of 8 working period record of the prior consecutive hours 7 days. Record can in berth and 2 be electronic or h in berth or handwritten, depends off-duty in any on motor carrierh,i combination
Medical 80 h per week 24 h + 6 h 10 consecutive One continuous ACGME ACGME residentsj,k averaged over 4 transition time hours 24 h rest period weeks (recommended per week but not required) aFederal Aviation Administration. 2008. Title 14: Aeronautics and Space, Part 121, Subpart Q, Â§ 121.471 (b):3 and Â§ 121.471 (a):3. This regulation applies to domestic flights that occur within the contiguous 48 states of the United States or the District of Columbia, or entirely within any State, territory, or possession of the United States. bFederal Aviation Administration. 2008. Title 14: Aeronautics and Space, Part 121, Subpart R, Â§ 121.481 (d) and Â§ 121.483 (a). This regulation applies to flights that occur between Alaska, Hawaii, or any territory or possession of the United States and any point outside these locations; or between any point within the 48 contiguous States of the United States or the District of Columbia and any point outside these locations; between any two points outside the United States. cIf a pilot receives less than 9 h of rest during a 24 h period, he or she must be compensated with at least 10 continuous hours of rest that begins no later than 24 h after the commencement of the reduced rest period. dUnited States Coast Guard Marine Safety Manual, Chapter 24, Section C. eFederal Railroad Administration. 2008. Title 49: Transportation, Subtitle VâRail programs, Part A, Chapter 211âhours of service, Â§ 21103. fFederal Railroad Administration, 2008. gFederal Motor Carrier Safety Administration. 2005. Federal motor carrier safety regulations 49 CFR, Part 395: Hours-of-service change. hFederal Motor Carrier Safety Administration. 2008. Part 395: Hours of service of drivers, Â§ 395.8. iFederal Motor Carrier Safety Administration. 2008. Part 395: Hours of service of drivers, Â§ 395.15. jACGME, 2007a. kNot a federal mandate; guidelines are set by ACGME. 67
68 RESIDENT DUTY HOURS ers have an 84-hour week. Only airline pilots have a significantly shorter workweek, but the limit is based on only one type of work (e.g., pilotsâ work on non-flying activities is not counted against the 30-hour weekly limit). Statutes and regulations often establish different duty hour limits for different types of work (e.g., truckers can work on non-driving activities for 3 hours beyond their daytime driving time limits). Paperwork is frequently done in all transportation modes outside duty hour limits. The allowable length of a single shift is considerably shorter in these other industries than the 24 + 6 hour extended duty period for residents (Table 2-6). Some medical and surgical specialties view the extended duty period as necessary to obtain unique patient care learning experi- ences. To preserve this aspect of residency training, while acknowledg- ing that residents have the same physiological needs for rest as other human beings, the committee examined ways to prevent and mitigate acute sleep deprivation when residents have extended duty periods (see Chapter 7). Need for Modernization Based on Sleep Science Although federally mandated hours of service still rely on a model that assumes the length of work time is the factor most relevant to fa- tigue, this is only one component of the relationship of fatigue to risk. Other factors can include the time of day work occurs in relation to oneâs circadian rhythm, the volume and intensity of work, and the amount of sleep obtained (Dinges, 1995; Drake et al., 2004; Folkard et al., 2005; Rosa, 2001; Van Dongen and Dinges, 2005). These factors are detailed in Chapter 7. Over the past century, federally mandated hours of service (HOS) for aviation, trucking, railroad, and marine workers have not kept pace with the extensive science on the biological causes, consequences, and prevention of fatigue, prompting the National Transportation Safety Board to urge re- peatedly that the relevant regulatory agencies set working hour limits based on fatigue research, circadian rhythms, and sleep and rest requirements (NTSB, 2007). Despite this, federal HOS in these industries have remained largely unchanged for decades and are seen as either âantiquatedâ (e.g., railroad; Boardman, 2007) or inadequate as prescriptive rules because they do not permit the operational flexibility increasingly required in modern systems (e.g., commercial aviation; Gilligan, 2007). Once promulgated as either statutes or regulations, hours of service in transportation modes have proven remarkably difficult to revise to incor- porate new scientific evidence on the biological causes of fatigue, due to lack of political consensus and legal challenges to even the smallest changes
current DUTY HOURS AND MONITORING ADHERENCE 69 (Boardman, 2007). As a result, federally mandated hours of service are often seen as a barrier to a modern evidenced-based approach to preventing fatigue in many industries. Moreover, it is increasingly recognized that even meeting the nominal requirements of current HOS rules is not sufficient to effectively manage fatigue. Additional efforts involving organizational com- mitment and allocation of resources for establishing and sustaining fatigue management are necessary (McCallum et al., 2003). Thus, the committee concludes that one goal for its recommendations is to combine scientifically based duty hour limits with adequate adherence. MONITORING DUTY HOURS The level of adherence to resident duty hour limits has raised ques- tions about the current approach to monitoring duty hours and whether the culture of expectation, if not overt intimidation, results in pressure on residents to work more than their assigned hours (AMA, 2006; Arora et al., 2008; CIR/SEIU, 2007). All hospitals, including teaching hospitals, are under pressure to increase revenue and manage their costs (Weissman et al., 2007). In a teaching environment, residents are relatively low-cost person- nel available to handle increasing admissions (AAMC, 2007; Kozak et al., 2006). Achieving the correct balance between providing service and meeting educational goals has been a long-standing issue since the first report on graduate medical education was issued in the 1940s (Ludmerer, 1999). ACGME Review of Adherence to Duty Hour Limits The ACGME metric is that a program must have at least substantial compliance with accreditation standards for institutions and programs, including duty hour limits. A program once cited for deficient educational practices or duty hour violations can remain accredited while the problems are remediated if the ACGME judges that these do not immediately jeop- ardize the overall performance of the program. In AY 2006-2007, ACGME issued 8,804 citations for the 2,589 programs under accreditation review; most citations (54 percent) were for educational deficiencies and only 2.9 percent were for duty hour compliance issues (ACGME, 2007a). Serious ed- ucational deficiencies might include the following: (1) less than 50 percent of an internal medicine residency programâs graduates pass the American Board of Internal Medicine exam on the first try, (2) the pediatric inpatient population lacks sufficient diversity and complexity for adequate training, â PublicCitizen, et al., v. Federal Motor Carrier Safety Administration, 374 F.3d 1209, 362 (U.S. App. D.C. 384).
