The contribution of genetic risk to the comorbidity of depression and anxiety: a multi-site electronic health records study

Importance: Depression and anxiety are common and highly comorbid, posing a clinical and public health concern because such comorbidity is associated with poorer outcomes. Objective: To evaluate association of genetic risk scores with depression and anxiety diagnosis either in isolation or comorbid with each other. Design: International Classification of Diseases (ICD) ninth and tenth edition codes were extracted from longitudinal electronic health records (EHR) from four EHR-linked biobanks with genetic data available. Data analysis was performed between February 2021 to October 2021. Setting: EHR-linked biorepository study. Participants: Across the four biobanks, 140947 patients (80601 female [57.2%] including 109592 European ancestry [77.8%], 22321 African ancestry [15.8%], and 9034 Hispanic [6.4%]) were included in the study. Main outcomes and measures: Polygenic risk scores (PRS) for depression and anxiety were computed for all participants. They were assessed for diagnosis of depression and anxiety using ICD9/10 codes. The primary outcome was a four-level depression/anxiety diagnosis group variable: neither, depression-only, anxiety-only, and comorbid. Estimated effect measures include odds ratios and the proportion of variance on the liability scale explained by the PRS. Results: 95992 patients had neither diagnosis (68.1%), 14918 depression-only (10.6%), 12682 anxiety-only (9.0%), and 17355 comorbid (12.3%). PRS for depression and anxiety predicted their respective diagnoses within each biobank and each ancestry with the exception of anxiety-PRS not predicting anxiety in any ancestral group from one biobank. In the meta-analysis across participants of European ancestries, both PRSs for depression and anxiety were higher in each diagnosis group compared to controls. Notably, depression-PRS (OR=1.20 per SD increase in PRS; 95% CI= 1.18-1.23) and anxiety-PRS (OR=1.07; 95% CI=1.05-1.09) had the largest effect size for the comorbid group when compared to controls. The confidence interval for the depression-PRS effect did not overlap across groups demonstrating a gradient of genetic risk across the four diagnosis groups. Conclusions and Relevance: The genetic risk of depression and anxiety make distinct contributions to the risk of comorbid depression and anxiety, supporting the hypothesis that the correlated disorders represent distinct nosological entities.