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.
1 Background and Overview After graduation from medical school, residency training provides an in- depth experience in learning the science and art of medicine for a specific discipline (e.g., general surgery, internal medicine, pediatrics). This chap- ter gives a brief history and overview of graduate medical training in the United States, with a focus on resident duty hours. Key issues related to patient safety, resident safety, and resident learning are introduced, includ- ing adherence to current duty hour limits, the relationship of duty hours to patient and resident safety, and the importance of training for achieving the competence to practice medicine independently at the end of residency. These issues are reviewed in greater detail in subsequent chapters. Based on evidence in the later chapters, the committee recommends adjustments to current duty hours and other steps to improve the education of future physicians, which should enhance the safety of residents and their current and future patients. The primary mission of graduate medical education (GME) is to train the next generation of physicians. To achieve this, graduates of medical school spend 3 to 7 additional years in residency training, becoming skilled doctors prepared for independent practice in different disciplines. This evolu- tion takes place in environments both exhilarating and exhausting, being a preparation for similar challenges and rewards later in their career Â(Ludmerer, 1999). Residency training is essential for professional development both edu- cationally and practically, since all state medical boards require at least 1 year of graduate medical training before a doctor is eligible for an Â unrestricted license to practice medicine (AMA, 2008; FSMB, 2007). 27
28 RESIDENT DUTY HOURS Early in the 20th century, graduate medical trainees, also known as residents or house staff, actually resided at the hospital where they pro- vided âon-callâ medical service as part of their training. Until the recent duty hour reforms, resident work schedules frequently totaled 90 hours or more per week, made up of 36-hour shifts separated by 12 hours or less of rest. Proposals to reduce resident duty hours have been met with concern within the medical community that physician training could be compro- mised. Given that residents provide a source of inexpensive labor for medi- cal institutions, questions have also been raised about whether duty hours are inflated to meet the service needs of hospitals without a commensurate increase in educational value for residents. The overall structure of funding for GME in the United States has promoted hospital-based training for resi- dents. Teaching hospitals and the public have come to depend on residents to deliver services around the clock, and providing substitutes for their time is expensive. Attending physicians (physicians who supervise the care provided by residents) also benefit from having residents onsite to facilitate patient care rather than always being present themselves. In 1984, the length of resident duty hours came under public scrutiny when Libby Zion died after being seen in an emergency room in New York City. A subsequent grand jury investigation highlighted the risks to patient care posed by inadequately supervised and fatigued residents. Following the recommendations of the Bell Commissionâs review of emergency care, New York State limited resident duty hours in 1989 to 80 hours a week (averaged over 4 weeks), affecting medical facilities statewide where ap- proximately 15 percent of all residents in the country trained (IPRO, 2007). The duty hour limits set in New York later became the basis for national re- form. In 2003, the Accreditation Council for Graduate Medical Education (ACGME) mandated an 80-hour weekly average for all residents along with implementing other minimum requirements for time off from the hospital; these are discussed more fully in Chapter 2 (ACGME, 2007b). This chapter outlines the scope of issues studied by the Institute of Medicine (IOM) Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety. The com- mitteeâs name highlights the groupâs task of looking at residentsâ total hours of work and the distribution of those hours (schedule) over a period of time. This chapter provides background on the general nature of residency training, areas of residency specialization, and number of residents and training programs in the country. To provide some context, the total work- week hours of residents are compared with the total duty hours of physi- cians in practice. Finally, the chapter highlights concerns with respect to duty hours that surfaced in the scientific and academic medicine literature and in testimony to the committee.
BACKGROUND AND OVERVIEW 29 CHARGE TO COMMITTEE Congress, through the Subcommittee on Oversight and Investigations of the House Committee on Energy and Commerce, requested that the Department of Health and Human Services (HHS) sponsor a study by the IOM to examine the relationship between resident duty hours and patient safety. The subcommittee had been investigating preventable medical errors and asked if the duty hours of physicians and residents are among the most serious threats to patient safety (Dingell et al., 2007). The IOM, under a contract with the Agency for Healthcare Research and Quality (AHRQ), HHS, agreed to form a consensus committee and conduct a study of residents to 1)â synthesize current evidence on medical resident schedules and health- care safety, and 2)â develop strategies to enable optimization of work schedules to improve safety in the healthcare work environment, .ââ.ââ.