The issue of work-related sleep loss and consequent fatigue in medical residents continues to cause concern. Some, but not all, studies point to postcall deficits in performance. In this prospective study of 34 pediatric residents, postcall performance was compared during a heavy call rotation (defined as every fourth or fifth night) with light call, light call combined with a blood alcohol concentration of 0.04 gm% to 0.05 gm% per 100 mL, and heavy call with a placebo. The study group included 18 women and 16 men with a mean age of 29 years who worked at an academic medical center. An hour-long test battery was administered during the final week of a light call rotation (nonpostcall); in the final week of a heavy call rotation (postcall); and after alcohol (in the light call group) or placebo (in the heavy call group). Each test session ended with a self-assessment of performance. Outcome measures included the Stanford Sleepiness Scale, a visual analog scale, the Psychomotor Vigilance Task, the Continuance Performance Test, and a simulated driving task.
Heavy and light call rotations averaged 90 and 44 hours per week, respectively. Nocturnal sleep averaged approximately 7–1/2 hours during light call and 6–1/4 hours during heavy call. Compared with light call, heavy call reaction times were 7% slower, commission errors were 40% higher, and performance on the driving simulator was substantially poorer. One driving parameter, speed variability, was greater in heavy call with placebo than in light call with alcohol. There were, however, no significant group differences in reaction time, omission errors, lapses, or off-road riving. Self-assessments correlated with actual performance in heavy call residents with respect to commission errors, lane variability, and speed variability, but not reaction time. In both heavy call groups, more participants rated their performance on the Psychomotor Vigilance Task to be poor. Performance on the Continuous Performance Test did not relate to training year. Residents on heavy call or light call with alcohol made more commission errors on this test, and self-ratings showed the same trend. Far more heavy call residents than light call residents rated their simulated driving performance as poor or worse.
Using a variety of performance tests, medical residents in this study were about as impaired postcall after 4 weeks on a heavy call rotation as were those on light call who had blood alcohol levels of 0.04 to 0.05 gm%. Residents must be aware of postcall performance impairment and the consequent risk to both themselves and their patients. Sleepy residents may not recognize that they are impaired. Residency programs should include discussions of sleep loss, fatigue, and effective countermeasures. Residency programs might consider adopting such measures as alternative call schedules and postcall quarters in which residents can nap.