By Tara Mancini
Increasingly, the idea of programs that aim to modify the health behavior of participants through the use of incentives has surfaced as a way to cut medical costs and improve health outcomes. However, there is a paucity of research on the effectiveness of such programs, particularly in Medicaid. Fortunately, the ACA authorized funding to states to test the value of incentive programs in Medicaid, and in September 2011, CMS awarded $85 million dollars in grants to ten states under the “Medicaid Incentives for Prevention of Chronic Diseases Model.” Under the initiative, states can “[P]rovide Medicaid incentives to Medicaid beneficiaries of all ages who participate in prevention programs and demonstrate changes in health risk and outcomes, including the adoption of healthy behaviors.”
While the evaluation is still ongoing for the ten states, an article in the current issue of Health Affairs looks at three existing Medicaid incentive programs that pre-date the ACA initiative. In the article, Blumenthal et al., evaluate what is known about three early programs in Idaho, Florida, and West Virginia, and suggest how other states may improve upon them.
In each of the three states profiled in Health Affairs, the structures and incentives of the programs differed. What they did share was a lack of evidence to indicate their effectiveness in modifying the behavior they sought to change, such as smoking cessation or weight loss. (The one exception was Idaho’s Wellness Incentive Program, which provided premium support to parents who stayed up to date on well-child visits). Common characteristics among the programs were low rates of patient participation, unclear incentive structures, and little program evaluation.
What lessons can the demonstration states learn from the experience of Idaho, Florida, and West Virginia? Blumenthal et al. observe that one of the biggest hindrances is the limited evaluation of the programs, for either the effectiveness of modifying behavior or lowering costs. This could be remedied in the future, the authors suggest, by implementing rigorous evaluations at the state level, as well as a cross state comparison.
Other problematic features of the programs included incentive design and low levels of program awareness, as our own Joan Alker has documented in her work on Florida. The programs had low award amounts, and in the case of Florida, were complicated to retrieve. Based on other work in behavioral economics, the authors suggest that states create an infrastructure that can quickly deliver awards after goals are completed. In addition, they note that research on the relationship between reward magnitude and effectiveness would be helpful. Awareness of the incentive programs was low among eligible Medicaid beneficiaries. To increase awareness, states would do better to use multiple avenues of communication with the patients, in which the information is presented simply and straightforward. Also, relying on health care providers to educate patients about incentive programs is not likely to be sufficient.
Hopefully, the awardee states will heed the experience of earlier programs. They should already have a leg up, as all ten of the awardee states have evaluations planned, and the Research Triangle Institute will report on the initiative as a whole. A report is due to Congress at the end of the Medicaid Incentive demonstration, but we will be waiting awhile as the program doesn’t end until 2016.