ââ. [and] Consider also evi- dence on the safety of the residents, the education and training experience of the residents, the quality of the interactions from both the resident and patient perspective, and other aspects of safety and quality of care such as care hand-offs and transitions. (See Appendix A for the complete Statement of Task.) Residency continues to consist of an apprenticeship approach to learn- ing through service to hospital inpatients and their clinic patients under the guidance of attending faculty physicians. Residents play a significant role in the healthcare system. They are frequently the primary frontline physi- cian-level staff on duty around the clock in teaching hospitals. Residents are exposed to disease pathology and learn effective management of both common and unusual illnesses and injuries. The residentâs dual roles of learner and provider of care mean that changes in duty hours have implica- tions for educational quality and access to care, as well as both short-term and long-term patient safety. Adjusting hours without other compensatory system changes can endanger these desirable outcomes. Because residents provide their valuable skills for a relatively mod- est taxable stipend (e.g., for 2007-2008 the mean stipend for a first-year resident was $44,747), reductions in duty hours also have economic im- plications (AAMC, 2007a). A reduction in residentsâ duty hours creates a demand for other clinicians (e.g., more attending physicians, more nurse practitioners, more residents, more support staff) to fill the resulting gaps in medical care, and there are already projected shortages of some clini- cians (AAMC, 2008b; ANSR, 2008; HHS, 2006). The committee therefore felt that its evaluation of the effects of resident duty hour changes should
30 RESIDENT DUTY HOURS be viewed in the broader context of the possible educational, safety, and resource consequences. In meeting its charge, the committee carefully examined a broad range of pertinent evidence. Since the 2003 ACGME duty hour limits for resi- dents have been in effect for only 5 years, there are limited data concerning the impact of these changes. The committee represented diverse points of view and vigorously debated the issues and relevance of studies. The ex- isting body of evidence was sufficient for the committee to find common ground and conclude that steps should be taken to strengthen resident education and establish an environment that would better protect residents from fatigue and patients from the potential for resident errors. The com- mittee sought to prevent acute and chronic sleep deprivation in residents when possible and, when it is not, to reduce the risks to patients and resi- dents resulting from residentsâ loss of sleep and fatigue. GRADUATE MEDICAL TRAINING Teaching hospitals provide a setting for the clinical education of resi- dents as well as other health profession students (e.g., nursing, pharmacy) and have different degrees of organizational integration with medical schools (e.g., shared ownership or affiliations but separate funding and governance). Residents primarily take care of inpatients and outpatients at these hospitals as well as spend some time in community-based settings, such as community health centers and physiciansâ offices, depending on their areas of study. These academic institutions often combine teaching, patient care, and medical research. There are approximately 1,100 hospi- tals participating in medical education, but three-fourths of all residency training takes place in about 275 hospitals and health systems in the United States. Highly specialized services (e.g., transplant services, interventional cardiology, neonatal intensive care units [ICUs], burn care units, regional trauma centers, AIDS services) are more concentrated into this smaller set of teaching institutions than other teaching and non-teaching hospitals. The median number of residents at each of the 275 institutions in 2005 was 173, â Bibliographic searches were conducted of the primary biomedical bibliographic databases, MEDLINE, EMBASE, CINHAL, and PsychInfo. The searches included articles from January 1980 to January 2008. The terms used for these searches, many in combination with each other, included resident(s), residency, internship, fatigue, sleep, sleep disorders, burnout, mood, depression, work schedule(s), work hours, 80-hour workweek, adverse events, medical errors, job satisfaction, handoffs, handovers, transitions, mortality, patient outcomes, patient safety, quality of care, medical education, graduate medical education, workload, and performance. Publications dated after January 2008 were added to the evidence base of this reportÂ as they became available or were brought to the attention of IOM staff.
BACKGROUND AND OVERVIEW 31 while other teaching hospitals have a median of 21 residents (AAMC, 2004, 2008a; Council of Teaching Hospitals and Health Systems, 2008). A resident is any physician enrolled in a GME program (ACGME, 2007a). First-year residents were, until recently, more often called interns, and they are also referred to as PGY-1 (postgraduate year 1) residents in many programs and research articles. Residents in the later years of practice are termed PGY-2, PGY-3, and so on to distinguish the year of postgradu- ate training. Since residents are in training, they must function under the qualified supervision of faculty and more senior residents. As residents demonstrate knowledge and skills appropriate to progressive levels of their postgraduate training, they are given increasing responsibility for the care of patients, larger patient loads, and greater authority to make final patient management decisions (ACGME, 2007a). A patient may not always be able to distinguish resident trainees from other physicians because those in residency have an M.D. or a D.O. degree. By tradition and as necessitated by the nature of their responsibilities, the length and rigor of formal education and training for physicians is among the most challenging of any job or profession in the United States (U.S. Department of Labor, 2007). Residency is a period of intensive su- pervised learning in a real-world environment where critical skills and competencies