Abstract Urbanization and aging populations challenge public health in developing cities like Shiraz, where environmental factors significantly influence old adults’ health. This study examined urban environmental impacts on older adults’ health in Shiraz and developed predictive machine learning models for health outcomes. A cross-sectional study was conducted from December 2024 to January 2025, involving 3,000 older persons aged 60 years and above across 11 municipal zones of Shiraz. Stratified random sampling was used. Environmental data (green space per capita, population density, waste production) were extracted from municipal records. Health outcomes (BMI, frailty, depression, anxiety, and life satisfaction) were assessed using validated tools (GDS-4, GAI-5, LSI-Z). Statistical analyses included regression models and machine learning (Decision Tree, SVM). The SVM model demonstrated superior predictive performance (R²=0.75 for frailty) compared to Decision Trees (R²=0.71). Key predictive relationships emerged: each 1 m² increase in green space per capita predicted a 0.8-point reduction in depression scores (95% CI: -1.2 to -0.4) and 0.3-point lower frailty index. Waste production exceeding 250 kg/capita was associated with 35% greater fall risk (OR = 1.35, 95% CI: 1.12–1.63). Population density showed nonlinear associations with outcomes in SVM models, with thresholds varying by health indicator. Environmental quality plays a critical role in older adults’ health. “Urban planning strategies that enhance green spaces and strengthen waste management systems may substantially improve health outcomes among older adults in urban settings.
Asadollahi et al. (Sat,) studied this question.