Abstract
BACKGROUND: Older adults are insufficiently physically active, despite its importance for healthy aging. To develop appropriate physical activity interventions, it is necessary to understand their physical activity. This study applies a theoretical perspective, the COM-B model, and a mixed-method design to examine what influences older adults' physical activity levels with three questions: (1) What individual and external factors predict older adults' physical activity levels? (2) What do older adults perceive as influencing their levels of physical activity? (3) To what extent do the quantitative results on older adults' physical activity levels agree and disagree with the qualitative findings on older adults' physical activity levels? METHODS: A convergent mixed-method design was used with questionnaire (n = 334) and interview (n = 14) data from adults 65 years and older. Regression analyses were used for quantitative measurements: physical activity, age, subjective socioeconomic status, health status, capability, opportunity, motivation, and depression. Content analysis was applied to the qualitative data. The two forms of data were then integrated to provide greater insights than would be obtained by either dataset separately. RESULTS: The regression analyses showed that previous physical activity, current motivation, health status, and age significantly predicted older adults' physical activity levels. The content analysis revealed that participants addressed all subcomponents of the COM-B model, indicating its pertinence in understanding how older adults discuss their current physical activity levels. The integrated findings showed convergent and divergent results. Overall results indicated that previous physical activity engagement, present motivation, capability, and opportunity influenced older adults' physical activity levels. CONCLUSIONS: This study is the first to use this mixed-methods design to examine factors influencing physical activity levels among older adults living in rental apartments with community hosts. The integrated result reveals convergence for findings on motivation and physical capability but divergence on psychological capability, opportunity, and previous physical activity. The findings underscore a complex interplay of factors influencing older adults' physical activity levels and indicate relevance for the COM-B model. The results can guide future research on theoretically informed interventions to promote physical activity and healthy aging. Future research should clarify the role of opportunity for older adults' physical activity.