Trauma type as a conditional risk factor for posttraumatic stress disorder in a referred clinic sample of adolescents.

INTRODUCTION: Traumatic experiences that are varied in type and severity may lead to the development of Posttraumatic Stress Disorder (PTSD). Some trauma types present a higher conditional risk for PTSD owing to their nature and impact on growth and functioning. Few studies have investigated the conditional risk of PTSD in clinic referred adolescents in low- and middle-income countries. The aim of the study was to determine the conditional risk for PTSD based on various trauma types (car accidents, other serious accidents, fires, witnessing a natural disaster, witnessing a violent crime, being confronted with traumatic news, witnessing domestic violence, physical abuse and sexual abuse) and to stratify risk by gender.

METHOD: Adolescents exposed to at least one Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) qualifying trauma were referred to a research clinic in Cape Town, South Africa (n = 216). PTSD status was assessed using a clinician administered interview. Conditional risk was determined using backwards stepwise multiple logistic regression analysis for 1) the whole sample, 2) females only and 3) males only. Gender differences in exposure to trauma types were determined using chi-square tests and cross-tabulation.

RESULTS: The prevalence of PTSD was 48.1% in the whole sample. Age (β = .16, p = .048, OR 1.17), fire exposure (β = 2.32, p = .036, OR 10.12) and sexual abuse (β = .93, p = .001, OR 2.54) were significant predictors of PTSD in the whole sample with the model explaining 12.4% of the variance in PTSD status. Age (β = .22, p = .041, OR 1.24) and sexual abuse (β = .87, p = .018, OR 2.39) were significant predictors of PTSD in female participants and explained 9.8% of the variance in PTSD status. Being a victim of a violent crime (β = .78 p = .100, OR 2.19) was the only remaining predictor of PTSD in male participants and showed a trend towards significance. The model explained 7% of the variance in PTSD status.

CONCLUSIONS: The findings underscore the importance of timely identification of trauma, particularly, sexual abuse and violence. Longitudinal tracking of adolescents exposed to different trauma types may identify those in need of treatment and enhance our understanding of the lasting impact of trauma.