Physical Activity, Functional Performance, and Healthy Aging Across Body Mass Index Categories: A Real-World Primary...

Background/Objectives: Physical activity (PA) and body mass index (BMI) are two central modifiable determinants of cardiometabolic health and functional performance across the adult lifespan. Excess adiposity and physical inactivity frequently coexist and may interact in determining functional aging trajectories. However, evidence from real-world primary care populations examining whether PA-cardiometabolic associations are consistent across BMI categories-including individuals with obesity-remains limited. This study evaluated the associations between self-reported PA categories and LDL-C concentrations and timed up and go (TUG) functional performance across three WHO-defined BMI categories and formally tested whether BMI category modified these associations. Methods: This cross-sectional observational study included 863 adult primary care patients stratified into normal weight (BMI < 25 kg/m², n = 241), overweight (BMI 25-29.9 kg/m², n = 348), and obese (BMI ≥ 30.0 kg/m², n = 274) groups. PA was categorized as low, moderate, or high based on World Health Organization recommendations. LDL-C and TUG were the primary outcomes. BMI-category-stratified analyses included one-way ANOVA with Bonferroni post hoc correction and multivariable linear regression adjusted for age, sex, arterial hypertension (AH), and diabetes mellitus (DM). An interaction term (PA × BMI category) formally tested BMI modification. Results: Age, LDL-C, TUG, sex, AH, DM, and PA distribution did not differ significantly across BMI categories (all p > 0.05). Higher PA categories were associated with significantly lower LDL-C in all three BMI categories (normal: F = 8.93, p < 0.001; overweight: F = 10.53, p < 0.001; obese: F = 9.35, p < 0.001) and with significantly better TUG performance (normal: F = 18.17, p < 0.001; overweight: F = 41.62, p < 0.001; obese: F = 21.14, p < 0.001). PA was the only consistently significant independent predictor of both LDL-C (all p < 0.001) and TUG (all p < 0.001) in all three BMI categories after multivariable adjustment. The interaction term (PA × BMI category) was not statistically significant for LDL-C (p = 0.972) or TUG (p = 0.475). The proportion of patients with TUG ≥12 s among low-PA patients increased across BMI categories: 33.7% in patients with normal weight, 37.5% in patients with overweight, and 42.3% in patients with obesity. Conclusions: PA-cardiometabolic and PA-functional performance associations were consistent across all three BMI categories, with no statistically significant BMI modification detected. These findings support the potential clinical utility of routine PA assessment as a BMI-independent indicator of cardiometabolic risk and functional health in primary care, including in individuals with obesity, and are relevant to healthy aging and maintaining mobility and functional independence across the adult lifespan.