UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The health resource utilization and economic burden of systemic autoimmune rheumatic diseases McCormick, Natalie 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2012_spring_mccormick_natalie.pdf [ 2.98MB ]
Metadata
JSON: 24-1.0105195.json
JSON-LD: 24-1.0105195-ld.json
RDF/XML (Pretty): 24-1.0105195-rdf.xml
RDF/JSON: 24-1.0105195-rdf.json
Turtle: 24-1.0105195-turtle.txt
N-Triples: 24-1.0105195-rdf-ntriples.txt
Original Record: 24-1.0105195-source.json
Full Text
24-1.0105195-fulltext.txt
Citation
24-1.0105195.ris

Full Text

THE HEALTH RESOURCE UTILIZATION AND ECONOMIC BURDEN OF SYSTEMIC AUTOIMMUNE RHEUMATIC DISEASES by Natalie McCormick BSc., The University of British Columbia, 2007 A THESIS SUBMITTED IN PARTIAL FUFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Pharmaceutical Sciences) The University of British Columbia (Vancouver)  April 2012 © Natalie McCormick, 2012  Abstract Background: SARDs (Systemic Autoimmune Rheumatic Diseases) are a group of rare, chronic conditions (systemic vasculitis, systemic lupus erythematosus, scleroderma, Sjogren's disease, and poly/dermatomyositis) associated with high health resource consumption. However, estimates of their healthcare burden are sparse, with most determined at tertiary centres over short periods. Studying them separately has also limited research progress. Here we grouped the SARDs, for the first time ever, to quantify their collective, longitudinal (twelve-year) burden at the population-level. Methods: A population-based cohort of SARDs cases was identified from the administrative database of BC’s single-payer health system (PopDataBC). A detailed algorithm, with time and specialist parameters, was used to enhance diagnostic specificity. From PopDataBC, all provincially-funded health services, and all prescriptions (regardless of funding source), consumed from 1996 -2007 were captured. Costs for outpatient services and prescriptions were summed directly from paid claims; case-mix methodology was used for most hospitalizations. To quantify their net burden, costs were summed for claims attributable (under broad and narrow definitions) to SARDs. Costs are reported in 2007 Canadian dollars. Results: 18,741 SARDs cases were identified, contributing 82,140 patient-years(PY). After inflation adjustments, the annual mean per-PY direct medical costs of SARDs averaged $6,954/PY, with $1,882/PY(27%) from outpatient, $3,551/PY(51%) from hospital, and $1,521/PY(22%) from prescriptions. Over twelve years, annual costs decreased by 32%, from $8,901/PY in 1996 to $6,087/PY in 2007. Outpatient costs and encounters decreased by 26% ($2,205$1,641/PY) and 19% (34-27/PY), respectively. Mean annual hospital costs decreased by half ($5,579-$2,776/PY), and admissions by 46% (0.89-0.48/PY). Despite these decreases, the annual mean number of dispensed prescriptions increased by 49% (23-34/PY), and their costs by 50% ($1,117-$1,670/PY). The annual net per-PY costs of SARDs, mainly from hospitalizations(18-43% of costs) and prescriptions(48-76%), averaged $2,011$3,202/PY. Conclusions: SARDs impart a substantial healthcare burden at the population level, and in 2007 were directly responsible for ≥44% of cases’ gross mean annual healthcare costs ($6,087/PY). Most costs have decreased over twelve years; however, medication costs are rising (by 4% annually, on-average), which suggests comorbidity burdens are too. As demand grows for expensive but potentially-better SARDs therapies, research to assess their impact on long-term comorbidity risk is needed.  ii  Table of Contents Abstract ................................................................................................................................................................ ii Table of Contents ................................................................................................................................................ iii List of Tables ....................................................................................................................................................... vi List of Figures.................................................................................................................................................... viii List of Nomenclature ............................................................................................................................................ x Acknowledgements ............................................................................................................................................ xiii 1 Introduction ....................................................................................................................................................... 1 1.1 Systemic Autoimmune Rheumatic Diseases.................................................................................................... 1 1.1.1 Epidemiology ........................................................................................................................................ 1 1.1.2 Pathogenesis.......................................................................................................................................... 2 1.1.3 Clinical Manifestations .......................................................................................................................... 3 1.2 SARDs Treatment ......................................................................................................................................... 4 1.2.1 Current Therapies .................................................................................................................................. 4 1.2.2 Emerging Therapies............................................................................................................................... 6 1.3 SARDs as a Group of Diseases ..................................................................................................................... 9 1.4 Significance of the Health Resource Utilization and Economic Burden of SARDs ........................................ 11 1.4.1 Rising Case Numbers .......................................................................................................................... 11 1.4.2 Increasing Drug Costs.......................................................................................................................... 12 1.5 Review of Available Literature .................................................................................................................... 13 1.5.1 Methods .............................................................................................................................................. 13 1.5.2 Results ................................................................................................................................................ 14 1.5.3 Results from Clinic-Based Studies ....................................................................................................... 15 1.5.4 Results from Population-Based Studies ................................................................................................ 16 1.5.5 Results from SARDs-Attributable Costs .............................................................................................. 16 1.5.6 Summary of Current Cost Estimates .................................................................................................... 17 1.5.7 Research Gaps ..................................................................................................................................... 17 1.6 Research Objectives ................................................................................................................................... 18 1.7 Study Significance ...................................................................................................................................... 19 1.7.1 Importance of Canadian Estimates ....................................................................................................... 19 1.7.2 Advantages of Administrative Data...................................................................................................... 21 1.8 Study Implications ...................................................................................................................................... 23 2 Methods ........................................................................................................................................................... 54 2.1 Case and Data Source-British Columbia Linked Health Database ............................................................... 54 2.1.1 Medical Services Plan Dataset ............................................................................................................. 55 2.1.2 Discharge Abstract Database ............................................................................................................... 56 2.1.3 PharmaNet .......................................................................................................................................... 57  iii  2.2 Inclusion Criteria ....................................................................................................................................... 57 2.2.1 Rationale ............................................................................................................................................. 58 2.3 Exclusion Criteria ...................................................................................................................................... 59 2.3.1 Diagnostic Specificity.......................................................................................................................... 59 2.3.2 Costing Specificity .............................................................................................................................. 59 2.4 Index Dates and Subgroups ........................................................................................................................ 60 2.4.1 Index Dates ......................................................................................................................................... 60 2.4.2 Subgroups ........................................................................................................................................... 60 2.5 Health Resource Utilization ........................................................................................................................ 61 2.5.1 Follow-Up Parameters ......................................................................................................................... 61 2.5.2 Outpatient Encounters ......................................................................................................................... 62 2.5.3 Prescriptions........................................................................................................................................ 62 2.5.4 Hospitalizations ................................................................................................................................... 62 2.6 Cost Calculation......................................................................................................................................... 63 2.6.1 Outpatient Encounters ......................................................................................................................... 63 2.6.2 Prescriptions........................................................................................................................................ 63 2.6.3 Hospitalizations ................................................................................................................................... 64 2.7 Attribution .................................................................................................................................................. 66 2.7.1 Narrow Definition. .............................................................................................................................. 67 2.7.2 Broad Definition .................................................................................................................................. 67 2.8 Statistical Analysis ..................................................................................................................................... 67 3 Results.............................................................................................................................................................. 73 3.1 Descriptive Statistics .................................................................................................................................. 73 3.2 Gross Overall Direct Medical Costs ........................................................................................................... 73 3.3 Gross Outpatient Consumption ................................................................................................................... 74 3.4 Gross Hospital Consumption ...................................................................................................................... 74 3.5 Gross Prescription Medication Consumption .............................................................................................. 