A Daily Study of PTSD and Interpersonal Processes

Sarah Burns Campbell

Major Professor: Keith D Renshaw, PhD, Department of Psychology

Committee Members: Timothy W. Curby, Todd B. Kashdan

Research Hall, #92
June 17, 2015, 02:00 PM to 11:00 AM


Posttraumatic stress disorder (PTSD) is associated with changes in relationship functioning, which are in turn hypothesized to influence the trajectory of PTSD symptoms. One focus of recent research is romantic partners’ accommodation of trauma survivors’ symptoms. In the context of PTSD, such accommodation may involve restricting noise to avoid provoking a startle response, limiting social engagements if survivors are nervous when in public, and/or limiting difficult discussions to avoid arguments or emotionally laden topics. Emerging research shows that partners’ accommodation may interfere with survivors’ response to treatment for PTSD, and it is associated with psychological and relationship distress in partners. To date, however, no studies have explored the transactional associations of PTSD symptoms and accommodation across multiple time points to determine the precise direction of effects. The current study used a daily diary format to explore daily associations of PTSD symptoms with partner symptom accommodation over a 2-week period. Daily diaries involve repeated quantitative assessment of the same phenomena across a specified period of time. In addition to allowing for tests of directionality in associations, daily diary studies greatly reduce retrospective bias on self-report measures.

In the current study, participants were 64 male military service members who had deployed at least once since September 11, 2001, and their female civilian romantic partners. Service members had subclinical or clinical levels of PTSD, and all couples were cohabiting. Cross-lagged autoregressive path analyses assessed the stability of both service members’ PTSD and partners’ accommodation across the 14 days, as well as the prospective associations of earlier PTSD with later accommodation and of earlier accommodation with later PTSD (i.e., cross-lag paths). After exploring these associations using total PTSD symptom severity, I also conducted four additional models evaluating each PTSD symptom cluster (intrusion, situational avoidance, emotional numbing, and hyperarousal) independently, instead of total PTSD. I used Bayesian estimation to obtain point estimates as well as Bayesian Credible Intervals (CIs) for paths of interest. In all models, total PTSD and individual PTSD clusters were highly stable across time. Accommodation was also highly stable across time, albeit less so than PTSD. In all models, earlier total PTSD and PTSD clusters were significantly and positively associated with later accommodation. However, earlier accommodation was not significantly associated with later PTSD or clusters, with one exception. The model assessing associations of situational avoidance with accommodation suggested a bidirectional effect, with significant positive associations from earlier situational avoidance to later accommodation and vice versa. Collectively, the results suggest that PTSD symptoms may lead to greater accommodating behaviors in romantic partners. However, partner accommodation seems to contribute only to greater future situational avoidance symptoms in trauma survivors. Broadly, the findings reinforce the notion that PTSD symptoms and relationship behaviors are associated over time, and that accommodation may act to sustain avoidant behaviors in particular over time. Clinicians should attend to romantic partners’ accommodating behaviors in assessing the interpersonal environment and planning exposures for survivors.