A nomogram for predicting oral frailty in older adults: a small-sample cross-sectional study in Anhui Province, China

Background With the population aging, oral frailty among older adults has become an increasingly prominent concern. Oral frailty is a condition that is highly prevalent among older adults and has a significant negative impact on their quality of life. The condition can exacerbate physical frailty among older adults, increasing the risk of disability or death. This study investigated the current status and influencing factors of oral frailty in older adults and identified the potential risk factors for oral frailty. Methods The oral frailty of older adults was measured using the Oral Frailty Index-8 (OFI-8) scale. At the same time, their nutritional status was assessed using the Mini Nutritional Assessment–Short-Form (MNA-SF), depressive status was evaluated using the Geriatric Depression Scale (GDS-5), eHealth literacy was measured using the eHealth Literacy Scale, and cognitive status was determined using the Subjective Cognitive Decline Questionnaire (SCD-Q9) scale. First, variables related to oral frailty were preliminarily screened using univariate analyses (the chi-square test and t -test). Subsequently, variables with a p -value of < 0.05 in the univariate analysis were incorporated into a multivariate binary logistic regression analysis. The forward stepwise selection method (likelihood ratio test) was used to determine the final predictive model to control for overfitting and ensure the model’s parsimony. Based on the final multivariate logistic regression model, an individualized prediction nomogram was constructed. This nomogram converts the regression coefficients of each predictor variable into a 0–100 point scoring system, allowing for intuitive visualization of oral frailty risk by mapping the total score to the predicted probability. Results The prevalence of oral frailty among older adults was 46.8% (1,433/3,061). Hospitalization within the past year ( p = 0.001), depressive symptoms ( p < 0.001), social isolation ( p < 0.001), malnutrition ( p < 0.001), and subjective cognitive decline ( p < 0.01) were highly correlated with oral frailty in older adults. eHealth literacy ( p < 0.001) was a protective factor against oral frailty. The area under the curve (AUC) value of the constructed oral frailty prediction model was 0.747 (95% CI: 0.729–0.764), with the calibration curve slope approximating 1. The calibration curve closely aligned with the ideal standard curve, and the quantitative analysis of the H–L value indicated a good fit of the nomogram model ( χ 2 = 7.965, p = 0.437). This indicates that the final oral frailty prediction model for older adults in Anhui Province demonstrates good predictive performance and can accurately assess the risk of oral frailty in older adults. Conclusion This study showed a high prevalence of oral frailty among older adults in China. Hospitalization within the past year, depressive symptoms, malnutrition, and subjective cognitive decline were found to be highly correlated with oral frailty in older adults. Additionally, eHealth literacy was identified as a protective factor against oral frailty in older adults. The government and medical institutions need to develop and implement oral health prevention and management strategies for older adults in China to help reduce the risk of oral frailty.