Differences in sleep between concussed and nonconcussed college students: a matched case-control study

To describe sleep 2-3 days postconcussion through symptom recovery and make comparison to well-matched nonconcussed controls.

Twenty college students were physician diagnosed with a concussion and compared with 20 nonconcussed controls matched on age, sex, physical activity, and sleep quality. A wrist-worn ActiGraph GT9X Link was provided during initial evaluation (within 72 hr postinjury for concussed) and worn continuously until symptom resolution (duration matched for nonconcussed). All participants completed a sleep symptom severity checklist, the Pittsburgh Sleep Quality Index (PSQI), and the Epworth Sleepiness Scale (ESS). Separate 2(group) × 3(time points) mixed model ANOVAs were conducted to compare actigraphy sleep outcomes (sleep onset latency [SOL], normalized wake after sleep onset [WASOnorm], total sleep time [TST], sleep efficiency, and number of awakenings) across recovery stages (2-3 days postinjury, mid-point, and end of recovery). Intraindividual coefficient of variation was calculated for each sleep outcome. Mann-Whitney U tests compared PSQI global score, ESS total score, and sleep symptom severity between groups (α = 0.05).

At 2-3 days postinjury, concussed individuals took longer to fall asleep compared with controls (p = 0.002). Greater intraindividual variability in WASOnorm (p = 0.017) and TST (p = 0.044) existed in concussed individuals across recovery. Poorer sleep quality (p < 0.001), excessive daytime sleepiness (p = 0.014), and worse sleep symptoms (p < 0.001) existed in concussed compared with controls.

Concussed individuals took longer to fall asleep 2-3 days postconcussion, experienced greater variation in sleep fragmentation and sleep time until symptom resolution, and reported worse sleep quality. Our preliminary findings may guide researchers interested in better understanding sleep postconcussion.