Prevalence of the dissociative subtype of post-traumatic stress disorder

William White, Aaron Burgess, Tim Dalgleish, Sarah Halligan, Rachel Hiller, Anna Oxley, Patrick Smith, Richard Meiser-Stedman

OAI: oai:purehost.bath.ac.uk:publications/663de9c2-cc38-49f1-b6a8-b2e82bc0226d • DOI: https://doi.org/10.1017/S0033291722001647

The dissociative subtype of post-traumatic stress disorder (PTSD-DS) was introduced in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), and is characterised by symptoms of either depersonalisation or derealisation, in addition to a diagnosis of post-traumatic stress disorder (PTSD). This systematic review and meta-analysis sought to estimate the point prevalence of current PTSD-DS, and the extent to which method of assessment, demographic and trauma variables moderate this estimate, across different methods of prevalence estimation. Studies included were identified by searching MEDLINE (EBSCO), PsycInfo, CINAHL, Academic Search Complete and PTSDpubs, yielding 49 studies that met the inclusion criteria (N = 8214 participants). A random-effects meta-analysis estimated the prevalence of PTSD-DS as 38.1% (95% CI 31.5–45.0%) across all samples, 45.5% (95% CI 37.7–53.4%) across all diagnosis-based and clinical cut-off samples, 22.8% (95% CI 14.8–32.0%) across all latent class analysis (LCA) and latent profile analysis (LPA) samples and 48.1% (95% CI 35.0–61.3%) across samples which strictly used the DSM-5 PTSD criteria; all as a proportion of those already with a diagnosis of PTSD. All results were characterised by high levels of heterogeneity, limiting generalisability. Moderator analyses mostly failed to identify sources of heterogeneity. PTSD-DS was more prevalent in children compared to adults, and in diagnosis-based and clinical cut-off samples compared to LCA and LPA samples. Risk of bias was not significantly related to prevalence estimates. The implications of these results are discussed further.