Alcohol use affects sleep duration among military couples

ABSTRACT

Alcohol misuse and sleep disorders are highly comorbid, prevalent among service members and their romantic partners, and affected by relationship interdependence. As most military health research focuses on either service members or their spouses, the current study examined dyadic effects of alcohol use on sleep cycle duration in dating and married military couples (N = 149 dyads), using data from the National Longitudinal Study of Adolescent Health. Person-level results from a series of multilevel path models showed partial support for our focal hypothesis implicating high alcohol use in shortened average sleep duration for the service members and their romantic partners. Specifically, partners of service members who drank more regularly had shorter average sleep durations, as did female service members who drank more alcohol per drinking occasion. At the couple-level, a partner effect indicated that service members’ lower depressive symptoms were associated with their partner’s shorter average sleep durations. In addition, when service members reported relatively high alcohol-related problems, their partner tended to have a shorter average sleep duration. In contrast, when service members’ partners reported relatively high alcohol-related problems, they had a longer average sleep duration. This suggests the consequences of problematic alcohol use for service members’ partners depended on each dyad member’s drinking patterns. Taken together, these findings underscore the importance of viewing military couples as a dyadic unit in research studies and clinical interventions.

KEYWORDS:

• Military couples
• alcohol use
• substance use
• alcohol abuse
• sleep cycles
• sleep disorders
• sleep
• young adulthood
• romantic relationships
• social influence

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. We compared drinking habits of people who were younger than the legal drinking age of 21 (n = 37) to people who were 21 or older (n = 261). There were no mean differences in the frequency, t(238) = .63, p = .53, nor the quantity to which the two groups drank, t (217) = 1.43, p = .15.

Additional information

Funding

This work was supported by the Kent State University.