Updated

A small study suggests surgeons in training are still tired enough to raise their risk of making significant errors, despite new guidelines limiting their work hours.

Researchers found that orthopedic surgical residents at two Boston-area hospitals were averaging five and a half hours of sleep a night, and so fatigued during waking hours that a quarter of the time their "impairment" was equivalent to being legally drunk.

The study was conducted in 2010 and throughout most of 2011, and while it was underway rules limiting the work hours of the novice surgeons were enacted by a graduate medical education body. But they may not be enough, researchers agreed.

"I think that this study suggests that it's more than just regulatory changes," said Dr. Frank McCormick, of the Harvard Combined Orthopaedic Residency Program in Boston and lead author of the new study.

McCormick and his colleagues focused on identifying when during a resident's waking hours they were most impaired by fatigue.

Knowing when surgical residents are most likely to be dangerously tired, the group explains in the Archives of Surgery, may help to target interventions that could prevent medical mistakes.

For their study, McCormick and his fellow researchers analyzed information collected from 27 orthopedic residents at Massachusetts General Hospital and Brigham And Women's Hospital.

At the outset, each resident was asked about his or her sleep and exercise habits and use of sedatives, alcohol and other stimulants.

The residents also kept daily sleep and work logs for two weeks while they wore a special "actigraphy" wristwatch that recorded movements to gauge the person's activity level.

Overall, residents got an average of 5.3 hours of sleep per day -- though the amount of sleep each resident got ranged widely, from 2.8 hours to 7.8 hours.

Based on the activity recordings, the researchers also determined that -- compared to data on a group of well-rested residents -- the study participants were only functioning at about 70 percent of their mental effectiveness during 27 percent of the time they were awake.

According to the researchers, 70 percent mental effectiveness is equivalent to having a 0.08 blood alcohol level, which is considered legally drunk in most states.

Based on that level of fatigue, McCormick and colleagues calculated the residents had a 22 percent greater risk of causing a medical error than alert, well rested doctors.

"Fatigue was higher than we expected, especially within specific subgroups," said McCormick.

For nearly half their waking hours -- 48 percent -- the residents were mentally fatigued, but functioning above the 70 percent level.

In general, night-shift workers were worse off than the day-shift workers.

Residents working nights slept on average about 5.1 hours a day and functioned at less than 70 percent of their mental effectiveness 32 percent of their time awake.

Residents on day shifts slept about 5.7 hours and functioned at or below 70 percent effectiveness 17 percent of the time.

Day shift residents had a projected 19 percent increase in their risk of making a medical error, while night shift residents had a 24 percent increased risk.

McCormick cautioned, however, that these risks of medical error are only predictions based on fatigue level. The researchers did not assess how many medical errors were actually caused by tired surgical residents in the study.

As a result, "It's difficult to actually quantify the amount of medical error that we're preventing," he told Reuters Health.

The study had other limitations as well, the authors note. Only 40 percent of the orthopedic residents at the two hospitals completed the study. And the researchers are not sure to what degree their findings can be applied to other types of medical residents.

In the same journal, Dr. Thomas Tracy Jr. of Hasbro Children's Hospital and Brown University in Providence, R.I. wrote in an editorial that discovering surgical residents are fatigued during certain periods "is not startling" but its pervasiveness is "a finding we simply cannot avoid and may have paid lip service to in the past."

Last July, the Accreditation Council for Graduate Medical Education, the body that oversees resident training, enacted new rules stating that first-year residents should work no more than 16 hours straight. The rules did not impact more experienced residents, however. They are allowed to work 28-hour shifts.

The Institute of Medicine estimates the rules would cost close to $1.7 billion dollars to implement -- costs stemming mostly from hiring additional staff.

McCormick suggested targeting tired residents for preventive interventions may be more cost effective, and he said there have already been changes at the hospitals involved in the study.

"We're currently switching around our schedule, eliminating our night flow and continuing to monitor fatigue," he told Reuters Health.