PURPOSE: Diabetic retinopathy (DR) is a significant comorbidity with Type 2 Diabetes Mellitus (T2DM), however, risk prediction for DR remains understudied in the elderly population. This study aimed to develop and validate a nomogram for identifying individuals at high risk of DR among elderly T2DM patients to guide early clinical intervention. PATIENTS AND METHODS: A retrospective cohort of 1912 T2DM patients (aged ≥60 years) was enrolled from 2018 to 2024. Sociodemographic, biochemical, and health-related variables were extracted. The cohort was randomly stratified into derivation (70%) and validation (30%) sets. Least Absolute Shrinkage and Selection Operator (LASSO) regression was applied to identify key predictors, followed by multivariate logistic regression to construct the nomogram. Model performance was evaluated via Receiver Operating Characteristic-Area Under the Curve (ROC-AUC), calibration plots, Hosmer–Lemeshow (H-L) tests, and decision curve analysis (DCA). External validation was performed using an independent cohort (n = 476). RESULTS: Among 1912 patients, 655 (34.3%) were diagnosed with DR. Independent predictors included T2DM duration, glycosylated hemoglobin (HbA1c), platelet-to-lymphocyte ratio (PLR), estimated glomerular filtration rate (eGFR), and neutrophil percentage to albumin ratio (NPAR) (all p < 0.05). The nomogram demonstrated robust discrimination, with AUCs of 0.823 (95% CI: 0.805–0.851) and 0.808 (95% CI: 0.770–0.846) in the derivation and internal validation sets, respectively. Calibration plots demonstrated strong agreement between predicted and observed risks (H-L test: p = 0.807 [derivation], p = 0.374 [validation]). DCA indicated favorable clinical utility across threshold probabilities, and external validation confirmed generalizability (AUC=0.788) and readiness for clinical deployment. CONCLUSION: This rigorously validated nomogram, integrating clinical accessible variables, provided a pragmatic tool for early DR risk stratification in elderly T2DM patients. Implementation of this model in clinical practice may enable personalized risk mitigation strategies to reduce DR incidence in this vulnerable population.