Residency Overwork and Changing Paradigms of Service

In this issue, Green [1] reviews key issues in residency overwork, including the effect of overwork on patients, on the residents themselves, and on the hospitals that sponsor the residents' training. He concludes that residency overwork is both prevalent and harmful and that the most persuasive argument against long hours is the negative effect they have on the professional and personal growth of residents at a critical time in their development.

Green reviews several studies that either support or reject the thesis that overwork harms patients. These studies include indirect measures of performance among sleep-deprived and rested residents and self-reported survey results attributing important clinical mistakes to fatigue [2, 3]. Green points out that, conflicting conclusions notwithstanding, intuition suggests that an exhausted physician does not function as effectively as a rested colleague. I agree. Most of us can probably identify many substantial mistakes we have made while caring for patients that resulted from inordinate fatigue. Mistakes of greatest concern, such as a preventable death or a permanent disability, are not likely to be shown by analytical studies. As with the relation between chloramphenicol and aplastic anemia, in which the risk for the disorder with any given dose of the drug is low, uncommon events like these are usually recognized by anecdote.

The issue of whether sleep deprivation harms patients may be moot. Patients trust their physicians and expect physicians to treat them effectively and safely. As with airline pilots, the public can understand and accept tiredness but not exhaustion; as a result, we have regulations governing the work hours of persons in such positions. Given an average U.S. work week of 40 hours, we can hardly expect most people to support the idea of training programs that require as much as 110 to 130 hours per week in the hospital, even if much of this time may be spent resting. Training programs involving such hours were repeatedly encountered during hearings conducted by New York State's Bell Commission, which investigated residency training conditions several years ago (personal observation).

Green also notes that the cheap labor provided by trainees, viewed by many as a practical necessity to help control costs in teaching hospitals, is used as a rationale for long hours. Replacing residents with physician's assistants or nurse clinicians would be costly, given the higher salaries of these replacements and the fewer hours they work. However, this argument is rapidly losing its justification. As teaching hospitals reduce the number of residency and fellowship positions to become more cost-competitive and to respond to the oversupply of specialists, many positions are not being filled by other providers. Attending physicians, who had become increasingly remote from direct patient care, are assuming more responsibilities, as are floor-based nurses and technical support staff.

The redistribution of responsibilities among providers of patient care in teaching hospitals is just one of the sweeping changes occurring to reduce costs, improve quality, and reform education. These changes will probably profoundly affect the lives of residents. If these changes are implemented thoughtfully, with due recognition of both patient care and education goals, the effect should be supportive, not destructive.

What are these changes? One is the use of approaches designed to reduce variation in medical practice. Variation often signals opportunities to reduce cost and improve quality. These approaches include care pathways, clinical guidelines, protocols, and case management. Many, perhaps most, teaching hospitals are now implementing them. Some educators argue that such initiatives lead to cookbook medicine and erode the problem-solving experiences of physicians in training. I disagree. After making recent visits to 10 academic medical centers to determine how they are responding to the rapidly changing external environment, I concluded that the thoughtful introduction of the strategies designed to improve quality and efficiency can benefit residents. This is particularly true if residents are involved in the development and implementation of such approaches. These strategies reduce the amount of time residents must spend on unproductive paperwork, transfer responsibility for discharge planning to appropriate staff, and educate staff about best practices. On this last point, one continues to hear the argument that residents learn best by making mistakes [4]. Isn't it better for residents to learn by being aided in avoiding mistakes? If the educational paradigm that residents learn best by being allowed to make mistakes ever had merit, such merit is difficult to sustain today. Mistakes will always be made by those in training. The trick is to encourage systems that prevent or at least recognize errors while still promoting resident responsibility.

A second organizational change evolving in teaching hospitals is the integration of residents into care teams that manage patients over time and that are not limited to the hospital phase of an episode of illness. The ability to track and participate in the longer-term management of many, if not all, patients (supported by the necessary information management infrastructure) will build on the educationally sound but more narrowly focused resident clinic system of the past. It should help to alleviate the problems associated with the hospital-based block assignment of residents. These problems include not only continuous duty but also an excessive pace, an equally important contributor to exhaustion not mentioned by Green.

Participation in long-term management will also strengthen residents' understanding of the importance of more relevant markers of treatment success than death and complications, including improved functional status, quality of life, and patient satisfaction. Finally, long-term management should provide residents with experiences across settings that better reflect academic medical centers of the future, that is, integrated health care delivery systems that provide a full array of services to defined populations in addition to continuing to function as tertiary referral centers for evaluating and treating complex problems. Hospital-based assignments will occupy a much smaller proportion of the total residency curriculum as technology permits more care to be provided safely and effectively in settings outside the hospital.

Green is correct in suggesting that the chief problem with resident overwork is that it interferes with the development of physicians as mature professionals. But, as he also points out, the problem goes beyond tired residents viewing patients with disdain, trying to avoid new ones, or getting rid of those they already have. These attitudes are not likely to be lifelong. Most of us know many physicians who spent exhausting years as residents but still reflect the attributes of the caring physician so eloquently articulated by Francis Peabody [5]. The larger problem is the failure of training programs to meet the needs of residents at a particularly vulnerable time in their development as physicians. Depression, suicide, the development of substance abuse, stunted marital relationships, and the failure to develop balance and perspective in their lives are some of the sad outcomes. Most are preventable, and none requires that we jettison the principle that hard work is always expected of dedicated professionals, all the more so during their training, when valuable experience can be gained.

The quality and intensity of medical student and residency training need not be compromised as teaching hospitals respond to pressures to reduce costs and maintain or improve quality. On the contrary, if educators and managers participate in breakthrough thinking and involve residents in the process, threats can be turned into opportunities for improved educational experiences. An important fringe benefit will be a reduction in the number of problems associated with residency overwork.

@copy; 1995 American College of Physicians

• Copyright ©2004 by the American College of Physicians