We identified 116,529 patients who met the study inclusion criteria. After exclusions (admissions that spanned the pre-pandemic and pandemic periods, no valid SARS-CoV-2 PCR test results in the pandemic era, or missing patient gender or race information), we matched 3090 patients with a positive test for COVID -19 to 11,004 patients with a negative COVID-19 test, and 5005 pre-pandemic patients in a sensitivity analysis (Fig. 1).

Demography

Our unmatched patient demographics showed 27,447 (of 47,519; 57.8%) women in the COVID-19 negative cohort compared to 2,281 (of 4838; 47.1%) women in the COVID-19 group. 19 positive (P P P P

The two comparison groups showed differences in the proportion of racial groups (P

Among known risk factors for AF, we found differences between unmatched positive and negative COVID-19 cohorts regarding pre-existing chronic kidney disease, mitral valve disease, obesity, diabetes mellitus, hypertension, peripheral vascular disease, hyperlipidemia, smoking, and history of paroxysmal disease. AF/AFl. When comparing the pre-pandemic cohort with the COVID-19 positive cohort, differences were found for mitral valve disease, congestive heart failure, COPD, history of myocardial infarction, obesity, diabetes mellitus, hypertension, peripheral vascular disease, hyperlipidemia, smoking, and history of AF/AFl (see Table 1).

Results

Among unmatched patients, AF occurred in 552 of 4838 COVID-19 positive patients (11.4%), in 4718 of 47,519 COVID-19 negative patients (9.9%), and in 2451 of 26,368 patients pre-pandemic (9.3%). After matching, 192 of 5005 pre-pandemic patients (3.8%) developed AF/AF on admission compared to 145 of 2283 COVID-19 positive patients (6.4%) leading to a ratio of crude odds 1.7 (95% CI 1.36, 2.12; PP= 0.007; see table 2). Comparing matched COVID-19 negative and positive patients, 626 of 11,004 (5.7%) COVID-19 negative patients developed AF/AF compared to 249 of 3,090 patients in the COVID-19 positive group (8, 1%), resulting in a crude odds ratio of 1.45 (95% CI 1.25, 1.69; PP= 0.0038).

Death on admission occurred in 544 of 26,368 unmatched pre-pandemic patients (2.1%), 1,139 of 47,519 COVID-19 negative patients (2.4%), and 496 of 4,838 COVID patients -19 positive (10.3%). After matching, 76 of 5,005 pre-pandemic patients died on admission (1.5%) versus 163 of 2,283 COVID-19 positive patients (7.1%) with a crude odds ratio of 4.99 (95% CI 3.78, 6.58; PPPPP

Multivariate regression model

After adjusting for patient demographics and comorbidities, COVID-19 was associated with 1.19 (95% CI: 1.00, 1.41) times the risk of developing AF (P= 0.0495; see Figs. 2) comparing matched positive and negative COVID-19 patients. Comparing COVID-19 positive patients with pre-pandemic patients yielded an OR of 1.57 (95% CI 1.23, 2; P= 0.0003) in our sensitivity analysis.

Multivariate logistic regression model and forest plots to determine atrial fibrillation risk after matching (blank rows due to insufficient data to analyze Hispanic vs. White and History of myocardial infarction variables; “Race: Other vs. White” includes Hispanic for COVID-19 positive compared to the pre-pandemic model; A Fatrial fibrillation, AFlatrial flutter, COPDchronic obstructive pulmonary disease).

History of paroxysmal AF or AF was associated with an OR of 8.25 (95% CI 5.57, 12.23; P > 0.001) for developing AF/AF during admission comparing COVID-positive patients -19 and pre-pandemic patients, whereas comparing COVID-19 positive to matched COVID-19 negative patients led to an OR of 5.01 (95% CI 3.92, 6.4; PP= 0.001), while gender resulted in an OR of 1.7 (95% CI 1.32, 2.18) and 1.81 (95% CI 1.55, 2.12) respectively (both P= 0.001). Asian and black race resulted in a reduced risk of developing AF in the COVID-19 positive to negative patient cohort with an OR of 0.46 (95% CI 0.21, 0.99; P= 0.0462) for Asian race versus Caucasian race and an OR of 0.63 (95% CI 0.44, 0.88; P= 0.00769) for blacks against whites. We further found differences for known risk factors for chronic kidney disease, mitral valve disease, congestive heart failure, history of myocardial infarction, obesity and hypertension (see Fig. 2 ).