Social Relationships and Physical Activity: Perspectives from Home Care Aides and Their Older Clients

Social relationships, such as positive support from behavioral change agents, are known to motivate older adults to engage in healthy activities, such as physical activity (PA). Previous PA intervention studies largely focus on the role of informal social support from family and friends reported by participants and provide limited information about how support from and interactions with formal care providers or behavioral agents may influence participants’ outcomes. To address this knowledge gap in the context of a PA intervention and home care for frail community-dwelling older adults, this dissertation examined exercise-related social support provided by home care aides (HCA) to their clients, as well as HCAs’ and clients’ perceptions and beliefs related to their interactions and PA in the context of HCA-client dyads. Using quantitative and qualitative data from a 4-month home-based PA intervention delivered by HCAs to their clients, this dissertation examined how client function (self-reported daily activity function and performance-based function) is related to: 1) baseline outcome expectations for exercise held by HCAs towards their client, partly through increased exercise-related social support provided by HCAs for their clients after the intervention; and 2) shared perceptions of their relationship or beliefs towards PA within HCA-client dyads at baseline. Two major findings emerged. First, mediation analysis results showed that higher baseline outcome expectations for exercise held by HCAs were related to higher level of exercise-related social support. However, unexpectedly, exercise-related social support was not related to client outcomes. Qualitative data suggested alternative factors, such as clients’ family beliefs in the intervention and clients’ experiences with the intervention participation, may explain the unexpected results. Second, results of linear mixed models showed, as expected, when clients and their HCAs both perceived their interactions positively, or both held positive outcome expectations for exercise at baseline, the clients showed better outcomes. Qualitative data revealed two potential explanations: 1) shared perceptions about HCA-client relationship and shared beliefs in PA may protect HCA-client dyads from stressful situations, such as clients’ health decline, conflicts or issues arisen during the intervention, etc.; 2) clients and HCAs were motivated by each other throughout the intervention and may both benefit from participating in the intervention. This dissertation expands the scope of social relationships and intervention outcomes to a dyadic perspective. It highlights the advantages of using dyadic data and mixed methods in understanding intervention process and outcomes. Finally, this dissertation adds to a growing body of studies that document the role of non-medical home care workers in delivering a health behavior intervention.