Pre-enlistment anger attacks and postenlistment mental disorders and suicidality among US Army soldiers
Smith, D.M., Meruelo, A., Campbell-Sills, L., Sun, X., Kessler, R.C., Ursano, R.J., ... Army STARRS Team
Objective To explore the associations of pre-enlistment anger attacks with postenlistment mental health.
Design, Setting, and Participants In this observational cohort study, the Army Study to Assess Risk and Resilience in Servicemembers (Army STARRS) New Soldier Study (NSS) surveyed soldiers entering basic training from April 2011 to November 2012, with a subsample recruited for wave 1 of the STARRS Longitudinal Study (STARRS-LS) (conducted September 2016 to April 2018). Participants were recruited from 3 US Army installations for the NSS survey. Those who were subsequently contacted for STARRS-LS completed the follow-up survey via web or telephone. Prospective analyses were based on a weighted NSS subsample included in wave 1 of STARRS-LS. Data were analyzed from May 22, 2020, to March 17, 2021.
Exposures History of anger attacks at baseline (NSS). Survey responses were used to classify new soldiers as having nonimpairing anger attacks (>2 attacks without interference in work or personal life), impairing anger attacks (>2 attacks with interference in work or personal life), or no significant history of anger attacks.
Main Outcomes and Measures Baseline analyses examined sociodemographic and clinical correlates of a history of anger attacks. Prospective logistic regression models estimated associations of baseline history of anger attacks with new onset and persistence of posttraumatic stress disorder, major depressive disorder (MDD), generalized anxiety disorder (GAD), panic disorder, mania/hypomania, substance use disorder, suicidal ideation, and suicide attempt at wave 1 of STARRS-LS.
Results Of the 38 507 baseline participants (83.0% male and 17.0% female; mean [SD] age, 20.97 [3.57] years), 6216 were selected for and completed wave 1 of the STARRS-LS. Baseline prevalence (SE) of nonimpairing and impairing anger attacks was 8.83% (0.16%) and 5.75% (0.15%), respectively. Prospective models showed that impairing anger attacks were associated with new onset of MDD (adjusted odds ratio [AOR], 1.98; 95% CI, 1.31-2.99), GAD (AOR, 2.39; 95% CI, 1.66-3.45), panic disorder (AOR, 2.02; 95% CI, 1.34-3.05), and suicidal ideation (AOR, 2.11; 95% CI, 1.45-3.07). When baseline psychiatric comorbidity was controlled for, impairing attacks remained associated with onset of GAD (AOR, 1.75; 95% CI, 1.19-2.58) and suicidal ideation (AOR, 1.62; 95% CI, 1.09-2.42). Anger attacks were not significantly associated with persistence of pre-enlistment mental disorders.
Conclusions and Relevance The findings of this study suggest that a pre-enlistment history of impairing anger attacks may be associated with elevated risk of developing GAD, MDD, and suicidality after enlistment. Detection of impairing anger attacks could aid in assessing psychiatric risk in new soldiers.