Sleepiness in medical residents: impact of mandated reduction in work hours

Sleep Med. 2007 Jan;8(1):90-3. doi: 10.1016/j.sleep.2006.08.018. Epub 2006 Dec 6.

Abstract

Objectives: To assess the sleepiness among Family Practice residents using subjective data, the Epworth sleepiness scale (ESS), as well as objective assessment, the multiple sleep latency test (MSLT).

Method: We conducted the study on Family Practice residents at Texas A&M University, Corpus Christi. Sixteen residents, aged 21-40 yrs, were recruited and divided into two groups. The study group was comprised of residents who were on night-float rotation for at least three days and underwent MSLT post-call, and the control group was comprised of residents who were not on night-float and were not post-call. Daytime sleepiness was evaluated using both the ESS and MSLT. Respondents also completed a questionnaire assessing their beliefs regarding effects of sleep loss.

Results: Sleep latencies <10 min were observed in 6 out of the 8 day shift-work interns, while only 1 out of 8 night-float residents had an average sleep latency <10 min (p=0.0195). Following night-float, despite less sleep, residents had sleep latencies which were higher [14.2 (+/-5.0) min] than those in the control group [8.4 (+/-5.4) min, (p=0.043)]. Despite the difference in objective sleepiness, subjective sleepiness (Epworth scores) did not differ. Over 80% of interns reported having driven while sleepy.

Conclusion: Sleepiness continues to be a significant issue among medical residents despite recently mandated reductions in resident work hours. MSLT values in the pathologic range are seen in the majority of residents studied, but in those who were post-night call there seemed to be a paradoxical improvement.

MeSH terms

  • Adult
  • Body Mass Index
  • Disorders of Excessive Somnolence / diagnosis
  • Disorders of Excessive Somnolence / epidemiology*
  • Female
  • Humans
  • Law Enforcement*
  • Male
  • Severity of Illness Index
  • Surveys and Questionnaires
  • Time Factors
  • Workload / statistics & numerical data*