So Mi Jemma Cho1, Hokyou Lee2, Jee-Seon Shim3, Yoosik Youm4, Sun Jae Jung5, Dae Jung Kim6, Hyeon Chang Kim7. 1. Department of Public Health, Yonsei University Graduate School, Seoul, South Korea; Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, South Korea. Electronic address: jemma.so.mi.cho@gmail.com. 2. Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, South Korea; Department of Internal Medicine, Yonsei University College of Medicine, Seoul, South Korea. Electronic address: hokyou.lee@yuhs.ac. 3. Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, South Korea; Cardiovascular and Metabolic Diseases Etiology Research Center, Yonsei University College of Medicine, Seoul, South Korea. Electronic address: shimjs@yuhs.ac. 4. Department of Sociology, Yonsei University, Seoul, South Korea. Electronic address: yoosik@yonsei.ac.kr. 5. Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, South Korea; Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA, United States. Electronic address: sunjaejung@yuhs.ac. 6. Department of Endocrinology and Metabolism, Ajou University School of Medicine, Suwon, South Korea. Electronic address: djkim@ajou.ac.kr. 7. Department of Preventive Medicine, Yonsei University College of Medicine, Seoul, South Korea; Department of Internal Medicine, Yonsei University College of Medicine, Seoul, South Korea; Cardiovascular and Metabolic Diseases Etiology Research Center, Yonsei University College of Medicine, Seoul, South Korea; Integrative Research Center for Cerebrovascular and Cardiovascular Diseases, Yonsei University College of Medicine, Seoul, South Korea. Electronic address: hckim@yuhs.ac.
Abstract
PURPOSE: Evidence suggests that physical activity participation is shaped through a myriad of structural aspects of social relationships. We examined the relationship between social network structure based on egocentric social network and physical activity. METHODS: From 6799 middle-aged Korean adults, we assessed the social network density and proportion of closed triads, using the name generator module. Self-reported physical activity for functional and leisure purposes was calculated in metabolic equivalent of task (MET)-min/week. We employed sex-stratified, multivariable linear regression to assess the association between each network structure variable and total physical activity, adjusting for age, network size, socioeconomic status, and comorbidities. We also examined the association with moderate-to-vigorous physical activity (MVPA) and wrist-worn accelerometer assessed physical activity. RESULTS: The network members of female participants were more likely to be of same sex and family member compared to those of males. There were no sex differences in average network size. Network density based on affiliation was sex-differentially associated with physical activity (male β -346.7, p 0.2221 and female β -528.6, p 0.0002). In parallel, the proportion of closed triads was negatively associated with physical activity only in females (male β -542.6, p 0.0551 and female β -641.51, p < 0.0001). However, network density and closed triads were insignificantly yet positively associated with MVPA in male (density β 229.7, p 0.3193 and closed triad β 109.21, p 0.6333). Network structure by contact frequency or emotional closeness and accelerometer-assessed physical activity showed inconsistent results. CONCLUSION: Understanding the role of social network structures can help to achieve ideal physical activity level in the context of primary prevention of cardiometabolic disorders.
PURPOSE: Evidence suggests that physical activity participation is shaped through a myriad of structural aspects of social relationships. We examined the relationship between social network structure based on egocentric social network and physical activity. METHODS: From 6799 middle-aged Korean adults, we assessed the social network density and proportion of closed triads, using the name generator module. Self-reported physical activity for functional and leisure purposes was calculated in metabolic equivalent of task (MET)-min/week. We employed sex-stratified, multivariable linear regression to assess the association between each network structure variable and total physical activity, adjusting for age, network size, socioeconomic status, and comorbidities. We also examined the association with moderate-to-vigorous physical activity (MVPA) and wrist-worn accelerometer assessed physical activity. RESULTS: The network members of female participants were more likely to be of same sex and family member compared to those of males. There were no sex differences in average network size. Network density based on affiliation was sex-differentially associated with physical activity (male β -346.7, p 0.2221 and female β -528.6, p 0.0002). In parallel, the proportion of closed triads was negatively associated with physical activity only in females (male β -542.6, p 0.0551 and female β -641.51, p < 0.0001). However, network density and closed triads were insignificantly yet positively associated with MVPA in male (density β 229.7, p 0.3193 and closed triad β 109.21, p 0.6333). Network structure by contact frequency or emotional closeness and accelerometer-assessed physical activity showed inconsistent results. CONCLUSION: Understanding the role of social network structures can help to achieve ideal physical activity level in the context of primary prevention of cardiometabolic disorders.