Phenotype data for ARIC samples came from measurements aggregated by the GENEVA substudy (pht000114.v2.p1), downloaded with permission from dbGaP [23]. Eligible participants were those with data on incident CAD, defined as myocardial infarction, fatal coronary heart disease, silent myocardial infarction detected by electrocardiography, or revascularization procedure. We additionally extracted data on sex, age at first visit, and time to event (CAD or censoring in controls), as well as first visit data on BMI, waist girth, smoking status, hypertension medication use, systolic and diastolic blood pressure, type 2 diabetes diagnosis, highest level of education, use of cholesterol-lowering medication and other medications that secondarily affect cholesterol, and blood measurements of HDL, LDL, and triglycerides (Supplementary Table 3). We defined hypertension by the variable hypertension medication use, or systolic blood pressure >130 mmHg, or diastolic blood pressure >80 mmHg [31]. MDD was not assessed in ARIC participants because ARIC was designed to study cardiovascular outcomes. Therefore, we were only able to test the MDD and loneliness polygenic scores for association with CAD but were not able to adjust those analyses for the presence of clinical MDD.
