Post-traumatic stress disorder rates in trauma-exposed children and adolescents: updated three-level meta-analysis.

Abstract

In the past decade, no meta-analytical estimates of the prevalence of post-traumatic stress disorder (PTSD) among children and adolescents have been published, despite a host of new prevalence studies and updated DSM-5 criteria. We set out to estimate the prevalence rates of PTSD in trauma-exposed children and adolescents on the basis of DSM-IV and DSM-5 criteria, and investigate differences in prevalence across trauma type, gender, time since exposure, type of informant and diagnostic measures. Studies identified in a previous meta-analysis were combined with more recent studies retrieved in a new systematic literature search, resulting in a total of 95 studies describing 64 independent samples (<i>n</i> = 6745 for DSM-IV, <i>n</i> = 12 644 for DSM-5) over a 30-year period. Three-level random-effects models were used to estimate prevalence for DSM-IV and DSM-5 criteria separately, and for testing coded variables as moderators. The DSM-IV meta-analysis estimated a PTSD prevalence of 20.3% (95% CI 14.9-26.2%) using 56 samples with age range 0-18 years, and revealed moderating effects of gender, trauma type and diagnostic interview type. The DSM-5 meta-analysis found an overall prevalence of 12.0% (95% CI 3.7-24.2%) using eight samples with age range 1-18 years. There was insufficient data for moderation analyses. Although most trauma-exposed children and adolescents do not develop PTSD, a significant proportion (20% under DSM-IV criteria and 12% under DSM-5 criteria) do, particularly girls and individuals exposed to interpersonal trauma. These findings highlight the urgent need of continuous efforts in prevention, early trauma-related screening, and effective diagnostics and treatment to address the substantial burden of PTSD.