75 3.5.1 Costs and Consumption by Drug and Drug Class ................................................................................. 76 3.5.2 Major Contributors Toward Twelve-Year Per-Patient-Year Consumption and Cost Increases ............... 79 3.6 Net Overall Direct Medical Costs ............................................................................................................... 80 3.6.1 Net Overall Costs - Narrow Definition ................................................................................................. 80 3.6.2 Net Overall Costs - Broad Definition ................................................................................................... 81 3.7 Net Prescription Medication Consumption .................................................................................................. 82 3.8 Net Outpatient and Hospital Consumption - Narrow Definition ................................................................... 82 3.8.1 Net Outpatient Consumption ................................................................................................................ 82 3.8.2 Net Hospital Consumption ................................................................................................................... 83 3.9 Net Outpatient and Hospital Consumption – Broad Definition ..................................................................... 83 3.9.1 Net Outpatient Consumption ................................................................................................................ 83 3.9.2 Net Hospital Consumption ................................................................................................................... 83 4 Discussion and Conclusion ............................................................................................................................ 151 iv  4.1 Summary of Key Findings ......................................................................................................................... 151 4.1.1 Burden of SARDs .............................................................................................................................. 155 4.1.2 Longitudinal Trends .......................................................................................................................... 156 4.2 Interpretation of Findings ......................................................................................................................... 157 4.2.1 Decreases in Mean Per-Patient-Year Outpatient Utilization and Costs ................................................ 157 4.2.2 Increases in Mean Per-Patient-Year Prescription Costs ....................................................................... 158 4.2.3 Potential for Improved Healthcare Efficiencies................................................................................... 161 4.2.4 Potential for Lower Costs and Better Outcomes from Greater Prescription Use ................................... 162 4.2.5 Potential for Rising Comorbidity Burdens and Complication Rates..................................................... 163 4.2.6 Implications for Patients .................................................................................................................... 164 4.3 Limitations ............................................................................................................................................... 166 4.3.1 Omitted Health Care Costs ................................................................................................................ 166 4.3.2 Underestimated Day Surgery Costs .................................................................................................... 167 4.3.3 Minor Costing Uncertainties .............................................................................................................. 168 4.3.4 Capture of Attributable Outpatient Encounters ................................................................................... 169 4.3.5 Accuracy of SARDs Diagnoses.......................................................................................................... 170 4.4 Strengths .................................................................................................................................................. 172 4.4.1 Maximum Case Ascertainment and Sample Size ................................................................................ 173 4.4.2 Minimally-Biased and Comprehensive Estimates ............................................................................... 173 4.4.3 Longitudinal Design .......................................................................................................................... 174 4.5 Contribution ............................................................................................................................................. 174 4.5.1 Knowledge Gaps ............................................................................................................................... 174 4.5.2 Future Cost-Effectiveness Analysis .................................................................................................... 175 4.6 Applications and Knowledge Translation .................................................................................................. 176 4.7 Future Work ............................................................................................................................................. 177 4.7.1 Discrete Longitudinal Analysis .......................................................................................................... 177 4.7.2 Updated Health Utilization Data ........................................................................................................ 178 4.7.3 Control Data ...................................................................................................................................... 178 4.7.4 Impact of Complications and Comorbidities ....................................................................................... 179 4.7.5 Significant Cost Predictors................................................................................................................. 179 4.8 Conclusion ............................................................................................................................................... 179 References......................................................................................................................................................... 182 Appendices........................................................................................................................................................ 212 Appendix A: Diagnostic Codes, Fee Items, and Drug Classes.......................................................................... 212 Appendix B: Additional Results ...................................................................................................................... 219  v  List of Tables 1.1 Prevalence and Incidence of SARDs, Stratified by Sex .................................................................................... 24 1.2 Prevalence of SARDs ..................................................................................................................................... 29 1.3 Drugs Used in the Treatment of SARDs .......................................................................................................... 33 1.4 Estimated Annual Per-Patient Costs of Standard and Emerging SARDs Therapies in Canada ........................... 35 1.5a Study Characteristics for Included Studies using Clinic-Based Data – Mean Annual Direct Health Care Costs, Per-Patient ....................................................................................................................................................... 36 1.5b Study Characteristics for Included Studies using Clinic-Based Data –Mean Annual Health Resource Utilization, Per-Patient ....................................................................................................................................................... 38 1.6a Study Characteristics for Included Studies using Administrative Databases – Mean Annual Direct Medical Costs, Per-Patient ............................................................................................................................................. 40 1.6b Study Characteristics for Included Studies using Administrative Databases – Mean Annual Health Resource Utilization, Per-Patient ..................................................................................................................................... 41 1.7 Studies Reporting a Greater SARDs Burden in Older Individuals .................................................................... 43 1.8 Studies Reporting an Increase in the Incidence of SARDs................................................................................ 48 1.9a Study Characteristics for Included Studies with Incremental/Attributable Mean Annual Direct Medical Cost Data, Per-Patient .............................................................................................................................................. 50 1.9b Study Characteristics for Included Studies with Incremental/Attributable Health Resource Utilization Data, PerPatient.............................................................................................................................................................. 52 1.10 Mean Annual Per-Patient Direct Medical Costs of Other Arthritides in Canada .............................................. 53 2.1 Comparison of Incidence, Prevalence, Sensitivity, and Specificity Calculated from Different Administrative Sources ............................................................................................................................................................ 69 2.2 Summary of Included Encounters and Costing Procedures ............................................................................... 70 2.3 Provincial (BC) CPWC Values ....................................................................................................................... 71 2.4 Summary of Encounters Considered for Each Utilization and Cost Analysis .................................................... 72 3.1 Number of Cases and Patient-Years Contributed ............................................................................................. 85 3.2 Cohort Characteristics..................................................................................................................................... 86 3.3 Crude Annual Overall Mean Per-Patient-Year Costs (2007 Canadian dollars) .................................................. 87 3.4 The Top Five Most-Frequent Outpatient Encounters, by Billing Specialty ....................................................... 88 3.5 Annual Mean Length-of-Stay per-Year and per-Admission (amongst hospitalized cases) (days)....................... 89 3.6 Top-Ten Most Frequently-Prescribed Drug Classes ......................................................................................... 90 vi  3.7 Top-Ten Most-Costly Drug Classes (2007 Canadian dollars) ........................................................................... 92 3.8 Top-Ten Most Frequently-Prescribed Drugs Overall ....................................................................................... 94 3.9 Top-Ten Most Costly Drugs Overall (2007 Canadian dollars) .......................................................................... 95 3.10 Distribution of Specific SARD Diagnoses Amongst Cases Dispensed a Biologic Therapy .............................. 96 3.11 Annual Total and Mean Per-Patient-Year Prescriptions and Costs for the Biologic Therapies, 2005-2007....... 97 3.12a Drug Classes Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Prescription Costs . 98 3.12b Drugs Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Prescription Costs ............ 98 3.13a Drug Classes Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions .................................................................................................................................................... 99 3.13b Drugs Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions ... 99 3.14 Twelve Year Changes in Prescription Quantity and Cost, Amongst Selected Drug Classes ........................... 100 3.15 Crude Annual Mean Net Overall Per-Patient-Year Costs – Narrow Definition ............................................. 101 3.16 Crude Annual Mean Net Overall Per-Patient-Year Costs – Broad Definition ................................................ 102 4.1 Changes in the Incidence and Prevalence Rates for SARDs in BC, 1996-2007 (1).......................................... 181  vii  List of Figures Figure 3.1a Twelve-Year Proportion of Gross Medical Costs by Component – SARDs........................................ 103 Figure 3.1b Twelve-Year Proportion of Gross Medical Costs by Component – SARDs-CTD............................... 104 Figure 3.1c Twelve-Year Proportion of Gross Medical Costs by Component – SARDs-VD ................................. 105 Figure 3.2a Annual Proportion of Gross Direct Medical Costs by Component – SARDs ...................................... 106 Figure 3.2b Annual Proportion of Gross Direct Medical Costs by Component – SARDs-CTD ............................. 