The effect of prior night sleep on simulated driving performance in medical residents

Indications of driving performance negatively affected by poor sleep often occur early in simulated driving experiments and are measured to progress over relatively large epochs of time. How driving performance changes over smaller increments of time as a function of not only sleep quantity but also sleep quality is largely unknown. The overall objective of this work in progress is to examine the trajectory of driving performance in medical residents as a function of the prior night's sleep quality using a high-fidelity driving simulator.

Thirty-two medical residents were enrolled and wore sleep tracking devices for up to 2 weeks. The residents drove a 16-min scenario in a high-fidelity driving simulator. A mixed effects model was used to estimate baseline intercept and slope of simulated driving performance over the course of the drive. The slope of driving performance over the drive and actigraphy-estimated sleep variables from the prior 24 h served as predictors.

Preliminary descriptive findings indicate a wide range of sleep quality metrics in the sample.

This study is among the first to focus on the trajectory of driving performance over small continuous epochs of time when simulated driving performance may first begin to degrade. Further, objective estimates of sleep using actigraphy as predictors of the next day's driving will enhance our understanding of the potential "dose-response" between low sleep quality and crash risk in the following 24 hours.