Are There Causal Associations Between Obsessive-Compulsive Disorder and Cardiometabolic Phenotypes? A Genetic Correlation and Bi-Directional Mendelian Randomization Study

Abstract

In epidemiological studies, obsessive-compulsive disorder (OCD) is robustly associated with increased risk of cardiometabolic disorders, including cardiovascular diseases, type 2 diabetes, and obesity. However, the mechanisms behind these associations are unclear. We conducted genetic correlation analyses to explore shared genetic etiology and bi-directional summary-level Mendelian randomization (MR) to explore potential causal effects between genetic liability to OCD and 14 cardiometabolic phenotypes (e.g., coronary artery disease, blood pressure, body mass index [BMI]). If causal effects were observed, we planned to conduct multivariable MR to explore indirect effects via health behaviors. We found no evidence for genetic correlations between OCD and any of the cardiometabolic phenotypes under study, except for a negative correlation with BMI (rG = −0.123, SE = 0.029, p < 0.001). Summary-level MR showed no evidence for causal effects. Therefore, multivariable MR was not conducted. We found limited evidence for shared genetic etiology or causal effects using the largest OCD GWAS to date. However, we were predominantly only powered to detect medium to large effects in the direction of OCD to cardiometabolic traits, leaving the possibility of smaller causal effects existing. Future studies with larger, more representative samples will help to further interpret findings.