Changes in alertness over consecutive workdays for internal medicine interns: a secondary analysis of the icompare trial

Abstract Introduction Little is known about the impact of cumulative workdays on medical residents' alertness. The purpose of this study was to examine changes in alertness over consecutive workdays following a day off for internal medicine interns. Methods This is a secondary report of a randomized non-inferiority trial of 12 internal-medicine residency programs assigned to either standard duty-hour (80h workweek/16h shifts) or flexible (80h workweek/no shift-length limit) policies. Interns were followed for 2 weeks during inpatient rotations. Each morning, alertness (number of Brief Psychomotor Vigilance Test [PVT-B] lapses) was assessed, and interns selected the type of shift worked (day-off, days, nights, beginning/ending extended overnights, or other). Sleep duration (actigraphy) was averaged each 24h day. For this analysis, interns were included if they had ≥1 day-off followed by at least 3 workdays, and had no flagged PVT-B results for non-adherence. To examine the longitudinal effect of consecutive workdays on alertness, a generalized linear mixed model with random intercept and slope, and Poisson distribution was used to determine the rate of PVT-B lapses for up to 4 work days following a day off, controlling work shift type, sleep duration, and policy, with sleep and shift type interaction, and linear spline to account for the change in slope after the 2nd workday. Results N=328 interns were included (mean age 27.8±2.2y, 49% males). Mean±SD number of PVT-B lapses were 3.4±4.5, 4.2±5.6, 5.3±6.6, 4.8±5.8, and 4.7±6.0, and mean±SD sleep duration was 9.0±1.9, 6.9±1.3, 6.5±2.1, 6.6±1.8, and 6.9±1.7 hours for a day off and workdays 1-4 respectively. Rate of lapses increased by 1.1 lapse/day from a day off to the 2nd workday (p=0.004; 95%CI: 1.03-1.18), and then significantly decreased from days 2-4 at a rate of 0.89 lapses/day (p<0.0001; 95%CI: 0.85-0.92). Patterns of change in the rate of lapses were similar to changes in sleep duration, where, from baseline, every 1h longer sleep duration was associated with 0.91 fewer PVT-B lapses (p<0.0001; 95%CI: 0.93-0.97). Conclusion Both sleep and subsequent alertness were negatively impacted when returning to work following a day off for interns in this study. After two workdays, sleep duration appeared to increase again, with observed improvements in alertness. Support (If Any) Funded by the NHLBI and American Council for Graduate Medical Education. M.C. is supported by NIH/NINR (K99 NR019862).