Abstract
Background: Studies have found increasing rates of excessive drinking among women and high levels of psychiatric comorbidity. This follow-up study replicated our predominantly male study to identify multimorbidity clusters among the subsample of military women with postdeployment at-risk drinking and to examine the association of cluster membership with health care utilization and military readiness. Methods: Participants consisted of Army women who screened positive for postdeployment at-risk alcohol use (n = 12,066). Latent class analysis (LCA) categorized participants into classes using 31 indicators of mental and physical health, alcohol, and tobacco use. Cox proportional hazard models estimated the relative hazards of classes with health care utilization and military readiness outcomes. Results: LCA identified a 5-class model as optimal: Class 1-relatively healthy/pain (50.6%), Class 2-pain/tobacco (16.7%), Class 3-mental health (MH)/pain/sleep (16.7%); Class 4-heavy drinking/MH/pain/tobacco/sleep (8.9%); and Class 5-heavy drinking/pain/tobacco (7.2%). Pain diagnoses (i.e., musculoskeletal, headache/migraine, and visceral/pelvis) and tobacco use were particularly elevated in certain classes and differed in nuanced ways from the full-sample study. Compared to Class 1, women in Classes 2-5 had increased risk for health care utilization and adverse military readiness outcomes. Class 4 had the highest odds of adverse readiness outcomes, including a higher hazard for self-harm/suicide attempt compared to the comparable class from the full-sample study. Conclusions: This study of women soldiers with at-risk postdeployment alcohol use found similarities with the predominantly male full-sample study, with some important differences, including a higher prevalence of pelvic pain and migraines and a greater hazard for self-harm/suicide attempt. Findings inform the need for targeted, integrated treatment for women soldiers at the highest risk for negative outcomes.