# Construction and validation of a risk prediction model for oral frailty in rural hypertensive patients

## TL;DR

This study identifies factors contributing to oral frailty in rural hypertensive older adults and builds a predictive model to assess risk.

## Contribution

A validated risk prediction model for oral frailty in rural hypertensive patients is developed and tested.

## Key findings

- The incidence of oral frailty in rural older adults with hypertension was 61.5%.
- The model achieved an AUC of 0.781 in internal validation and 0.810 in external validation.
- Key risk factors include age, education, living alone, polypharmacy, and comorbidities.

## Abstract

To explore the influencing factors of oral frailty in rural hypertensive patients in Zhuzhou City, Hunan Province, and construct a risk prediction model.

A randomized sampling method was employed to investigate 538 older adults hypertensive patients in a rural area of Zhuzhou City, Hunan Province, using general data questionnaires, the Oral Hypertension Index-8 (OHI-8), and the EAT-10 scale. Among them, 377 cases were assigned to the modeling group, with internal validation conducted through 1,000 repeated Bootstrap sampling. An independent external validation group of 161 cases was also established.

The incidence of oral frailty in rural older adults hypertensive patients was 61.5%. Logistic regression analysis showed that age, education level, living alone, polypharmacy, smoking, dysphagia, xerostomia, and co-morbidities were influencing factors for oral frailty in rural older adults hypertensive patients (P<0.05). The modeling results showed an area under the curve of 0.781 (95%CI: 0.735–0.827). The Hosmer-Lemeshow goodness-of-fit test showed χ2 = 13.736, p = 0.089. The areas under the ROC curves for internal and external validation were 0.769 (95%CI: 0.754–0.779) and 0.810 (95%CI: 0.745–0.875), respectively, and the p values for the Hosmer-Lemeshow test were 0.089 and 1.133, respectively. The calibration plot showed a high degree of overlap between the predicted curve and the ideal curve, and the DCA plot indicated good applicability of the model in clinical decision-making.

The incidence of oral frailty is high in rural older adults hypertensive patients. Its influencing factors include age, education, living alone, polypharmacy, smoking, dysphagia, xerostomia, and comorbidities. The oral frailty risk prediction model constructed based on these influencing factors has high predictive and discriminative power and can be used as a screening tool for the risk of oral frailty in rural older adults hypertensive patients.