Evidence of peer influence on the healthfulness of food choices in an employee population

Research suggests that obesity might be transmitted through social networks, but mechanisms are not well-understood. Unhealthy food choice is an important driver of obesity, but research examining the relationship of food choices and social influence has been limited to narrowly-defined samples and single meal occasions. We sought to determine associations in food choices among a population of diverse employees of a large workplace. Employee social connections were inferred using a validated model incorporating the time and location of food purchases, as well as employee characteristics. Data were drawn from 3 million encounters where pairs of employees made purchases together from Jan. 2015-Dec. 2016. The healthfulness of food items was defined by “traffic light” labels – green (healthy), yellow (less healthy), red (unhealthy) – based on USDA dietary guidelines. We examined year-long cross-sectional associations between the food purchases of all connected employees during each calendar year using simultaneously autoregressive models. In addition, generalized estimating equations were used to test for longitudinal (lagged) associations resolved to the second-by-second level between food purchases of employee/coworker dyads over 24 months. Cross-sectional analyses revealed that proportions of both healthy and unhealthy items purchased were positively associated between connected employees. Longitudinal models also found positive associations between an employee’s current food purchase and the most recent food purchase a coworker made together with the employee. Evidence that healthy food choices may be driven by workplace social connections, together with increasing availability of data to infer social connections, suggests that peer-based interventions may be effective for promoting healthy eating and reducing obesity. This work is collaborative with colleagues at Massachusetts General Hospital and Dartmouth College.