Prediction of Late-Onset Psychiatric Disorder in Survivors of Severe Injury: Findings of a Latent Transition Analysis

Objective: There is a growing body of evidence indicating that late or delayed onset of psychiatric disorder following traumatic injury and other psychological trauma is common. This research, however, has not examined factors that pose risks for delayed development of different types of psychopathology or at different time points. Such research has considerable implications for the development of screening, assessment, and intervention practices. This article investigates risk factors for late-onset disorders up to 72 months after a severe injury.

Methods: In this 6-year longitudinal study, 1,167 hospitalized patients with severe injury recruited between April 2004 and February 2006 were analyzed with repeated measures at 3, 12, and 72 months after injury. The Mini-International Neuropsychiatric Interview (MINI) and Clinician-Administered PTSD Scale (CAPS) were employed to complete diagnoses according to DSM-IV. Latent transition analyses with continuous covariates (injury severity, social support, recent life events, and pain) and 1 dichotomous covariate (presence/absence of a psychiatric disorder before injury) were conducted to identify risk factors for transitioning out of a No Disorder class and into one of 3 previously reported psychopathology classes (PTSD [posttraumatic stress disorder]/Depression, Alcohol/Depression, and Alcohol only) between 3 and 12 months (transition 1) and between 12 and 72 months (transition 2) postinjury.

Results: Movement into the PTSD/Depression class was predicted by injury severity at transitions 1 (P = .003) and 2 (P = .017) and social support (P = .006) at transition 1. Past psychiatric history increased the likelihood of moving into the PTSD/Depression class, with anxiety or mood disorders specifically implicated in transition 1. Movement into the Alcohol/Depression class was predicted by social support at transitions 1 (P = .008) and 2 (P < .001) and also by injury severity (P < .001) and pain (P < .001) at transition 2. Movement into the Alcohol class was predicted only by pain (P = .011) at transition 2. A history of a substance use or alcohol use disorder before injury was implicated in movement into both of the alcohol-based classes.

Conclusions: Predictors of developing a delayed-onset psychiatric disorder after severe injury differed by duration after injury and class of disorder. These findings highlight the need to offer targeted screening based on these risk factors to severe injury survivors up to 12 months postinjury, even when they present without disorder at 3 months.