Medical Errors Decreased When Work Schedules for Interns Were Limited, NIOSH- and AHRQ-Funded Studies Find

October 28, 2004
NIOSH Update:

Contact: Fred Blosser (202) 401-3749

The rate of serious medical errors committed by first-year doctors in training (interns) in two intensive care units (ICUs) at a Boston hospital fell significantly when traditional 30-hour-in-a-row extended work shifts were eliminated and when interns’ continuous work schedule was limited to 16 hours, according to two complementary studies funded by the National Institute for Occupational Safety and Health (NIOSH) and the Agency for Healthcare Research (AHRQ).

The studies were published in the October 28, 2004, issue of the New England Journal of Medicine.

Interns made 36 percent more serious medical errors, including five times as many serious diagnostic errors, on the traditional schedule than on an intervention schedule that limited scheduled work shifts to 16 hours and reduced scheduled weekly work from approximately 80 hours to 63. The rate of serious medication errors was 21 percent greater on the traditional schedule than on the new schedule.

In the first research of its kind on the impact of lack of sleep on the safety of hospital care, researchers at Brigham and Women’s Hospital in Boston eliminated the traditional schedule that required interns—doctors who have completed medical school and are finishing their medical training by working in the hospital—to work “extended duration work shifts” of approximately 30 consecutive hours every other shift. Under the traditional schedule, interns in hospital ICUs were scheduled to work approximately 80 hours per week. Under the intervention schedule that was tested in the studies, the “extended duration work shift” was eliminated, and weekly scheduled work hours were decreased by approximately 20 hours. Interns also were encouraged to sleep on their time off and to take naps before night shifts.

“As NIOSH works with hospital administrators, physicians, nurses, and other partners to assess the impact of long working hours on health and performance, studies such as these will help us better identify steps to promote the health and well-being of health professionals, as well as the health and well-being of their patients,” said John Howard, M.D., NIOSH director.

“The impact of sleep deprivation on performance has been well documented in other industries, but studies like these are providing evidence of its impact in health care,” said Carolyn M. Clancy, M.D., AHRQ’s director. “This research clearly demonstrates that changing the design and structure of the systems in which clinicians practice is essential to improving patient safety.”

In this study, “Effect of Reducing Interns’ Work Hours on Serious Medical Errors in Intensive Care Units,” Christopher P. Landrigan, M.D., M.P.H., Director of the Sleep and Patient Safety Program at Brigham and Women’s Hospital, and his colleagues randomly assigned 24 interns to work either the traditional schedule in the cardiac care unit and the intervention schedule in the medical intensive care unit or the converse from July 2002 to June 2003.

The change in work schedule did not diminish interns’ role in ICUs or shift the burden of work to more senior staff, according to the study authors. The number of medications ordered and tests interpreted by interns did not differ significantly. In addition, the error rates for more senior residents and other staff did not increase during the study.

The other study, “Effect of Reducing Interns’ Weekly Work Hours on Sleep and Attentional Failures,” examined the impact of the new work schedule on interns’ sleep patterns and “attentional failures,” characterized by nodding off while on duty, even while providing care to patients. Steven W. Lockley, Ph.D., and his colleagues studied 20 interns each in two 3-week ICU rotations under both the traditional and intervention work schedules. Interns worked an average of 84.9 hours per week on the traditional schedule and 65.4 hours per week on the new schedule. They completed daily sleep and work logs that were validated through observation by study staff. In addition, interns were monitored using polysomnography, a device that can objectively document sleep and attentional failures.

The study found that under the new schedule interns worked 19.5 hours per week less, slept 5.8 hours per week more, and had typically slept more in the previous 24 hours when working. The percentage of work hours preceded by more than 8 hours of sleep in the traditional schedule was 17 percent as compared with 33 percent for the new schedule. Overall, the rate of attentional failures was twice as high at night on the traditional schedule than on the intervention schedule.

The study concludes that interns who worked the intervention schedule were less sleep deprived at work and were able to sleep longer at home, which led to them having less cumulative and acute sleep deprivation. Interns on the new schedule were encouraged to take naps in the afternoon before overnight shifts to mitigate the effects of sleep deprivation on their ability to provide care.

Charles A. Czeisler, Ph.D., M.D., the senior author of both papers and Professor of Sleep Medicine at Harvard Medical School says, “While sleep experts advocate 8 hours of sleep per 24-hour period, it has historically been difficult to achieve in medicine as patient care is an around-the-clock effort. These are the first studies to demonstrate clinically that reducing work shifts and tackling sleep deprivation will help increase attentiveness and reduce medical errors.”

Issues and research needs related to overtime and extended work shifts as potential factors for work-related stress were addressed by NIOSH in a 2004 technical document available on the web at Additional information on NIOSH research pertaining to work-related stress is available at or from the NIOSH toll-free information number, 1-800-35-NIOSH (1-800-356-4674). Copies of the new studies co-funded by NIOSH and other information on sleep deprivation and safety are available at