Abstract / Summary
This study aimed to explore factors associated with oral frailty in older adults with dental implants and to develop and internally validate a nomogram-based screening model for identifying prevalent oral frailty. A total of 466 older adults visiting the Implantology Department of Nantong Stomatological Hospital were recruited via convenience sampling from July 2025 to February 2026. Participants were randomly assigned to a training set ( n = 326, 70%) and an internal validation set ( n = 140, 30%). LASSO regression and logistic regression were used to identify factors associated with oral frailty, and R software was used to establish a nomogram model for identifying prevalent oral frailty. Model performance was evaluated using the receiver operating characteristic (ROC) curve, Hosmer-Lemeshow (H-L) test, calibration curve, and decision curve analysis (DCA). The final model variables included age, education level, the number of natural teeth, smoking status, the number of dental implants, cognitive decline, oral health status, oral self-efficacy, oral health literacy, social isolation and frailty. In the training set, the area under the ROC curve (AUC) was 0.931 (95% CI: 0.905–0.957), with the optimal cutoff value of 0.456. The model achieved an accuracy of 0.859, sensitivity of 0.863, and specificity of 0.855. Internal validation indicated excellent model fit via Hosmer-Lemeshow test (χ²=3.789, P = 0.876), with calibration intercept − 0.218, slope 1.009 and Brier score 0.115. The developed nomogram showed good discrimination, calibration, and potential screening utility for identifying prevalent oral frailty in this study sample. Prospective studies and independent external validation are still required before it can be used for prediction or clinical decision-making. Not applicable.