Prior-night sleep as an event-level predictor of alcohol-induced memory loss

Alcohol-induced memory loss (also known as "blackout") is prevalent and confers risk for other adverse drinking outcomes. Sleep health is implicated in memory and cognitive functioning broadly, but proximal impacts of sleep on next-day memory loss are understudied. We hypothesized that worse sleep health the night before drinking (fewer hours, misaligned timing, worse sleep quality, greater sleepiness) would (a) increase the odds of next-day memory loss and (b) moderate day-level associations between estimated blood alcohol concentration (eBAC) and blackouts, such that lower eBACs would be required to experience memory loss following nights of worse sleep.

Heavy-drinking young adults (N = 203, 57 per cent female) in the United States completed baseline self-report measures, followed by 28 days of ecological momentary assessments.

Memory loss was reported on 15.4 per cent (n = 366) of drinking days (n = 2380). Multilevel models, controlling for eBAC, indicated that fewer hours of sleep (OR = 0.84 [95% CI: 0.76, 0.92], p < .001), more deviation than usual from mid-sleep timing (OR = 1.43 [1.25, 1.65], p < .001), and worse sleep quality (OR = 0.72 [0.60, 0.85], p < .001) on the night before drinking significantly increased the odds of next-day memory loss. Sleepiness at the start of drinking events was unrelated to memory loss [OR = 0.91 (0.81, 1.03), p = .13]. Sleep parameters did not significantly moderate associations between eBAC and memory loss (p ≥ .08).

Prior-night sleep health increases risk for alcohol-induced memory loss. These behavioral data extend experimental findings to young adults' real-time drinking outcomes. We encourage prevention and intervention efforts highlighting poor sleep as a potential contributor to alcohol-induced memory impairment.