A Randomized Controlled Trial of Financial Incentives for Home-Based Health Monitoring

Home wireless device monitoring could play an important role in improving the health of patients with poorly controlled chronic diseases, but daily engagement rates among these patients may be low. Financial incentives have been successfully used to promote many healthy behaviors, but no studies have tested the effects of incentives on home-based health monitoring. We tested the effectiveness of two different magnitudes of financial incentives for improving adherence to remote-monitoring regimens among patients with poorly controlled diabetes.

We conducted a 24-week randomized controlled trial (12-week intervention, 12-week follow-up) in which participants were provided with biometric devices to monitor weight, blood pressure, and glucose at home and randomized to one of three groups: (1) lottery with an expected value of $2.80 a day for daily use of all three devices, (2) lottery with an expected value of $1.40 a day for daily use of all three devices, or (3) no financial incentive. Participants in both lottery arms had a daily 1 in 5 chance of a small reward ($10 or $5, respectively) and a 1 in 100 chance of a large reward ($100 or $50, respectively). The primary outcome was daily use of all three devices during the intervention period; secondary outcomes were change in hemoglobin A1c over the intervention period and adherence during the follow-up period.

Seventy-five participants from a Primary Care Medical Home practice at the University of Pennsylvania between age 18 and 80 years with a baseline hemoglobin A1c value greater than or equal to 7.5 percent.

Incentive arm participants used devices on a higher proportion of days relative to control (81% low incentive vs. 58% control, p-value 0.007; 77% high incentive vs. 58%, p-value 0.02) during the three-month intervention period. There was no difference in adherence between the two incentive arms (p-value 0.58). Adherence in the high incentive arm, however, fell significantly following discontinuation of incentives while usage among low incentive arm participants remained relatively high. In month 6, the low incentive group had an adherence rate of 62% compared to 35% in the high incentive group (p-value 0.015) and 27% in the control group (p-value 0.002) while rates in the high incentive and control groups were insignificantly different (p-value 0.45). There were no significant changes in hemoglobin A1c levels during the intervention period though point estimates suggest larger decreases in the low and high incentive arms (-1.5; 95% CI (-2.4, -0.5) and -1.2; (-2.1, -0.3), respectively) than in the control arm (-0.7; (-1.4, 0.1)).

Deploying wireless technologies to patient populations with high rates of non-adherence is unlikely to singlehandedly improve medical management.  A daily lottery incentive could be a cost effective way of lowering risk among high-risk patients for insurers and health systems. We find that a daily lottery incentive worth an average of $1.40 a day had similar effectiveness to a lottery incentive worth twice that amount while incentives were in place and significantly better efficacy once incentives were removed.