70 RESIDENT DUTY HOURS or (3) residents perform an insufficient or excessive number of orthopedic procedures (ACGME, 2008b,e). The sanctions that ACGME currently applies involve program proba- tion or withdrawal of accreditation. The adverse action rate for ACGME is about 8 percent (this includes probationary accreditation, withdrawal of accreditation, withholding of requests for new program accreditation, and forced reductions in resident complement); most actions result in programs being placed on probation. During AY 2007-2008, 10 programs had their accreditation withdrawn or were on track for expedited withdrawal; 3 of these had duty hour violations. Training facilities must maintain systems and documentation (e.g., rota- tion schedules, call rosters, sign-in/sign-out systems) to assure ACGME that staff hours are under the required limits. ACGME looks at this documen- tation during scheduled onsite visits and analyzes responses to its annual resident survey to see what residents report about compliance with duty hours, including whether there is any undue pressure to work more than the required hours. Typically, the annual survey involves half of the residents in the country. ACGME interviews another 12,000 or so residents during site visits (ACGME, 2008c). Complaints about training program quality including violations of duty hours can be made to ACGME, but those with a complaint are directed before filing a formal complaint to discuss the is- sues with the local program director or the supervising institutional official responsible for all graduate medical education (ACGME, 2004, 2007a). If the issue detailed in the complaint is egregious enough, there will be an immediate site visit (ACGME, 2007c). Institutions pay ACGME $2,750-$3,500 per residency program for the accreditation review (including assessment of duty hours). If an academic medical institution has 20 programs, it could cost $54,000 to $70,000 (ACGME, 2008d). New York State Monitoring Unlike ACGMEâs scheduled accreditation visits, IPRO reviews are un- announced. The review team examines 3 months of schedules and further validates that residents are staying within the limits by collecting about 9 days of detailed data through interviews, direct observation and review of chart notes, operating room logs, clinic records, and test orders. This in- tensive audit is designed to ensure that residents are not coming in before their scheduled hours to prepare patient data for rounds or procedures, or staying after their hours. For example, a facility schedule may say that a â Personal communication, Ingrid Philibert, ACGME, July 22, 2008. â Personal communication, Ingrid Philibert, ACGME, August 2, 2008.
current DUTY HOURS AND MONITORING ADHERENCE 71 resident starts at 7 a.m. but he or she actually comes in at 5 a.m. to pre- pare for rounds, or the schedule says the resident is off on Saturday but other documents show the resident was in the operating room that day. An IPRO team will be onsite for 1 to 2 weeks. The IPRO overall sample size of a facilityâs resident trainees is 50-60 percent, with a nearly 100 percent sample in areas that have proven to have more violations over years of reviewâsurgery and intensive care units. After the audit, IPRO presents facilities with a detailed exit report list- ing any deficiencies to make sure violations are clearly documented by time and dates. Facilities must submit a plan for correction within 30 days of being notified of deficiencies. IPRO tries to work with facilities to resolve issues before the state gets to the stage of fining the hospital. New York State imposes fines for residency programs that persist in noncompliance ($6,000 for a first offense escalating to $50,000 for a third offense). When a deficiency persists, facilities are not able to receive any certificate-of-need approvals from the state (e.g., to expand capacity). In New York, the state, not the institution, pays for the reviews of duty hour compliance. For the IPRO contract year 2008-2009, the cost for duty hour review alone is $2.9 million annually, averaging about $24,000 for each of the 124 hospitals under review. New York State has approximately 15 percent of the graduate medical trainees in the country (IPRO, 2007b). For 2006-2007, 16 percent of the facilities in New York State had violations of some duty hour rule (IPRO, 2007b). Future Approach to Monitoring Hours The committee concludes that violations of duty hours are frequent and underreported and that more intensified monitoring is necessary im- mediately to ensure adherence. Achieving adherence to existing duty hour rules is an established and essential first step to which stakeholders have al- ready agreed. The next step would be adherence to the committeeâs recom- mended duty hour parameters. The committee considered carefully whether ACGME should remain the body that sets and monitors duty hour limits, or whether an alternative organization and approach are warranted. The goals of the committee are to have an effective monitoring process under a substantial compliance model and documentation of when and why viola- tions occur in order to guide institutions in reconfiguring their scheduling and workload and provide a better understanding of the circumstances when exceptions to duty hour limits might be permitted. The main monitoring alternatives include: (1) the status quo, con- Personal communication, V. Wilbur, IPRO, January 30, 2008. â Personal communication, V. Wilbur, IPRO, January 30, 2008.
