The association between sleep disturbances and alcohol relapse: A 12-month observational cohort study.

Background: Sleep disturbances are extremely common in alcohol recovery. Systematic research into the relationship between alcohol relapse and sleep disturbances using validated scales and accounting for potential confounders is lacking.

Methods: Patients admitted to a 1-month residential addiction treatment program were administered the Pittsburg Sleep Quality Index (PSQI) at admission/discharge. In addition, the Alcohol Use Disorders Identification Test (AUDIT), Patient Health Questionnaire-9 (PHQ-9), and Pennsylvania Alcohol Craving Scale (PACS) were administered. Patients were contacted every 3 months over 1 year following discharge. Associations of clinical factors with time until relapse were examined using univariate Cox proportional hazard models.

Results: One-hundred and nineteen patients with alcohol use disorders met inclusion criteria (mean age 50.6 ± 13.2 years, 57% male), relapse data were available for 81 patients. Eighty percent of subjects had other psychiatric diagnoses, 66.3% had sleep disturbances at the time of admission, and 57.1% were using hypnotics; 49.1% of patients had sleep disturbances at discharge. Sleep disturbances at admission and discharge were not associated with alcohol relapse at 12 months (OR = 1.00, 95% CI = 0.89-1.13; p = 0.95 and OR = 0.97, 95% CI = 0.86-1.09; p = 0.61). The PSQI sub-scale scores were also not associated with relapse at 12 months. The use of alcohol to help fall asleep (OR = 3.26, 95% CI = 1.33-7.95; p = 0.008), hypnotic use at admission (OR = 4.03, 95% CI = 1.63-9.97; p = 0.002) and age (OR = 1.03, 95% CI = 1.00-1.06; p = 0.035) were associated with relapse over 12 months.

Conclusion: In patients completing a residential treatment program, sleep disturbances as measured by the PSQI were not associated with alcohol relapse at 12 months. Alcohol use as a hypnotic and hypnotic use at admission were associated with subsequent relapse.