Personality disorder is a severe health issue. However, the epidemiology of personality disorders is insufficiently described and surveys report very heterogeneous rates.
We aimed to conduct a meta-analysis on the prevalence of personality disorders in adult populations and examine potential moderators that affect heterogeneity.
We searched PsycINFO, PSYNDEX and Medline for studies that used standardised diagnostics (DSM-IV/-5, ICD-10) to report prevalence rates of personality disorders in community populations in Western countries. Prevalence rates were extracted and aggregated by random-effects models. Meta-regression and sensitivity analyses were performed and publication bias was assessed.
The final sample comprised ten studies, with a total of 113 998 individuals. Prevalence rates were fairly high for any personality disorder (12.16%; 95% CI, 8.01–17.02%) and similarly high for DSM Clusters A, B and C, between 5.53 (95% CI, 3.20–8.43%) and 7.23% (95% CI, 2.37–14.42%). Prevalence was highest for obsessive–compulsive personality disorder (4.32%; 95% CI, 2.16–7.16%) and lowest for dependent personality disorder (0.78%; 95% CI, 0.37–1.32%). A low prevalence was significantly associated with expert-rated assessment (versus self-rated) and reporting of descriptive statistics for antisocial personality disorder.
Epidemiological studies on personality disorders in community samples are rare, whereas prevalence rates are fairly high and vary substantially depending on samples and methods. Future studies investigating the epidemiology of personality disorders based on the DSM-5 and ICD-11 and models of personality functioning and traits are needed, and efficient treatment should be a priority for healthcare systems to reduce disease burden.
Declaration of interest