The impact of different monetary incentives on the individual ability to enhance physical activity and goal attainment: A within-subject field experiment

Previous research has shown that monetary incentives often have little or no effect on subjects' health behaviors. This lack of effect in the aggregate is despite the fact that many people believe incentives help them, and contrary to the fact that many have also been shown to adjust their behavior. Also, as more and more health programs involving money are offered, we see it as a necessity to find out which individuals will increase their physical activity if they can receive small amounts of money upon reaching an individual step goal.

To this end, we have developed a within-subject randomized controlled trial design in which each participant will serve as his/her own control. About 500 participants, aged 18 to 85 years with daily steps ranging from 1,285 to 84,942, will receive four experimental treatments and a control treatment during the same 5-weeks session in a randomized order (i.e. each condition on 7 days). For each individual, we want to learn the optimum incentive design (bonus or deposit contract) and whether the prospect of higher pay leads subjects to increase their daily step count beyond the individual target (i.e., median) toward the value of the best 40% of days or even best 25% of days. We are also interested in finding out which types of subjects do not respond to these incentives at all.

We run the experiment within an ongoing 365-day long study whit subjects who all aim at improving their physical activity. All those participants have been positively health screened, are using a smartphone app (ActiVAtE Behavior) to transmit their steps (main performance measure) in a timely manner and have already provided extensive individual data at time of intervention. While all participants indicated in the application questionnaire that they were eager to walk more steps per day, there is a large variation in ex-ante daily steps submitted via a smartphone app over the past 9 months. Since the individual daily goal is based on the median (minus data below 1,000 steps/day), there is also respective variation in individual goals, ranging from 1,285 to 21,690 steps/day. Furthermore, we have a very rich dataset on each individual that includes not only activity data before, during, and after the intervention, but also a large variety of preferences (measured via economic laboratory experiments) as well as sociodemographic data and individual attitudes and self-reported behavior (measured via questionnaires).