107 Figure 3.2c Annual Proportion of Gross Direct Medical Costs by Component – SARDs-VD ............................... 108 Figure 3.3a Crude Annual Mean Overall Gross Direct Medical Costs, Per-Patient-Year (2007 Canadian dollars) ............................................................................................................................. 109 Figure 3.3b Crude Annual Mean Overall Gross Direct Medical Costs, Per-Patient-Year (2007 Canadian dollars) – SARDs and SARDs-CTD only .......................................................................... 110 Figure 3.4 Crude Annual Mean Gross Direct Medical Costs, Per-Patient Year – Overall and By Component, all SARDs cases (2007 Canadian dollars) ........................................................................................................ 111 Figure 3.5 Crude Annual Mean Gross Outpatient Costs, Per-Patient-Year (2007 Canadian dollars) ...................... 112 Figure 3.6 Crude Annual Mean Gross Hospital Costs, Per-Patient-Year (2007 Canadian dollars) ......................... 113 Figure 3.7 Annual Mean Length of Stay Per-Admission, Amongst Hospitalized Cases (days) .............................. 114 Figure 3.8 Annual Mean Length of Stay Per-Year, Amongst Hospitalized Cases (days) ....................................... 115 Figure 3.9 Crude Annual Mean Gross Prescription Costs, Per-Patient-Year (2007 Canadian dollars).................... 116 Figure 3.10 Gross Crude Annual Mean Number of Prescriptions Dispensed, Per-Patient-Year ............................. 117 Figure 3.11a Top-Ten Most Frequently-Prescribed Drug Classes – SARDs ......................................................... 118 Figure 3.11b Top-Ten Most Frequently-Prescribed Drug Classes – SARDs-CTD ................................................ 119 Figure 3.11c Top-Ten Most Frequently-Prescribed Drug Classes – SARDs-VD .................................................. 120 Figure 3.12a Top-Ten Most Frequently-Prescribed Drugs – SARDs .................................................................... 121 Figure 3.12b Top-Ten Most Frequently-Prescribed Drugs – SARDs-CTD ........................................................... 122 Figure 3.12c Top-Ten Most Frequently-Prescribed Drugs – SARDs-VD ............................................................. 123 Figure 3.13a Top-Ten Most Costly Drug Classes – SARDs ................................................................................. 124 Figure 3.13b Top-Ten Most Costly Drug Classes – SARDs-CTD ........................................................................ 125 Figure 3.13c Top-Ten Most Costly Drug Classes – SARDs-VD ......................................................................... 126 Figure 3.14a Top-Ten Most Costly Drugs – SARDs ............................................................................................ 127 Figure 3.14b Top-Ten Most Costly Drugs – SARDs-CTD ................................................................................... 128 Figure 3.14c Top-Ten Most Costly Drugs – SARDs-VD ..................................................................................... 129 Figure 3.15a Drug Classes Contributing Most, On-Average, to the Twelve-Year Increase in................................ 130 Figure 3.15b Drugs Contributing Most, On-Average, to the Twelve-Year Increase in .......................................... 131 Mean Per-Patient-Year Prescription Costs .................................................................................................. 131 Figure 3.16a Drug Classes Contributing Most, On-Average, to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions ............................................................................................................................. 132 Figure 3.16b Drugs Contributing Most, On-Average, to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions ............................................................................................................................. 133 Figure 3.17a Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, ................................. 134 Narrow Definition – SARDs ...................................................................................................................... 134 Figure 3.17b Twelve-Year Proportion of Overall Net Direct Medical Costs by Component,................................. 135 Narrow Definition – SARDs-CTD.............................................................................................................. 135 Figure 3.17c Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, ................................. 136 Narrow Definition – SARDs-VD ................................................................................................................ 136 Figure 3.18a Annual Proportion of Gross Overall Direct Medical Costs Attributable to SARDs – all SARDs cases .................................................................................................................................................................. 137 Figure 3.18b Annual Proportion of Gross Overall Direct Medical Costs Attributable to SARDs – SARDs-CTD... 138  viii  Figure 3.18c Annual Proportion of Gross Overall Direct Medical Costs Attributable to SARDs – SARDs-VD ..... 139 Figure 3.19a Twelve-Year Proportion of Net Direct Medical Costs by Component, Narrow Definition – SARDs . 140 Figure 3.19b Twelve-Year Proportion of Net Direct Medical Costs by Component, Narrow Definition – SARDsCTD .......................................................................................................................................................... 141 Figure 3.19c Twelve-Year Proportion of Net Direct Medical Costs by Component, Narrow Definition – SARDs-VD .................................................................................................................................................................. 142 Figure 3.20 Crude Annual Mean Attributable Overall Direct Medical Costs, Per-Patient-Year – Narrow Definition (2007 Canadian dollars) ............................................................................................................................. 143 Figure 3.21a Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Broad Definition – SARDs ...................................................................................................................................................... 144 Figure 3.21b Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Broad Definition – SARDs-CTD ............................................................................................................................................. 145 Figure 3.21c Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Broad Definition – SARDs-VD ................................................................................................................................................ 146 Figure 3.22a Twelve-Year Proportion of Net Direct Medical Costs by Component, Broad Definition – SARDs ... 147 Figure 3.22b Twelve-Year Proportion of Net Direct Medical Costs by Component, Broad Definition – SARDs-CTD .................................................................................................................................................................. 148 Figure 3.22c Twelve-Year Proportion of Net Direct Medical Costs by Component, Broad Definition – SARDs-VD .................................................................................................................................................................. 149 Figure 3.23 Crude Annual Mean Attributable Overall Direct Medical Costs, Per-Patient-Year – Broad Definition (2007 Canadian dollars) ............................................................................................................................. 150  ix  List of Nomenclature ACE-II  Angiotensin-II Converting Enzyme  ALC  Alternate Level of Care  ANA  Anti-Nuclear Antibody  ANCA  Anti-Neutrophil Cytoplasmic Antibody  ATC-2  Anatomical Therapeutic Chemical Classification System, Second Level  BAFF  B-Cell Activating Factor  BC  British Columbia  BCLHD  British Columbia Linked Health Database  B-Lys  B-Lymphocyte Stimulator Protein  CI  Confidence Interval  CIHI  Canadian Institute for Health Information  CMG  Case-Mix Group  CPI  Consumer Price Index  CPWC  Cost-Per-Weighted-Case  CSD  Churg-Strauss Syndrome  CTD  Connective Tissue Disorders  DAD  Discharge Abstract Database  DIN  Drug Identification Number  DM  Dermatomyositis  DPG  Day Procedure Group  DPG-RIW  Day Procedure Group Resource Intensity Weight  FFS  Fee-For-Service  GC  Systemic Glucocorticoids  GCA  Giant Cell Arteritis  GERD  Gastrooesophageal Reflux Disease  H2RA  Histamine-II Receptor Antagonist  x  HAQ  Stanford Health Assessment Questionnaire  ICD  International Classification Of Diseases  ICD-10  International Classification Of Diseases, 10th Revision  ICD-9  International Classification Of Diseases, 9th Revision  IL  Interleukin  JIA  Juvenile Idiopathic Arthritis  LN  Lupus Nephritis  LNN  Lupus Nephritis-Negative  LOS  Length-Of-Stay  MSP  Medical Services Plan  NSAID  Non-Steroidal Anti-Inflammatory Drug  OA  Osteoarthritis  PAH  Pulmonary Arterial Hypertension  PAN  Polyarteritis Nodosa  PDE-5  Phosphodiesterase-Five  PHN  Personal Health Number  PM  Polymyositis  PM/DM  Poly/Dermatomyositis  PopDataBC  Population Data BC  PPI  Proton Pump Inhibitor  PPP  Purchasing Power Parity  PY  Person-Year/Patient-Year  RA  Rheumatoid Arthritis  RAAS  Renin Angiotensin Aldosterone System  RIW  Resource Intensity Weight  SARDs  Systemic Autoimmune Rheumatic Diseases  SARDs-CTD  Connective Tissue Disorders  xi  SARDs-VD  Vasculitic Disorders/Systemic Vasculitides  SjD  Sjogren’s Disease  SLE  Systemic Lupus Erythematosus  SMR  Standardized Mortality Ratios  SSc  Systemic Sclerosis/Scleroderma  STD  Standard Deviation  TA  Takayasu’s Arteritis  TNF-α  Tumour Necrosis Factor-Alpha  UK  United Kingdom  USA  United States Of America  VD  Vasculitic Disorders/Systemic Vasculitides  Wegener’s  Wegener’s Granulomatosus  xii  Acknowledgements I am most grateful for the feedback and guidance I received from both my supervisors, Dr. Carlo Marra and Dr. Antonio Avina-Zubieta, during this project and throughout my Master’s program. The research advice and support provided by the rest of my supervisory committee, and the staff at CORE, ARC, and CHEOS were also essential to this project. I am very appreciative of the many helpful, skilled, and friendly faculty, staff, and students I have gotten to know at the Faculty of Pharmaceutical Sciences, and especially for the interest and support from my parents, extended family, and ZQ.  xiii  1 Introduction  1.1 Systemic Autoimmune Rheumatic Diseases Systemic Autoimmune Rheumatic Diseases, or SARDs, refers to a group of related chronic rheumatic disorders. They are divided, on the basis of the affected tissues, into connective tissue (SARDs-CTD) and vasculitic (SARDsVD) disorders. SARDs-CTD includes systemic lupus erythematosus (SLE), systemic sclerosis/scleroderma (SSc), Sjogren’s disease (SjD), polymyositis (PM), and dermatomyositis (DM). SARDs-VD, known collectively as systemic vasculitides, are characterized by inflammation in one or more types of blood vessels throughout the body – ranging from the smallest capillaries to the aorta. This results in fibrosis, narrowing, stenosis, and restricted blood flow (2). SARDs-VD includes polyarteritis nodosa (PAN), Wegener’s granulomatosis (Wegener’s), giant cell arteritis (GCA), Takayasu’s arteritis (TA), and Churg-Strauss syndrome (CSD).  1.1.1 Epidemiology SARDs are seen in those of all ages, including children, but the incidence of most SARDs peaks in the fourth and fifth decades. Two exceptions are TA and SLE, which arise during the childbearing years (3,4). All SARDs-CTD predominantly affect females: in a recent population-based Canadian study SARDs-CTD affected about four-to-six times more females overall, and 1-in-100 females over 45 years (5). SARDs-VD affect males more than females, with male-to-female incidence ratios of 1.09 (6) and 1.43 (7) reported. More sex-stratified incidence and prevalence estimates are listed in Table 1.1. Within Canada SARDs are more prevalent among those of Asian and Aboriginal ethnicities (8,9), as compared to Caucasians. Two Canadian population-based studies have shown that Aboriginals, particularly females over 45 years of age (9), may be affected twice as often as non-Aboriginals (10).  SARDs are considered rare - a widely-used definition of rare disorders includes those affecting less than one per2,000 (11), or 50 per-100,000, individuals – and as illustrated in Table 1.2, most SARDs meet this definition. Because of this rarity prevalence estimates for Canada are limited (8). However in the first and only (known todate) population-based investigation of the epidemiology of each SARD in adults, the prevalence of SLE, SSc, SjD, PM, and DM in the province of British Columbia (BC), Canada, was 113.9, 21.3, 21.3, 5.4, and 9.2, respectively, per-100,000 adult in 2007(1). While these estimates are low, they were determined with high sensitivity and case  1  capture was not limited to a single algorithm or institution. This investigation showed how SARDs-VD are even rarer, with their prevalence ranging from just 0.2 per-100,000 for TA to 17.3 per-100,000 for GCA in BC (1). In comparison, the prevalence of rheumatoid arthritis (RA) in Canada is approximately 1,000 per-100,000 adults (12) while the adult prevalence of osteoarthritis (OA) in BC is ten-times greater at approximately 11,000 per-100,000 (13,14).  