are developed, including the needed professionalism that is the hallmark of a caring, competent, and dedicated physician. Residency training has periods during which prolonged duty hours are perceived as necessary to achieve the educational goalsâthis is more the case for some specialties than others. Trainees spend years preparing for the opportunity to train as resident physiciansâthrough 4 years of a premedical curricu- lum in college and 4 more years of challenging medical school study and testing. Only about half of the applicants to medical school are accepted (AAMC, 2008d), but more than 90 percent of the students accepted will graduate (AAMC, 2007b). It is at this point that most graduates will enter a residency in their chosen area of specialization. Two main tracks exist for preparing doctors in the United States: graduates of allopathic schools receive an M.D. (doctor of medicine) degree and graduates of osteopathic schools receive a D.O. (doctor of osteopathic medicine) degree. Graduates of both types of schools pursue graduate medi- cal training before being licensed to practice independently. In 2003, 99 percent of U.S. allopathic and 43 percent of osteopathic school graduates undertook allopathic residencies (Shannon, 2007). The focus of this report is on the allopathic residencies accredited by the ACGME. For academic year 2008-2009, 15,242 U.S. medical school seniors were matched to one of the 22,240 available first-year U.S. residency posi- tions available across the United States (about 68 percent of the available first-year positions) based on applicant preferences and how the training
32 RESIDENT DUTY HOURS program ranked them (AAMC, 2008e; National Residency Match Pro- gram, 2008). An insufficient number of U.S. medical school graduates are available to fill all residency positions. In 2006-2007, 66.5 percent of all graduate medical trainees graduated from U.S. medical schools, 26.9 percent were international medical graduates (non-Canadian), 6.3 percent graduated from schools of osteopathy, and 0.3 percent were from Canadian medical schools (Brotherton and Etzel, 2007). Efforts are under way to increase the number of U.S. medical school graduates (AAMC, 2008d). Total Number of Residents and U.S. Training Programs Nearly 105,000 graduate medical trainees were at various stages of their residency training in the 2007-2008 academic year. Residents work in public and private, teaching and community hospitals across the country, affiliated with more than 8,500 distinct accredited residency programs. ACGME reviews and evaluates each residency program on average every 3 to 4 years with site visits and resident interviews to examine the content of training and to ensure compliance with educational and duty hour require- ments (ACGME, 2007a, 2008). Although residency programs are regulated by this private sector organization, they have grown and are maintained with substantial federal and state funding support, particularly with money from the Medicare program, the Department of Veterans Affairs, the Health Resources and Services Administration, the Department of Defense, and various state and local programs. In 2006-2007, of the total number of residents there were 89,269 in medical specialty programs, the first stage of graduate medical training (Brotherton and Etzel, 2007). âSpecialtyâ medical training programs are available in the fields listed in Table 1-1, and many of these specialties are combined in 19 additional residency designations, such as internal medicine and pediatrics, internal medicine and psychiatry, and internal medicine and emergency medicine. After having completed a specialty residency, 15,610 residents in 2006-2007 trained in advanced subspecialty programs, also called fellowships, and these resident trainees are sometimes referred to as âfellows.â The 26 types of specialty training programs listed in Table 1-1 offer subspecialty fellowships in about 100 areas, ranging from vascular surgery to geriatrics to pediatric endocrinology. ACGME duty hour rules apply to both specialty and subspecialty residents. Resident Learning Environment At the best of times, residency training provides daily intellectual stimu- lation and gratification in solving complex problems and making a differ- ence in patientsâ lives. Delivering a baby, teaching skills to medical students,
BACKGROUND AND OVERVIEW 33 TABLE 1-1â U.S. Resident Training Programs by Specialty and Resident Physicians on Dutya Number of Number of Resident Accredited Specialty Physicians Programs Internal medicine 22,099 386 Family medicine 9,456 464 Pediatrics 7,964 201 Surgery (general) 7,651 251 Anesthesiology 4,970 131 Obstetrics and gynecology 4,739 250 Psychiatry 4,613 181 Emergency medicine 4,379 140 Radiology, diagnostic 4,368 188 Orthopedic surgery 3,187 152 Pathology 2,310 150 Neurology 1,507 122 Otolaryngology 1,292 104 Ophthalmology 1,225 117 Physical medicine and rehabilitation 1,167 79 Dermatology 1,069 112 Urology 992 118 Neurological surgery 881 97 Plastic surgery 609 89 Radiation oncology 556 79 Preventive medicine 285 74 Surgery (thoracic) 282 85 Allergy and immunology 274 71 Nuclear medicine 143 61 Medical genetics 77 47 Surgery (colon and rectal) 71 45 aAsof December 1, 2006. SOURCE: Brotherton and Etzel, JAMA, 2007 298:1081-1096. Copyright Â© 2007, American Medical Association. removing a gall bladder, performing a lumbar puncture that confirms a case of treatable meningitis, diagnosing an unusual ailment, or running a code blue to resuscitate a patient in cardiac arrest before the attending a Â rrivesâall can provide a sense of accomplishment. Surgical residents gain confidence as they learn surgical principles and perfect their technique. Residents learn how to talk with patients and families both when the news is good and when it is not. There are also quiet momentsâholding the hand of a dying patient or waiting by the bedside to see if a patient is responding to treatment. Each specialty will have a different complement of illnesses and injuries, tests and procedures, but they have in common
