Background: In patients with coronary heart disease (CHD), atrial fibrillation (AF) is associated with increased morbidity and mortality. We investigated the associations between clinical risk factors and biomarkers with incident AF in patients with CHD. Methods and results: Around 13,153 patients with optimally treated CHD included in the STabilization of Atherosclerotic plaque By Initiation of darapLadIb TherapY (STABILITY) trial with plasma samples obtained at randomization. Mean follow-up time was 3.5 years. The association between clinical risk factors and biomarkers with incident AF was estimated with Cox-regression models. Validation was performed in 1,894 patients with non-ST-elevation acute coronary syndrome included in the FRISC-II trial. The median (min-max) age was 64 years (range 26-92) and 2,514 (19.1%) were women. A total of 541 patients, annual incidence rate of 1.2%, developed AF during follow-up. In multivariable models, older age, higher levels of NT-proBNP, higher body mass index (BMI), male sex, geographic regions, low physical activity, and heart failure were independently associated with increased risk of incident AF with hazard ratios ranging from 1.04 to 1.79 (P ≤.05). NT-proBNP improved the C-index from 0.70 to 0.71. In the validation cohort, age, BMI, and NT-proBNP were associated with increased risk of incident AF with similar hazard ratios. Conclusions: In patients with optimally treated CHD, the incidence of new AF was 1.2% per year. Age, NT-proBNP as a marker of impaired cardiac function, and BMI were the strongest factors, independently and consistently associated with incident AF. Male sex and low physical activity may also contribute to the risk of AF in patients with CHD.