72 RESIDENT DUTY HOURS tinuing with the ACGME and its current approach, (2) a new agency to conduct the monitoring, or (3) the ACGME with changes. The committee concluded that neither of the first two options was acceptable. The commit- tee preferred the third alternative, strengthening the ACGME procedures and providing additional oversight and evaluation by other organizations to ensure that duty hours are considered in the context of quality improve- ment and patient safety. Below the committee discusses the advantages and disadvantages of each option. Alternative 1: The Status Quo The ACGMEâs approach to monitoring duty hours through site visits and national surveys of residents is described earlier in this chapter. Sev- eral advantages have been cited supporting the continuation of ACGMEâs current monitoring procedures. ACGME would be able to respond more easily and quickly than a new organization to implement the committeeâs recommendations since it already has a monitoring process in place, trained field staff, and relationships with all the training institutions and programs. It would not require new legislation. Also, ACGME could adapt readily if future scientific evidence prompts fine tuning or adjustments to the duty hours that the committee now recommends without the time needed to enact legislation and develop governmental regulations. ACGMEâs current monitoring of duty hours is a relatively inexpen- sive add-on because it is embedded in its overall accreditation and survey processes, onsite monitoring happens infrequently for individual residency programs, and it uses volunteers to a large extent, a financial advantage. Additionally, both the ACGME and the AAMC advocate a continued role for ACGME in monitoring as well as establishing duty hours, and train- ing institutions seem more comfortable with the status quo, also (AAMC, 2008; ACGME, 2008c). Since ACGME reviews the quality of the residency programs seeking their accreditation, it can readily integrate duty hour compliance data with assessments of educational programs. For example, when ACGME was reviewing a surgical program that wanted to expand the size of its training program, it discovered duty hour violations and would not allow expansion because of a pattern of such violations (Kowalczyk, 2008). Additionally, ACGME presented to the committee national data on the positive correla- tion between substantial violations of duty hours and other undesirable educational program characteristics (problems with faculty, teachers, resi- dent intimidation, excessive service obligations) as reported by residents (p = .0001). Residents in these 115 outlier programs (3.8 percent) report that they are less likely to participate in scholarly activities (38 percent vs. 58 percent) and more likely to be required to provide support services (45
current DUTY HOURS AND MONITORING ADHERENCE 73 percent vs. 22 percent). Resident survey data from previous years show that 91 percent of programs with educational program citations are also in the most noncompliant quartile for duty hours (ACGME, 2008c). ACGME now uses a substantial compliance threshold for assessing adherence to duty hour limits, so that no program is punished for isolated, individual incidents when there is not a pattern of abuse. This indicates an appropriate effort to focus monitoring attention on the more serious cases and an attempt to avoid unfairly punishing programs for isolated events. There are disadvantages associated with continued ACGME monitor- ing. Its monitoring processes have not been effective in bringing adherence to the duty hour limits that have been in place since 2003. Suboptimal adherence to current limits means their expected positive effect on resident fatigue and patient safety may be less than anticipated to date, making it difficult to assess the national impact of the 2003 duty hour rule changes on patient outcomes, as discussed in Chapter 6. ACGME assurances that reasonable duty hour limits have been set and are being followed and that ACGME can detect the full extent of viola- tions have been met with skepticism. Violation data from multiple other sources find higher levels of violations although there are methodological differences in reporting and timeframe that may account for some of the differences. Some groups suggest the membership of ACGME has been slow to accept duty hour limits and that some members may have a conflict of interest in enforcing limits on resident duty hours because of the costs of replacing resident labor with other personnel (CIR/SEIU, 2007; Sleep- deprived doctors, 2002). The long time between accreditation visits (e.g., 3.35 years even for programs with citations) (ACGME, 2008c) leaves opportunity for duty hour violations to escalate between visits. The current average time between visits can be longer than a complete residency period for some trainees. When the monitoring visit does occur, it is announced by ACGME and expected by the training institution, which also diminishes its value in un- covering problems. The current ACGME procedures for residents to report violations of the duty hour limits and undue pressure to work beyond the limits are a deterrent to whistle-blowing since the residents are expected to report through their residency program director first, before taking a complaint to the ACGME. Residents are concerned that the process will identify them to their program directors or senior colleagues, whose recommenda- â American Board of Medical Specialties, American Hospital Association, American Medi- cal Association, Association of American Medical Colleges, Council of Medical Specialty Societies.
74 RESIDENT DUTY HOURS tions can determine a residentâs opportunities for fellowship training or employment. Current ACGME monitoring data and studies that simply report vio- lation rates are insufficient for policy purposes. They do not provide a complete picture of actual hours worked and when exceptions to the rules might be necessary for educational or patient safety reasons. Aside from limited national data on mean hours worked in a week and consecutively by interns in 2003-2004, the committee does not know for certain how many hours over or under the time limits residents are working nationally and by specialty, or the frequency of different reasons that might push resi- dents over the time limits (e.g., unstable patient, lack of system supports, workload). Such data from a nationally representative sample of institu- tions and specialty programs would have helped the committee determine whether there are other reasonable adjustments to duty hours that would help achieve the training goals of each specialty and provide safe working conditions for residents and patients, beyond the committeeâs recommenda- tions in Chapter 7. The status quo is unacceptable to the committee, although ACGME is an attractive option because it links the education and duty hour policy development and monitoring. Many of the shortcomings of the ACGME monitoring process could be corrected with some additional effort and re- sources. Also, neutral organizations could provide oversight of the ACGME process to provide assurance to the public, patients, and residents. Alternative 2: A New Organization An organization unrelated to ACGME, such as a government or an independent agency, could take over the responsibility for duty hours moni- toring. This might take the form of an existing organization new to resident hours monitoring, a newly formed and purpose built organization, or one with experience monitoring duty hours at the state level but not nationally. One existing organization that fits the latter category by already performing duty hour monitoring is IPRO in New York State. Its monitoring processes are discussed earlier in this chapter. IPRO has the expertise to perform duty hour monitoring as demonstrated over their multiple years of experience. The advantage of having a different organization take on the functions of duty hour monitoring currently performed by ACGME would be that it could have a clean slate without the perceived biases of ACGME. There would be disadvantages to using a new organization or even having IPRO take on the function nationwide. These include a major expense to create any new agency from scratch, and whether a new agency or an expanded IPRO, major delays because of the need to get legislation or some other acceptable authority to delegate the responsibility to conduct such monitor-
current DUTY HOURS AND MONITORING ADHERENCE 75 ing and determine a way to provide the necessary funds. There would be major disruptions to all parties while new working relationships are estab- lished. The new agency would need to establish credibility with the public, Congress, physicians, and residents. Another drawback of public agencies would be the difficulty of adapting quickly to changing circumstances, as evidenced by historical impediments to updating existing legislation and regulations on hours of service requirements in the transportation arena to use newer scientific evidence (see discussion earlier in this chapter) ( Â Boardman, 2007; Gilligan, 2007; NTSB, 2007). The committee decided that the expense and delays involved with creat- ing a new organization were unacceptable. Among existing organizations, OSHA was an obvious option because it has responsibility for enforcing work hours in other industries. However, when Public Citizen requested OSHA to set duty hour limits for residents in 2001, it declined in favor of ACGME because the issues involved patient as well as worker safety and because others were taking action on duty hours (Department of Labor, 2002). The committee also considered whether IPRO or other QIOs (Quality Improvement Organizations) could fulfill the monitoring role in a man- ner that would be acceptable to all the involved parties. The advantage of IPRO fulfilling this function nationwide would be its expertise in duty hour monitoring and providing education to facilities on how to better achieve compliance. The main disadvantage would be that duty hour monitoring would be separate from review of the educational program. Additionally, the IPRO approach is quite expensive per institution ($24,000 annually covering all programs in an institution) within New York State; more ex- tensive travel requirements likely would mean even higher costs even if they could develop sufficient staff capacity to perform the function. Currently, IPROâs duty hour monitoring function falls under state authority and other than in New York and Puerto Rico there has been limited interest in passing duty hour limits at the state level. This implies it might not be a priority for funding in all states. In addition, New York State training facilities still have significant levels of violations despite years of intensive monitoring by IPRO, as noted earlier in this chapter. Possibly all QIOs or a selected few could have duty hours monitoring added to their portfolio of duties in the contracts they negotiate with CMS, but they would need time to develop staff expertise and procedures since such monitoring is not consistent with their other responsibilities and they would also need to establish working relationships with all the residency programs. The committee did not find any of these options preferable to the third alternative, a change in ACGME practices along with oversight and evaluation.