34 RESIDENT DUTY HOURS the need to learn how to communicate with patients and other members of their medical teams. Residents train as a team of doctors, with supervision provided by residents further along in training and by attending physicians (senior physicians, most often faculty of medical schools, who have completed residency training and are ultimately responsible for patient care decisions). Attendings provide the daily instruction during teaching rounds by guid- ing residents through consideration of possible diagnoses and management plans, and in the case of surgery, they spend much of their time in direct instruction and observation of procedures. Attendings have other âteach- able momentsâ during the day when residents present summaries about newly admitted patients. Residents have other sources of learning, including the medical literature, lectures on important concepts, grand rounds, edu- cational conferences, and interactive skills-training exercises. Residents are also part of a larger team or system of care that includes nursing, adminis- trative, and other staff who are integral to the hospitalâs care mission (e.g., pharmacy, laboratory, transport, social work, nutrition, administration) and provide other valuable sources of information for resident learning. Residents rotate through a variety of teams or services, usually spend- ing 4 weeks per rotation. It is important for the reader of this report to recognize that each rotation as well as each specialty has different duty hour demands. An ambulatory care rotation might have five 8-hour shifts total- ing a 40-hour workweek. On an emergency department (ED) rotation, there might be five 12-hour shifts (day or night) in the ED plus another 12 hours for instructional activities, totaling 72 hours. An ICU rotation might have some duty periods that are 10 hours long alternating with 30-hour duty periods, totaling 80 hours or more over the course of a week. Each year of training has a different set of rotations, and residents have greater choice of rotations as they progress through training, allowing them to concentrate on areas that they plan to pursue at the end of training or supplementing their knowledge in areas where improvement is needed. DUTY HOUR DEMANDS IN THE MEDICAL PROFESSION Graduate medical training programs have a tradition of requiring long hours. In 1998-1999, residents in surgical specialties were still regularly clocking more than 100 hours per week in their PGY-1 and PGY-2 training years. First-year residents in 8 of 12 specialties surveyed at that time had an average workweek of more than 80 hours, and the average across all 12 types of programs for these interns was 83 hours (Baldwin et al., 2003). Medical educators expect that ample duty hours provide residents with the needed time to think through diagnoses, manage patient plans, and gain a
BACKGROUND AND OVERVIEW 35 rich educational experience as they follow patients closely during the criti- cal first day of admission, learn from exposure to a great variety of patient cases, increase technical skills as they participate in many procedures, and develop communication skills. Long hours during training have also become a rite of passage to the profession, testing residentsâ stamina, resil- ience, and dedication and reinforcing the idea that medical practice requires a certain mental toughness. In 2003, ACGME set an 80-hour workweek average as the limit for all graduate medical specialties, although a few programs received exemptions to have an 88-hour week average. The best available nationwide data on how many hours residents work under the new ACGME limits come from the initial year of implementation and apply only to first-year residents. On average, PGY-1 residents (interns) in more than 13 types of programs reported a mean of 66.6 hours weekly in 2003-2004 (Landrigan et al., 2006). This was a decline of 4.1 hours from 2002-2003 when interns reported 70.7 hours on average (Baldwin et al., 2003). Interns, those with the least experience, tend to work longer hours than residents in the same field but in later years of specialty training. An important part of graduate medical training is that it exposes residents to the demands of real-life practice, apart from the training envi- ronment, including the necessity to attend to ill patients at all hours even when away from the training environment. This is often inconvenient and involves night and weekend work. According to the U.S. Bureau of Labor Statistics, many physicians work long and unpredictable hours around the clock once they finish their graduate medical trainingâlonger hours than most other workers in the United States. The Bureau of Labor Statistics estimates that more than one-third of fully licensed physicians work 60 hours or more per week (U.S. Department of Labor, 2007). Physicians in office-based practice report an average workweek of 60 hours for surgical specialties and 50 hours for primary care (e.g., family practitioners, inter- nists), but the workweek is as long for some internal medicine specialties as for surgeons (e.g., mean of 60 hours per week for cardiologists and nephrologists) (Weiss, 2006). Certain types of specialties are more likely to report working 80 hours per week or more. Of the 23 specialties examined in 2005, the following reported 15 percent or more of their office-based practice members working more than 80 hours: urologists (15 percent), obstetricians-gynecologists (OB/GYNs; 16 percent), pulmonologists (16 percent), hematologists-oncologists-immunologists (17 percent), infectious disease specialists (17 percent), general surgeons (19 percent), cardiolo- gists (20 percent), neurosurgeons (23 percent), and thoracic surgeons (33 percent) (Weiss, 2005). Similarly, residency programs have different work â This national sample of interns was self-selected.
