Electronic Theses and Dissertations

Date of Award


Document Type


Degree Name

Ph.D. in Pharmaceutical Sciences


Pharmacy Administration

First Advisor

David J. McCaffrey

Second Advisor

Donna West-Strum

Third Advisor

Sarahmona Przybyla

Relational Format



Objectives the objectives were to 1) examine adherence to multiple medications prescribed for a chronic disease (intra-disease multiple medication adherence) and that of multiple chronic diseases (inter-disease multiple medication adherence); 2) determine appropriate measurement paradigm from different intra-disease multiple medication adherence measurement approaches; 3) identify optimal cut-point for a dichotomized composite measure. Methods a retrospective study design was used. The subjects came from the marketscanâ® commercial claims and encounters data 2002-2003 and filled both sulfonylurea (su) and thiazolidinedione (tzd). Adherence was measured by proportion of days covered (pdc) over each period of 30 or 90 days and cumulatively. Random components from multivariate multilevel models were analyzed to examine multiple medication adherence relationships, including associations of evolutions of adherence. Survival analysis was performed on any-cause or diabetes-related emergency services (er) utilization. Concordance statistics were computed to compare different measurement approaches. Results intra-disease multiple medication analysis demonstrated strong and significant (p<0.05) relationships between overall adherence estimates for su and tzd and changes in adherence estimates over time. Patients who were receiving lipid or hypertension medications, or both in addition to su and tzd shostrong and significant (p<0.05) relationships between overall adherence to cross-disease medications or cross-disease adherence slope estimates. However, such results were not observed in diabetic subjects who were prescribed nitrates for angina. Each of six composite measures of intra-disease multiple medication adherence significantly predicted hazard (hazard ratio <1.0) of all-cause or any diabetes-related er utilization. Although each concordance statistic was significant (p<0.05), there were no differences among concordance statistics produced by these measurement approaches. The average and all approach shosome superiority. The optimality of cut-point for categorizing adherence based on a composite measure of intra-disease multiple medication adherence ranged from 75-85%. Conclusion the study population demonstrated good but not optimal levels of adherence to multiple chronic disease medications. Factors that affect adherence to individual medications appear to be related and should be targeted for intervention. Efficacy of a composite measure of intra-disease multiple medications may depend on intervention goals. Further research needs to identify a composite measurement approach that demonstrates superiority in predictive and discriminatory power consistently.


Emphasis: Pharmacy Administration



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.