King's College London

The aim of this study was to examine the trajectories of posttraumatic stress disorder (PTSD) symptoms over a 14 years period and the risk factors associated with each trajectory.1885 UK military personnel provided information at four time points since 2002. The PTSD Check List-Civilian (PCL-C) was used in all time points. Growth mixture models (GMM) were estimated to examine whether individuals could be clustered into discrete groups with similar trajectories. Multinomial logistic regressions were carried out to investigate factors associated with class membership. The three-class GMM was the most parsimonious solution. This included 90.2% in the resilient class, 4.1% in the improving class and 5.7% in the deteriorating class trajectories. Both the deteriorating and improving classes were associated with childhood adversity (odds ratios (OR) 3.9 (95% CI 2.3, 6.7) and 3.3 (95% CI 2.1, 5.0) respectively) and antisocial behavior (OR 2.8 (95% CI 1.9, 4.2) and 3.7 (95% CI 2.4, 5.8) respectively), alcohol misuse (OR 3.5 (95% CI 2.4, 5.1) and 3.3 (2.1, 5.2) respectively) and longer time since leaving Service in comparison to the resilient group. Those in the youngest group and those in a combat role (OR 0.32, 95% CI 0.19, 0.54) were more likely to belong to the deteriorating trajectory. 10% of the cohort had symptoms of PTSD, of those up to half were symptomatic for most of the follow up period. Those whose score improved did not reach the low scores of the resilient group. Younger age and a combat role were associated with worse prognosis of PTSD.