1.1.2 Pathogenesis The aetiology and exact pathogenesis of most SARDs is unknown; however, they do share an autoimmune origin. All SARDs elicit an immune response against the body’s own cells and tissues, instead of a foreign body or pathogen, that lead to inflammation and potentially to organ failure. There may be shared aetiologies or environmental triggers too, with solvents linked to the development of SSc and SARDs-VD (8), silica dust to SLE, SSc, and SARDs-VD (8), and Epstein-Barr virus to SLE and SjD (15). This abnormal immune activity can be cellmediated or humoral, though these two mechanisms work hand-in-hand. When cell-mediated, B-lymphocytes (Bcells) produce antibodies (auto-antibodies in this case) against antigen proteins on the surface of body’s own cells (16). Antibodies may be produced against nuclear components like DNA (referred to as antinuclear antibodies, or ANA) (15) or against the cytoplasmic components of white blood cells (referred to as antineutrophil cytoplasmic antibodies, or ANCA) (17). These are present in multiple SARDs: ANA in SLE and PM/DM (15,17), and ANCA in PM/DM (15,17) CSD (18), and Wegener’s (19). Cases diagnosed with one SARD, such as SjD, may even test positive for auto-antibodies usually produced in other SARDs (20).  When these auto-antibodies bind to ‘self’-antigens, immune complexes are formed and deposited in sites around the body (4,15). When they deposit in organs – like the skin and kidneys in SLE (4) – injury occurs directly at those sites. They can also deposit in blood vessels where they promote further systemic inflammation and damage by activating the complement pathway (15). Humoral immunity involves another type of white blood cell, the T-cell, which stimulate B-cells to produce antibodies. T-cells enhance the immune response by releasing pro-inflammatory substances called cytokines (16), most often interleukins (IL) (15).  One T-cell of particular importance in  autoimmunity and chronic inflammation is the TH17 cell, which produces IL-17 (15). Usually T-cells are only  2  activated by foreign antigens, but in autoimmune disorders they are activated by the body’s own antigens, and carry out and mediate the attack of its own cells and tissues (16).  Different cytokines may be involved in this process depending on the specific SARD, such as type-I interferons in SLE (15), IL-4 in SSc (21), and IL-2 and IL-6 in GCA (22). Still, this general process of autoimmune-mediated attack, followed by systemic inflammation and dysfunction and damage in organs, blood vessels, and connective tissues, links all the SARDs together (15). The individual SARDs-VD are differentiated by the size and/or type of blood vessel affected, but all result from autoimmune-mediated inflammation and degradation of these vessels, often from the formation of lesions (2). Different tissues may be targeted amongst SARDs-CTD (15) – the epithelia of the salivary and lachrymal glands in SjD (20), muscle fibres in PM (23), and blood vessels in DM (15,24) – but the clinical implications of this chronic inflammation are similar throughout the spectrum of SARDs. On top of this, many individuals are actually afflicted with more than one SARD: SjD often develops in those with SLE or PAN (20,25), and about 25% of SSc cases also have PM/DM (23). These commonalties amongst the SARDs will be emphasized in the next section when describing their many clinical manifestations.  1.1.3 Clinical Manifestations We have just described how a similar autoimmune-mediated mechanism underlies the pathology of all the SARDs. The systemic inflammation characterizing these disorders leads to a diverse spectrum of manifestations. SLE is one of the most well-known SARDs and considered the “prototypic systemic autoimmune rheumatic disease” (15). In addition to systemic symptoms and haematological problems - like anaemia and thromboses (17,26) - a multitude of organ-specific effects are seen. SLE can affect nearly every organ system, but the skin, and renal, central nervous (CNS), and musculoskeletal systems (particularly the joints) are affected most (17,26). Renal disease (or lupus nephritis (LN)), which is very common, can result in hypertension (4,26) that further increases the already elevated risk of coronary artery disease and stroke (27,28) in these patients.  Given their common pathology all SARDs share some generalized or systemic features. These include headache (4,29,30), fatigue (2,4), fever (2,4), muscle pain (18,23-25), joint swelling and/or pain (2,4,18,19,25,29), peripheral neuropathy (18,20,25), hair loss (4,24), and rashes (4,24,25) and other skin ulcerations (19,24,25). Many different  3  SARDs can damage the same internal organs, including those of the cardiovascular, respiratory, gastrointestinal, renal, and nervous systems, via inflammation. Some of the adverse cardiovascular effects include stroke (4,29), heart failure (4,18), and cardiomyopathy (23), while the respiratory morbidities include reactive airway disease (18), interstitial lung disease/fibrosis (20,29), and pulmonary hypertension (21,29). In addition to LN, renal failure (19,25) is common in other SARDs too, as are impaired digestion and gastrooesophageal reflux disease (GERD) (20,21). Some of these manifestations (ie. neuropathy and nephritis) are quite progressive and others can resolve, but each of the SARDs generally relapse and remit over the patient’s lifetime. It is proposed these relapses stem from the immune system encountering the same self-antigen that first triggered the disease and, often in conjunction with an environmental trigger, launching another damaging autoimmune response (15). Unfortunately these subsequent responses tend to be stronger than the initial ones (15).  Particularly in SLE, these relapses are referred  to as flares.  1.2 SARDs Treatment 1.2.1 Current Therapies Consistent with their shared pathogenesis and manifestations, the different SARDs are treated with many of the same medications, the details of which are outlined in Table 1.3. Systemic glucocorticoids (GC), as they decrease both inflammation and the abnormal immune response (4), are a mainstay for treating every SARD (4,18-21,2325,29,31). However their long term use is associated with multiple adverse effects including weight gain/obesity, cataracts, glaucoma, adrenal suppression, glucose intolerance/hyperglycaemia/diabetes, hyperlipidaemia, hypertension, and avascular necrosis of bone (17,26,27,32). Because of these, GC are often given in conjunction with an immune-modulating or -suppressing agent to minimize the dose required and therefore the associated side effects (32). The powerful but toxic immunosuppressants include cyclophosphamide, methotrexate, azathioprine, cyclosporine, and mycophenolate (19). Methotrexate has anti-inflammatory and immune-modulating properties while azathioprine works to reduce the number of immune cells (4). Cyclophosphamide, being associated with infertility, an increased susceptibility to infections, and bladder and haemotologic cancers (19,25), is considered the most toxic of these; therefore, it is generally reserved for severe or unresponsive cases (4,20). Because of this, mycophenolate has become a popular alternative to cyclophosphamide and azathioprine. It has been used successfully in many studies as an induction (33-37) and maintenance therapy in LN (38-52) with fewer adverse 4  events like opportunistic infections (33,40,41,53,54). It appears to be well tolerated (55) and unlike cyclophosphamide, does not decrease fertility although it cannot be taken during pregnancy (56). The antimalarial hydroxychloroquine, an immune-modulator, is also less toxic than the conventional immunosuppressants (32). It is employed in SLE (4), SjD (20), and DM (24). Some other drugs used for multiple SARDs include non-steroidal anti-inflammatory drugs (NSAIDs) to alleviate joint and muscle pain (4,20,28,57), aspirin to improve blood flow (29,57) and prevent stroke and vision loss (22,28), and calcium channel blockers for vasospasm in SSc (21,57), calcinosis in DM (24), and hypertension (29). Angiotensin-II converting enzyme (ACE-II) inhibitors, which are anti-hypertensives with additional renal benefits, are used to manage LN (27) and prevent renal disease in SSc (17), amongst other SARDs. Immune globulins are another option for SLE flares (4), and in PM/DM (23,24), Wegener’s (19), and CSD (18).  These drugs are intended to treat the manifestations of SARDs and have definitely improved survival (26). Immunosuppressants have been credited with improving the five-year survival rate in LN from close to zero in the 1950’s (before these therapies were available) to around 85% now (27). Amongst more than 9,500 SLE cases in seven countries (including Canada), standardized mortality ratios (SMRs) decreased by over 60% from 1970-1979 to 1990-2001. In this study longitudinal mortality decreases specifically for many lupus-related causes, like infections and renal disease (58), further illustrate the positive impact of these therapies and other improvements in care. In another Canadian SLE cohort, SMRs not only decreased from 1970-1994, but these decreases were 3-10 times greater than those in the general population (59).  However despite their benefits these drugs can lead to further morbidities requiring additional therapy. GC are associated with myopathy (23), a particular problem for PM/DM cases (17). They can cause gastrointestinal haemorrhage, so histamine-II receptor antagonists (H2RAs) and proton-pump inhibitors (PPIs) are prescribed both to prevent this adverse effect (18) and address GERD, a primary SARD symptom (21,23). GC can also cause osteoporosis, so calcium, vitamin D, and bisphosphonates are needed as preventive therapies (22,23,25). Prophylactic antibiotics may also be prescribed to prevent opportunistic infections, like pneumonia, that can develop during corticosteroid and/or cyclophosphamide treatment (19,25). Since cyclophosphamide is also associated with haemorrhagic cystitis, an agent called mesna is given before and following infusion as a preventive measure (17)  5  1.2.2 Emerging Therapies In addition to these longstanding regimens, there are a variety of other treatments being introduced to market or otherwise increasingly indicated for SARDs. Of particular note are mycophenolate and the biologic immune modulators rituximab and belimumab. Some vasodilators with other treatment indications, including bosentan, sildenafil, and epoprostenol, are increasingly being used to address symptoms of SSc. For many of these emerging therapies, more and larger-scale studies of their safety and efficacy are required for their use to become widespread in SARDs. But they are also much more expensive than conventional treatments (as shown in Table 1.4, about tento-fifty times greater) which presents another barrier to their use.  Mycophenolate Mycophenolate inhibits the production of lymphocytes and antibodies, suppressing both humoral and cell-mediated immunity (60,61). Its lower toxicity when compared to cyclophosphamide and azathioprine is likely from it targeting T- and B-cells more selectively than the older therapies (62). It is officially indicated to prevent organ rejection after transplant (63) but has become a common treatment for LN. Now, its use in SLE is expanding from renal manifestations to treating those of the dermatological, haematological, and musculoskeletal systems (56,60,61,64-67). Studies have suggested it may be more effective than azathioprine (68) and at least as effective as cyclophosphamide (37) for inducing remission in these systems and reducing the number of new SLE flares. Mycophenolate is emerging as an option for other SARDs too, showing promise in SjD (69), DM (70-73) and SARDs-VD (49,74-87). With it increasingly recommended as a treatment for SLE and other SARDs, some may no longer consider mycophenolate - which costs approximately $1,000-$2,000 CDN per-patient annually (88) - as an emerging therapy. However, though available in Canada since 1995 (89) it is still not approved in BC for the management of SARDs.  Rituximab Rituximab modulates cell-mediated immunity by reducing the number of B-lymphocytes. Specifically, it induces Bcell death by binding to the CD20 antigen on its surface (90). It was first approved in Canada in 2000 as a cancer therapy (89) but with evidence from many published reports (91-115) of its efficacy, it was officially indicated for  6  use in Wegener’s in late 2011 (116). Some investigations and reviews have concluded rituximab – which costs about $9,500 per-patient (32) - to be beneficial for SLE, often inducing good renal responses(117-123) and partial or complete disease remission (92,93,114,124-135) or other benefits (91,115,136-163). Initial studies suggest it may also be beneficial in SSc (91,136,164-166), SjD (91,93,125,167-175), PM/DM (93,176-184), and, along with Wegener’s, other SARDs-VD (185-193) including CSD (92,194-197), PAN (198), and GCA (136).  