76 RESIDENT DUTY HOURS Alternative 3: ACGME with Changes The committee recommendation is to retain ACGMEâs current role in establishing duty hour limits and monitoring, in part because of the impor- tance of maintaining the link between the residency programsâ quality of education and duty hour compliance. It is essential to design educational programs in concert with duty hour schedules. The committee urges the ACGME to foster not only changes in scheduling and staffing patterns in response to duty hours but innovations in education and ways to measure competency while ensuring patient safety. Chapter 4 discusses educational considerations associated with duty hours in more depth. The changes to the ACGME monitoring process that are built into Alternative 3 are designed to correct some of the shortcomings identified in the discussion above. The committee recommends: â¢ Strengthening the ACGME monitoring process. Increasing the fre- quency of duty hour audits from the more than 3 year interval for programs with citations and making unannounced visits would allow observation of operations under normal circumstances with- out advance preparation by the facility for a visit. This expansion of monitoring would require ACGME to raise additional funds to cover the review costs whether using its own staff, voluntary reviewers, or contractors for these additional reviews of duty hour adherence. The costs might be borne by the institutions to be au- dited through additional visit fees and by organizational members of ACGME. Being mindful of the potential costs, the committee does not expect that unannounced visits would be needed yearly to every institution as New York State requires. Currently, ACGME makes separate visits to institutions for each program review; a team on site for one specialty programâs accreditation review could build in an unannounced look at some other programâs scheduling practices and compliance in order to minimize transportation costs incurred for more frequent reviews. Given that institutions often have 30 or more residency programs, there would be numerous opportunities for unannounced visits to one or more programs. â¢ Creating robust whistle-blower protections and alternative viola- tions reporting procedures. Encouraging residents (1) to complain about duty hour violations directly to the ACGME without first requiring them to go through their program director and the des- ignated institutional officer for graduate medical education or (2) to complain to their local institutional compliance office could help address residentsâ concerns that by reporting onerous working
current DUTY HOURS AND MONITORING ADHERENCE 77 conditions they place their own career in jeopardy. Institutional compliance offices have a broader role than just graduate medical education and exist to ensure compliance with laws, regulations, and policies that govern medical facility operations (e.g., worker safety, Health Insurance Portability and Accountability Act, re- search subjectsâ protection, billing practices). They investigate com- plaints in a confidential manner and develop plans of corrective action. The institutionâs compliance office might be able to provide a more immediate response to a duty hours problem than a national organization even when direct reporting is allowed. â¢ Gathering useful data to drive policy and evaluate progress. Duty hour compliance audits by ACGME could be made more useful than they are now. They provide opportunities not only to ensure adherence to rules but also to gather data on how long residents are really working (by specialty and rotation), why they violate limits, and when they violate limits (e.g., night shifts but not day shifts). Such data may illuminate when exceptions might be permissible and how to target fatigue mitigation strategies and staffing. Some program directors and residents have complained that the limits are inflexible and sometimes interfere with professional obligations and important educational opportunities (Fletcher et al., 2008; Lin et al., 2006). Better collection and analysis of monitoring data could document such problems. These changes recommended by the committee should greatly improve adherence to the duty hour limits. The committee, however, concluded that oversight of the ACGME process was also needed. Future Approach to Providing Oversight of ACGME Monitoring The committee considered different ways to provide assurance to the public and Congress that ACGMEâs discovery of the extent of duty hour violations is accurate, that residency programs move more quickly towards full adherence, and that residents can safely report violations when neces- sary. Oversight is recommended to provide such assurance. This function could be assigned to a new organization, a government agency or to an existing organization. The advantages and disadvantages of these options are discussed below. Alternative 1: A New Organization for Oversight Creating a new organization to provide oversight would have the ad- vantage of independence from all the existing stakeholders, if structured
78 RESIDENT DUTY HOURS appropriately, and a clean slate with no perceptions of bias. The main dis- advantages to doing this are similar to those mentioned for a new organiza- tion to conduct monitoring: delays in getting authority and funds to create such an organization and the need to establish public credibility. It would also need to design and establish a mechanism and procedures for providing oversight. The committee concluded that the delays and expense involved with creating a new oversight organization were not justified. Alternative 2: A Government Agency for Oversight Some of the advantages of using a public federal agency are that it is less likely to be co-opted by the profession than are private bodies, it can be tough and authoritative, and may already enjoy a measure of public trust. The most obvious government agencies that might conduct oversight of duty hour monitoring are the Centers for Medicare and Medicaid Services (CMS) or the Agency for Healthcare Research and Quality (AHRQ). The specific pros and cons of using CMS and AHRQ are discussed below. The role envisioned for CMS would be to help ensure the accuracy and reliability of ACGME procedures, data, and reports by supporting periodic evaluations of duty hours that would look not only at compliance but also examine the reasons behind violations and to suggest when exceptions to rules might be necessary to promote patient safety and under what circum- stances (e.g., direct supervision) a resident might be able to stay beyond his or her hour limit to participate in an unusual learning opportunity. This overview of the exceptions process as well as duty hour adherence is important in light of the committeeâs recommendations on adjustments to duty hours and provisions for exceptions in Chapter 7. CMS has an Office of Clinical Standards and Quality (OCSQ) that serves as a focal point for all quality and safety issues and it has direct access to funds from the Medicare Trust Fund that support contracts for research and evaluation related to quality and safety. A very small percent- age of those funds could support periodic contracted evaluations of duty hours and their monitoring and their relationship to quality of care, patient safety, resident safety, and educational outcomes. CMS could either con- tract for studies of duty hour compliance and manage the contracts directly or it could support research managed by another federal agency, such as AHRQ. Alternatively, OCSQ also has ongoing contracts with private qual- ity improvement organizations in each state, such as IPRO in New York, and could support one or more of them to conduct an evaluation of the outcomes of ACGME monitoring on adherence to rules. There are several reasons why CMS would be the most appropriate agency to take on some of the necessary evaluative responsibilities, but also reasons why that might not be desirable. On the positive side, CMS
