Abstract

Objective:
This study aimed to analyze the influencing factors associated with low grip strength in the western China population and to develop a nomogram prediction model, thereby providing a scientific basis for the early prevention of grip strength decline. Methods: A total of 810 participants were enrolled in a cross-sectional survey conducted in Lanzhou, Gansu Province, as part of the National Epidemiological Survey on Thyroid Disorders, Iodine Nutrition, and Diabetes. Participants were randomly divided into a training set (n = 568) and a validation set (n = 242) in a 7:3 ratio. Univariate and multivariate logistic regression analyses were performed using SPSS version 27.0 to identify significant influencing factors. A visual nomogram prediction model was constructed using R version 4.4.2. Model validation was conducted through 1,000 bootstrap resampling iterations. The model’s performance was assessed using the Hosmer-Lemeshow (H-L) goodness-of-fit test and the receiver operating characteristic (ROC) curve. External validation was performed using data from rural residents in Longnan, Gansu Province. Results: The prevalence of low grip strength in the training set was 18.8% (107/568). Univariate analysis identified eight variables with significant associations: age, height, weight, smoking status, free triiodothyronine (FT3), free thyroxine (FT4), urinary creatinine (UCr), and urinary iodine concentration (UIC). Multivariate logistic regression revealed that female sex (95%CI: 0.199–0.800), height (95%CI: 0.893–0.971), and urinary creatinine (95%CI: 0.992–0.999) were protective factors, while advanced age (≥75 years, 95%CI: 3.208–31.903) and smoking (95%CI: 1.392–6.990) were independent risk factors for low grip strength. The area under the ROC curve (AUC) for the internal and external validation of the prediction model was 0.748 (95%CI: 0.666–0.829) and 0.719 (95%CI: 0.672–0.765), respectively, indicating good discriminatory power. Brier scores for internal (0.137) and external (0.128) validation demonstrated good calibration of the model. The H-L test confirmed a good fit of the model (internal: χ² = 7.80, P = 0.82; external: χ² = 3.58, P = 0.89). Conclusion: The nomogram model developed in this study effectively identifies individuals at high risk of low grip strength. It serves as a quantitative tool for early screening and intervention planning and supports the formulation of personalized prevention strategies for grip strength decline in the western China population.