PURPOSE: Concerns about medical errors due to the effects of sleep deprivation in residency training have led to sweeping changes in work schedules mandated by the ACGME (Accreditation Council for Graduate Medical Education). Assessments of performance or memory impairment associated with sleep duration are lacking in this population. We investigated the differences in working memory capacity (WMC) and sleep duration of residents between on-call and call-free rotations.
METHODS: Internal Medicine interns and residents were studied during one 4 week call-free rotation and one 4 week call rotation requiring 30 hour shifts every 4th day. We measured WMC using a standardized test - an automated version of the operation span task (Ospan). WMC testing was completed daily between 11AM and 2PM. We also obtained subjective sleepiness scores (0=alert to 10=drowsy) at the time of Ospan testing as well as daily sleep logs to assess sleep duration.
RESULTS: Analysis of results from 12 interns and 11 residents who completed both rotations found significantly more math errors (4.6 vs. 4.1, p = 0.003), more accuracy errors (3.6 vs. 3.2, p = 0.002) and greater sleepiness scores (3.5 vs. 2.5, p = 0.001) when on-call. Total sleep time per day assessed from sleep logs was greater during call-free rotations (32.0%. vs. 28.7%, p = 0.0001). OSPAN scores in the 11 residents were greater during call free rotations but the difference was not statistically significant (59.4 vs. 56.6, p = 0.14).
CONCLUSION: Less sleep time during on call rotations caused more math errors and was associated with more self-reported sleepiness. Sleep debt that occurs during on call rotations, when not “paid off”, is associated with more math errors.
CLINICAL IMPLICATIONS: Studies are needed to evaluate the link between measurable decline in cognitive function due to sleep deprivation and medical errors.
DISCLOSURE: Ashraf Gohar, No Financial Disclosure Information; No Product/Research Disclosure Information