current DUTY HOURS AND MONITORING ADHERENCE 79 has had an intimate relationship with teaching hospitals since 1965 con- cerning graduate medical education (GME) funding and resident educa- tion. It expends more than $8 billion in funds annually related to GME and associated patient care. CMS is the main federal agency responsible for assuring healthcare quality, paying for the care of millions of patients in teaching hospitals, as well as auditing the facilities. In addition, CMS has had ongoing relations with the VA and DOD health systems over the years concerning Medicare-eligible veterans and retirees, which might facilitate cooperative oversight of resident hours in those systems. CMS has the resources to conduct the evaluative studies envisioned for their oversight role. Having CMS involved in the oversight of duty hours monitoring pro- vides additional possibilities for increasing adherence to the rules. If the changes in ACGME monitoring practices and whistle-blower protections do not prove sufficient to have institutions comply, financial levers should be considered in addition to the threat of ACGME accreditation withdrawal or placing a residency program on probationary status. For example, in New York State, fines for duty hour violations are levied on institutions. The committee suggests that ACGME and CMS explore this and other options related to Medicareâs program rules for institutions receiving direct or in- direct GME funds. CMS would want to have confidence in the monitoring process before leveling such fines and having conducted an evaluation of the process would be critical. Additionally, ACGME through its Committee on Innovations may discover that certain carrot-and-stick approaches will foster adherence and these should also be considered (Volpp and Landrigan, 2008). On the negative side, CMS is a large bureaucracy that has not done such oversight of the GME program in the past and that function may not be a top priority for funding and attention in the organization. Since it is a federal agency, its policies and staff could potentially change significantly from one administration to another. Some people might object to giving a government agency oversight over a private organizationâs monitoring of duty hours as has been evidenced in opposition to previous attempts to regulate duty hours in HHS through legislation. While AHRQ might also be an appropriate option for the evaluative studies, it would likely have more difficulty obtaining needed funds than would CMS, and it does not have the leverage over training institutions that CMS has. The committee expects that AHRQ would play a significant role in implementing the recommendation concerning future research and evaluation, discussed in Chapter 9, and AHRQ would benefit from having a neutral role when working with other research-oriented parties planning a research agenda rather than direct oversight responsibilities. Addition- ally, the newly initiated AHRQ program on Patient Safety Organization
80 RESIDENT DUTY HOURS reporting could yield complementary information on whether residents and/or fatigue contribute to reported events if resident status and fatigue are included in those reporting requirements (AHRQ, 2008). The committee recommends that CMS should provide evaluative over- sight of ACGMEâs monitoring of duty hours and the possible effects of violations on quality of care and patient safety. That oversight function would be enhanced by complementary oversight by an existing private organization as well. Alternative 3: An Existing Private Organization for Oversight An existing private organization conducting related functions could have certain advantages over both a new organization and a public one. It could move quickly and readily update its procedures, have stature and recognition among the profession and the public, and a focus on quality and safety. The Joint Commission, which currently accredits hospitals, could play a complementary role to CMSâs oversight of the duty hours monitoring. The oversight role for the Joint Commission would differ from that of CMS and should fit consistently with its own accreditation process, which focuses on patient safety and quality during periodic, unannounced visits to institutions by a team of surveyors. Testimony by the Joint Commission on its approach to monitoring quality of care and safety indicated that rather than monitoring whether resident duty hours meet ACGME limits within an institution, the Joint Commissionâs approach could be to deter- mine whether residents or other staff were involved in patient safety events examined through patient-centered tracer cases and whether fatigue was a contributing cause (Joint Commission, 2008). For the Joint Commission to take on this systematic oversight function, it would likely need to adjust its policies and procedures to include a stronger focus and guidance on fatigue, safety, and work hours, although it already has raised the issue through its publications. Since their tracer case process as well as preliminary data analysis related to each hospitalâs accreditation visit include a wide variety of data and record checks, the marginal increase in work and costs based on cases with fatigued residents to assure proper monitoring policies and pro- cedures would likely not be great. The surveyors would not have to check adherence documentation for all the residency programs in an institution, just those related to programs in a tracer case. There are advantages to including the Joint Commission in the over- sight process. The Joint Commission currently accredits 97.5 percent of major teaching hospitals and 93.6 percent of minor teaching hospitals (Joint Commission, 2008). An oversight role would place adherence to duty
current DUTY HOURS AND MONITORING ADHERENCE 81 hours and prevention of fatigue within institution-specific quality and safety efforts and highlight the role of residents and their importance to patient safety. Joint Commission accreditation affects the entire hospital not just the educational programs and, through its recommendations for systems improvements, receives the attention of institutional administrators. The accreditation process is used to identify areas for correction and improve- ment in a hospital (i.e., Requirements for Improvement that facilities must address specified by surveyors based on findings of deficiencies), and if a problem concerning adherence to duty hours is uncovered, it should be treated as other similar violations by the Joint Commission. The loss of accreditation is rare and occurs because of large, serious, and persistent problems. Disadvantages associated with the inclusion of the Joint Commission in the oversight process include the need for the Joint Commission to expand its survey process to include some specific attention to resident fatigue and to adjust its working relationship with other organizations to recognize the duty hour limits set by ACGME. The Joint Commissionâs priorities are on patient safety and quality issues. Their complaint process, which receives approximately 12,000 complaints per year, gets only 5-8 related to resident work hours and their voluntary sentinel events reporting system rarely finds resident fatigue mentioned in the root cause analyses, indicating that currently the resident fatigue issue does not demand much attention (Joint Commission, 2008). It could happen that the Joint Commissionâs use of its tracer case method will reveal very few patient events related to resident fatigue and duty hours, but that would not necessarily negate the value of its oversight role. The committee concludes that the advantages of a strengthened ACGME monitoring process along with external oversight by both CMS and the Joint Commission would help assure the public that programs would be more likely to adhere to the rules, problems with duty hours com- pliance would be uncovered and dealt with properly, and there would be more rapid implementation of the committeeâs recommended adjustments to duty hours. CMS, the Joint Commission, and ACGME should discuss how their functions could complement each other and what information can be shared. The recommended oversight functions discussed for CMS and the Joint Commission are designed to be practical, derive from existing functions, and not be overly burdensome. Also, residents are more likely to report problems when they arise if probation of their educational program was not the only lever. The stature of existing relationships of both CMS and Joint Commission with teaching institutions would significantly add weight at the institutional level to ACGME processes.
