Research News

Protected sleep time for US interns

BMJ 2012; 345 doi: (Published 05 December 2012) Cite this as: BMJ 2012;345:e8224


In 2009 general medical interns at two US hospitals had to work one night in four. They began work at 7 am one morning and went home about 1 pm the following day, an overnight shift of roughly 30 hours. They got little sleep during shifts, so researchers conducted trials (one at each hospital) to test a new schedule that gave on-call interns five hours of protected sleep between 12.30 am and 5.30 am. Interns slept about one hour longer during protected nights than they did during control nights (2.86 v 1.98 hours at one hospital and 3.04 v 2.04 hours at the other; P<0.001 for both comparisons). They had significantly fewer nights with no sleep, did better than controls on tests of vigilance in the morning, and were less sleepy during the day. The authors couldn’t tell if protected sleep helped interns make fewer mistakes. It had no discernible effect on patient outcomes, including deaths.

Protecting interns’ sleep proved feasible in both hospitals, although one had to employ an extra resident to cover protected nights. Both hospitals had to pay incumbent residents extra for holding interns’ mobile phones between 12.30 am and 5.30 am.

In 2011, the US regulator of graduate medical education had to decide how to tackle harmful long shifts for doctors in the first year of training. The accreditation council chose shorter shifts (16 hours maximum) over protected sleep time. This new evidence justifies a head to head trial comparing the two, say the authors, and it must be big enough to establish which option is better for patients, as well as doctors.


Cite this as: BMJ 2012;345:e8224