Relationship between Socioeconomic Status, Physical Activity, and Health Outcomes: National Health and Nutrition Examination Survey



Journal Title

Journal ISSN

Volume Title



According to the Social Determinants of Health, social factors such as education, income, and employment (i.e. factors that comprise socio-economic status, SES) impact adult health and associated health behaviors such as physical activity (PA). The purpose of this three-paper dissertation was to better understand the inter-relationship of SES, PA, sedentary behaviors (SB), and health (including metabolic syndrome and overweight/obesity) among a nationally representative sample of U.S. adults. Specifically, the three aims addressed the following 1) examined the relationship between three SES indicators: education, income, and employment status with non-leisure time physical activity (non-LTPA), 2) evaluated the relationship of occupational physical activity (OPA) and metabolic syndrome and its components, 3) assessed the role of LTPA and SB in the income-overweight/obesity relationship. To do this, all three aims utilized four waves of publicly available data from The National Health and Nutrition Examination Survey (NHANES) (2007– 2014), which included a total of 15,376 non-pregnant, non-older adults (aged 20-59 years). The sample was reduced to only include individuals who met the criteria and without missing data on the variables of interest for each aim (Aim 1: n=11,985, Aim 2: n= 3,253, Aim 3: n =10,348). Descriptive statistics, as well as weighted linear and logistic regression analyses were conducted using STATA version 15.0 statistical software (Aim 1 and 2). Structural equation modeling was conducted in Mplus version 8.3 (Aim 3). Survey procedures were used in all analyses to account for the NHANES sampling design. Aim 1: When examining the relationship between three SES indicators: education, income, and employment status with non-LTPA, findings indicated that only education and employment were related to non-LTPA. Having less than a high school education [OR = 1.44 (0.18), p < .01] and having a high school education [OR = 1.43 (0.12), p < .001] were associated with increased odds of meeting PA guidelines from non-LTPA, compared to a college degree. Part-time employment was associated with increased odds of meeting PA guidelines from non-LTPA [OR= 1.28 (0.12); p < .01], compared to full-time employment. Aim 2: When evaluating the relationship of OPA with metabolic syndrome and its components, findings suggest that OPA was not associated with metabolic syndrome, nor its components (p >.05). Further, the relationships did not differ between women and men (interaction term p >.05). Aim 3: When assessing the role of LTPA and SB in the income-overweight/obesity relationship, income indirectly influences overweight/obesity through its association with LTPA and SB. Greater income was negatively associated with overweight/obesity (Total effect: B=-0.046; 95%CI=-0.07,-0.02). Income indirectly influenced overweight/obesity through LPTA (Indirect effect: B=-0.005; 95%CI=-0.01,-0.003) and through SB (Indirect effect: B=0.008; 95%CI=0.005,0.01), but in opposing directions. The direct effect from income to overweight/obesity remained statistically significant (Direct Effect: B=-0.049; 95%C =-0.07;-0.02). LTPA partially accounted for the negative relationship between income and overweight/obesity; SB reduced the strength of the negative relationship between income and overweight/obesity. Aim 1 provides a comprehensive understanding of how SES is related to non-LTPA. Consequently, it raises awareness of the need to consider non-LTPA among low SES populations. Practitioners attempting to increase PA should consider these complexities and assess non-LTPA in addition to LTPA. Aim 2 indicated that there were no substantial associations between OPA and cardiovascular health indicators among a U.S. nationally representative cross-sectional sample. This contrasts findings from non-US-based samples which identified OPA as a risk factor for cardiovascular disease, especially among males (i.e. PA Health Paradox), Future prospective, longitudinal studies are needed to understand the long-term effects of OPA on the risk of experiencing metabolic syndrome among the U.S. population. Aim 3 suggests that targeted behavior approaches for weight management by income may be beneficial. Increasing LTPA among adults with lower income and decreasing SB among adults with higher income may provide some overweight/obesity protection. Taken together, these findings illustrate the complexities of the inter-relationships of SES, PA, SB, and health.



non-leisure time physical activity, leisure time physical activity, occupational physical activity, sedentary behaviors, metabolic syndrome, overweight, obesity