New Requirements for Resident Duty Hours

Many physicians recall their residency experiences as the best of times and the worst of times. The best refers to the camaraderie and shared values, the exponential growth in knowledge and skills, and lifelong friendships formed. The worst includes the demanding nature of the work, the difficulty of dealing with illness and death, and last, but certainly not least, the long hours. Extensive duty hours are a necessary component of resident education and a public symbol of a profession that requires hard work and dedication.

In June 2002, the Accreditation Council for Graduate Medical Education (ACGME) granted preliminary approval to new duty-hour standards for residency programs in all specialties.¹ The ACGME accredits 7800 US residency programs in 118 specialties and subspecialties. It establishes educational requirements and assesses compliance with them through its accreditation process. The new standards, which become effective in July 2003, will limit hours in all specialties to 80 a week. They will also require a rest period between duty periods. Continuous duty will be limited to 24 hours, with an added period of up to 6 hours for continuity and transfer of care and didactic activities.

The decision to set a global limit on duty hours builds on the ACGME's 20-year history of addressing the issue of resident hours. In the early 1980s, the ACGME requirements for programs in internal medicine and pediatrics began to address resident duty hours, with statements such as: "Hospital duties should not be so pressing or consuming that they preclude ample time for other important phases of the training program or for personal needs."² In 1987, the ACGME adopted several requirements for all accredited specialty programs.³ These have been incorporated into the current duty-hour standards and include at least 1 day in 7 free of patient care responsibilities and in-hospital call scheduled no more frequently than every third night. Five specialties have set more restrictive requirements that include a weekly duty-hour limit.

Three developments convinced the ACGME in 2002 of the need for a more comprehensive set of duty-hour standards for all specialties. The first is a changing health care delivery system, with increasingly ill patients and the resulting greater demands on residents. Second, there is a growing public opinion that long duty hours compromise patient safety and resident well-being. The members of the Work Group were concerned that governments could decide to regulate residents' duty hours if the medical education community failed to address these issues. Third, research has resulted in better data about the effects of sleep deprivation on residents' clinical and educational performance.^{4 - 8} Together, these factors prompted the ACGME to charge a Work Group with formulating common standards for resident duty hours and providing recommendations for their enforcement and related activities.

The 14-member Work Group included representatives from emergency medicine, family practice, internal medicine, obstetrics-gynecology, pediatrics, psychiatry and surgery, 2 members of the public, and 2 residents. The group sought input from academic and clinical communities and the public in its deliberations and in its consultations with constituent groups as the report evolved. Following preliminary approval of the standards in June 2002, the ACGME again solicited comments on the proposed standards as part of the customary approach of refining and clarifying the standards before final approval. This process will continue through the July 2003 implementation date.

The exchange of ideas across multiple constituencies highlighted the gulf between the perceptions of the resident education community and the general public. Two guiding principles emerged from the Work Group's efforts to build a consensus between educators and the public. First, standards need to be flexible and sensitive to the educational and patient care needs of the 26 core specialty disciplines that the ACGME accredits. Second, the public is concerned about the negative effects of sleep deprivation on patient safety and resident well-being, which has resulted in demands for regulation of work hours. The public could interpret a lack of action as the medical profession's abrogating its responsibilities, disregarding public opinion, and ignoring the scientific evidence on sleep and performance.

The Work Group was sensitive to the tension between time to allow residents to participate in learning experiences in clinical settings and duty-hour limits to allow them to be alert and able to benefit from the experiences. In comparing resident duty hours and regulation of hours in other industries, Markel recently stated: "What has not changed is that learning how to be a doctor is a far cry from doing a shift at a factory. You need to be there for long periods of time to see the progress of an illness."⁹ We recognized that, although shorter hours will make residents more alert and better able to learn, they will also result in less exposure to patients and the course of their illnesses.

Unfortunately, there are no scientific answers to resolve the tension between the optimal hours for education and hours for rest. Research has not established an exact number of hours a week below which residents may safely and effectively learn and participate in patient care. Thus, the Work Group chose 80 hours as the number that best reflected the consensus of its members. The group also recognized the need for flexibility to allow residents to participate in educational activities. Although there is no universal consensus regarding limits on resident hours in the medical community, the Work Group felt that a sizable segment supported some weekly limit. A limit of 80 hours a week has been common in position statements as well as calls for regulatory and legislative intervention.^{10 - 12}

Although such a limit may result in diminished continuity of care and clinical exposure, this must be weighed against the need to avoid sleep deprivation and to give residents time for learning, recovery, and personal needs. The Work Group found that research on sleep and performance supported the benefits of a schedule that puts a reasonable limit on continuous duty and thus allows for enough rest to avoid long-term, progressive sleep debt. The recommended standards address these goals with required rest between duty periods, call scheduled no more frequently than every third night, and at least 1 full 24-hour duty-free period a week. The work group believed that with a reasonable workday and a limit on continuous duty periods, accomplishing these goals should be possible within an 80-hour weekly limit. New York, which accounts for approximately 15% of all US residents, and internal medicine, the largest accredited specialty representing more than 20% of all physicians in residency (Accreditation Council for Graduate Medical Education, unpublished data, July 2002), currently limit resident hours to 80 a week. Thus, there is considerable experience with the educational and patient care aspects of this approach.

