Medicaid Incentives for the Prevention of Chronic Diseases: Diabetes Prevention

Overview

The relative effectiveness of incentives based on process (e.g. medication adherence) vs. outcome (improvements in blood pressure) is unknown, leading to the key research question: Which approach is more effective? The incentive structure for this initiative is based on best practices in the use of process and outcome measures to address this fundamental question. A series of incentive designs will be conducted to examine the relative effectiveness of equivalent value incentives based on process (e.g. attending smoking cessation counseling sessions), outcomes (e.g. quitting smoking), or a combination of process and outcomes incentives (e.g. attending smoking cessation counseling sessions and quitting smoking). This will also provide an overarching framework for assessing the relative importance of process versus outcome incentives in different contexts and for different populations.

Although great potential exists to promote healthy behaviors through financial incentives, few studies have compared the efficacy and effectiveness of incentives in a Medicaid population. This is important because although financial incentives structured as rewards to individuals substantially improve rates of healthy behaviors, the absolute proportions of people adopting healthier behaviors remain low. Early approaches to financial incentives generally have reflected all-or-nothing thinking by showing that providing incentives is better than not providing incentives, and by assuming that incentives will work similarly across different types of people. But basic research in behavioral economics suggests that how you pay and whom you pay may be critical factors. Furthermore, the relative effectiveness of incentives based on processes vs. outcomes is unknown, and indeed, is one of the most fundamental unresolved questions in the incentive literature. Economists would argue that outcome-based incentives are likely to be more effective because they allow each person to figure out the most efficient path to achieve a desired result. On the other hand, behavioral economists would argue in favor of incentivizing processes because they are simpler and more concrete. Additionally, some ethicists argue that incentivizing processes is fairer because they are effort dependent, and because in some cases the ability to achieve improvements in outcomes may be related to factors people cannot control, such as their environment or their genes. Ultimately, the key question from a health and economic policy standpoint is which approach is more effective. In this regard, improvements in outcomes seem appropriate to incentivize directly because such results are required for incentive programs or other interventions to improve health and/or reduce health care spending. And indeed, there is some limited evidence suggesting that incentives for weight loss, for example, may be more effective than incentives for process measures (eg, attendance) in obesity programs. The goal of the New York State Medicaid Incentives Plan is to improve clinical outcomes and decrease health expenditures by increasing smoking cessation, lowering high blood pressure, preventing diabetes onset, and enhancing diabetes self-management among Medicaid enrollees in New York State. Several incentive strategies will be explored to promote the use of under-utilized Medicaid benefits and regional resources. The New York State (NYS) Medicaid Incentives Program will target four prevention goals: 1) smoking cessation; 2) lowering high blood pressure; 3) diabetes onset prevention; and 4) diabetes management. For each prevention goal, four treatment arms have been defined. One treatment arm will receive incentives for process activities; one treatment arm will receive incentives for achieving desired outcomes; one treatment arm will receive incentives for both process activities and outcomes; and one arm will serve as a control, receiving no incentives. The smoking cessation incentive program will be piloted in western New York where smoking rates are higher than other regions of the state. Participants will receive direct cash payments for participating in smoking cessation counseling (process), filling nicotine replacement therapy prescriptions (process), and quitting smoking (outcome). 2,332 participants will be recruited for this study. The blood pressure control incentive program will be piloted in New York City where stakeholders are highly engaged and a large population of people at risk for inadequate blood pressure control reside. Participants will receive direct cash payments for attending primary care appointments (process), filling antihypertensive prescriptions (process), and decreasing or maintaining a decreased systolic blood pressure by 10mmHg or achieving another clinically appropriate target (outcome). 488 participants will be recruited for this study. The diabetes management incentive program will be piloted in New York City where the capacity of diabetes self management educators is the greatest. Participants will receive direct cash payment for attending primary care appointments (process), attending diabetes self-management education sessions (process), filling diabetes prescriptions (process), and decreasing their HbA1c by 0.6% or maintaining a level of 8.0% or less (outcome). 660 participants will be recruited for this study. The diabetes onset prevention incentive program will be piloted in western New York and New York City, where the capacity of YMCA Diabetes Prevention Programs is the greatest and well-integrated with stakeholders. Participants will receive lottery tickets for attending YMCA Diabetes Prevention Program sessions (process) and losing or maintaining a reduced weight (outcome). 596 participants will be recruited for this study. Other incentive approaches and research questions will be explored through rapid cycle evaluation.