Early and Recent Military and Nonmilitary Stressors Associated with Posttraumatic Stress Symptoms Among Military Service Members and Their Spouses

Background and purpose: With some exceptions, prior work with military couples has focused on the effects of military stressors on mental health outcomes, downplaying the influence of stress exposures not specific to military service. Using a stress process framework, this study considered the association of early stressors (e.g., adverse childhood experiences [ACEs]), military stressors (e.g., deployment), and recent nonmilitary stressors (e.g., financial difficulty) with posttraumatic stress symptoms (PTSS) among service members (SM) and their spouses (SP). The study was guided by four hypotheses: 1) ACEs, nonmilitary and military stressors would be independently associated with PTSS for SM and SP; 2) ACEs would be the strongest predictor of PTSS and would magnify the effect of other stressors on PTSS; 3) Stressors experienced by each member of the dyad would be associated with their partner’s PTSS; and 4) Estimated effects of stressors on PTSS would be larger for those with higher levels of PTSS.

Methods: This study employed Millennium Cohort Family Study (MCFS) data from 3,314 SM and SP dyads. The MCFS is a longitudinal computer-based survey of married military couples, with an initial panel enrolled from 2011 to 2013. Accounting for covariates, multiple linear regression, dominance analysis, and quantile regression were employed to consider the effects of SM and SP ACEs, military stressors, and recent nonmilitary stressors on their own and their partner’s self-reported PTSS. Dominance analysis is a state-of-the-art regression method for determining predictor relative importance. Quantile regression explored the differential effect of stressors on PTSS as a function of where on the PTSS distribution scores occurred with a focus on the 20th, 50th and 80th quantiles.

Results: For both SM and SP, the most dominant predictor of PTSS was their own ACEs (dominance index=52.04 and 64.35 for SM and SP respectively), followed by nonmilitary stressors (dominance index=24.09 and 22.66 respectively), far exceeding the effects of military stressors. Couple crossover dynamics were evident as SP ACEs and nonmilitary stressors significantly predicted SM PTSS, and SM ACEs predicted SP PTSS. ACEs also multiplied the effects of SM (product term b = 0.92, p < .05) and SP (product term b = 0.81, p < .05) nonmilitary stressors on PTSS. Lastly, quantile regression results suggested that stressors had less impact on PTSS at the lower end of the PTSS distribution, but stronger effects at the upper end of the distribution.

Conclusions and implications: Results underscore the importance of capturing overall stress burden in military families to understand their risk for adverse mental health outcomes. A narrow focus on the effects of military stressors alone would likely significantly underestimate mental health risk in this population. Findings emphasize the essential role that exposure to early adversity plays in understanding PTSS among both SM and SP, as ACEs strongly predicted PTSS and exacerbated the effects of other stressors on PTSS. Providers should assess for early adversity among both SM and SP and consider the provision of services at the couple or family level, given the potential for the transmission of stress within couples.