82 RESIDENT DUTY HOURS Recommendation 2-1: ACGME and residency programs should ensure adherence to the current limits now, and to any new limits when imple- mented, by strengthening their current monitoring practices. To provide additional support, the Centers for Medicare and Medicaid Services and the Joint Commission should take an active oversight role: â¢ ACGME should maintain responsibility for duty hour moni- toring and should enhance its procedures by including unan- nounced visits for monitoring duty hours and regular collec- tion of sufficient data to understand when and why limits are violated. â¢ Sponsoring institutions should provide for confidential, pro- tected reporting of duty hour violations by residents through their compliance office or by an entity above the program level that does not have direct responsibility over the residency programs. â¢ ACGME should strengthen its complaint procedures to provide more confidentiality and protection to persons reporting viola- tions of duty hours, as well as other violations of residency rules. â¢ The Centers for Medicare and Medicaid Services should as- sess the reliability of ACGME procedures and data and should sponsor periodic independent reviews of ACGMEâs duty hour monitoring to determine the characteristics of and reasons for violations. â¢ The Joint Commission should seek to ensure that duty hour monitoring is linked to broader activities to improve patient safety in hospitals, including the use of ACGMEâs adherence data as part of the Joint Commissionâs hospital surveys and ac- creditation actions. Service demands on residents and educational expectations can create pressures for longer hours of service than are necessary for achieving edu- cational competence alone. In instituting the 2003 duty hour reforms, the ACGME indicated that training programs needed to âdecouple notions of professionalism from the number of hours workedâ (AAMC, 2003). Going forward, professionalism should not just mean staying long hours. Edu- cational leaders, hospital administrators, and residents themselves should recognize that ensuring adequate sleep for residents is part of responsible behavior to promote safe conditions for both residents and patients. This chapter has focused on the need for increased monitoring of resident duty hours and increasing transparency of why rules are violated. The commit- teeâs ultimate intent is not to establish a burdensome and costly monitoring
current DUTY HOURS AND MONITORING ADHERENCE 83 process that must be continued forever, but to ensure that there is a change in practice and that we learn from its implementation. REFERENCES AAMC (Association of American Medical Colleges). 2003. Adapting to the new duty hour requirements: GME directors meet to compare notes. http://www.aamc.org/newsroom/ reporter/nov03/dutyhours.htm (accessed September 9, 2008). âââ. 2007. AAMC Survey of house staff stipends, benefits and funding: Autumn 2007 report. Washington, DC: Association of American Medical Colleges. âââ. 2008. Resident duty schedules and resident education. Presentation by Debra ÂWeinstein to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, May 8, 2008, Washington, DC. ABMS (American Board of Medical Specialties). 2008. Presentation by Kevin Weiss to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Sched- ules to Improve Patient Safety, May 8, 2008, Washington, DC. ABNS (American Board of Neurological Surgeons). 2008. Presentation by H. Hunt Batjer to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. ACGME (Accreditation Council for Graduate Medical Education). 2003. The ACGMEâs ap- proach to limit resident duty hours: The common standards and activities to promote adherence. http://acgme.org/acWebsite/GME_infor/history/GME.pdf (accessed December 27, 2008). âââ. 2004. The ACGMEâs approach to limit resident duty hours 12 months after imple- mentation: A summary of achievements. http://www.acgme.org/acWebsite/dutyHours/ dh_dutyhoursummary2003-04.pdf (accessed December 27, 2007). âââ. 2006. The ACGMEâs approach to limit resident duty hours 2005-06: A summary of achievements for the third year under the common requirements. http://www.acgme. org/acWebsite/dutyHours/dh_achieveSum05-06.pdf (accessed December 27, 2007). âââ. 2007a. The ACGMEâs approach to limit resident duty hours 2006-07: A summary of achievements for the fourth year under the common requirements. http://www.acgme. org/acWebsite/dutyhours/dh_achieve0607.pdf (accessed December 27, 2008). âââ. 2007b. The ACGMEâs approach to limiting resident duty hours. Presentation by Paul Friedmann to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. âââ. 2007c. Procedures for addressing complaints against residency programs and spon- soring institutions. http://www.acgme.org/acWebsite/resIfo/ri_complaint.asp (accessed March 29, 2008). âââ. 2008a. Accreditation Council for Graduate Medical Education: Glossary of terms. http://www.acgme.org/acWebsite/about/ab_ACGMEglossary.pdf (accessed July 17, 2008). âââ. 2008b. The accreditation process and the role of the ACGME site visitor. http://www. acgme.org/acWebsite/fieldStaff.fs_siteRole.asp (accessed March 26, 2008). âââ. 2008c. The ACGME duty hour standards: One element for promoting good learning, safe patient care and resident well-being: Data on compliance and effect. Presentation by Ingrid Philibert and Thomas Nasca to the Committee on Optimizing Graduate Medi- cal Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. âââ. 2008d. Fees for evaluation and accreditation. http://www.acgme.org/acWebsite/GME_ info/gme_feesAccred.asp (accessed March 30, 2008).