36 RESIDENT DUTY HOURS requirements depending on the specialty but now have common duty hour limits. SCOPE AND ORGANIZATION OF REPORT Oral presentations before the committee and additional submitted tes- timony from stakeholders and experts raised a broad array of issues and perspectives for consideration relative to resident duty hours and patient safety. It became obvious early in the study process that appropriately bal- ancing these issues would present a challenge, as would reaching a consen- sus within the committee. All of the testimony provided a useful guide for the committeeâs research, indicating areas for in-depth study and potential sources of data and evidence. The major topic areas raised included the following: â¢ Current duty hours and adherence to them â¢ Educational needs â¢ Resident safety and well-being â¢ Patient safety â¢ Economic implications The committee heard from diverse speakers: patient advocates; an ethicist; residents in training; resident and medical school student represen- tatives; residency program directors of several specialties; administrators in charge of all graduate medical training programs at their institution; hospital executives and financial officers from institutions with residents; scientists who study sleep, fatigue, and human performance; physician specialty societies; the president of the Royal College of Surgeons; repre- sentatives of national organizations involved in GME, including matching medical school graduates to residencies, and accreditation of programs; major funders of GME; and the Joint Commission. The presentations of the speakers that appeared before the committee are available on the project website, www.iom.edu/residenthours (see Appendix F for the public agenda for committee meetings). Current Duty Hours and Adherence to Them Currently, the ACGME sets standards for GME including duty hour limits, and it monitors how well residency programs adhere to those rules and the educational standards set for each specialty. Questions surfaced about whether there was a scientific basis for the 2003 ACGME rules now in place and how the length of the workweek and the number of consecutive duty hours compared with the experiences of other countries
BACKGROUND AND OVERVIEW 37 and other industries. Some speakers were in favor of changing certain ele- ments of the existing duty hour limits, especially the extended duty period of 30 hours and averaging provisions (CIR/SEIU, 2007; Landrigan, 2007; Public Citizen, 2007), while others thought any further change in hours premature (AAMC, 2008c; ABNS, 2008; ACGME, 2007a, 2008; ACS, 2008). Moreover, there was uncertainty about the actual number of hours that residents currently work per week, how often the limits are violated, and the reasons for those violations (e.g., patient care needs, excessive workload) and whether residents were staying of their own volition versus program or institutional pressures (ACGME, 2008; AMA Resident/Fellow Section, 2007; American Medical Student Association, 2007; Arora, 2007; CIR/SEIU, 2007; Resident Panel, 2008; Vidyarthi, 2007). The committee was asked by some presenters to determine whether the current ACGME procedures for assessing adherence to duty hours were sufficiently rigorous (CIR/SEIU, 2007; Public Citizen, 2007). The ACGMEâs position was that hours should not be viewed in isolation but as one of many pieces of information integral to assessing the quality of an educational program (including supervision and institutional support) and the quality of patient care delivered (AAMC, 2008c; ACGME, 2007a, 2008). Some or- ganizations argued that ACGMEâs data collection methods were insufficient to adequately enforce work hour limits, resulting in identification of too few violations of duty hours (CIR/SEIU, 2007; Public Citizen, 2007). Ad- ditionally, speakers pointed to disincentives to accurate duty hour reporting among residents who feared that their training program could lose accredita- tion or that they would personally face retribution (AMA Resident/Fellow Section, 2007; American Medical Student Association, 2007). Chapter 2 examines past and present duty hour limits in the United States and what is known about the monitoring and compliance issues outlined above. Further, it briefly examines the duty hour limits set for other safety-sensitive industries such as aviation and trucking and the ef- forts at regulation in those work environments. Appendix C draws lessons from the experiences of other countries that have mandated significantly reduced resident duty hours (e.g., by 2009, Europe will reduce duty hours to 48 hours per week). Lessons drawn from the experiences of other coun- tries relate to the period needed for phase-in of requirements, workforce implications, scheduling adaptations, and the need to modify educational programs to adapt to reduced duty hours. Educational Needs The quality of the educational experience of residents today is a âde- terminant of patient safety and health care quality for decades to comeâ (AAMC, 2008c). Educators asked the committee to consider the long-term
38 RESIDENT DUTY HOURS goal of ensuring patient safety by producing competently trained physicians versus the potential risks to patients that they treat during residency train- ing. The sparsity of data on educational outcomes since duty hour reform is partly due to the fact that the first cohorts of residents fully trained under the 2003 limits are now completing their residencies (ABMS, 2008; AMA Resident/Fellow Section, 2007; American Orthopaedic Association, 2008). In considering the initial data, it is important to recognize that adaptations to the limits may have evolved over time and conclusions drawn from early implementation studies may not paint an accurate picture of the current situation. Duty hours were not uniformly implemented at the same rate or in the same manner across all programs; some programs have modified their schedules several times over the past few years before moving to full compliance. Reducing hours means having fewer residents available for duty at any single time. The committee heard testimony that this has led to increased resident workload as sufficient substitutes for residents were not always added (AAMC, 2008c; Arora, 2007; Bellini, 2008; Vidyarthi, 2007). Al- though the intensity of work for residents may have increased during the hours they are on service, there are also questions about the educational value of some of the work they currently perform (e.g., blood drawing, routine scheduling of appointments) (Bellini, 2008). Consequently, the committee heard that when residents make mistakes, heavy workload and inadequate supervision are factors as well as fatigue (Bellini, 2008; Con- sumers Union of the United States, 2008). The 2003 ACGME duty hour limits resulted in some residents hav- ing less opportunity to observe patientsâ care from beginning to end and to observe changes in the course of their illness and recovery, all of which are considered by many educators to be essential for quality patient care and effective education (AAMC, 2008c). Therefore, the challenge to the committee was, on the one hand, to suggest ways to minimize any risks of extended duty hours for patients and residents, while, on the other hand, suggesting ways to maximize the presumed educational and patient safety benefits of uninterrupted continuity of resident involvement (especially early in the course of illness or immediately after surgery). Chapter 3 examines the work and learning environment of residents, types of strategies used to try to conform to the 2003 rules, and how the content of resident work and the patient caseload have consequences for adhering to duty hour limits. Chapter 4 draws upon the educational research literature on how people learn, basic tenets underlying GME (as- sumption of responsibility, time for reflection, continuity of care), and the importance of supervision for resident training. It also considers the neces- sity for new models of instruction and assessment for GME within reduced duty hours.
