WASHINGTON Five years after the establishment of across-specialty rules to limit resident work hours, the issue of trainee schedules in teaching hospitals is again under the microscope as a continuing threat to patient safetyand this time an Institute of Medicine committee has been forewarned that specific "workable" solutions are needed.
The schedules in teaching hospitals "belie virtually all the tenets of providing good health care. How can we profess to provide the best possible quality of care when we know we have staff members who are operating at levels of sleep deprivation so severe that they are similar to someone driving under the influence of alcohol?" Dr. Carolyn Clancy, director of the federal Agency for Health Care Research and Quality, asked at a meeting sponsored by the Institute of Medicine.
"If we don't give the members of Congress some workable solutions, they will come up with their own," she told members of the IOM's Committee on Optimizing Graduate Medical Trainee Hours and Work Schedules.
The committee, which held the first of four workshops in December, was formed at the request of Rep. John D. Dingell (D-Mich.) and colleagues on the House Committee on Energy and Commerce as part of an investigation into preventable medical errors. The IOM will publish a report including strategies and actions for implementing safe work schedules in February 2009.
The issue of resident work hours received relatively little attention in the IOM's landmark 1999 report on medical errors, experts said at the workshop, despite several decades of research on the effects of sleep deprivation on human performance and research more specifically showing an impaired ability of interns to read EKGs after long shifts.
Since thenand especially within the past several yearsvarious studies have demonstrated the effects of sleep deprivation in medical residents and have shown that reductions in work hours can reduce errors, physicians told the committee.
A prospective, national survey of more than 2,700 interns, for instance, showed that residents were seven times more likely to report a harmful fatigue-related error when they worked five or more 24-hour shifts in a month than when they worked no 24-hour shifts. They were four times more likely to report a fatal error.
And in a randomized controlled trial, residents had twice as many EEG-documented attention failures at night when working the traditional schedule of 24- to 30-hour shifts than when working an "intervention" schedule of a 16-hour maximum. Both studies were led by researchers at Harvard University.
Dr. Christopher P. Landrigan, who directs the Sleep and Patient Safety Program at Brigham and Women's Hospital, Boston, said the Harvard research has also shown that residents working 24- to 30-hour shifts make five times as many serious diagnostic errors as do those scheduled to work 16 hours or less. They're also twice as likely to crash their cars, and they suffer 61% more needlestick injuries, he told the IOM committee.
Limits instituted by the American Council on Graduate Medical Education in 2003 mark shifts of 2430 hours as acceptable. The council's "common duty hour standards" call for a 24-hour limit on continuous duty, with an additional 6 hours allowed for continuity and the transfer of care, as well as an 80-hour weekly limit averaged over 4 weeks. Programs can request an increase of up to 8 hours a week and can apply for further exemptions.
Residents must also have a minimum rest period of 10 hours between duty periods, 1 in 7 days free from patient care responsibilities, and in-house call no more than every third night, averaged over a 4-week period, the standards say.
The American Council on Graduate Medical Education says that it has issued citations to individual programs for duty hour violations and has done resident surveys that demonstrate a compliance rate of 94%.
Others argue, however, that enforcement is inadequate and that an independent body is needed to ensure compliance with the rules. Culture and tradition are so entrenched, they say, that too little has changed and that residents routinely underreport hours for fear of retaliation.
"I'm a resident who said one thing on a survey and did another thing in real life," Dr. Sunny Ramchandani, past chair of the AMA Residents and Fellows Section, told the IOM committee. "I'd have a 30-hour shift, work at least 34 hours, and report 16."
Part of the problem is that residents' workloads tend to remain the same even when shifts are shortened. Surveys of faculty and program directors taken by the American College of Physicians indicate that, even under the current rules, there is often less time for both formal and informal education, less time for ambulatory training, less time at the bedside, and a loss of continuity in care.