Medical residents routinely work for 24 hours or longer at a stretch, although their work hours were reduced in July 2003. Many doctors say they can work fine without much sleep because their training emphasizes the ability to make decisions under intense pressure. Researchers at Harvard looked at how interns, residents in their first year of training after med school, did when they worked on a modified schedule designed to increase their sleep time.

What the researchers wanted to know: Are sleep-deprived interns more prone to make serious medical errors than interns on a more sleep-friendly schedule?

What they did: Before the new work-hour limits went into effect, Harvard ran a study in the coronary care and medical intensive-care units of Brigham and Women's Hospital in Boston. Critical-care units have more medical errors, according to some studies, and residents who work in them have particularly long hours—in this case, 30 hours at a stretch. The researchers designed a schedule in which each intern worked relatively short hours: They were scheduled for not more than about 16 hours at a stretch and 60 to 63 hours a week. For the year that ran July 2002 to June 2003, interns rotated through different parts of the hospital as usual. Each was randomly assigned to do the modified schedule in the intensive-care unit and the traditional long-hour schedule in the coronary care unit, or vice versa. Interns filled out sleep logs, and researchers followed them continuously in the hospital to watch for errors.

What they found: Interns slept more on the schedule designed to be sleep friendly—an average of 7.4 hours a day of sleep, compared with 6.6 hours a day when they were on the traditional schedule. They also did better at paying attention on the modified schedule, according to tests that measured eye movements. Of course, none of this matters to patients if interns perform just as well when they're sleep deprived. But interns made 36 percent more serious medical errors on the traditional schedule than they did on the schedule that gave them more time to sleep.

What the study means to you: Sleep-deprived interns make medical decisions all the time; this research suggests that they may not be doing as well as they think they are. The rules limiting hours of work should improve chances of getting sleep for interns across the country.

Caveats: This experiment was carried out at one hospital with an excellent reputation; conditions may be different in other places. Also, now that residents work shorter shifts, they have to hand off their patients more often. Many doctors worry that those transfers produce many more opportunities for bad communication between doctors‑and thus more chances for patients to fall through the cracks.

Find out more: The Accreditation Council for Graduate Medical Education has information on the rules about resident duty hours. The American Medical Student Association has been very critical of residents' long work hours.

Are surgery residents less likely to burn out when they work shorter hours?

Read the articles: Lockley, S.W., et al. "Effect of Reducing Interns' Weekly Work Hours on Sleep and Attentional Failures." New England Journal of Medicine. Oct. 28, 2004, Vol. 351, No. 18, pp. 1829–1837.

Landrigan, C.P., et al. "Effect of Reducing Interns' Work Hours on Serious Medical Errors in Intensive Care Units." New England Journal of Medicine. Oct. 28, 2004, Vol. 351, No. 18, pp. 1838–1848.

Abstracts online: Lockley et al.: http://content.nejm.org/cgi/content/abstract/351/18/1829

Landrigan et al.: http://content.nejm.org/cgi/content/abstract/351/18/1838