Determining Resident Sleep During and After Call With Commercial Sleep Monitoring Devices

Objective: To demonstrate that commercial activity monitoring devices (CAMDs) are practical for monitoring resident sleep while on call. Studies that have directly monitored resident sleep are limited, likely owing to both cost and difficulty in study interpretation. The advent of wearable CAMDs that estimate sleep presents the opportunity to more readily evaluate resident sleep in physically active settings and "home call," a coverage arrangement familiar to urology programs.

Methods: Twelve urology residents were outfitted with Fitbit Flex devices during "home call" for a total of 57 (out of 64, or 89%) call or post-call night pairs. Residents were surveyed with the Stanford Sleepiness Scale (SSS), a single-question alertness survey. Time in bed (TIB) was "time to bed" to "rise for day." Fitbit accelerometers register activity as follows: (1) not moving; (2) minimal movement or restless; or (3) above threshold for accelerometer to register steps. Total sleep time (TST) was the number of minutes in level 1 activity during TIB. Sleep efficiency (SE) was defined as TST divided by TIB.

Results: While on call, 10 responding (of 12 available, 83%) residents on average reported TIB as 347 minutes, TST as 165 minutes, and had an SE of 47%. Interestingly, SSS responses did not correlate with sleep parameters. Post-call sleep demonstrated increases in TIB, SE, and TST (+23%, +15%, and +44%, respectively) while sleepiness was reduced by 22%.

Conclusion: We demonstrate that urologic residents can consistently wear CAMDs while on home call. SSS did not correlate with Fitbit-estimated sleep duration. Further study with such devices may enhance sleep deprivation recognition to improve resident sleep.