Home of Your Leadership Life Coach
Physicians often struggle to help patients change their health behaviors. Patients may know that they need to quit smoking, lose weight, or exercise more, but summoning the will to change is hard. It’s particularly difficult for the highest-risk patients who may have life circumstances — challenges such as unemployment or homelessness — that make it harder for them to focus on the long-term. But combining behavioral economics and “gamification” — putting game elements such as points and achievement levels into non-game contexts — holds promise for driving behavior change when a doctor’s advice, and patient’s good intentions, are not enough.
Other industries have long used game elements that leverage behavioral science to drive desired customer behavior (think airline loyalty programs that award points and status for miles traveled). And indeed gamification is increasingly being incorporated into health insurance design and wellness programs. However, despite its growing use, there’s only limited evidence of its effectiveness in health care, and in particular whether existing gamification makes the best use of behavioral economic principles. Members of our group recently evaluated 50 of the most popular smartphone applications for health and fitness and found that while nearly two-thirds of the apps used game elements in their design, none incorporated several key insights from behavioral economics that could effectively influence desired actions and address predictable barriers to behavior change.
A central challenge for all health-related gamification programs is engaging participation, particularly among high-risk patients. Several design elements commonly found within gamified health and wellness programs could be made more engaging by incorporating behavioral insights. For example, most programs invite patients to join, framing their choice as an opt-in decision. But we have found that opt-out framing significantly improves participation. In a randomized trial, our group tested how to engage adults with uncontrolled diabetes in a remote-monitoring program. In the traditional, opt-in approach, only 13% signed up. But when the introductory letter framed the program as standard care, but allowed patients to opt out if they wished, enrollment rates nearly tripled to 38%. We’ve found similar results when testing ways to engage patients in a medication adherence program after a heart attack.
Another common feature of gamification programs is goal-setting. The traditional approach is to assign everyone the same goal (for example, taking 10,000 steps per day) and ask to them to strive for it immediately. However, this is probably overly ambitious for many and not sufficiently ambitious for others. We have found that it is more effective for programs to establish a baseline for each individual and then engage him or her in personalized goal-setting, with goals that gradually become more demanding and that adapt to ongoing performance. For example, in a randomized trial of patients with heart disease, we combined financial incentives of $2 per day for each day step goals were met with personalized goal-setting in a program that used wearable devices to measure activity. Half the patients were assigned to use the device’s preset 10,000 step per day goal that began immediately. The other half established a personal baseline step count which increased for the first two months and then remained steady for four months. During the 6-month trial, the patients who had been assigned the preset goal had no overall change in activity. But those with personalized step goals increased their activity significantly, walking about 100 miles more than the patients in the control group. They even remained more active than the control group for two months after the incentives stopped.
In another test of gamification, we partnered with the Framingham Heart Study, which has followed the health of generations of families but never previously participated in an intervention study. Two hundred participants over 18 years old enrolled with their families and used either a smartphone or wearable device to track their physical activity as measured by how many steps they took each day. We established a baseline activity level for each person, and then each selected an increased step-count goal to shoot for. Families in the control arm received daily feedback on how they were doing relative to their goal, but no other interventions. The plan undertaken by families in the gamification program, on the other hand, incorporated several behavioral economic and game principles.
First, they signed a pre-commitment pledge in which they agreed to try their best to achieve their goal — a simple technique that’s known to help people stick with goals. Each time they signed onto the platform they would be shown the contract and be reminded of their pledge.
Second, they received points that were allocated up front which they stood to lose if they failed to achieve their goal. This element harnesses the concept of loss-aversion — people’s tendency to be more motivated to avoid losing something they already have than to gain an equivalent new benefit.
Third, since we know people in these types of programs can occasionally fall off the wagon, we replenished participants’ points each week to give families a fresh start, which leverages the tendency to be more motivated to pursue a goal when the prompt or decision is anchored at a “landmark” time, such as the beginning of the week. (New Year’s resolutions are another example of this effect.)
Finally, we incorporated social incentives. Each day one member of a family was selected at random to represent the entire family. If that person achieved his or her goal, all family members kept their points; otherwise, everyone lost points. This design encourages collaboration, accountability, and peer support. It also leverages the principle of “anticipated regret” — fear of falling short of a goal and letting others down.
During the three-month study, family members in the gamification program walked on average nearly a mile farther each day (1,661 steps) than they had at baseline — about 1,000 steps more each day than people in the control group. What’s more, while their activity dropped somewhat after the game ended, they continued to walk more than members of the control group. We’ve tested a similar approach for weight loss and found that groups lose more weight when the team members live together, further demonstrating how social incentives can be harnessed for behavior change.
Gamification is already being used widely to encourage healthy behaviors. However, many current designs cater to “super users” who already like games and are motivated to improve their fitness. These programs are unlikely to engage people at higher risk who could benefit most from changing their behavior. Our group is using gamification to help patients with uncontrolled diabetes, heart disease, and cancer. We have found that incorporating principles from behavioral economics is not hard or expensive, but instead requires attention to detail. Subtle changes to program design and communications can have an outsized impact on how patients behave. That’s why embedding behavioral insights into gamification could represent a significant opportunity to improve health and wellbeing.