BACKGROUND AND OVERVIEW 39 Resident Safety and Well-Being The overall sense from testimony to the committee from a wide spec- trum of sources was that reduced duty hours had improved resident quality of life (AAMC, 2008c; ABMS, 2008; ACGME, 2008). Residents them- selves said that ACGME 2003 duty hours allowed them to spend more time with family, catch up on personal chores (e.g., do laundry, pay bills), and participate in more leisure activities (e.g., exercise, social events with friends). Some of the committee members heard residentsâ testimony that the time off afforded under current limits was not necessarily spent sleeping because time off is still relatively limited (Resident Panel, 2008). Concerns remain about personal safety issues for residents due to fatigue, including driving incidents and needlestick injuries (AMA Resident/Fellow Section, 2007; CIR/SEIU, 2007; Landrigan, 2007; Public Citizen, 2007). Chapter 5 examines what is known about risks for residents, associated with working long hours and having limited sleep, in terms of their physical and mental health, personal relationships, and professional interactions. Patient Safety The committee appreciates that a complex set of issues is associated with considering the short- and long-term safety implications of making any adjustments to resident duty hours (Cohn, 2008). Carolyn Clancy, direc- tor of AHRQ and the sponsor of the IOM study, crystallized the challenge before the committee from her perspective (Clancy, 2007): At some point .ââ.ââ. we have to acknowledge the fact that a human being can work only so long without sleep deprivation becoming a factor. Research shows that we do not do well in transitions of care .ââ.ââ. but limiting these transitions by having duty hours that are not compatible with human physiology is not the answer. The public perception of an appropriate number of hours for doctors to work often differs from the current reality of residency training (Public Citizen, 2007). One speaker quoted findings from a National Sleep Foun- dation phone poll of 1,010 Americans in 2001 who were asked about the likelihood of their actions if they learned that âthe doctor that is about to perform their surgery has been on duty for 24 consecutive hours.â Sixty- five percent indicated they would very likely feel anxious about their safety, and 45 percent indicated they would very likely ask for another doctor (National Sleep Foundation, 2002; Public Citizen, 2007). Thus, the benefits for residents of long continuous duty hours thought desirable by educators are not always clear to the public and those being treated. Another study at three institutions found that nearly one-quarter of internal medicine
40 RESIDENT DUTY HOURS inpatients surveyed were concerned about resident fatigue and about dis- continuity of care due to patient handovers; these patients tended to be the same set with worries about their care. Patients reported their perception of how many hours residents work in a week (60 hours) and how long they should work (51 hours). In actuality, residents at the three institutions studied worked 67 to 69 hours per week (Fletcher et al., 2007). All who spoke to the issue agreed that shorter duty hours have resulted in more handovers of care, which have been associated with increased risks to patient safety particularly due to poor communication of essential infor- mation (AAMC, 2008c; ACGME, 2007a). This risk may be due as much to the fact that handover techniques are not standardized or optimized for their intended purpose, rather than the fact that handoffs are inherently a systemic hazard. Handovers can also be viewed as a time to reassess patient care and catch previous errors. As is the case for fully trained physicians in practice, residents clearly cannot be on duty 24/7. Handovers of patient care to competent colleagues at appropriate intervals are essential, and strengthening these transfers of patients from one clinician to another is necessary. Several researchers presented information on schedule changes, in- creases in the number of handovers, error rates, and mortality data for the committee to consider (Arora, 2007; Czeisler, 2007; de Virgilio, 2008; Landrigan, 2007; Vidyarthi, 2007; Volpp, 2007). Various observers ques- tioned whether there was sufficient evidence to link resident duty hours to direct harm for patients and whether enough is known about the offset in the risks of more handovers versus less fatigued residents (AAMC, 2008c; AMA Resident/Fellow Section, 2007; Volpp, 2007). Furthermore, a sleep scientist indicated that experience in other industries suggests that limiting duty hours alone is not sufficient to ensure safety. Multiple factors related to time worked, recovery sleep, frequency of shift changes, night work, and the length of continuous work all contribute to the level of human perfor- mance. One type of schedule is unlikely to fit the needs of all residency pro- gram operations (Rosekind, 2008). Whether advocating for shorter shifts or not, speakers encouraged the committee to examine information from sleep science when considering any adjustments to duty hours (AAMC, 2008c; ACGME, 2007a; Bellini, 2008; CIR/SEIU, 2007; Czeisler, 2007). Chapter 6 examines what is known about errors in hospital care at- tributable to fatigued residents and the effects of the 2003 duty hours on patient outcomes. A set of interrelated studies on resident hours of work and sleep are examined in depth to determine what lessons might be learned about resident error and patient safety. Chapter 7 covers what is known about preventing acute or chronic sleep loss and its effects on making errors and what the implications would be for the redesign of resident duty hours and schedules, and the chapter includes the committeeâs recommendations
