Hilltop Publishes Work on Propensity Score Matching as Method to Study Medicare Part D

In collaboration with a team that includes researchers at Johns Hopkins and UMBC’s Department of Public Policy, researchers at The Hilltop Institute have published a new peer-reviewed article in the AcademyHealth journal eGEMstitled Estimating Causal Effects in Observational Studies Using Electronic Health Data: Challenges and (some) Solutions. Researchers used Medicaid and Medicare administrative data to study the impact of the Medicare prescription drug program, Part D (first implemented in 2006), on persons with serious mental illness (SMI). Such a study is important because persons with SMI are medically vulnerable and because a large proportion of their prescription drug health care coverage shifted from state to federal responsibility as a result of the Part D program. The published study describes propensity score methods to “match” persons with Medicare and Medicaid coverage (dual eligibles) to persons with similar demographic and health profiles who are only engaged in Medicaid. Such matching is necessary to balance the selected samples so that they are comparable for inference testing. The suggested citation for this work appears below, and the full article can be viewed here: http://repository.academyhealth.org/egems/vol1/iss3/4. For more information about this work, please contact Hilltop Senior Policy Analyst, Michael Abrams at mabrams@hilltop.umbc.edu.

Stuart, E. A., DuGoff, E., Abrams, M., Salkever, D., & Steinwachs, D. (2013). Estimating causal effects in observational studies using electronic health data: Challenges and (some) solutions. eGEMs (Generating Evidence & Methods to improve patient outcomes), Vol. 1, Iss. 3, Article 4. DOI: 10.13063/2327-9214.1038

December 18, 2013