Promoting Employee Health Through The Worksite Food Environment

Adults in the United States gain an average of 1-2 pounds a year. Interventions to prevent weight gain at the population level are needed to reverse the rising prevalence of obesity. Although individual-level interventions can result in large weight changes among small groups of individuals, achieving changes in the population will require long-term strategies that create healthier food environments, establish new social norms, and improve motivation and skills for healthy lifestyle behaviors. The worksite is ideal for interventions to address weight and lifestyle behaviors because a majority of adults are employed, and provisions in the Affordable Care Act encourage worksite wellness. Our research team at Massachusetts General Hospital (MGH) has demonstrated that behavioral economics strategies, including traffic-light labels, choice architecture, social norms, and financial incentives, improve employees' healthy food choices. The proposed project will address the critical next phase of this research to determine if a worksite intervention delivered through the food environment can prevent weight gain and reduce cardiovascular risk of employees. This project builds on the established traffic-light labeling system at MGH and tests an intervention that aims to increase nutrition knowledge, motivate change in lifestyle behaviors, and promote socially normative behavior for healthier lifestyles among employees. The intervention will be integrated into the flow of the work day, thus lowering burden to employees and the employer. Study Design: In a randomized controlled trial, 600 MGH employees will be assigned to: 1) an intervention arm with automated, personalized feedback about (a) worksite food purchases and calorie and physical activity goals (weekly emails) and (b) social norm feedback plus small financial incentives for healthy food purchases (monthly letters) or 2) a control arm (standardized monthly letters). Study outcomes will be assessed at 1 year (end of intervention) and 2 year follow-up. The primary outcome is change in weight at 1 year. Secondary outcomes are cardiovascular risk factors, worksite food purchases, and dietary intake (as measured by the Healthy Eating Index). A novel exploratory outcome will be healthy food purchases of co-workers who are socially connected to study subjects. Aim 1 is to determine if employees assigned to the intervention have less weight gain and lower cardiovascular risk factors than the control group at 1 year and 2-year follow-up. Aim 2 is to determine if employees assigned to the intervention group make healthier food choices than the control group at 1 year and 2-year follow-up. Exploratory Aim 3 is to determine if employees socially connected to the intervention group make healthier worksite food choices over 1 year than employees connected to the control group. Implications: This innovative strategy utilizing personalized feedback, social norms, and financial incentives will provide a scalable and sustainable model that could be adopted in other worksite, institutional, and retail settings to prevent obesity at the population level.

The overall objective of the ancillary studies added on to this project is to examine the psychological traits, cognitive skills, and genes that may influence the impact of a behaviorally-informed intervention on dietary choices, weight, and other objective health indicators. This research will expand on the randomized trial by examining psychological traits (impulsivity, self-control, social acceptance), cognitive skills (numeracy, health literacy), and genes (97 known BMI loci) that are associated with obesity and poor health and are specifically targeted by the intervention. We will use validated measures to assess traits and skills and well-established methods for genotyping and calculating genetic risk scores. Aim 1 will determine if psychological traits moderate the behavioral intervention effects on diet and weight. Aim 2 will determine if cognitive skills moderate the behavioral intervention effects on diet and weight. Aim 3 will determine if genetic risk for obesity moderates the intervention effect on weight. In secondary analyses, potential mediators of diet and weight outcomes, including dietary intent, self-efficacy, reward sensitivity, perceived norms, and perceived stress, will be assessed. Implications: Results of this research will l will inform the future design and implementation of more effective, tailored, and sustainable population approaches for obesity prevention.