Objective: Low physical activity increases cardiovascular disease (CVD) risk. Social context, operationalized through social networks, has been shown to drive health behaviors. This study examined the association between personal social networks and moderate-to-vigorous leisure-time physical activity (LTPA) among South Asian (Indian, Pakistani, Bangladeshi, Sri Lankan, Nepalese) immigrants, a group with high CVD rates.
Methods: This study used cross-sectional data from an ancillary study of social networks (2014-2017) in the Mediators of Atherosclerosis in South Asians Living in America (MASALA) study cohort. Participants, free from CVD at baseline and living in the San Francisco Bay-area, CA and Chicago, IL, were administered a detailed social networks questionnaire and physical activity questionnaire adapted from the Cross-Cultural Activity Participation Study. Participants reported on the exercise behaviors of each social network member and if they exercised with the network member. Network members who exercised with a participant were categorized as exercise partners. Moderate-vigorous LTPA was calculated as Metabolic Equivalent of Task (MET) minutes per week. Sex-stratified, linear regression models were used to examine associations between social network characteristics and MET-min/week of LTPA, independent of age, marital status, and network size. The effect of having an exercise partner in the network, above simply having network members who exercised, was tested using a partial F-test to compare nested models.
Results: Among the 700 participants, this analysis only included the 89% who reported any LTPA (n=623, 43% female). These individuals reported a median of 1335 MET-min/week of LTPA (IQR=735-2212 MET-min/week) and had an average of 4 network members (SD +/- 1). The proportion of network members who exercised was 0.89, and the proportion of exercise partners was 0.28. Exercise partners were most commonly spouses (56%) and friends (20%). Among South Asian men who exercised, having a social network member who exercised instead of having a non-exercising network member, significantly increased LTPA by 310 MET-min/wk (95% CI=152-470). For men, having a social network member who was an exercise partner instead of a non-exercising network member, was associated with an additional 520 MET-min/wk of LTPA (95% CI= 344-696). The effect on LTPA of having an exercise partner in the network was significantly greater than the effect of simply having a network member who exercised (p-value < 0.001). Results were similar for women, but not statistically significant (p-value=0.05).
Conclusions: Among South Asian immigrants, having an exercise partner in one’s personal social network was associated with significantly more LTPA. Social network support, in the form of an exercise partner, may be an effective component of interventions to promote LTPA in South Asians.