Incentives for Physical Activity - an RCT with Older Adults

Inactivity is the fourth leading risk factor for global mortality, leading to chronic diseases (e.g., heart disease, diabetes) and contributing to the obesity epidemic. Much of the world's population is inactive and older adults are at highest risk. Inactivity in older adults is linked to age-related diseases and cognitive decline. Inactivity also imposes social costs through increased medical expenses, which are already high among the growing older adult population in the US.

Incentive-based interventions have gained popularity among behavioral scientists and policymakers as a tool for improving health-related behaviors. But there are drawbacks - first, monetary incentives are often not cost-effective, and therefore scalability is limited. Second, behaviors often return to baseline when monetary incentives are removed, i.e., healthy habits are hard to maintain when incentives are limited in duration. Third, there is a concern that monetary incentives crowd out intrinsic motivation to engage in health-promoting behaviors.

In light of the limited success of incentive-based behavior change programs, the investigators propose to design and evaluate alternative incentives that address these challenges of scalability, habit formation and crowd-out. The investigators will aim to encourage physical activity through alternative incentives - by making a meal donations on behalf of participants (prosocial incentives) and give participants monetary incentives (personal incentives). Both types of incentives have underpinnings in behavioral economics.

Meal donations harness prosocial preferences, which may be more powerful and less likely to reduce intrinsic motivation than equivalent monetary incentives.

The overall aim is to evaluate the impact of alternative incentives on step count of older adults in the short-term and long-term. Exploratory analysis will also evaluate the impact on physical and mental health. The investigators will recruit 200 older adults and randomize half of them to receive additional prosocial and personal incentives for their walking behavior. The other half will not receive these incentives. The investigators will track step count of these two groups for 8 weeks using a Fitbit device. Under the treatment functionality, participants accrue a meal donation and a point for Feeding America for each day that they meet the step goal.