Alcohol Use Predicts Longer But More Fragmented Sleep: A Daily Diary Study of Alcohol, Sleep, and PTSD in Nurses.

Abstract

Due to the demanding nature of their profession, nurses are at risk of experiencing irregular sleep patterns, substance use, and fatigue. Evidence supports a reciprocal relationship between alcohol use and sleep disturbances; however, no research has examined such a link in a sample of nurses. One factor that may further impact the dynamic between alcohol and sleep patterns is posttraumatic stress disorder (PTSD) symptoms. We investigated the daily bidirectional associations between alcohol use and several sleep domains (i.e., self-report and actigraphy-determined sleep), and moderation by baseline PTSD symptom severity. Over a 14-day period, 392 nurses (92% female; 78% White) completed sleep diaries and actigraphy to assess alcohol use and sleep patterns. Within-person bidirectional associations between alcohol and sleep were examined using multilevel models, with symptoms of PTSD as a cross-level moderator. Daily alcohol use (i.e., ≥ 1 alcoholic beverage; 25.76%) was associated with shorter self-reported sleep onset latency (b = -4.21, p = .003) but longer self-reported wake after sleep onset (b = 2.36, p = .009). Additionally, days with any alcohol use were associated with longer self-reported sleep duration (b = 15.60, p = .006) and actigraphy-determined sleep duration (b = 10.06, p = .037). No sleep variables were associated with next-day alcohol use. Bidirectional associations between alcohol consumption and sleep were similar regardless of baseline PTSD symptoms. Our results suggested that on days when nurses drank alcohol, they experienced longer but also more fragmented sleep.