DIRECT COST OF OSTEOARTHRITIS IN CANADA: AN APPLICATION OF MICROSIMULATION MODELING WITH UNCERTAINTY ANALYSIS by Behnam Sharif M.Sc., The University of Manitoba, 2007 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Population and Public Health) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2014 ©Behnam Sharif, 2014 Abstract Introduction: While OA is a debilitating disease with an immense economic burden on the Canadian society, there is a lack of understanding about OA’s direct costs and its future trend in Canada. Objectives: The overall goal of this thesis is to illustrate the application of population- based disease microsimulation (PDMS) modeling in estimating the economic burden of a disease by performing the direct cost analyses for osteoarthritis (OA) using Population Health Microsimulation Model for OA (POHEM-OA). Specific objectives were: 1) To estimate the average direct costs of OA from 2003 to 2010 in Canada; 2) To estimate the future direct cost of OA from 2010 to 2031 in Canada; 3) to estimate the uncertainty around the prevalence and total cost of OA in future years. Methods: I used administrative health data from the province of British Columbia (BC), Canada, a survey of a random sample of BC residents diagnosed with OA (Ministry of Health of BC data), Canadian Institute of Health Information (CIHI) cost data and literature estimates to perform a bottom-up cost of illness (COI) study for OA. I then implemented the results of the COI study into POHEM-OA and constructed cost profiles for each individual. Finally, I developed a framework and adapted an ANOVA-based approach for performing uncertainty analysis (UA) for OA outcomes. Results: I showed that the average cost increased from $735 to $811 between 2003 and 2010 (in 2010 $CAD). From 2010 to 2031, while the prevalence of OA increases from 13.8% to 18.6%, the total direct cost of OA is projected to increase from $2.9 billion (95% uncertainty interval (UI): $2.4-$3.1 billion), to $7.6 billion ($6.2-$9.1 billion), an almost 2.6-fold increase (in 2010 $CAD). From the highest to the lowest, the cost ii components that will constitute the total direct cost of OA in 2031 are hospitalization cost, outpatient services, drugs, and out-of-pocket cost categories. Conclusions: By further developing a PDMS model of OA, I were able to project trends in the cost of OA and identify the key cost drivers, while predicting significant shifts in distribution of cost in the future. iii Preface In this thesis, I used the Population Health Microsimulation model of Osteoarthritis (POHEM-OA), a microsimulation model previously developed by Kopec et al. (1) and Statistics Canada. I further developed this simulation model by incorporating cost modules as described in Chapter 3 and Chapter 4. After implementing the coding into POHEM-OA, it was then corrected and validated by Mr. Bill Flanagan, chief of microsimulation modeling at Statistics Canada, Health Analysis Division. A section of this thesis has been published as a multi-authored paper in a refereed journals. Details of the co-authors’ contributions are provided below. The section in Chapter 3 on the Cost of illness (COI) study for per patient-year cost was performed as part of a prior study, the Canadian Osteoarthritis Simulation Team (COAST) by Dr. Jacek Kopec, Dr. Aslam Anis, Dr. Nick Bansback, and Mr. Mushfiqur Rahman. Drs. Kopec and Bansback designed this section of the study and Mr. Rahman performed the statistical analysis, using provincial population-based administrative data of British Columbia (BC). I have performed additional analysis on the procedure costs for inpatient cost components, surgery costs, rehabilitation costs, alternative care costs and out-of-pocket-costs. I have designed these analyses to estimate the per patient-year cost of OA as a function of individual characteristics and implemented the results of the aforementioned COI study into POHEM-OA, and performed the writing and revisions of Chapter 3. Dr. Kopec, Dr. Anis, and Dr. Bansback helped me in design, analysis and revisions of this Chapter. A version of Chapter 5 has been published in the journal Epidemiology Research International (Behnam Sharif, Jacek A Kopec, Hubert Wong, Philippe Finès, Douglas G iv Manuel, David L. Buckeridge, Uncertainty analysis in Population-based disease microsimulation models, Epidemiology Research International, July 2012). With the assistance of Drs. Kopec, Wang, and Fines, I conceptualized and designed the study. I conducted the analysis with support from the Canadian Arthritis Network grant. Dr. Kopec contributed to the interpretation and discussion of the study. Dr. Wang and Dr. Fines contributed to the statistical analysis. I prepared the manuscript and submitted it for publication. All co-authors contributed feedback on my drafts of the manuscript and I completed all revisions. Ethics approval for the studies described in Chapter 3 and 4 of this thesis was obtained from the University of British Columbia Behavioral Research Ethics Board (certificate number H06-04000). v Table of Contents Abstract ........................................................................................................................... ii Preface ........................................................................................................................... iv Table of Contents …..……………………………………..………………………………….vi List of Tables ................................................................................................................. ix List of Figures ................................................................................................................ xi Abbreviations .............................................................................................................. xiii Acknowledgements .................................................................................................... xvi Dedication ................................................................................................................... xvii Chapter 1: Introduction ......................................................................................................... 1 1.1 Thesis Organization ..................................................................................................... 1 1.2 Overview ...................................................................................................................... 3 1.3 Rationale .................................................................................................................... 11 1.4 Study Objectives ........................................................................................................ 13 Chapter 2: Background ....................................................................................................... 14 2.1 Osteoarthritis .............................................................................................................. 14 2.2 Cost-of-Illness Studies ............................................................................................... 24 2.3 Economic Burden of OA ............................................................................................. 33 2.4 Discussion .................................................................................................................. 54 Chapter 3: Application of a Simulation-Based Cost-of-Illness Study to Estimate Average Direct Cost of OA From 2003 to 2010 .................................................................. 56 3.1 Introduction ................................................................................................................. 56 3.2 Methods ...................................................................................................................... 58 vi 3.3 Analysis of Input Parameters for Cost Components .................................................. 65 3.4 The Population Health Microsimulation Model (POHEM) .......................................... 83 3.5 Description of the Cost Algorithm ............................................................................... 89 3.6 Scenario Analysis for Drivers of the OA Direct Cost Burden ...................................... 92 3.7 Results: Average Costs From 2003 to 2010 .............................................................. 95 3.8 Discussion ................................................................................................................ 106 Chapter 4: Projecting the Total Direct Cost Burden of Osteoarthritis Patients in Canada From 2010 to 2031 Using the POHEM-OA Model ............................................... 111 4.1 Introduction ............................................................................................................... 111 4.2 Methods .................................................................................................................... 112 4.3 Results ..................................................................................................................... 127 4.4 Discussion ................................................................................................................ 139 Chapter 5: Uncertainty Analysis in Population-Based Disease Microsimulation Models……………………………………………………………………………………………….143 5.1 Introduction ............................................................................................................... 143 5.2 Uncertainty