Financial incentives to patients are increasingly being used, and they can be effective in driving better health behaviors, but their design is crucial to their success in motivating change, said Kevin G.M. Volpp, MD, PhD, Director of the Center for Health Incentives at the University of Pennsylvania, Philadelphia, during the 2012 ADA annual meeting.
The use of incentives by large employers to improve health has increased from slightly more than 30% of large employers in 2009 to approximately 70% in 2012. But the design of the reward programs is important, said Dr Volpp.
“Rewards once a year ignore myopia,” he said, citing one employer’s offer to rebate $150 to employees who join and regularly use a fitness center. Such one-time incentives, especially if they occur at the end of a calendar year, may not prompt long-term change, because people tend to focus on the present. In addition, an incentive in the form of a discounted employee insurance premium will not be as effective in changing behavior as a direct payment.
These lessons are gleaned from studies of behavioral economics, in which immediacy is important. The impact tends to be greater with immediate rewards, he said. The use of frequent, small rewards tends to work better than single awards to promote positive behavior. The amount of the reward is not as critical as immediacy either.
As an example, he used his own randomized, controlled study in which “long-term smoking-cessation rates tripled in the incentive group.” The study followed 878 General Electric employees from 85 worksites. Half of the employees were offered $100 to complete a community-based smoking-cessation program, another $250 if they refrained from smoking for 6 months after finishing the program, and $400 more if they remained smoke-free for an additional 6 months. The other half was asked to enroll in a smoking-cessation program without incentives.
Enrollment in the smoking-cessation program was more than double in the group that was offered incentives. Furthermore, smoking cessation rates at 12 months were 14.7% in the incentive group and 5% in the controls. At the last follow-up (at 15-18 months), 9.4% in the incentive group remained smoke-free compared with 3.6% of the controls.
Playing the Lottery
Other types of incentive programs have been examined for their effectiveness in promoting behavior change, and lottery systems appear to be a winner, said Dr Volpp. A study that he coauthored showed that a Dutch lottery system added to a wellness program motivated individuals to complete health risk assessment (HRA) questionnaires.
In this study, 3 types of incentives were compared. The first was a $25 cash award to an employee for completing an HRA, the second was a $25 cash award combined with a $25 grocery gift card, and the third was assignment to workforce groups of 4 to 8 members with a chance to win prizes (a Dutch lottery). Among those in the Dutch lottery, 1 group was chosen randomly each week. Each member of the chosen group who had completed an HRA received $100. The groups in which at least 80% of the members had completed HRAs had their cash prizes increased to $125.
The study was conducted among 1299 employees from 14 offices of a healthcare management company. Some 64% of the employees in the lottery groups completed HRAs compared with 42% in the $50 cash or grocery gift card group and 40% in the $25 cash–only group.
Similar success was achieved with enhancing compliance to warfarin, using a lottery in which patients could win money daily if they had taken their warfarin the previous day.
Because humans discount the future, they are not as motivated by far-off rewards, said Dr Volpp. This concept was illustrated in a study of weight-loss programs. The patients who were randomly assigned to a deposit contract in which they had to put up their own money, which was matched by researchers monthly if they met their weight-loss goal, lost an average of 14.0 lb during the study. Those who were assigned to a lottery lost 13.1 lb, and those assigned to a control group lost only 3.9 lb.
Loss aversion was a powerful incentive for those in the deposit group, because they stood to lose money, Dr Volpp pointed out.
Peer Mentoring Works Well
Peer mentorship worked well to improve glucose control, more so than financial incentive or usual care. Over 6 months, patients assigned to speak with a peer mentor at least weekly had a mean hemoglobin A1c level decline of >1% from baseline to 6 months compared with a 0.46% decline in the financial-incentive group and 0.01% in the usual-care group.
Because Americans spend only 1 to 2 hours annually with their physicians, monitoring their behaviors beyond the examination room may be key to their engagement in healthy behaviors.
Automated hovering is the principle of using wireless or electronic technologies to engage patients while they are away from a healthcare setting, offering constant reinforcement. In the study of warfarin mentioned above, electronic pill boxes were used to automatically enter patients into a lottery each time they took their warfarin. The use of the electronic pill dispenser increased the time spent in the therapeutic range and decreased the number of incorrect doses.