Etanercept (a fusion protein (17)) and infliximab (a monoclonal antibody (17) ) are two other biologics that were approved in Canada in 2001 (89). There is evidence etanercept may be useful in GCA (199-202) and PM/DM (203). Infliximab has been associated with improvements in PM/DM (203-206) and SjD (207,208), and has been shown to reduce inflammation and induce disease remission in SARDs-VD (209), including GCA (201,202,210-213) and Wegener’s (214-224). Still interest in these two very expensive therapies (the annual per-patient cost for etanercet is approximately double that of rituximab (225)) is very guarded Although both bind and inhibit tumour necrosis factor (TNF)-α, but one cytokine whose levels are elevated in SLE (226), its role in the pathogenesis of SLE is not well-defined, with suggestions it may actually be protective or otherwise beneficial in these patients (227,228). Shockingly, etanercept and infliximab may even induce SLE (17,226-228)! They have been used extensively in other forms of arthritis such as RA, ankylosing spondylitis, and psoriatic arthritis (229), but there are reports of such cases developing auto-antibodies and SLE or a lupus-like syndrome after exposure to these therapies (230-243).  Belimumab Belimumab is brand new and has generated much publicity since it is considered to be the first specific SLE treatment to come out in nearly 50 years (244). It is a monocolonal antibody that binds the B-lymphocyte stimulator protein (B-LyS), also known as B-cell activating factor (BAFF), which SLE patients often have higher levels of (245). This action prevents the usual binding of B-LyS to B-cells, thus indirectly inhibiting their survival and differentiation into antibody-producing plasma cells (245). Being so new – it was only formally approved in the United States of America (USA) in March 2011 (246), and just four months later in Canada (244), fewer studies have been published on it, but its efficacy, particularly in relation to standard therapy, has been demonstrated in some clinical trials (247-249). Even more, these trials also suggest the drug is well-tolerated (226,250,251), but any  7  benefits from this therapy will come at a significant price, with annual Canadian per-patient costs for belimumab estimated at $20,000 (32).  Vasodilators When, in SSc, excess collagen is deposited in blood vessel walls these become narrowed and spastic (57) which increases blood pressure and can lead to pulmonary arterial hypertension (PAH) (17). Blood flow to the fingers, toes, and skin is also restricted (this is referred to as Raynaud’s phenomenon (21)), causing ulcers (252), reduced oxygenation to these tissues, and potentially even more fibrosis (57). Calcium channel blockers have traditionally been employed to reverse this spasticity but interest is growing in several other classes of vasodilators (253). Bosentan (an endothelin-1 receptor antagonist) prevents this peptide from binding to vascular smooth muscle cells (252), and may also directly inhibit collagen production (57). It was approved in Canada in 2001 (89) and though officially indicated to treat PAH (254), is used off-label to treat digital ulcers and Raynaud’s phenomenon too (255). Of these emerging SARDs therapies, bosentan is by-far the most expensive (Table 1.4), costing over $40,000 CDN per-patient each year (256). Still, its efficacy for treating and preventing digital ulcers has been demonstrated in studies (257-259), including some RCTs (260,261), and it is approved in Europe for this purpose (252).  Epoprostenol, a prostaglandin analogue, also costs about $40,000 CDN per-year (256), but both it and bosentan may be better treatments for PAH since they are more selective than calcium channel blockers for regulating pulmonary blood flow (57). Sildenafil promotes vascular smooth muscle dilation by inhibiting the phosphodiesterase (PDE)-5 enzyme (262). It was approved in Canada for erectile dysfunction (262) in 1999 (89) but this agent and other PDE-5 inhibitors may be useful for Raynaud’s (263,264), digital ulcers (265,266), and PAH (267). It is cheaper than the vasodilators just described but still five-to-fifteen times more expensive than calcium channel blockers used for this purpose (Table 1.4). Currently its use in PAH is recommended by the European League Against Rheumatism (EULAR) and it has been officially approved in the US for this purpose (253) but not Canada.  Implications These emerging therapies are not suitable for all patients, and may induce further morbidities. Because the biologics are proteins themselves, the body may recognize them as foreign bodies and launch an acute immune response. This  8  is referred to as an infusion reaction, and can result in fever and rash (17). Not only may infliximab and etanercept actually induce SLE, but by inhibiting TNF, these therapies also make patients more vulnerable to infections, especially tuberculosis, and sepsis (17). Rituximab may only be a superior therapy in cases of severe refractory SLE (226) with two very recent RCTs finding it not efficacious in LN (268,269). Rituximab may also increase susceptibility to infections and has been associated with the development of a rare but disabling CNS disorder called progressive multifocal leukoencephalopathy (PML) (226,251). More studies are needed before belimumab can be used in those with CNS lupus or LN (245,250,270) and although SLE is more common in those of black African descent, the efficacy of this drug in such patients was not demonstrated in the initial studies (245,246,270,271). The new vasodilators carry risks as well: bosentan is associated with liver toxicity and being a teratogen (253) cannot be taken during pregnancy.  Aside from their clinical limitations, there are significant costs associated with these therapies. We have just described how the prices of these drugs alone are very high – at least five-to-fifty-times those of longstanding therapies - with some costing $10,000 or more per-patient each year (Table 1.4). On top of this, some (infliximab, rituximab, belimumab, and epoprostenol) must be administered by intravenous infusion, an additional expense not required for oral formulations of hydroxychloroquine, calcium channel blockers, GC, or methotrexate (32,272). Plus the administration of epoprostenol requires the implantation of a central venous line (272), increasing costs and the risk of infection. Still, given the adverse effects we described for some current therapies, and their limited ability to control the progression of these diseases and development of complications, both clinicians and patients are excited at the prospect of these new options.  1.3 SARDs as a Group of Diseases In sections 1.1 and 1.2 of this chapter we established how SARDs have shared pathogenesis, manifestations, and treatments, but despite these commonalities they have traditionally been studied as separate disorders. Now there is increasing recognition, particularly following the Canadian Arthritis Society’s multinational Consensus Conference on SARDs in 2005, of the benefits that could come from studying them together (8). In section 1.1.1 we described how SARDs are quite rare, with most of the individual disorders affecting less than 50 per-100,000 adults. This means focussing on a single disorder severely limits the number of cases available for study: to illustrate, the  9  aforementioned BC SARDs prevalence study identified only 711 SSc cases and 710 SjD cases in 2007 (1). These numbers are miniscule when compared to BC’s entire adult population, with each disorder affecting only 0.02% of it (273), but with 12,966 collective SARDs-CTD cases identified in 2007 (1), the SARDs-CTD together affected twenty-times more BC adults (0.4%) (273). When conducting research, small samples limit the ability to identify trends, and attribute these trends to actual phenomena instead of chance alone. But when the SARDs are grouped, larger samples are available and statistical power is gained.  Apart from this methodological advantage, grouping the SARDs also emphasizes the public health impact of these disorders. It was noted at the Consensus Conference these individually-rare disorders can be considered “common” when grouped (8), and this is evident when comparing the single-SARD prevalence estimates in BC with the combined: the prevalence of SARDs-CTD together was 388.6 per-100,000 adults (more than three-times the prevalence of SLE alone), and the prevalence of the SARDs-VD combined was 31.9 per-100,000 adults (1). Being rarer, this transformation is especially important for SARDs-VD. There were only 135 diagnoses of PAN, and 58 diagnoses of TA, located in BC in 2007 (1), and researchers, sponsors, and the general public may see little benefit in studying a disorder that affects just one hundred or-so individuals. But they may think differently when thousands (all SARDs-VD combined) and tens-of-thousands (all the SARDs combined) (1) are affected.  Increasing the public profile of SARDs is important because of the high economic costs, both direct (in terms of actual medical costs) and indirect (the value of productivity losses from time spent seeking healthcare or unable to work) incurred annually by the average SARD patient in Canada. In 2007 the annual costs of SSc averaged $18,543 CDN (95% CI: $16,598-$20,308) per-patient (274). Canadian estimates for SLE are also high, costing approximately $60,590 2007 CDN (275) per-patient cumulatively over four years, and approximately $22,352 2007 CDN (276) per-patient in 1990. Per-patient cost estimates of the direct medical costs of many SARDs are detailed in section 1.5 and Table 1.5-1.6 of this chapter. Combined, SARDs-CTD and SARDs-VD affect about 0.4% of the adult population (1) while RA affects more than twice as many adults, about 1% (12). However these annual perpatient cost estimates for SARDs have exceeded multiple Canadian estimates for RA (all in 2007 Canadian dollars): $8,248 (277) annually from 1990-1994, $13,029 (278) in 2000, and $16,552 (279) in 2002. Therefore the total  10  economic burden of SARDs in Canada is clearly disproportionate to the number of individuals affected, and may be similar to that of RA, a more well-known disorder.  1.4 Significance of the Health Resource Utilization and Economic Burden of SARDs As detailed in section 1.2, the health resources consumed in the management of SARDs are vast and diverse. Their diagnosis and follow-up entail many diagnostic radiologic and laboratory tests, and consultations with multiple physicians such as rheumatologists, respirologists, gastroenterologists, nephrologists, pathologists, and dermatologists, among others. Along with the complex medication regimens, frequent monitoring laboratory tests (for example, monthly during mycophenolate treatment (63), and every three months in SLE (4,21)) and other consultations - annual ophthalmologist visits when taking anti-malarials (32) - are needed to monitor for adverse effects and track disease activity and progression. As SARDs are chronic and often relapse, this heavy utilization can be lifelong. If renal involvement progresses to end-stage renal disease (ESRD), dialysis or renal transplant are required (4). Additionally, cases face an increased risk of comorbidity either as complications of the disease itself or secondary to the treatments used in their management. For example cancer can be directly associated in SSc (21), SjD (20), PM/DM (23,24), and SLE (280-282), or could be related to the use of immunosuppressive therapy (19,20,25), such that vigilant screening is required. Any malignancies that do develop naturally impart an additional burden on the healthcare system. Altogether the health care needs of SARDs patients are great and this high health resource consumption entails considerable costs, particularly in a publically-funded healthcare system. With the number of new SARDs diagnoses increasing, and many SARDs disproportionally affecting seniors (Table 1.7), the very population that is currently expanding, the overall burden of these disorders on the healthcare system may continue to grow. When combined with the very expensive drugs emerging for SARDs, the cost of treating all these cases could be considerable.  1.4.1 Rising Case Numbers As large as the burden of SARDs may be now, it may even get larger since the incidence of SARDs, or number of new diagnoses relative to the total population, may actually be increasing. This is supported by epidemiological evidence from many countries, as listed in Table 1.8. We must emphasize that each increase listed in Table 1.8 occurred within the context of the same investigation. This means they cannot be wholly explained by differences in 11  the makeup of the study populations or methods of case ascertainment between different investigations. Although increases in clinical recognition (283,284) and the sensitivity of diagnostic tests (285) have likely contributed to these additional diagnoses, this may not explain 100% of the rise. A true increase in the incidence of CSD is biologically plausible since it may be triggered by one asthma therapy, leukotriene receptor antagonists, whose use has recently increased (18). As exposure to environmental pollutants may contribute to the development of SARDsVD (286), longitudinal rises in pollution levels may also have led to a true rise in case numbers. Whatever the reason for the increasing incidence rates, the implications for the health care system are great: more diagnoses, even if just of milder or earlier-onset cases, could equate to health resource consumption that would not have occurred otherwise.  