Research & Insights

By Renette Youssef, November 14, 2014

Data Science Musings: We Make Better Health Choices When We Think We’re in Control

Obviously, there are few topics that are as morbid as our own morbidity, and it is for this reason that we often avoid thinking about it. However, as it turns out, how we view the probability of the causes and the timing of our own demise is very important. Why? It is because these views are subconsciously affecting how we approach health related decisions that we are making every day.

Think of it this way. If you believed that you would likely die from an asteroid attack next week, would you bother to continue your low calorie diet today? Your answer is probably no, right? Similarly, you might not put much effort into regular-strenuous-exercise to avoid heart disease if you thought it likely that atmospheric pollution was going to shorten and end your life a few decades from now anyway. These two examples reveal a common truth: Any awareness of uncontrollable variables linked to your own demise might push you away from healthy choices in favor of more instantly gratifying behavior. That is exactly what was seen in this study titled, “Out of control mortality matters: the effect of perceived uncontrollable mortality risk on a health-related decision.”

In the study, PHD Student Gillian V. Pepper and behavioral scientist Daniel Nettle used surveys administered through CrowdFlower, a data enrichment platform, to collect the data and measure this effect. It looked at how the “perceived controllability of mortality risk would affect a health-related decision.” In the case of this study, on UK citizens, that decision was between choosing chocolate or fruit to eat.

Specifically the study used prompts, in an online survey environment, that convinced the reader that people in their area were dying early. One group was told they were dying early from a lack of good diet and exercise, and the other group were told they were dying early from air pollution caused by traffic emissions. At the end of the survey, they were told that they were being entered into a contest, and they could choose the prize of either chocolate or fruit.

This experiment was also reenacted ‘in the field’ at a shopping center where the participants interacted with someone who led them through the process of entering a contest with a prize of the participant’s choice, fruit or chocolate.

Two things they found:

  1. Something the researchers already knew, prior to the experiment, was that fear-tactics do work to promote healthy choices (but only if accompanied with explicit suggestions on how to fix the fear inducing threat). However, they did not know that this was only true for actual decisions, and it would not affect self-reporting of behavior (e.g., checking a box that says I plan to drink more water next week, etc.).

    mortality.png

  2. 31% of the uncontrollable short-life scenario participants chose fruit (i.e., the good health choice). This is in contrast to the 57% of the controllable long-life scenario participants who chose fruit as a prize.

    mortality1.png

 

Why it Matters:

Health interventions can be life saving, and understanding how perceptions on mortality risks are affecting people’s health choices is key to designing effective interventions. It all comes back to the concept (mentioned in the research) called the health-locus-of-control, “the extent to which a person believes that their health is determined by the actions of individuals, rather than by chance.” If someone has a high locus-of-control, they have a good chance of making the right choices and staying in control of their health. What do you think? How would you encourage this, and how would you design a health intervention based on this study?