Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Health Outcomes and Policy Research

Research Advisor

Teresa M. Waters, PhD


Lawrence M. Brown, PharmD, PhD Cyril F. Chang, PhD Christa M. George, PharmD Song H. Hong, PhD David K. Solomon, PhD


cardiovascular disease, Medicare Part D, medication adherence, MEPS, racial disparities


This dissertation sought to explore the relatively understudied area of racial disparities in adherence to cardiovascular medication regimens among the elderly. Black and Hispanic seniors are well documented to have lower rates of adherence to their prescribed cardiovascular medications, relative to their white counterparts. This disproportionately lower adherence places these minority groups at higher risk for worse cardiovascular prognosis and premature cardiovascular death. The Medicare Program, which covers healthcare predominantly for elderly Americans, offers an interesting laboratory to study these disparities and their response to policy changes. Using nationally representative data from the Medical Expenditure Panel Survey on white, black, and Hispanic Medicare seniors, this dissertation was an endeavor to: 1) evaluate the impact on these disparities of the introduction of Medicare Part D, the prescription drug benefit, in 2006, 2) explore the heterogeneity of these disparities at various locations in the adherence distribution, signifying population subgroups with potentially distinct behavioral patterns, and 3) systematically estimate the extent to which the inequality in the determinants of adherence, such as socioeconomic status, insurance coverage, access to care, and experience with providers, contributes to the particularly significant and consequential black-white adherence differential. To pursue these aims, this dissertation used some of the recent advances in econometric techniques for the study of inequality, including the rank-and-replace procedure to adjust for health status in non-linear models, unconditional quantile regression, and distribution-wide Oaxaca-Blinder decomposition, respectively. The Institute of Medicine’s framework on racial healthcare disparities constituted the primary basis for defining and empirically estimating adherence disparities. These investigations contribute threefold to the literature: substantively, conceptually, and empirically. The substantive findings can improve our understanding of what works and what does not work for disparity reduction, bring our attention to disparities among subpopulations potentially deserving priority intervention, and systematically quantify the roles of clinical, social, and health-system factors in perpetuating adherence disparities. The various conceptualizations of adherence disparities employed in this dissertation, along with the state-ofthe-art empirical approaches to implement them offer much needed examples to guide future research on disparities in general and medication-related disparities in particular.




One year embargo expired May 2015