With the Canadian population growing older - by 2036, those 65 years and older are expected to make up 24% of BC the population, versus 15% in 2010 (287) - a further increase in the number of new SARDs diagnoses over and above general population growth is anticipated. Plus with survival improving (58,59,288,289) the prevalence of many SARDs is likely to increase too. National healthcare spending on seniors (44% of total healthcare spending in 2009 (290)) is already disproportionate to their share of the Canadian population (14% in 2009 (290)), and when combined with the additional healthcare consumption necessitated by SARDs, a great demand may be imparted on the healthcare system going forward.  1.4.2 Increasing Drug Costs It is exciting to think the therapies emerging for SARDs could improve outcomes and cause fewer adverse effects. However as shown in Table 1.4, their annual per-patient costs (around $10,000 CDN) are about twenty-times greater than those of established therapies. Being so new (244), the cost of belimumab in Canada is less certain but is predicted to be $20,000 CDN (32). Already, biologics are believed responsible for a substantial increase in the direct medical costs of RA in Canada (278), and as their use in SARDs grows, they could certainly impact the costs for SARDs in the same way (26). With the complications of SARDs also having the potential to strain healthcare budgets (291), these additional costs may be worthwhile over the long-term, but detailed breakdowns of the current healthcare costs of SARDs and types of health resources consumed in these disorders are needed for such an analysis.  12  1.5 Review of Available Literature We have established that the diagnosis, treatment, and ongoing care for SARDs mandate a large and diverse consumption of healthcare resources. However, details of this consumption and the associated costs are difficult to find. Therefore a comprehensive literature review was undertaken to locate and evaluate such estimates, and identify gaps in the literature and ongoing research needs in this area.  1.5.1 Methods An extensive literature search was undertaken to locate any studies reporting on the direct health care costs and/or health resource utilization for the SARDs as a whole, or for any group of SARDs or individual disorder. The MEDLINE database (1948 to present with daily update) was searched in July 2011. Keywords and MeSH subject headings relating to either SARDs (including each disorder individually) or health care costs were compiled by mapping the associated subject headings and perusing those assigned to a preliminary set of SARDs costing studies we had located. The disease terms included “Lupus Erythematosus, Systemic”, “Lupus Nephritis”, “Sjogren’s Syndrome”, “Scleroderma” (localized, systemic, limited, and diffuse), “CREST Syndrome”, “Polymyositis”, “Dermatomyositis”, “Connective Tissue Diseases”, “Vasculitis”, “Polyarteritis Nodosa”, “Giant Cell Arteritis”, “Takayasu Arteritis”, “Wegener Granulomatosis”, “Churg-Strauss Syndrome”, and “Microscopic Polyangitis”. Some of the terms for healthcare costs included “Costs and Cost Analysis”, “Health Care Costs”, “Health Expenditures”, “Direct Service Costs”, “Hospital Costs”, “Hospital Charges”, “Drug Costs”, “Fees, Pharmaceutical”, and “Prescription Fees”. This allowed us to collect all possible search terms relating to each of these two concepts (SARDs and healthcare costs), and searches combining the two concepts were executed.  Additional studies were located through the “Cited By” and “Find Similar” features in MEDLINE, Web of Science, and Google Scholar. Economic and Canadian search parameters were also used to increase the specificity of our results. The reference lists of publications initially located, including systematic reviews, were also examined. Studies exclusively involving paediatric patients were excluded, as were those reporting on arthritis or musculoskeletal conditions in general but not any SARD specifically. Commentaries, editorials, and materials with previously presented work, including subgroup and secondary analyses, were also excluded. Foreign currencies 13  were converted to Canadian dollars based upon the historical exchange rate on December 30-31 of the reporting year (292). From this, all Canadian costs were inflated to 2007 Canadian dollars using the BC Health Care Component of the Canadian Consumer Price Index (293).  1.5.2 Results The MEDLINE searches retrieved a total of 91 records, of which 13 were duplicates. After their removal, 6 additional non-English records were excluded, as were 18 that were not primary research articles. The titles and abstracts of the remaining records were screened for relevance, and 23 studies reporting primary data on the direct health care costs and/or utilization of at least one SARD were selected from MEDLINE for review. An additional 10 articles were located through handsearch, for a final total of 33. Some overlap existed between cohorts and investigative groups – such as longitudinal follow-ups of cross-sectional studies, and the cost and utilization reports from one investigation being split between two papers – but only papers with a unique primary objective, and/or those with otherwise unavailable primary data were included.  There were many inconsistencies amongst these 33 studies, which illustrated many limitations in the current knowledge. For one, no studies were found that reported on the SARDs collectively. Instead, almost all examined just a single SARD, with the exception of one analysis of PM and DM (294), two very similar disorders, and an abstract reporting on the costs of SARDs-VD together (295). Not only have the SARDs never been studied together, but with more than two thirds (23 of the 33) of the selected papers reporting on SLE, the other disorders have been practically ignored. Amongst most studies, the period of follow-up for each patient was brief, often twelve months or less. Only three of the studies were longitudinal analyses, and just two of these produced annual cost estimates.  Most reports, 24 of the 33, concerned North America (Canada, USA, and Puerto Rico) and/or the United Kingdom (UK), with five for Canada alone (274,276,294,296,297), thirteen for the US alone (295,298-309), one for Puerto Rico alone (310), two for the UK alone (311,312), one comparing Canada and the US (313), and two (314,315) reporting on all three countries. There were four reports from Europe (316-319), and five from Asia (320-324). As detailed in section 1.7.1 many factors, including currency conversion, and transnational differences in health  14  insurance, health care delivery, and unit prices, limit the applicability of international estimates in informing Canadian health care policy.  This large group of studies can first be divided based on the data source, either clinic-based or from administrative databases. With clinic-based sources, utilization data are obtained directly from patients (usually by completion of a questionnaire, often a modified version of the economic portion of the Stanford Health Assessment Questionnaire (HAQ)) and/ or from their actual medical records. The cost of each health care service/resource is then estimated from multiple sources. With administrative databases, utilization data are obtained through the healthcare claims collected for billing purposes by governments or health insurance providers. Cases are identified on the basis of having a health encounter with an International Classification of Diseases, 9 th (ICD-9) or 10th (ICD-10) revision, diagnostic code for the SARD in question recorded with it, and all claims billed for that case are included. Such studies are population-based since investigators have no direct contact with the patients, or their records or healthcare providers.  1.5.3 Results from Clinic-Based Studies The majority of the publications located (21 of 33), including all but one of the Canadian studies, used clinic-based data. Cases consisted of patients admitted to the study hospital, attending specialty clinics, and/or enrolled in a regional or national disease registry. Clinical examinations were conducted to confirm the diagnosis in accordance with disease-specific criteria. For instance SLE cases needed to meet at least four of the American College of Rheumatology criteria (276,302,309,312-314,317,322,323). The characteristics of these studies and their main findings are listed in Tables 1.5a, 1.6a, and 1.9a. The annual mean overall direct per-patient healthcare costs ranged, in 2007 Canadian dollars, from $5,038 (274) – in a Canadian SSc study – to $17,413 (313) – in a US study on SLE. All estimates referenced in this section pertain to gross healthcare costs – the costs for all health resources consumed by SARDs cases, and not just those resources related to SARDs care. Gross mean annual outpatient costs ranged from $223 per-patient for SSc cases in Hungary (319) to $8,098 per-patient for the same US SLE cases that accounted for the highest overall estimate (313). Annual per-patient gross medication costs spanned $241 for SjD cases in the UK (311) to $4,979 for Canadian LN cases (297). Hospital costs were provided inconsistently, with some studies only reporting per-patient estimates for total hospital costs ($1,670-$8,723), which included day  15  surgeries, ER visits, and/or rehabilitation stays, as well as inpatient admissions. Other studies just reported average annual inpatient costs, ranging from $951-$7,996 per-patient.  1.5.4 Results from Population-Based Studies The literature search found 12 population-based studies. From these only one was conducted in Canada, where administrative data was used to study PM/DM. The characteristics of these studies and their main cost and utilization findings are listed in Tables 1.5b, 1.6b, and 1.9b. In these, the mean annual overall direct per-patient gross healthcare costs ranged from $1,399 (324) to $28,312 (304) per-year for SLE cases in Taiwan and the USA, respectively. The Taiwanese study did not include medications. The minimum average annual per-patient gross cost estimate for any SARD amongst those studies that included all three components (outpatient, hospital, and medication) was $8,685 (300). This was the primary estimate in a US study on SSc. Gross mean outpatient costs for SARDs ranged from $713 per-patient for SLE cases in Taiwan (324) to $18,183 per-patient for LN cases each year in the USA (301). Drug costs ranged from $1,882 for US SSc cases (300) to $7,256 per-patient for US LN cases (301) but were only reported in three of these population-based studies. Again, there were inconsistencies when reporting hospital costs: total hospital costs were $2,964-$3,871, with inpatient costs of $682-$11,067 (all in 2007 Canadian dollars). In general the population-based studies produced higher annual per-patient estimates of gross costs, but five of the seven costing studies were from the USA. One explanation may lie in the healthcare prices found in this nation which, as outlined in section 1.7.1 of this chapter, are typically higher than other countries’.  1.5.5 Results from SARDs-Attributable Costs Of course SARDs are not the responsible for every health resource a SARD case consumes, and the per-patient estimates cited above definitely encompass some costs incurred by individuals without a SARD. To accurately quantify the net burden imparted by these disorders on the healthcare system we sought published estimates of direct medical costs that could be attributed to SARDs. Unfortunately we only found eight studies providing any estimate of the incremental costs of SARDs or reporting on some aspect of attributable costs or utilization (Table 1.9). Three compared the gross healthcare costs of SARDs cases to those of a matched control group (299,304,311), one population-based study separately tabulated the costs of just claims with an SLE diagnostic code (301), and another 16  collected data on SARDs-attributable utilization exclusively (316). From these, the annual incremental direct medical costs of SARDs (being the additional healthcare costs that SARDs are responsible for, above baseline consumption) ranged (in 2007 Canadian dollars) from $2,964 per-patient (301) to $14,964 per-patient (299). Both of these estimates were for SLE in the USA. The proportion of gross medical costs that were attributable to SARDs ranged from 15% to 63%. In a sixth study, the gross per-patient cost estimates for SLE cases were compared to the average per-capita healthcare costs amongst the general US population. Here, 66% of costs, or $10,566 per-patient annually, could be attributed to SLE (302). Incremental utilization, which was only measured by four groups, is listed in Table 1.9b. In one report, 70% of hospital admissions for SLE cases could be attributed to SLE either directly or indirectly (325). In another it was determined SLE cases made an average of 4.9 additional outpatient visits each year than did the general Quebec population (276).  1.5.6 Summary of Current Cost Estimates Estimates for the annual healthcare costs of SARDs clearly run a wide spectrum, with mean per-patient estimates of overall costs ranging (in 2007 Canadian dollars) from nearly $1,400 to over $28,000 per-year. In Canada we found the annual per-patient costs for PM/DM cases were $4,006 (294) and for SSc cases they were $5,038 , while the annual per-patient costs of SLE cases ranged from $6,210 (315) to $12,122 (297). With some of these estimates exceeding Canadian reports of the direct medical costs of other forms of arthritis (Table 1.10), the burden of SARDs is clearly significant, and potentially greater than that of these better-studied disorders. Additional reports from the UK (311), USA (299), and Hungary (319) comparing the direct costs of RA to some SARDs further supports this revelation.  