Functional fitness, perceived barriers and exercise health belief domains as correlates of physical activity in older adults attending a Third Age University: a cross-sectional study

Physical activity in later life may depend on exercise-related beliefs, perceived barriers, and functional capacity. This study examined the associations of functional fitness, perceived barriers and exercise health belief domains with self-reported physical activity in older adults attending a Third Age University. This cross-sectional study included 214 adults aged 60 years and older enrolled in a Third Age University in Antalya, Türkiye. Physical activity was assessed with the Physical Activity Scale for the Elderly (PASE). Perceived barriers (subjective appraisals of factors making activity harder) were measured with the Physical Activity Barriers (PAB) Questionnaire; exercise health beliefs with the Exercise Health Belief Model Scale (EHBMS), which has no self-efficacy subscale; and functional fitness with the Senior Fitness Test (SFT). Pearson correlations with Benjamini-Hochberg FDR correction and a prespecified four-block hierarchical regression were used. Mean PASE score was 67.2 ± 29.8. Personal barriers showed a weak negative association with physical activity, the only psychosocial correlate significant after correction ( r = − .171, p = .012). EHBMS total was unrelated to physical activity ( r = − .007, p = .917) and added negligible variance beyond demographics (ΔR² = 0.001). In the final model ( N = 193), functional fitness added 16.2% points (ΔR² = 0.162, p < .001), with 2-minute step and chair sit-and-reach independently associated with physical activity. The model explained 25.3% of variance (adjusted R² = 20.8%). In this selected sample, functional fitness, particularly aerobic endurance and lower-limb flexibility, accounted for the largest share of explained variance, whereas EHBMS-assessed beliefs added negligible variance. Because no analysis formally compared predictor domains, this reflects relative incremental contribution, not a tested difference. The null belief findings warrant caution because self-efficacy was unmeasured and several subscales showed ceiling effects.