Construction and validation of a nursing dependency risk prediction model for elderly patients with benign prostatic hyperplasia

Objective To screen risk factors for nursing dependency in elderly patients with benign prostatic hyperplasia (BPH) based on social ecology theory, and to construct and validate a risk prediction model for nursing dependency.

Methods A convenient sampling method was used to select 183 elderly BPH patients as the study subjects. General demographic questionnaires, international prostate symptom score (IPSS), activities of daily living (ADL) scale, health literacy management scale (HeLMS), FRAIL scale, and nursing dependency assessment scale were used to collect patient-related data. Binary logistic regression analysis was used to determine the independent risk factors; the ROC curve was used to evaluate the model's discrimination, and the Hosmer-Lemeshow test was used to evaluate the model's fit, followed by the construction and validation of a nomogram.

Results Among the 183 elderly BPH patients, the incidence of nursing dependency was 68.85%. Age, history of falls, FRAIL, IPSS, ADL, and HeLMS were all influencing factors for the occurrence of nursing dependency. The logistic regression equation was: Logit(P) = 5.011 + 1.401 × age + 1.490 × falls + 0.917 × IPSS + 1.002 × FRAIL – 0.187 × ADL score –0.064 × HeLMS score. ROC curve analysis showed that the area under the curve (AUC) of the model was 0.889 (95%CI: 0.835–0.942), with a sensitivity of 80.2% and a specificity of 84.2%, indicating good discrimination (0.8 ≤ AUC ＜ 0.9) of the model. When the cutoff value was set at 0.705, the Youden's index of the model was 0.644, indicating optimal predictive performance of the model, which could better identify patients with positive and negative nursing dependency; the Hosmer-Lemeshow test result was χ² = 8.28, P = 0.41, indicating good fit of the model. The nomogram based on independent risk factors allowed for rapid calculation of the probability of nursing dependency risk through assignment, making it easy to operate.

Conclusions The risk prediction model for nursing dependency in elderly BPH patients constructed in this study has good discrimination and calibration, and the nomogram can achieve rapid assessment of nursing dependency risk, providing a quantitative tool for identifying high-risk populations, effectively preventing the occurrence of nursing dependency, and significantly reducing family and medical burden.