1.5.7 Research Gaps Our review showed that SARDs potentially impart an outstanding healthcare burden but also revealed many gaps in the current knowledge of this burden, particularly pertaining to Canada. For one there are no Canadian estimates concerning any of the SARDs-VD, let alone the SARDs together. As two of the SLE estimates (276,315) were calculated in the 1980’s and 90’s, they are unlikely to reflect current healthcare prices and utilization patterns. As well nearly every estimate was cross-sectional, allowing little context in which to interpret the results, and neither of the two longitudinal investigations we located had produced annual Canadian cost estimates. Greatly needed are 17  estimates of the net medical costs stemming from SARDs but there are few available from any country, and again, none from Canada. Most importantly, the Canadian estimates were limited by all but one (294) being clinic-based, since population-based administrative health databases permit the study of a much larger and unbiased sample of cases over many years. They also provide detailed data on each health encounter and its cost, making cost estimates more precise. The one population-based Canadian estimate did not even include medication costs. We intend to fill these gaps in our forthcoming research by using population-based Canadian data to produce the largest, longest, and most comprehensive estimates of the healthcare burden (including the net burden) of SARDs to-date  1.6 Research Objectives The primary objective of this study is estimate the health resource utilization and direct medical costs of SARDs at the population level, and longitudinally.  This will be accomplished using twelve years of patient-level, administrative health claims data from the Canadian province of British Columbia (BC). The quantities and types of health care services consumed by a cohort of SARDs cases in each year will be captured. These include all provincially-funded hospitalizations, surgeries, outpatient visits and investigations, and all prescription medications, regardless of funding. We will sum the costs of these services, annually and cumulatively, and with these sums calculate the average annual healthcare costs incurred by each case.  Once these estimates are obtained, subsequent study objectives are to:  a) Determine the net health resource utilization and economic burden that SARDs impart on the provincial healthcare system  b) Identify any longitudinal trends in cost and utilization  18  1.7 Study Significance This study will be like no other. Our literature review confirmed the paucity of estimates available, particularly from Canada, of the direct medical costs of SARDs, with none determined for the SARDs as a group. Even fewer were the number of population-based or longitudinal estimates for any SARD, and with many SARDs including SLE following an “unpredictable course” (26) and characterized by periods of flares and remissions, cross-sectional estimates may not encompass the long-term burden of disease. Instead our analysis will be population-based and longitudinal, and by spanning twelve-years, be the longest of any previous. This will add another dimension to our work, enabling us to identify trends and observe the impact of any new technologies, drugs, or policies in this time. Multiple years of study can also help predict areas of future consumption and costs, which could be used by healthcare administrators to better prepare for ongoing healthcare needs.  We located no incremental cost estimates from Canada in our review, and of the just eight studies reporting on attribution, all but one (311) reported on SLE alone. The reports themselves were inconsistent and limited (Table 1.9). By quantifying the additional costs incurred by SARDs patients, which for most SARDs are currently unknown, we shall emphasize the impact of these disorders and provide more incentive for reducing their costs and morbidities.  1.7.1 Importance of Canadian Estimates Only eight studies have reported cost and/or utilization estimates from Canada and unfortunately many factors limit the usefulness of foreign estimates in Canadian health policy decisions. In addition to different currencies, these include transnational differences in health insurance, and health care prices and delivery. Precision is always lost when converting cost estimates between currencies, but this was a bigger problem with investigations from Taiwan (324) and Hong Kong (322,323) that reported their estimates in US dollars instead of their respective national currencies. It meant two currency conversions took place between the original calculations and final Canadian dollar estimates we used for comparison, making these less reliable. Even when expressed in the same currency, there are variations in the unit prices for health services between countries, which reduces the comparability of total cost estimates. As shown in Tables 1.5 and 1.6, the annual per-patient estimates from the US, where health care costs are usually high, were generally the largest. However, in the two investigations where US cost estimates were  19  calculated with Canadian unit prices (314,315), these estimates were much lower and differed little from the Canadian. In contrast, other countries may have lower health care prices that drive down the totals, and reliance on these would underestimate the burden of SARDs in Canada. This may explain the comparatively low hospital perpatient estimate from Taiwan (at most, $835 2007 CDN annually (324) ), and overall per-patient estimate from Hungary ($5,177 2007 CDN) (319).  Purchasing power parity (PPP) can help standardize international estimates by adjusting for transnational differences in the unit prices of the same goods and services. In theory, after conversion, any residual difference in average perpatient cost estimates would be from the patients in each country consuming different quantities of the same health service, and not the unit price of each service (314). However, PPP conversions do not make international estimates fully comparable to Canadian ones: the conversion factors are general, and may not sufficiently standardize healthcare expenditures (314).  Costs of care are also influenced by practice patterns within various countries. In the Hungarian study, over 98% of SSc cases were hospitalized in one year while outpatient costs made up only 4% of the total. The authors attributed this to national reimbursement practices that favour inpatient care over outpatient, and transportation inefficiencies that make day trips difficult for patients (319). Even utilization estimates from the US may not be applicable to Canada, with reports indicating Canadian SLE cases consume more hospital resources (313,315) but see fewer specialists (276) and undergo fewer diagnostic tests (276,314,315) than American. This implies that strategies to reduce hospitalizations in SARDs would have a greater financial impact in Canada, and this would not be as apparent if only examining US figures. These differences in practice patterns are another limitation of PPP conversion. As Clarke et al note, this procedure assumes transnational price differences are just due to macroeconomics, but medical care is not provided identically throughout the world. They propose that PPP adjustment makes the actual costs of smaller-scale, more basic services - like blood tests - comparable between countries, but not necessarily more complex ones like hospitalizations (315). For example, between two countries an admission for the same purpose may use different resources and entail a different length-of-stay (LOS), and thus a different cost (315).  20  A final difference relates to health insurance, as Canada has a single-payer, public system, while there are multiple providers in the US, some of them for-profit. US health care costs are higher than Canada’s, regardless of the insurance provider, but the added influence of for-profit companies, whose costs are often higher than those from government-based insurance (302), can further reduce the comparability of any cost estimates. In addition to the provider (302,306,308,326,327), the type of plan (capitated versus fee-for-service) can also influence health care costs in the US (299,301,309), but is not relevant in Canada. This makes comparing cohort mixes and cost estimates between the two countries even more difficult.  1.7.2 Advantages of Administrative Data Most estimates of the healthcare costs of SARDs were produced using clinic-based data. This limited their accuracy and external validity. Below we describe the many advantages of population-based data and how it will help us overcome these earlier limitations. Population-based data has only been used in only one other Canadian SARDs costing study. In that study, the costs of only two disorders (PM/DM) were investigated and medication use was not included, which will make ours the first in Canada to include all three health resource components: outpatient encounters, hospitalizations, and every dispensed prescription.  Large and Minimally-Biased Study Population With clinic-based data, cases are recruited from patient populations attending at these tertiary rheumatology clinics, but population-based data allows for study on a much larger scale. Retrospective data from multiple years can be accessed, thus permitting the identification of trends. While recruitment from tertiary centres may bias the study population toward cases whose disease is severe enough (326) to warrant referral there, using population-based data allows nearly every case in the province to be followed. Considering the well-documented association between greater disease activity and/or lesser health status (including nephritis) in SARDs and greater healthcare costs (274,276,291,294,297,299,301,302,304,305,311-314,316,322,323,328) the healthcare costs of tertiary-clinic patients would not reflect that of milder cases. Patients attending specialized rheumatology clinics may also consume more resources than patients treated in other settings (297) simply as a result of being seen by specialists with specific interest in, and knowledge of, their disorder (311). Therefore estimates derived from these clinic populations may misrepresent actual mean per-patient costs.  21  In contrast, Canadian databases like ours provide a less-biased and more representative cohort regardless of the urban/rural residence, income level, employment status, or race of cases. Since these factors have also been shown to impact costs (297-299,308), this source will make our cost estimates even more reliable and representative of most cases. Selection bias aside, the generalizability of the results from clinic-based studies is often limited by the small number of cases available for study (this ranged from 67 to 812 in our review). These cases may not adequately reflect the full spectrum of disease severity and health resource utilization in SARDs (304), but since we identified approximately 13,000 SARDs cases in just one year of our data (1), we expect to overcome this limitation.  Comprehensive and Precise Health Resource Utilization The questionnaires used to collect utilization data from clinic patients, especially the HAQ, have been used extensively in costing studies. However the onus is on the patient to report their utilization and they may do so inaccurately (329), leading to incomplete and potentially biased estimates of health resource use. This is more likely the longer the recall period, with some investigations inquiring, at one time, about all consumption in the previous twelve months (302). Instead administrative databases record every health encounter in a centralized and systematic manner, which eliminates this problem. A particular advantage of BC data are its records of all dispensed prescriptions, regardless of funding source, which are not available in many other Canadian provinces. Together, this shall equate to more accurate cost estimates and more detailed and comprehensive accounts of utilization.  Precise Costing With clinic data, a unit price is assigned to each service that is consumed, but these prices may not reflect the actual unit costs. Prices derived from the fee schedule at an urban, tertiary care hospital, as used in one study (312), would not reflect the lower costs of care seen in a smaller hospital or town. Prescription prices have particularly lacked precision in previous studies, with some investigators using wholesale prices (thus omitting dispensing fees), or approximating a therapy duration of three (311,312) , six (302), or nine (316) months in their calculations when details were not available. This incorporates some uncertainty into the final cost estimates from these studies, but our estimates will be more precise since each outpatient and prescription claim in the BC database specifies what the  22  health ministry reimbursed for it. Given the longitudinal nature of our study, with payment arrangements and actual fees changing over time, having these precise costs is especially valuable.  