Our first aim was to identify medical morbidities associated with these polygenic scores using a phenome-wide association study. We analyzed 882 phecodes with at least 100 cases, and used a Bonferroni-corrected phenome-wide significance threshold of 0.05/882 = 5.67 × 10–5. This threshold, however, is likely over-conservative because it incorrectly assumes independence between phecodes. The phenome-wide association study of the polygenic score for MDD identified phenome-wide significant associations with “Mood disorders” (odds ratio [OR], 1.10 [95% CI, 1.06-1.15]; P = 1.83 × 10−6), and “Depression” (OR, 1.10 [95% CI, 1.05–1.15]; P = 1.13 × 10−5), parent codes for MDD, as expected (Fig. 2a; interactive plots available at: https://sealockj.shinyapps.io/mdd_loneliness_cad_interactive/). The remaining six phecodes that were significant after Bonferroni correction were related to acute and chronic heart diseases and their associated risk factors, including “Ischemic heart disease” (OR, 1.09 [95% CI, 1.05–1.14]; P = 8.28 × 10−6) and “Coronary atherosclerosis” (OR, 1.10 [95% CI, 1.05–1.14]; P = 1.08 × 10−5). The phenome-wide association study of the polygenic score for loneliness (Fig. 2b) also found significant associations with “Mood disorders” (OR, 1.09 [95% CI,