84 RESIDENT DUTY HOURS âââ. 2008e. Frequent citations of orthopaedic surgery residencies. Orthopaedic RRC Newsletter (Winter 2008):1-2. âââ. 2008f (unpublished). The ACGME limits on resident duty hours: Standards, promot- ing compliance and effect. May 2008. ACS (American College of Surgeons). 2008. Presentation by Gerald Healy to the Commit- tee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. AHRQ (Agency for Healthcare Research and Quality). 2008. The patient safety and quality improvement act of 2005. http://www.ahrq.gov/qual/psoact.htm (accessed November 13, 2008). AMA (American Medical Association). 2005. Medical students and residents work-hour survey. Chicago, IL: American Medical Association, Division of Market Research and Analysis. âââ. 2006. Issues of importance to resident physicians and fellows highlights. Chicago, IL: American Medical Association, Division of Market Research and Analysis. âââ. 2007. Duty hours: The resident perspective. Presentation by Sunny Ramchandani to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. American Medical Student Association. 2007. Presentation by Michael Ehlert to the Commit- tee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. âââ. 2008. Principles regarding resident and student work hours. http://www.amsa.org/ about/ppp/rwh.cfm (accessed February 10, 2008). Andrews, A. 1995. I was juror no. 6, the lone dissenter in the Libby Zion case. New York Times, February 21. Arora, V., C. Dunphy, V. Y. Chang, F. Ahmad, H. J. Humphrey, and D. Meltzer. 2006. The effects of on-duty napping on intern sleep time and fatigue. Annals of Internal Medicine 144(11):792-798. Arora, V., D. B. Wayne, R. A. Anderson, A. Didwania, and H. J. Humphrey. 2008. Participa- tion in and perceptions of unprofessional behaviors among incoming internal medicine interns. JAMA 300(10):1132-1134. Asad, M., A. K. Shah, and F. Ehrlich. 2006. Electronic submission of resident work hours: A convenient way of exchange of information. American Surgeon 72(8):735-738. Associated Press. 2003. Hospitals face limit on resident hours. http://www.paworld.net/ pasmeetneed.htm (accessed August 27, 2008). Baldwin, D. C., Jr., S. R. Daugherty, R. Tsai, and M. J. Scotti, Jr. 2003. A national survey of residentsâ self-reported work hours: Thinking beyond specialty. Academic Medicine 78(11):1154-1163. Bell, B. M. 2003. Reconsideration of the New York State laws rationalizing the supervi- sion and the working conditions of residents. Einstein Journal of Biological Medicine 20(1):36-40. Boardman, J. H. 2007. Written statement to the Subcommittee on Surface Transportation and Merchant Marine Infrastructure, Safety, and Security, Committee on Commerce, Science, and Transportation, U.S. Senate. Administrator, Federal Railroad Administration, U.S. Department of Transportation. Brotherton, S. E., F. A. Simon, and S. I. Etzel. 2002. U.S. graduate medical education, 2001- 2002: Changing dynamics. JAMA 288(9):1073-1078. Carpenter, R. O., J. Spooner, P. G. Arbogast, J. L. Tarpley, M. R. Griffin, and K. D. Lomis. 2006. Work hours restrictions as an ethical dilemma for residents: A descriptive survey of violation types and frequency. Current Surgery 63(6):448-455.
current DUTY HOURS AND MONITORING ADHERENCE 85 Chang, L. W., A. R. Vidyarthi, and R. J. Kohlwes. 2006. Baseline duty hours recorded with time-cards: A pre-regulation study of internal medicine residents. Medical Education 40(7):662-666. Chao, L., and M. K. Wallack. 2004. Changes in resident training affect what you can expect from your next partner. Bulletin of the American College of Surgeons 89(9):12-15. CIR/SEIU (Committee of Interns and Residents-Service Employees International Union Health- care). 2007. Presentation by L. Toni Lewis to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, De- cember 3, 2007, Washington, DC. Council on Accreditation. 2008. Understanding the Hague accreditation and approval standards and the substantial compliance system. http://coanet.org/front3/page. cfm?sect=54&cont=4251 (accessed March 24, 2008). Cull, W. L., H. J. Mulvey, E. A. Jewett, E. L. Zalneraitis, C. E. Allen, and R. J. Pan. 2006. Â Pediatric residency duty hours before and after limitations. Pediatrics 118(6): e1805-e1811. Czeisler, C. A. 2007. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. DeBuono, B. A., and W. M. Osten. 1998. The medical resident workload: The case of New York State. JAMA 280(21):1882-1883. Department of Labor-Occupational Safety and Health. 2002. Letter from John L. Henshaw, Assistant Secretary for Occupational Safety and Health to Sidney M. Wolfe, Direc- tor, Health Research Group, Public Citizen. http://www.citizen.org/documents/ACFC2B. pdf. Dinges, D. F. 1995. Sleep deprivation, impairment of resident performance, and medical error. Journal of Bone and Joint Surgery 87a(11):2581-2582. Douglas, R. G. 1995. Zion case verdict vindicates training system. New York Times, Febru- ary 14. Drake, C. L., T. Roehrs, G. Richardson, J. K. Walsh, and T. Roth. 2004. Shift work sleep disorder: Prevalence and consequences beyond that of symptomatic day workers. Sleep 27(8):1453-1462. Federal Railroad Administration. 2008. National Rail Safety Action Plan final report 2005- 2008. Cambridge, MA: U.S. Department of Transportation. Fletcher, K. E., V. Parekh, L. Halasyamani, S. R. Kaufman, M. Schapira, K. Ertl, and S. Saint. 2008. Work hour rules and contributors to patient care mistakes: A focus group study with internal medicine residents. Journal of Hospital Medicine 3(3):228-237. Folkard, S., D. A. Lombardi, and P. T. Tucker. 2005. Shiftwork: Safety, sleepiness and sleep. Industrial Health 43(1):201-223. Gilligan, P. 2007. Statement of Peggy Gilligan, Deputy Associate Administrator for Aviation Safety. Testimony before the Congressional Committee on Transportation and Infrastruc- ture, Subcommittee on Aviation on the Most Wanted List of the National Transportation Safety Board. Washington, DC: Federal Aviation Administration, U.S. Department of Transportation. Goldstein, M. J., B. Samstein, A. Ude, W. D. Widmann, and M. A. Hardy. 2005. Work Hours Assessment and Monitoring Initiative (WHAMI) under resident direction: A strategy for working within limitations. Current Surgery 62(1):132-137. GovTrack.us. 2005a. H.R. 1228â109th Congress: Patient and Physician Safety and Protection Act of 2005. http://www.govtrack.us/congress/bill.xpd?bill=h109-1228&tab=summary (accessed March 29, 2008).