BACKGROUND AND OVERVIEW 41 for adjusting duty hour limits. Chapter 8 discusses strategies to incorporate residents into an enhanced culture of safety that promotes resident learn- ing about error prevention and about improved communication during handovers. Economic Implications Further adjustments to duty hour limits or the content of resident work will have economic implications for institutions with resident training pro- grams just as the 2003 rules did (Arora, 2007; Daschbach, 2008; Dyne, 2008; Hara, 2008; Liekweg, 2008; Noah, 2008; Opas, 2008). Addition of other personnel to substitute for resident work time was a frequently used but costly strategy employed by academic medical centers in response to the ACGME mandate (ACGME, 2007a). On some services, especially surgical services, resident duty hours were reduced by 20-25 percent as they adapted from workweeks of 100 hours to 80 hours (ACS, 2008). More modifications to the 2003 duty hour limits raised questions of where the additional workforce would come from to substitute for resident- delivered care, and who would bear these costs. Without additional outside funds, many institutions will have a difficult time adapting and continuing to provide care to the same number of patients (AAMC, 2008c; Opas, 2008). The economic benefit that society has derived from the long duty hours of residents working at relatively low wages is substantial, and speak- ers felt that additional funding would be necessary to implement further changes to duty hours or workload (CIR/SEIU, 2007; Daschbach, 2008; Liekweg, 2008; Noah, 2008; Opas, 2008). Chapter 9 concludes the report with recommendations for funding and evaluation. The chapter includes a summary of an economic analysis com- missioned by the committee to estimate the order of magnitude of costs for substituting current resident duty hours with those of other personnel or additional residents according to various scenarios for changes in the duty hour and workload requirements of residents. Finally, the committee is aware of the possibility that even well- c Â onsidered recommendations might have unintended consequences, some of which will be discovered only after they are implemented. It urges an evaluation strategy incorporating data gathering and analysis from initia- tion of the recommended changes with the aim of detecting and minimizing unintended consequences on patient safety or resident education. Limits on Scope of the Study Some important issues necessarily fall outside the purview of this study because of time and other resource constraints. Thus, the committeeâs
42 RESIDENT DUTY HOURS report does not attempt to reorganize the whole healthcare workforce or reform the healthcare system relative to the issue of patient and resident safety but examines the system as it works now. Specifically, while Medi- care funding for GME flows to hospitals rather than other training sites or medical schools and reimbursement rates are highly variable from institu- tion to institution despite a shared teaching mission, it was beyond the scope of this committee to analyze the effects and offer explicit alternative funding strategies. Similarly, the committee does not make a judgment on the concentration of training programs in certain geographic areas or ad- dress the development of integrated delivery systems or other models of care. Furthermore, the committeeâs focus is on residents, as charged, not on students in medical school or physicians in practice. Fellows (doctors who have completed their specialty residency and are continuing training in a subspecialty) are included only to the extent that data concerning them are embedded within the studies of residents. Nonetheless, much of the discus- sion and recommendations may be applicable to fellows as well as residents. The study also does not address the fatigue problems of physicians who continue to work long hours well after their training or of nurses or other healthcare providers who work long hours, because they are not covered by the ACGME duty hour limits. The statement of task for the project did not require detailed cost estimates of every recommendation. Establishing conditions that promote patient safety involves many interacting variables. Duty hours are but one element. The committee necessarily has narrowed its focus with respect to patient safety given the limits of its charge and restricted its exploration to related contextual issues such as supervision, workload, handovers, and teamwork. References AAMC (Association of American Medical Colleges). 2004. The handbook of academic medi- cine: How medical schools and teaching hospitals work, edited by W. T. Mallon et al. Washington, DC: Association of American Medical Colleges. âââ. 2007a. AAMC survey of housestaff stipends, benefits and funding: Autumn 2007 report. Washington, DC: Association of American Medical Colleges. âââ. 2007b. Medical school graduation and attrition rates. Washington, DC: Association of American Medical Colleges. âââ. 2008a. AAMC data book: Medical school and teaching hospitals by the numbers, ed- ited by K. Brandenburg, S. Gaillard, W. Geraci, P. Vassev, and J. Youngclaus. Washington, DC: Association of American Medical Colleges. âââ. 2008b. Help wanted: More U.S. doctors projections indicate America will face short- age of M.D.s by 2020. http://www.aamc.org/workforce/helpwanted.pdf (accessed June 4, 2008). âââ. 2008c. Resident duty schedules and resident education. Presentation by Debra Weinstein to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, May 8, 2008, Washington, DC.
BACKGROUND AND OVERVIEW 43 âââ. 2008d. U.S. medical school enrollment projected to rise 21 percent by 2012: Both new and existing schools will fuel growth. http://www.aamc.org/newsroom/pressrel/ 2008/080501.htm (accessed May 2, 2008). âââ. 2008e. U.S. medical school seniors enjoy most successful âmatch dayâ in 30 years. http://www.aamc.org/newsroom/pressrel/2008/080320.htm (accessed March 21, 2008). 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 Surgery). 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). 2007a. The ACGMEâs approach to limiting resident duty hours. Presentation by Paul Friedmann to the Com- mittee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. âââ. 2007b. Common program requirements. http://acgme.org/acWebsite/dutyHours/dh_ dutyhoursCommonPR07012007.pdf (accessed March 23, 2008). âââ. 2008. 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. 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. AMA (American Medical Association). 2008. State medical licensure requirements and statis- tics, 2008. Chicago, IL: American Medical Association. AMA Resident/Fellow Section. 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. American Orthopaedic Association. 2008. Executive summary of the American Orthopaedic Association (AOA) resident work hours survey. Rosemont, IL: American Orthopaedic Association. ANSR (Americans for Nursing Shortage Relief). 2008. Testimony regarding FY 2009 appro- priations for Title VIIIânursing workforce development programs. https://www.ncsbn. org/FY09_Senate_Testimony.pdf (accessed August 26, 2008). Arora, S. 2007. Duty hours: Past, present and future: The University of Chicago internal medicine experience. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. 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. Bellini, L. 2008. Beyond duty hour reform: Redefining the learning environment. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, May 8, 2008, Washington, DC.
