Do obsessive-compulsive symptoms increase the risk of developing psychosis? A systematic review and meta-analysis

Background and Hypothesis There is a phenomenological overlap of obsessive-compulsive symptoms (OCS) and psychosis, leading to discussion about whether OCS increase the risk of developing psychosis. Previous reviews have focused on clinical high-risk (CHR) cohorts which may misestimate risk due to sampling biases. The current systematic review and meta-analyses aimed to determine the risk of OCS on developing psychosis in individuals classified at CHR and at the population level. Study Design A total of 2081 articles were screened, with 11 studies meeting criteria for inclusion. Two separate meta-analyses were conducted for CHR cohorts and for register-based cohorts to estimate the risk ratio of OCS on developing psychosis. Study Results In CHR cohorts, the meta-analysis found no significant difference in the risk of developing psychosis between individuals with and without OCS (RR = 0.99, 95% CI [0.71, 1.38], p=.95) across 8 studies which observed low heterogeneity (I2 < .000%, 95% CI [0%, 67.03%]). In register-based cohorts, OCS were associated with a 15-fold risk increase of developing psychosis (RR = 15.01, 95% CI [8.36, 26.93], p<.001), although this estimate was derived from 3 studies with significant heterogeneity (I2 = 85.1%, 95% CI [37.14%, 99.67%]). Conclusions The limited number of studies and high heterogeneity in the population-level cohort limit any firm conclusions. However, findings indicate that OCS may increase the risk of developing psychosis in the register-based cohorts, but not in CHR cohorts. The discrepancy may reflect shared underlying vulnerabilities present in OCS and psychosis that are obscured by samples enriched for psychosis risk.