Unraveling COVID-19 relationship with anxiety disorders and symptoms using genome-wide data.

Background: There is still a limited understanding of the dynamics contributing to the comorbidity of COVID-19 and anxiety outcomes.

Methods: To dissect the pleiotropic mechanisms contributing to COVID-19/anxiety comorbidity, we used genome-wide data from UK Biobank (up to 420,531 participants), FinnGen Project (up to 329,077 participants), Million Veteran Program (175,163 participants), and COVID-19 Host Genetics Initiative (up to 122,616 cases and 2,475,240 controls). Specifically, we assessed global and local genetic correlation and genetically inferred effects linking COVID-19 outcomes (infection, hospitalization, and severe respiratory symptoms) to anxiety disorders and symptoms.

Results: We observed a strong genetic correlation of anxiety disorder with COVID-19 positive status (rg = 0.35, p = 2×10) and COVID-19 hospitalization (rg = 0.31, p = 7.2×10). Among anxiety symptoms, "Tense, sore, or aching muscles during worst period of anxiety" was genetically correlated with COVID-19 positive status (rg = 0.33, p = 0.001), while "Frequent trouble falling or staying asleep during worst period of anxiety" was genetically correlated with COVID-19 hospitalization (rg = 0.24, p = 0.004). Through a latent causal variable analysis, we observed that COVID-19 outcomes have statistically significant genetic causality proportion (gcp) on anxiety symptoms (e.g., COVID-19 positive status→"Recent easy annoyance or irritability" │gcp│ = 0.18, p = 6.72×10). Conversely, anxiety disorders appear to have a possible causal effect on COVID-19 (│gcp│ = 0.38, p = 3.17×10). Additionally, we also identified multiple loci with evidence of local genetic correlation between anxiety and COVID-19. These appear to be related to genetic effects shared with lung function, brain morphology, alcohol and tobacco use, and hematologic parameters.

Conclusions: This study provided insights into the pleiotropic mechanisms linking COVID-19 and anxiety outcomes, suggesting differences between dynamics related to anxiety disorders and those related to anxiety symptoms.