44 RESIDENT DUTY HOURS Brotherton, S. E., and S. I. Etzel. 2007. Graduate medical education, 2006-2007. JAMA 298(9):1081-1096. CIR/SEIU (Committee of Interns and Residents-Service Employees International Union). 2007. Healthcare. 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. Clancy, C. 2007. Remarks by Carolyn M. Clancy, M.D., director, Agency for Healthcare Research and Quality (AHRQ). Presentation to the Committee on Optimizing Gradu- ate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. Cohn, F. 2008. Ethical decision making: A framework for considering resident work hours. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, May 8, 2008, Irvine, CA. Consumers Union of the United States. 2008. Presentation by Elizabeth Imholz to Commit- tee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Council of Teaching Hospitals and Health Systems. 2008. AAMC teaching hospitals and health systems. Washington, DC: Association of American Medical Colleges. 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. Daschbach, C. C. 2008. IOM panel: Resident duty hours cost impact estimates 2003-2008. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. de Virgilio, C. 2008. The 80-hour resident workweek: A perspective from a university- affiliated Los Angeles County surgical training program. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, May 8, 2008, Irvine, CA. Dingell, J. D., J. Barton, B. Stupak, and E. Whitfield. 2007. Letter to William Munier, acting director, Agency for Healthcare Research and Quality. Washington, DC: U.S. House of Representatives Committee on Energy and Commerce. Dyne, P. L. 2008. Emergency medicine: Shift workers and safety net. Presentation to the Com- mittee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Fletcher, K. E., F. C. Wiest, L. Halasyamani, J. Lin, V. Nelson, S. R. Kaufman, S. Saint, and M. Schapira. 2007. How do hospitalized patients feel about resident work hours, fatigue, and discontinuity of care? Journal of General Internal Medicine 23(5):623-628. FSMB (Federation of State Medical Boards). 2007. State-specific requirements for initial medi- cal licensure. http://fsmb.org/usmle_eliinitial.html (accessed March 17, 2008). Hara, J. 2008. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. HHS (U.S. Department of Health and Human Services). 2006. Physician supply and de- mand: Projections to 2020. http://bhpr.hrsa.gov/healthworkforce/reports/physiciansupply demand/physiciansupplyprojections.htm (accessed March 23, 2008). IPRO. 2007. Working hours and conditions post-graduate trainees annual compliance assess- ment: Contract year 6 10/1/06-9/30/07. 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.
BACKGROUND AND OVERVIEW 45 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. Liekweg, R. 2008. Resident work hours: A look at the University of California experience. Pre- sentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. 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. National Residency Match Program. 2008. Advance data tables, 2008 main residency match. http://www.nrmp.org/data/advancedatatables2008.pdf (accessed March 26, 2008). National Sleep Foundation. 2002. 2002 âSleep in Americaâ poll. http://www.kintera.org/atf/ cf/%7BF6BF2668-A1B4-4FE8-8D1A-A5D39340D9CB%7D/2002SleepInAmericaPoll. pdf (accessed October 16, 2008). Noah, M. 2008. Resident duty hour restrictions: Cost impact. Presentation to the Commit- tee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Opas, L. 2008. IOM workshop: Cost impact of duty hour change. Presentation to the Com- mittee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Public Citizen. 2007. Presentation by Peter Lurie to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, De- cember 3, 2008, Washington, DC. 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. Rosekind, M. R. 2008. Managing alertness and performance in 24/7 settings: Lessons learned. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, March 4, 2008, Irvine, CA. Shannon, S. C. 2007. Graduate medical education. http://www.aacom.org/advocacy/ policy%20Statement/06-28-07%20-%20AACOM%20Policy%20on%20Graduate%20 Medical%20Education.pdf (accessed March 21, 2008). U.S. Department of Labor, Bureau of Labor Statistics. 2007. Occupational outlook hand- book, 2008-09 edition, physicians and surgeons, on the Internet. http://www.bls.gov/ oco/ocos074.htm (accessed March 17, 2008). Vidyarthi, A. 2007. Findings and strategies from research literature. Presentation to the Com- mittee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. Volpp, K. G. 2007. Impact of work hour regulation on mortality in teaching hospitals: Some findings. Presentation to the Committee on Optimizing Graduate Medical Trainee (Resident) Hours and Work Schedules to Improve Patient Safety, December 3, 2007, Washington, DC. Weiss, G. G. 2005. Exclusive survey: Productivity takes a dip. Medical Economics 82(22):86- 87, 89, 91-93. âââ. 2006. Exclusive surveyâProductivity: Work hours up, patient visits down. Medical Economics 83(21):57-58, 60, 62-63.