86 RESIDENT DUTY HOURS âââ. 2005b. S. 1297â109th Congress: Patient and Physician Safety and Protection Act of 2005. http://www.govtrack.us/congress/bill.xpd?bill=h109-1297&tab=summary (ac- cessed March 29, 2008). IOM (Institute of Medicine). 2003. Keeping patients safe: Transforming the work environment of nurses. Appendix C. Washington, DC: The National Academies Press. IPRO. 2007a. Graduate medical educationâPast, present and future. Resident Times 1(1):2. âââ. 2007b. Working hours and conditions post-graduate trainees annual compliance as- sessment: Contract year 6 10/1/06-9/30/07. Albany, NY: IPRO. âââ. 2008. IPRO milestonesâ20 years of quality improvement. http://www.ipro.org/ index/milestones (accessed January 29, 2008). Jagsi, R., D. F. Weinstein, J. Shapiro, B. T. Kitch, D. Dorer, and J. S. Weissman. 2008. The Accreditation Council for Graduate Medical Educationâs limits on residentsâ work hours and patient safety: A study of resident experiences and perceptions before and after hours reductions. Archives of Internal Medicine 168(5):493-500. Joint Commission. 2008. Presentation by Paul Schyve to the Committee on Optimizing Gradu- ate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, May 8, 2008, Washington, DC. Kennedy, R. 1998. Residentsâ work hours termed excessive in hospital study. New York Times, May 1, p. 1. Kowalczyk, L. 2008. Beth Israel cited for residentsâ long hours. Boston Globe, September 7. Kozak, L. J., C. J. DeFrances, and M. J. Hall. 2006. National Hospital Discharge Survey: 2004 annual summary with detailed diagnosis and procedure data. Hyattsville, MD: National Center for Health Statistics. Lamberg, L. 2002. Long hours, little sleep: Bad medicine for physicians-in-training? JAMA 287(3):303-306. Landrigan, C. P. 2007. Effects of traditional 24-hour work shifts on physician and patient safety. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resi- dent) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Wash- ington, DC. Landrigan, C. P., L. K. Barger, B. E. Cade, N. T. Ayas, and C. A. Czeisler. 2006. Internsâ com- pliance with Accreditation Council for Graduate Medical Education work-hour limits. JAMA 296(9):1063-1070. Landrigan, C. P., A. M. Fahrenkopf, D. Lewin, P. J. Sharek, L. K. Barger, M. Eisner, S. Edwards, V. W. Chiang, B. L. Wiedermann, and T. C. Sectish. 2008. Effects of the Ac- creditation Council for Graduate Medical Education duty hour limits on sleep, work hours, and safety. Pediatrics 122:250-258. Lieberman, J. D., J. A. Olenwine, W. Finley, and G. G. Nicholas. 2005. Residency reform: Anticipated effects of ACGME guidelines on general surgery and internal medicine resi- dency programs. Current Surgery 62(2):231-236. Lin, G. A., D. C. Beck, and J. M. Garbutt. 2006. Residentsâ perceptions of the effects of work hour limitations at a large teaching hospital. Academic Medicine 81(1):63-67. Ludmerer, K. M. 1999. Time to heal: American medical education from the turn of the century to the era of managed care. New York: Oxford University Press. McCallum, M., T. Sanquist, M. Mitler, and G. Krueger. 2003. Commercial transportation operator fatigue management reference. Washington, DC: U.S. Department of Transpor- tation Human Factors Coordinating Committee. New York State Department of Health. 2002. State health department cites 54 teaching hospitals for resident working hours violations. http://www.health.state.ny.us/press/ releases/2002/resident_working_hours.htm (accessed May 26, 2008).
current DUTY HOURS AND MONITORING ADHERENCE 87 New York Supreme Court. 1986. Report of the fourth grand jury for the April/May term of 1986 concerning the care and treatment of a patient and the supervision of interns and junior residents at a hospital in New York County. NTSB (National Transportation Safety Board). 2007. NSTB most wanted listâTransportation safety improvements: 2008. Washington, DC: National Transportation Safety Board. PAIRO (Professional Association of Interns and Residents of Ontario). 2008. Principles of cross-coverage. http://www.pairo.org/Content/Default.aspx?pg=1177 (accessed July 21, 2008). Public Citizen. 2001. Petition to the Occupational Safety and Health Administration request- ing that limits be placed on hours worked by medical residents. Washington, DC: Public Citizen. âââ. 2007. Presentation by Peter Lurie to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2008, Washington, DC. Reiter, E. R., and D. R. Wong. 2005. Impact of duty hour limits on resident training in oto- laryngology. Laryngoscope 115(5):773-779. Resident Panel. 2008. Presentation to some members of the Committee on Optimizing Gradu- ate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Rosa, R. R. 2001. Examining work schedules for fatigue: Itâs not just hours of work. In Stress, workload, and fatigue, edited by P. A. Hancock and P. A. Desmond. Mahwah, NJ: Law- rence Erlbaum Associates, Inc. Pp. 513-528. Royal College of Surgeons of England. 2008. Lessons from surgery. Presentation by Bernard Ribeiro to Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Sleep-deprived doctors. 2002. New York Times, June 14. Steinbrook, R. 2002. The debate over residentsâ work hours. New England Journal of Medi- cine 347(16):1296-1302. Van Dongen, H. P. A., and D. F. Dinges. 2005. Circadian rhythm in sleepiness, alertness and performance. In Principles and practice of sleep medicine, 4th ed., edited by M. H. Kryger, T. Roth, and W. C. Dement. Philadelphia, PA: Elsevier Saunders. Pp. 435-443. Volpp, K. G., and C. P. Landrigan. 2008. Building physician work hour regulations from first principles and best evidence. JAMA 300(10):1197-1199. Weissman, J. S., J. M. Rothschild, E. Bendavid, P. Sprivulis, E. F. Cook, R. S. Evans, Y. Kaganova, M. Bender, J. David-Kasdan, P. Haug, J. Lloyd, L. G. Selbovitz, H. J. Murff, and D. W. Bates. 2007. Hospital workload and adverse events. Medical Care 45(5):448-455.