Background: There are an estimated 760,000 endometrial cancer survivors alive in the US today. We previously reported on increased heart disease (HD) risk among endometrial cancer survivors from our population-based cohort study. Although there are many risk prediction models for the risk of endometrial cancer, there are none to our knowledge for endometrial cancer survivors. Methods: We identified 2,994 endometrial cancer patients in the Utah Population Database, which links data from multiple statewide sources. We estimated hazard ratios with the Cox proportional hazards model for predictors of five-, ten- and fifteen-year risks. The Harrell’s C statistic was used to evaluate the model performance. We used 70% of the data randomly selected to develop the model and the rest of the data to validate the model. Results: A total of 1,591 patients were diagnosed with HD. Increased risks of HD among endometrial cancer patients were observed for older age, obesity at baseline, family history of HD, previous disease diagnosis (hypertension, diabetes, high cholesterol, COPD), distant stage, grade, histology, chemotherapy, and radiation therapy. The C-statistics for the risk prediction model were 0.69 for the hypothesized risk factors for HD, 0.56 for clinical factors, and 0.71 when statistically significant risk factors were included. With the final model selected, as one example, the absolute risks of HD were 17.6% at 5-years, 24.0% at 10-years and 32.0% at 15 years for a woman diagnosed with regional stage, grade I endometrial cancer in her fifties, was white, was obese at cancer diagnosis, had a family history of HD but no previous history of HD herself, had hypertension, but no history of diabetes or high cholesterol or COPD, and had radiation therapy treatment but no chemotherapy. The AUCs were 0.79 for the 5-year, 0.78 for the 10-year and 0.78 for the 15-year predictions. Conclusions: We developed the first risk prediction model for HD among endometrial cancer survivors within a population-based cohort study. Risk prediction models for cancer survivors are important in understanding long-term disease risks after cancer treatment is complete. Such models may contribute to management plans for treatment and individualized prevention efforts.