Background <p>To explore the risk factors for atrial fibrillation (AF) in patients with heart failure with preserved ejection fraction (HFpEF) and to construct a predictive model for AF in this population.</p> Method <p>Approximately 334 patients with HFpEF who were treated in the Department of Cardiology at Fuyang People’s Hospital from January 2022 to April 2024 were included. Patients were followed up for one year to observe whether they developed AF. All patients were randomly divided into a training set and a validation set in a 7:3 ratio, with the training set used for model construction and the validation set used for model performance evaluation. The Boruta algorithm was used for feature selection, and a predictive model was constructed using multivariate logistic regression analysis. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), Hosmer-Lemeshow goodness-of-fit test, calibration curve, and decision curve analysis (DCA).</p> Results <p>Logistic regression analysis revealed that coronary heart disease (CHD), body mass index (BMI), diastolic pressure (DP), free triiodothyronine (FT3), glomerular filtration rate (GFR), and sodium-glucose cotransporter 2 inhibitor (SGLT2i) were independent risk factors for the occurrence of atrial fibrillation in patients with heart failure with preserved ejection fraction (<i>P</i> &lt; 0.05). The AUC was 0.871 in the training set and 0.825 in the validation set; the Hosmer-Lemeshow goodness-of-fit test yielded χ² = 5.3976 (<i>P</i> = 0.7144); decision curve analysis (DCA) indicated that the predictive model could provide superior clinical net benefit for patients with heart failure with preserved ejection fraction.</p> Conclusion <p>A predictive model constructed using six predictive variables—CHD, BMI, DP, FT3, GFR, and SGLT2i—can effectively predict the risk of atrial fibrillation in patients with HFpEF and aid in early risk stratification.</p>