Social contexts are related to health behaviors: mEMA findings from the SPARC study

Understanding how social context is associated with behaviors can indicate circumstances when behaviors are more likely to occur. In this study 29,995 ecological momentary assessment (EMA) surveys from 805 participants were analyzed to determine the social context of when eating, drinking, physical activity behaviors, and none of those behaviors (not eating, drinking, or being physically active, "none") occurred. Within-person results indicated that participants had lower odds of reporting eating (OR = 0.81) and drinking (OR = 0.79), and higher odds of reporting physical activity behaviors (OR = 1.20) and "none" (OR = 1.11) when they were alone, compared to when they were with friends. Within-person results also indicated that when participants were off-campus they were more physically active (OR = 1.73) and had lower odds of being reported "none"(OR = 0.78) than when they were on campus. Surprisingly, within-person results also suggested participants reported less physical activity on the weekend (OR = 0.78) than during weekdays, and instead had higher odds of reporting "none" (OR = 1.11). Between-person results suggested participants who were alone had higher odds of reporting "none"(OR = 1.64). When stratified by demographics the findings were similar. These findings show that student behaviors are different when they are alone vs. not, and that social context matters when understanding student behaviors. Interventions aimed to improve dietary intake and promote physical activity may be more effective if a social context aspect is included in their design.