Suicidality in patients with obsessive-compulsive and related disorders (OCRDs): A meta-analysis

Abstract

Introduction Previous meta-analyses showed that OCD is associated with a substantial risk of suicidal behaviours. Conclusive rates of suicidal ideation (current and lifetime) and suicide attempts based on pooled prevalence rates have not so far been calculated using meta-analysis for the other DSM-5 Obsessive-Compulsive Related Disorders (OCRDs). Objectives This meta-analysis aims to separately calculate the pooled prevalence rates of lifetime suicide attempts and current or lifetime suicidal ideation in BDD, Hoarding Disorder (HD), Skin Picking Disorder (SPD) and Trichotillomania (TTM) and to identify factors associated with increased suicide rates. Methods Our protocol was pre-registered with PROSPERO (CRD42020164395). A systematic review and meta-analysis following PRISMA reporting guidelines was performed by searching in PubMed/Medline, PsycINFO, Web of Science and CINAHL databases from the date of the first available article to April 20th, 2020. Stata version 15 was used for the statistical analysis. Given the small number of studies in TTM and SPD, the two grooming disorders were grouped together. Meta-analyses of proportions based on random effects (Der-Simonian and Laird method) were used to derive the pooled estimates. Results Thirty-nine studies (N=4559 participants) were included: 23 for BDD, 8 for HD, 7 for Grooming Disorders. For BDD, the pooled prevalence of lifetime suicide attempts, current and lifetime suicidal ideation was, respectively 35.2% (CI:23.4-47.8), 37.2% (CI:23.8-51.6) and 66.1% (CI:53.5-77.7). For HD, the pooled prevalence of lifetime suicide attempts, current and lifetime suicidal ideation was 24.1% (CI:12.8-37.6), 18.4% (CI:10.2-28.3) and 38.3% (CI:35.0-41.6), respectively. For Grooming Disorders, the pooled prevalence of lifetime suicide attempts and current suicidal ideation were 13.3% (CI:5.9-22.8) and 40.4% (CI:35.7-45.3), respectively (no data available for lifetime suicidal ideation). Conclusions The OCRDs as a group are associated with relatively high rates of suicidal behaviour. Through indirect comparisons, we infer that BDD has the greatest risk. Comorbid substance abuse, possibly reflecting poor underlying impulse control, is associated with higher rates of suicidal behaviour in BDD. Our data emphasize the need for clinicians to consider the risk of suicidal behaviour in the management of patients presenting with all forms of OCRDs.