Background/Objectives: Frailty in older adults limits social participation. We aimed to predict social participation in older individuals undergoing frailty health checkups using three machine learning (ML) models and identify key predictive factors through deep neural network (DNN) analysis. Methods: Overall, 301 older individuals were enrolled; 295 were included in the final analysis. The survey measured 18 attributes, including demographic, physical, cognitive, and social factors. Logistic regression (LR), nonlinear support vector machine (NLSVM), and DNN were used for prediction, with precision, accuracy, sensitivity, specificity, F1 score, and area under the curve (AUC) calculated as evaluation metrics. Results: Among 295 participants, 236 (80%) engaged in social activities, whereas 59 (20%) did not. The three models demonstrated complementary strengths: DNN provided the most balanced performance with superior sensitivity for detecting social participants; NLSVM showed the best overall discriminative ability but with higher false positive rates; and LR achieved the highest precision for correctly identifying participants but missed detecting social participants. AUC values ranged from 0.776 to 0.795 across models, indicating moderate discriminative performance. Contribution analysis revealed information-collection ability as the strongest predictor of social participation, followed by walking speed and number of cohabitants. Conclusions: ML models achieved moderate discriminative performance for predicting social participation among frailty-screened older adults. The DNN provided the most balanced performance. Each model exhibited distinct characteristics suitable for different screening purposes, with information-collection ability emerging as a key factor. The findings suggest that models must be carefully selected based on specific community health screening objectives.
Efficient and accurate data entry is essential for ensuring high-quality clinical registries; however, conventional electronic data-capture (EDC) interfaces impose a substantial workload on healthcare professionals. This study compared a wizard-style graphical user interface (WS-GUI) with an exploratory–selective graphical user interface (ES-GUI) for registry data-entry tasks. Twenty physical and occupational therapists completed standardized simulated registry data-entry tasks using both interfaces in a randomized, within-participant crossover design. Data-entry time, input accuracy, and perceived workload were evaluated using task completion time, error counts, and the National Aeronautics and Space Administration Task Load Index (NASA-TLX), respectively. The evaluated WS-GUI implementation was associated with shorter data-entry times, fewer errors, and lower perceived workload than the evaluated ES-GUI implementation. The largest workload differences were observed for mental demand, physical demand, and effort. Because the interfaces also differed in navigation and input methods, these findings are specific to the tested implementations, participant group, and short simulated tasks and do not establish the general superiority of wizard-style interfaces.
Objective: We aimed to identify predictors of the transition to pre-frailty among older adults undergoing frailty health checkups and to evaluate their relative importance. Methods: A longitudinal analysis was conducted using data from a frailty health checkup project involving 830 older adults participating in community exercise classes in Matsumoto City, Nagano Prefecture, Japan. Among them, 502 individuals classified as robust (healthy) at baseline were followed for 1 year. Predictors of transition to pre-frailty were examined using multivariate logistic regression analysis. Given the low follow-up rate (43.4%), a sensitivity analysis using inverse probability weighting (IPW) was performed. Model performance was internally validated using bootstrap resampling and calibration assessment. Results: Of the 218 participants who completed the follow-up period, 45 (20.6%) transitioned to pre-frailty. Increasing age was a significant risk factor (OR = 1.13, 95% CI: 1.06–1.20), whereas higher health literacy was a protective factor (OR = 0.51, 95% CI: 0.29–0.86). In the IPW sensitivity analysis, the association with age remained stable, while health literacy demonstrated borderline significance (p = 0.08). The model showed acceptable discrimination (area under the curve [AUC] = 0.737; 0.728 after optimism correction) and good calibration (Hosmer–Lemeshow test, p = 0.124). Conclusions: Age and health literacy were independent predictors of transition to pre-frailty. As a modifiable factor, health literacy may represent a promising target for interventions aimed at preventing the progression to pre-frailty.