The new ACGME standards seek to preserve the strengths of the current system and balance the negative aspects of sleep loss on residents' performance against the time spent with patients, which is necessary for learning. They recognize the findings on sleep and performance by calling for program directors, faculty, and residents to familiarize themselves with the practical implications of this research, including recognizing the signs of sleep deprivation and taking action to relieve residents when fatigue may affect patient care or learning. The standards offer some flexibility to let residents participate in educationally valuable activities beyond the 80-hour limit, if these are vital to their learning. If programs have a sound educational rationale, they may seek approval from the sponsoring institution and the residency review committee to extend duty hours by up to 10%. Program directors are expected to demonstrate that all hours in the given week—as should all hours in any organized program of education—contribute to resident learning and that they will not compromise the safety of the residents or their patients.

As Work Group members engaged in deliberations, we were frequently reminded that residents are vital to the provision of patient care in a system that faces workforce shortages. Any effort to reduce duty hours must address the critical issue of providing clinical services with a reduced number of resident hours. To help implement the new standards, the ACGME has strengthened institutions' responsibility for oversight and monitoring of duty-hour practices, and it is committed to consistent enforcement. Essential elements of the enforcement protocol are surveying residents, collecting detailed information related to duty-hour compliance during the accreditation site visits related to duty hours, and shortening the amount of time for programs with excessive duty hours to come into compliance. As necessary, the ACGME will conduct special on-site inspections to verify the success of their efforts to reduce duty hours. Failure to comply may adversely affect programs' accreditation. We believe that use of adverse accreditation actions against programs with compliance deficits produces improvement. Probation can hamper recruitment, and withdrawing accreditation—the ultimate adverse accreditation action—has significant consequences for the sponsoring institution. Promoting adherence to the standards is crucial, yet the success of the effort will depend on the willingness of programs and institutions to make changes in their systems of care to create a balance of learning and patient care.

Unfortunately, there are no ready solutions for how programs and institutions can provide a given volume of clinical services in a system in which resident hours are constrained. In the United States and other countries, there are institutions where resident hours are limited. In the United States, New York has regulated resident hours since 1989, prompted in part by the death of a young woman in a New York City hospital in 1984.¹³ New York's experience has demonstrated that regulation is not a perfect solution and that reducing resident hours is not simple, even years after the promulgation of regulation. In June 2002, the State Department of Health cited 54 of the state's 82 teaching hospitals for some degree of violation of the duty-hour standards.¹⁴

Most models of resident-hour reductions have replaced residents with other practitioners. This approach is costly and may be complicated by shortages in health care workers. A rethinking of care in teaching settings and the roles residents play in it may be necessary. The ACGME, along with other organizations in the academic community, plans to foster the development of new models for providing care in a system in which resident duty hours are capped and to share these approaches with programs and sponsoring organizations. The issue goes beyond replacing residents as providers of clinical services. Residency is a multiyear experience that combines exposure to patients and didactic learning with other learning modes in a vastly changed delivery system. It also has some of the attributes of graduate education in other fields, including the need to comprehend and contribute to a rapidly growing body of knowledge. Duty hours, the elements of the clinical environment, and the learning model are inexorably linked, which needs to be considered in the effort to reduce resident hours.

The effective date of July 2003 will give institutions 1 year to identify new approaches to care that preferably can be used in other settings also. There are efforts under way in this area, spearheaded by the Association of American Medical Colleges. Also, resident hours cannot simply be limited without consideration of supervision, curricula, and how hours in a given rotation contribute to resident learning.

The problem of duty hours is complex, and any intervention must consider several interrelated factors. Although residents will still need to work long hours, such a practice may adversely affect their family life, attitudes toward the profession, moral and social development, and their safety and the safety of their patients. Making good use of the time and capabilities of all health care professionals, including residents, should be a central goal of ongoing reform. We believe that limits on duty hours can enhance resident education, if they are applied wisely and with consideration for the primary aim of residency as a time for learning.