Sleep disturbances can accompany alcohol use disorders during various phases of the disease. This analysis utilized a mixed methods approach to assess whether sleep-related beliefs and/or behavior of individuals who are alcohol dependent were associated with sleep quality both pre- and postdischarge from a clinical research facility providing inpatient alcohol rehabilitation treatment. Individuals with higher self-efficacy for sleep (SE-S) reported better sleep quality at both time points. Individuals with fewer dysfunctional beliefs about sleep had poorer sleep quality at both time points. Individuals with higher unhealthy sleep-related safety behaviors had poorer sleep quality at both time points. In a linear regression model, only the difference in SE-S scores from pre- to postdischarge (β = -.396, p = .01) and the postdischarge Penn Alcohol Craving Score (β = .283, p = .019) significantly predicted the change in sleep quality. Thus, those whose SE-S scores increased and those with lower postdischarge craving scores were more likely to experience a decrease on Pittsburgh Sleep Quality Index scores from pre- to postdischarge even after controlling for covariates. References to behavior or personal factors were often discussed during the qualitative interviews in tandem with the environment. Participants reported both (1) self-medicating anxiety with alcohol and (2) self-medicating the inability to fall asleep with alcohol. Given the success of behavioral sleep interventions in various populations and the unique potential contributions of mixed methods approaches to examine sleep and alcohol use, assessing sleep-related cognitions and behaviors of individuals with severe alcohol use disorders may be important in understanding sleep quality and subsequent relapse.
Keywords: alcohol dependence; alcohol use disorder; mixed methods; self-efficacy; sleep disturbances; social cognitive theory.