Metric development for medication compliance using a distribution approach
LE3 .A278 2006
Master of Science
Mathematics and Statistics
Mathematics & Statistics
This thesis evaluates and compares three metrics of medication compliance: CMA, mSSAI and SSAI. The CMA metric is used in current research, while mSSAI and SSAI are newly proposed metrics. DiÂ®erent aspects of each metric are discussed. Next, these three metrics are used to calculate compliance levels from clinical trial study data on patients with Heart Failure (HF) illness. The HF patients in the study took medications from four diÂ®erent drug classes during the course of their treatment. Both parametric and nonparametric approaches are made use of in order to explore the distributional properties of the CMA, mSSAI and SSAI metrics. A nonparametric approach utilizes kernel density estimation as a preliminary explo- ration for comparison of the drug classes' density curves. In addition, a parametric approach, the Pearson system of distributions, is used to eÂ®ectively identify a para- metric family for each HF drug class for all three metrics. As a result, the beta and Laplace family of distributions are identiÂ¯ed in this process.
The author grants permission to the University Librarian at Acadia University to reproduce, loan or distribute copies of my thesis in microform, paper or electronic formats on a non-profit basis. The author retains the copyright of the thesis.