Degree
Master of Science
Program
Computer Science
Supervisor
Dr. Michael Bauer
Abstract
Despite a large volume of research on the prognosis, diagnosis and overall burden of multimorbidity, very little is known about socio-demographic characteristics of multimorbid patients. This thesis aims to analyze the socio-demographic characteristics of patients with multiple chronic conditions (multimorbidity), focusing on patient groups sharing the same combination of diseases. Several methods were explored to analyze the co-occurrence of multiple chronic diseases as well as the associations between socio-demographics and chronic conditions. These methods include disease pair distributions over gender, age groups and income level quintiles, Multimorbidity Coefficients for measuring the concurrence of disease pairs and triples, and k-modes clustering to examine the demographics of patients with the same chronic condition. The experiments suggest that patient income quintile is not associated with multimorbidity rates, although gender and age group may play an important role in prevalence of multimorbidity and diagnosis of certain disease combinations.
Recommended Citation
Megerdichian Azad, Annette, "Mining of Primary Healthcare Patient Data with Selective Multimorbid Diseases" (2017). Electronic Thesis and Dissertation Repository. 4574.
https://ir.lib.uwo.ca/etd/4574
Included in
Applied Statistics Commons, Categorical Data Analysis Commons, Databases and Information Systems Commons, Numerical Analysis and Scientific Computing Commons