Abstract Background Chronic kidney disease (CKD) is a risk factor for end-stage renal disease and contributes to increased risk of cardiovascular disease morbidity and mortality. We aimed to develop a risk prediction score and equation for future onset of CKD using large-scale health checkup data. Methods This retrospective cohort study included 58,423 participants without baseline CKD who were randomly assigned to Derivation (n=38,948) and Validation cohorts (n=19,475) at a ratio of 2:1. The predictors were anthropometric indices, life style, and blood sampling data. In the Derivation cohort, we performed multivariable logistic regression analysis and obtained the standardized beta coefficient of each factor that was significantly associated with new-onset CKD and assigned scores to each factor. We created a score and an equation to determine the risk of developing CKD after 5 years and applied them to the Validation cohort to assess their reproducibility. Results The risk prediction scores ranged from 0 to 16, consisting of the seven indicators, including age, sex, hypertension, dyslipidemia, diabetes, hyperuricemia, and estimated glomerular filtration rate (eGFR). From the receiver operating characteristic (ROC) curve predicting CKD incidence, the area under the curve (AUC) was 0.78. A score of ≥8 showed the highest Youden index in the Derivation cohort, with a sensitivity of 0.90 and specificity of 0.52. The CKD incidence gradually and constantly increased as the score increased from ≤6 to ≥14 (Figure). The risk prediction equation consisted of aforementioned seven indicators: 1/(1 + exp[−(9.4876 + 0.0311×age + 0.2400×sex + 0.3470×hypertension + 0.0893×dyslipidemia + 0.3444×diabetes + 0.0832×hyperuricemia + (−0.1980)×eGFR]). The median probability obtained from the Derivation cohort was 0.018 (interquartile range 0.002–0.084), and the AUC value of the ROC curve for the development of CKD after 5 years was 0.88, with a sensitivity of 0.84 and a specificity of 0.78 at a cutoff value of 0.077. The Validation cohort analysis yielded similar results. Conclusion We developed a clinically useful risk score and equation to predict the CKD incidence after 5 years in the general Japanese population. These models have reasonably high predictability and reproducibility. Funding Acknowledgement Type of funding sources: None.