Gain Without Pain: The Extended Effects of a Behavioral Health Intervention

By Daniel Mochon

This week’s research talk is from Daniel Mochon, Assistant Professor in A. B. Freeman School of Business, Tulane University. The talk given by him took place in Tepper School of Business at Carnegie Mellon University, the photo of which is displayed above (source of the photo). In spite not doing research in social sciences, I personally have interest in such kind of studies inspecting the factors stimulating healthy life. Hence, I preferred this talk to attend this week. Below, you see the brief summary of the speech I constructed from the notes as far as I could take during the talk. All comments are welcomed to correct my faults, if there exist.


Unhealthy behaviors do not only damage the individuals, but also result in considerable amount of cost to the US Government, approximately $150 billion each year. Hence, the government and private companies have started to incentivize healthy behaviors. Having mentioned the importance of motivating people to engage in healthy behaviors, he continued with which sort of research questions they aimed to answer througout their study. Different from prior studies, they did not only consider the one point effect, effect behind a specific behavior, motivating people to engage in healthy behaviors. However, they inspected overall (extended) effects via research questions such as : 1)Does the intervention cause the target behavior to persist or reverse after it ends? 2) Does the intervention lead to negative or positive health spillovers? 3) Does the intervention lead to negative or positive customer loyalty spillovers? These are the questions both the government and firms try to answer when they make regulations. Private companies may also interest in incentivizing healthy behaviors, especially health insurance companies.

In order to study the extended effects of a health intervention, they collaborated with a wellness program, Discovery Vitality Program, in South Africa. Discovery incentivize healthy behaviors (like going a dietician, going to the gym, purchasing helthy food items) by offering discounts on British Always Flight. They recruited 4073 households (among 6570 who replied to invitation email) for participation of Healthy Food Benefit, which offers %25 cash back on the purchase of healthy foods. Those households were the shoppers of a grocery store, Pick n Pay, in South Africa. They required precommitment (commitment by participants to the duration and the amount of money they would spend during the study) to the program and the duration of  commitment was 6 months with an opt-out option after first month. They randomly assigned 1/3 of the participants to the control group, and the 2/3 of them to the treatment group.

The results reveal that the precommitment improves the engagement in healthy behaviors in both control and treatment groups. They also inspected the post-intervention status of the households in order to answer what happens after the intervention ends. Do committers continue to behave better or return to their old habits etc? They saw that, surprisingly the individuals continue to buy healthy food items after the 6-month commitment ends. In addition, they investigated the negative health spillover effects of the incentivization. This effect is  to look for whether the consumers compensete for healthy behaviors in one domain by behaving worse in another domains. Therefore, they would understand whether people give up exercising to compensete healthy food buying. In order to evaluate this, they tracked participants’ exercise behaviors. Results of this experiment point to non-significant difference meaning no spill-over effects!


Professor mentioned about the generalizability of the results  of the study to other healthy behaviors rather than only considering the healthy food item buying. He even noted that their models should be used as an overall evaluator of all kinds of behaviors. I accept that their work is worthy of respect, but this may be the oversimplification of the problem. There might be other “hidden “factors which may threaten the internal validity of the study. For example, they did not have the demographics of the participants. Hence, the results may be attributed to specific socio-economical status of the participants. Furthermore, a specific grocery store may have appealing characteristics making them to shop there such as cheerful staff, convenience of the place of the store, discounts on healthy food etc. Therefore, if the study would conducted in another store, they may not see the effect of incentive.

In fact, I was expecting a talk discussing the incentives people can apply on their own to stimulate their engagement in healthy behaviors, such as if I do exercise at least three times this week I can eat a small amount of chocolate so I should keep on exercising (foregetting that this was a business school talk). However, the talk really interested me and if the claims of the Professor are valid why not to conclude that the noneconomic factors (such as the one I give as an example) may be used as incentives to stimulate healthy behaviors.






Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s