Intelligent Precision Exercise Integrated Health Management: Construction of Innovative Technology and Management Models and Clinical Evidence Research in Elderly Communities.

This study was conducted in collaboration with a community-based retirement facility to evaluate the clinical effectiveness of a smart, technology-assisted circuit exercise intervention for community-dwelling older adults. Traditional exercise prescriptions for older adults often lack real-time monitoring of individual physiological responses, which can increase the risk associated with exercise participation. This program integrated real-time physiological monitoring (including heart rate reserve, HRR) to guide individualized exercise intensity within a moderate-intensity range (50%-60% HRR). The study was conducted in two stages with different allocation designs: Stage 1 (Randomized Controlled Design): Participants were randomly assigned to either an intervention group or a control group. The intervention group received the smart circuit exercise program, while the control group did not receive the exercise intervention and underwent pre- and post-assessment only. This stage allowed for between-group comparison of intervention effects. Stage 2 (Single-Group Extension): Following completion of Stage 1, the facility extended the exercise program to a broader group of community residents as part of a real-world implementation and scale-up initiative, in order to evaluate the program under routine practice conditions and to provide equitable access to the intervention across the community. A concurrent control group was not maintained during this stage, consistent with its focus on real-world effectiveness evaluation rather than efficacy comparison. Participants in both stages received circuit-based exercise training three times per week (24 sessions total over approximately 8 weeks), with intensity controlled at 50%-60% heart rate reserve (HRR). Outcome measures assessed before and after the intervention period included: cardiac autonomic activity and hemodynamic function (including blood pressure and heart rate variability), arterial stiffness, body composition (including skeletal muscle mass, body fat mass, and body fat percentage), functional fitness (including lower limb muscle strength and dynamic balance), clinical blood biochemistry indicators (including HbA1c), depressive symptoms (CESD-10), and sleep quality (Pittsburgh Sleep Quality Index, PSQI). Statistical analyses varied by stage and publication. For Stage 1 between-group comparisons, linear mixed models (LMM) with participant as a random intercept were used to test Time × Group interactions, adjusting for age, sex, and height. For broader cohort analyses, two-way mixed-design analysis of variance and paired-samples t-tests were used. Statistical significance was set at alpha = .05 for all analyses.