Zhao, Joanna - Differences in Prescription Drug use Among 5-year Survivors of Childhood, Adolescent, and Young Adul...

View the thesis

This project has been submitted to the Library for purposes of graduation, but needs to be audited for technical details related to publication in order to be approved for inclusion in the Library collection.
Term: 
Summer 2017
Degree: 
M.Sc.
Degree type: 
Project
Department: 
Department of Statistics and Actuarial Science
Faculty: 
Science
Senior supervisor: 
Rachel Altman
Thesis title: 
Differences in Prescription Drug use Among 5-year Survivors of Childhood, Adolescent, and Young Adult Cancer and the General Population in British Columbia, Canada
Given Names: 
Joanna
Surname: 
Zhao
Abstract: 
In this project, we analyze the prescription drug use of childhood, adolescent, and young adult cancer survivors identified by the CAYACS program in BC. Understanding the patterns of prescription use and factors associated with the tendency to be on prescriptions is important to policy and health care planners. Since data on actual prescription usage are not available, we use prescription dispensing data as a proxy. We examine the differences in prescription use between survivors and matched controls selected from the general population, and assess the impact of age and other clinical and sociodemographic factors on prescription use. Specifically, we model subjects' on-/off-prescription status by a first-order Markov transition model, and handle the between-subject heterogeneity using a random effect. Our method captures the differences in prescription drug use between survivors and the general population, as well as differences within the survivor population. Our results show that survivors tend to exhibit a higher probability of going on prescriptions compared to the general population over the course of their lifetime. Further, females appear to have a higher probability of going on prescriptions than males over the course of their lifetime. A simulation study is conducted to assess the performance of the estimators of the model.
Keywords: 
Health Services Utilization; Longitudinal Analysis; Markov Model; Prescription Drugs
Total pages: 
49