1.8 Study Implications SARDs impart a tremendous physical and psychological impact on patients and their families. A complex medication regimen is required to alleviate symptoms, slow disease progression, and prevent complications, but the current treatments remain suboptimal. The clinical manifestations can significantly limit participation in family life and career opportunities, while the associated health resource consumption imparts a great economic burden on government and society. But given the rarity of these disorders, the role they play in driving healthcare costs, especially compared to other forms of arthritis, is not well recognized. This has limited the research funding and support they receive, and in turn, the quality of care available to people living with these conditions.  We hope to turn this around and establish SARDs as a research priority by producing comprehensive estimates of their current health resource utilization and economic burden at the population-level. Our analysis will be the first of its kind in the world, allowing us to fill many existing gaps in research on the direct medical costs of SARDs. In undertaking this innovative work we hope other researchers will follow in our footsteps and investigate the health outcomes and pharmacoeconomics of these disorders in more detail. With the number of SARDs cases rising, and some very expensive therapies emerging to treat these disorders, their healthcare burden is expected to rise. As such, current estimates of these costs are especially needed by health policymakers.  This chapter has provided an introduction to the SARDs, including the current and emerging treatment options, and described the potential benefits of studying them together. A review of the literature available on the direct health care costs of SARDs was also presented, and gaps in current knowledge identified. From this, the current study and its objectives were outlined, along with its significance and potential impact. The next chapter will detail the data source and methods that were used in undertaking this analysis.  23  1.9 Tables Table 1.1 Prevalence and Incidence of SARDs, Stratified by Sex Disorder Study Location Year(s) Prevalence (95% CI) per-100,000  SARDsCTD (all)  SLE  Incidence (95% CI) per-100,000  Female  Male  Female  Male  Avina-Zubieta 2011 (1)  British Columbia  2007  591.9  176.3  67.5  20.1  Bernatsky 2011 (5)  Manitoba Nova Scotia Quebec  2003  -  Greece  British Columbia  19822001 19872006 2007  120 (110-130) 100 (80.0-140) 80.0 (80.0-90.0) 9.46 (6.14-12.78) 5.8 (2.0-12.0) 30.4  -  Alamanos 2003 (330) Alonso 2011 (331) Avina-Zubieta 2011 (1)  690 (660-720) 420 (360-570) 420 (410-440) 69.27 (65.90-72.64) 29.2 (20.0-40.7) 193.9  5.9 (4.9-7.0) 24.2  1.1 (0.7-1.7) 3.6  Chakravarty 2007 (332)  Pennsylvania California  2000  Govoni 2006 (333) Gudmundsson 1990 (334) Hochberg 1987 (335) Hopkinson 1993 (336) Lopez 2003 (337) McCarty 1995 (338) Naleway 2005 (339) Nightingale 2006 (340) Nived 1985 (341) Nossent 1992 (342) Nossent 2001 (343) Uramoto 1999 (288)  Italy  19962002 19751984 19811982 19891990 19982002 19851990 19912001 19921998 19811982 19801990 19781996 19501992  253 (248.3-257.7) 184.2 (181.4-187.0) 100.1  38.7 (36.8-40.7) 25.5 (24.5-26.6) 12.0  62  7.2  5.8  0.8  12.5 (7.6-19.3) 45.4 (37.6-53.1) 57.91 (51.61-64.21)  2.0  -  -  3.7 (1.5-6.0)  6.5 (3.5-9.4)  8.33 (5.84-10.82)  3.64 (2.93-4.35) 3.5 (whites) 9.2 (blacks) 8.5 (5.5-10.9) 5.30 ( 4.75-5.86) 7.6  1.5 (0.02-2.9) 0.54 (0.26-0.82) 0.4 (whites) 0.7 (blacks) 1.9 (0.6-3.3) 0.65 ( 0.45-0.85) 2.0  7.86 (2.3-13.2) 4.6 (3.6-5.8) 5.11  1.13 (0.9-3.1) 0.6 (0.3-1.3) 0.91  Italy  Iceland UK UK Spain Pennsylvania Wisconsin UK Sweden Curaco Norway Minnesota  131.5 (95.5-167.5) -  24.8 (9.4-40.2) -  64.8  11.7  83.8 (65.8-101.8) 89.3 (78.9-100.2) -  8.5(2.8-14.2) 9.7 (6.9-12.6) -  24  Disorder  SLE  SSc  Study  SjD  Year(s)  Prevalence (95% CI) per-100,000  Incidence (95% CI) per-100,000  Female  Male  Female  Male  37.9 (29.8-47.5) 100 (19.8-179.3)  -  -  2.5  4.7 (2.2-9.0) Females and males combined: 53.6 (12.2-95.0) 28.2  1.9  0.2  14.17  2.98  -  -  7.0 (2.4-14.4) – 9.9 (4.8-18.2)  1.8 (1.2-2.5) 3.5 (2.3-3.9)  6.4  5.6  0.7 (0.3-1.2) 1.0 (0.5-1.4) 1.5  Voss 1998 (344)  Denmark  1995  Ward 2004 (345)  USA  2000  Alamanos 2005 (346) Allcock 2004 (347) Arias-Nunez 2008 (348)  Greece  19812002 2000  Spain  19882006  Avina-Zubieta 2011 (1) Bernatsky 2009 (349)  British Columbia Quebec  2007  22.2 (14.2-33.1) 44.4 (32.8-58.9) 35.6  2003  -  -  74.4 (69.3-79.7)  13.3 (10.2-14.8)  Geirsson 1994 (350) Mayes 2003 (351) Rosa 2011 (352)  Iceland  19751990 19891991 19992004  11.9  1.5  0.7  0.05  38.98 (35.3-43.0) 47.7 (30.9-70.4)  8.41 (6.8-10.4) 2.8 (0.7-15.7)  2.85 (1.97-4.11) -  0.90 (0.47-1.73) -  1.33 (1.09-1.56) 2.76 (2.39-3.13)  0.55 (0.38-0.71) 0.88 (0.66-1.1)  8.4  females and males: 35.0 (4.0-127.0) 177.4  10.1  0.5  36.4  5.5  6.8  1.7  300 (110-660) – 490 (220-930) 25.6  0-50 (0-280)  -  -  1.9  -  -  6.9 (5.0-8.8) -  0.5 (0.0-1.2) -  11.0 (10.6-11.4)  1.1 (1.0-1.2)  Steen 1997 (353)  SjD  Location  UK  Michigan Buenos Aires, Argentina Pennsylvania  Valter 1997 (354)  Estonia  Alamanos 2006 (355) Avina-Zubieta 2011 (1) Birlik 2009 (356)  Greece  Miyasaka 1995 (357) Pillemer 2001 (358) Thomas 1998 (359) Weng 2011 (360)  19631972 19731982 ?  19822003 2007  British Columbia Turkey  2000?  Japan  1993  Minnesota  19761992 ?  UK Taiwan  20052007  54.0 (7.0-197.0)  4100 (3000-5500) -  2500 (1600-3700) -  25  Disorder  PM only  DM only PM/DM  SARDsVD (all)  Wegener`s  CSD  Study  Location  Year(s)  Incidence (95% CI) per-100,000  Female  Male  Female  Male  10.3  8.0  1.8  1.1  whites: 0.61 (0.037-1.183) blacks: 1.71 (0.292-3.128)  Avina-Zubieta 2011 (1) Oddis 1990 (361)  British Columbia Pennsylvania  Avina-Zubieta 2011 (1) Bernatsky 2009 (362)  British Columbia Quebec  2007  7.5  3.2  1.6  whites: 0.29 (0.0080.572) blacks: 1.14 (0.13-2.15) 0.4  2003  -  British Columbia Spain  2007  young rural: 2.7 (1.6-4.1) 21.1  -  Avina-Zubieta 2011 (1) Gonzalez-Gay 2003 (6) Mohammad 2007 (363) Mohammad 2009 (364) Reinhold-Keller 2000 (365)  older urban: 70.0 (61.3-79.3) 42.2  6.9  3.7  1.25 (0.84-1.85) -  1.36 (0.94-1.98) -  2.26 (1.74-2.77) -  2.10 (1.60-2.61) -  Reinhold-Keller 2002 (366)  Germany  1998 1999  4.84 (2.7-7.0) – 5.12 (2.9-7.3)  4.03 (2.1-5.9) – 5.29 (3.1-7.5)  Reinhold-Keller 2005 (367)  Germany  19982002  Watts 2000 (7)  UK  Avina-Zubieta 2011 (1) Gonzalez-Gay 2003 (6) Mohammad 2007 (363) Mohammad 2009 (364) Watts 2001 (368)  British Columbia Spain  19881997 2007  4.20 (2.90-5.50) 5.94 (4.40-7.40) 1.64 (1.14-2.28) 2.5  3.70 (2.50-4.90) 4.89 (3.50-6.30) 2.35 (1.73-3.13) 2.0  0.335 (0.165-0.68) -  0.27 (0.13-0.55) -  Gonzalez-Gay 2003 (6)  Spain  0.93 (0.60-1.26) 0.98 (0.62-1.47) 0.61 (0.24-1.26) 0.13 (0.09-0.21)  1.04 (0.68-1.39) 1.14 (0.74-1.68) 0.36 (0.10-0.93) 0.13 (0.09-0.18)  Sweden Sweden Germany  Sweden Sweden UK Spain  2007  Prevalence (95% CI) per-100,000  19631982  19882001 2003 19972006 1994  19882001 2003 19972006 19811998  19882001  29.0 (20.2-37.8) -  30.7 (21.7-40.0) -  25.6 (18.9-32.2) 27.5 (20.7-34.2)  12.8 (7.9-17.7) – 15.2 (10.0-20.4)  -  -  -  -  10.7  10.0  13.8 (7.7-19.8) -  18.2 (11.2-25.2) -  -  -  26  Disorder  CSD  PAN  GCA  Study  Location  Year(s)  Prevalence (95% CI) per-100,000  Incidence (95% CI) per-100,000  Female  Male  Female  Male  19972006 19811998  -  -  -  -  2007  5.0  3.0  0.12 (0.00-0.24) 0.21 (0.07-0.50) 0.09 (0.00-0.49) 0.8  0.06 (0.00-0.15) 0.41 (0.19-0.77) 0.09 (0.00-0.51) 0.4  13.8 (7.7-19.8) -  18.2 (11.2-25.2)  0.10 (0.09-0.10) -  0.08 (0.05-0.14) -  -  -  25.6  8.5  0.12 (0.00-0.24) 0.72 (0.42-1.16) 0.61 (0.24-1.26) 3.8  0.06 (0.00-0.15) 1.23 (0.81-1.78) 0.64 (0.25-1.31) 1.6  all ages: 35.7 50+ years: 120.8 -  all ages: 7.4 50+ years: 28.0  36.0 (29.3-44.3) all ages: 55.5 50+ years: 188.6  -  11.00 (8.97-13.54)  18.0 (12.8-24.5) all ages: 20.4 50+ years: 77.1 9.57 (5.56-13.58)  Mohammad 2009 (364) Watts 2001 (368)  Sweden  Avina-Zubieta 2011 (1) Gonzalez-Gay 2003 (6) Mohammad 2007 (363) Mohammad 2009 (364) Watts 2001 (368)  British Columbia Spain  Avina-Zubieta 2011 (1) Baldursson 1994 (369) Boesen 1987 (370)  British Columbia Iceland  Gonzalez Gay 2001 (371)  Spain  19811998  Gonzalez Gay 2007 (372)  Spain  19812005  -  -  10.23 ( 8.60-12.08)  9.92 (8.19-11.89)  Gonzalez-Gay 1997 (373)  Spain  19861990 19911995 19871994  6.33  10.53  -  -  8.94  12.14 all ages: 53.4 50+ years: 177.6  19761995  -  -  29.8  all ages: 27.7 50+ years: 99.5 12.5  urban: 54.5 (49.9-59.1) – 60.2 (55.4-65.0) rural: 14.1 (11.8-16.4) – 29.6 (8.26-33.0)  urban: 8.2 (6.4-10.0) – 8.6 (6.8-10.4) rural: 8.4 (6.6-10.2) – 12.5 (10.3-13.7) -  -  UK Spain  Sweden Sweden UK Spain  Denmark  Gran 1997 (374)  Norway  Petursdottir 1999 (375) Reinhold-Keller 2000 (365)  Sweden  19882001 2003 19972006 19811998  2007 19841990 1982  Germany  1994  -  27  Disorder  GCA  TA  Study  Location  Year(s)  Prevalence (95% CI) per-100,000 Female  Male  Female  Male  4.58 (3.2-5.9) 3.54 (2.4-4.7) 2.82 (1.40-4.20) – 6.90 (4.60-7.60) 24.2 (19.5-28.9) 24.4 (20.3-28.6) 12.1 (8.8-15.4) 0.5  3.24 (2.3-4.2) 1.76 (1.1-2.5) 1.09 (0.70-1.50) – 2.56 (1.60-3.60) 8.2 (4.8-11.6) 10.3 (6.9-13.6) 7.7 (5.0-10.5) 0.2  Reinhold-Keller 2002 (366)  Germany  1998 1999  -  -  Reinhold-Keller 2005 (367)  Germany  19882002  -  -  Salvarani 1995 (376) Salvarani 2004 (377) Sonnenblick 1994 (378) Avina-Zubieta 2011 (1)  Minnesota  19501991 19501999 19801991 2007  -  -  2.8  0.6  Minnesota Jerusalem British Columbia  Incidence (95% CI) per-100,000  28  Table 1.2 Prevalence of SARDs Location  Disorder  Study  SARDs-CTD (all)  Avina-Zubieta 2011 (1) Bernatsky 2011 (5)  British Columbia  2007  Manitoba Nova Scotia Quebec  2003  410 (390-430) 270 (230-340) 260 (250-270)  Alamanos 2003 (330) Al-Arfaj 2002 (379) Alonso (331) Anagnostopoulos 2010 (380)  Greece  2001  39.51 (37.70-41.62)  Saudi Arabia Spain Greece  ? 2006 2007-2008  19.28 17.5 (12.6-24.1) 110  Avina-Zubieta 2011 (1) Bernatsky 2007 (381) Bossingham 2003 (382) Chakravarty 2007 (332) Gourley 1997 (383)  British Columbia  2007  113.9  Quebec  2003  44.7 (37.4-54.7)  Australia  1996-1998  45.3  California Pennsylvania Northern Ireland  2001  Govoni 2006 (333) Gudmundsson 1990 (334) Hart 1983 (384) Helve 1985 (385) Hochberg 1987 (335) Hopkinson 1993 (336) Hopkinson 1994 (386) Johnson 1995 (387) Laustrup 2009 (388) Lopez 2003 (337) Maskarinec 1995 (389) Naleway 2005 (339) Nived 1985 (341) Nossent 1992 (342) Nossent 2001 (343) Serdula 1979 (390) Stahl-Hallengren 2000 (391) Uramoto 1999 (288) Voss 1998 (344) Ward 2004 (345)  Italy Iceland  2002 1984  107.6 (106.1-109.2) 149.5 (146.9-152.2) 21.7 (19.7-23.8) – 25.4 (22.1-28.7) 57.9 35.9  New Zealand Finland UK  1980 1978 1981-1982  17.62 28.0 6.5  UK  1989-1990  24.6 (20.6-28.7)  UK  1991  24.7 (20.7-28.8)  Birmingham, UK Denmark Spain Hawaii  1992 2002 2002 1989  27.7 (24.2-31.2) 28.3 (23.2-34.2) 34.12 (30.63-37.61) 41.8  Wisconsin Sweden Curaco Norway Hawaii Sweden  2001 1982 1990 1996 1970-1975 1986 1991 1993 1995 2000  78.5 (59.0-98.0) 39.0 (30.0-48.0) 47.0 (34.1-54.1) 49.7 (44.3-55.0) 15.3 42.0 68.0 122.0 (97.0-147.0) 21.7 (17.3-26.8) 53.6 (12.2-95.0) – 241 (130-152)  SLE  Minnesota Denmark USA  Year  1993  Prevalence (95% CI) per 100,000 388.6  29  Disorder  Study  Location  Year  SSc  Airo 2007 (392) Alamanos 2005 (346) Allcock 2004 (347) Arias-Nunez 2008 (348) Arnett 1996 (393)  Italy Greece  ? 2002  UK Spain  2000 2006  Oklahoma  1990-1994  Avina-Zubieta 2011 (1) Bernatsky 2009 (349) Geirsson 1994 (350) Kaliterna 2010 (394) Le Guern 2004 (395) Lo Monaco 2011 (396) Maricq 1989 (397) Mayes 2003 (351)  British Columbia  2007  8.21 (6.35-10.07) 14.9 (10.1-21.2) 27.7 (21.1-35.84) 9.5 (5.8-14.6) 469.0 (203.0-930.0) 21.3  Quebec  2003  44.3 (41.1-47.6)  Iceland Croatia Paris, France Italy  1990 2007-2009 2001 1999-2007  South Carolina Michigan  1985 1989-1991  Robinson 2008 (398)  USA  2001-2002  7.1 15.6 (11.8-19.4) 15.83 (12.9-18.7) 25.4 (22.2-28.6) 34.1 (30.4-37.8) 28.6-113.0 24.2 (21.3-27.4) 27.6 (24.5-31.0) 30.0-50.0  Rosa 2011 (352)  Buenos Aires, Argentina Ontario  1999-2004  23.8 (22.0-25.7)  1996?  Estonia Greece  ? 2003  7.1 (3.4-10.8) – 28.0 (9.7-46.4) 35.0 (4.0-127.0) 58.0  Greece  2003  92.8 (83.8-102.5)  Greece  2007-2008  230 (220-750)  British Columbia  2007  21.3  Turkey  2000?  Bowman 2004 (400)  UK  ?  Dafni 1997 (401)  Greece  1992  Goransson 2011 (402) Jacobsson 1989 (403) Kabasakal 2006 (404)  Norway  2009  160 (60-350) – 280 (130-510) 140 (17-510) 400 (40-1320) women: 600 (190-1390) 50.0 (48.0-52.0)  Sweden  ?  2700 (1000-4500)  Turkey  2001-2002  Thomas 1998 (359) Tomsic 1999 (405)  UK Slovenia  ? ?  women: 720 (330-1570) 1560 (920-2660) 3500 (2500-4800) 600 (70-2160)  SjD  Thompson 2002 (399) Valter 1997 (354) Alamanos 2006 (355) Alamanos 2006 (355) Anagnostopoulos 2010 (380) Avina-Zubieta 2011 (1) Birlik 2009 (356)  Prevalence (95% CI) per 100,000 33.9 (15.5-52.3) 15.4 (12.0-18.8)  30  Disorder  Study  Location  Year  SjD  Whaley 1972 (406) Zhang 1995 (407) Avina-Zubieta 2011 (1) Avina-Zubieta 2011 (1) Bernatsky 2009 (362) Avina-Zubieta 2011 (1) Haugeberg 1998 (408) Mahr 2004 (409) Mohammad 2007 (363) Ormerod 2008 (410)  UK China British Columbia  ? ? 2007  Prevalence (95% CI) per 100,000 elderly: 3300 330-700 9.2  British Columbia  2007  5.4  Quebec  2003  21.5 (19.4-23.9)  British Columbia  2007  31.9  Norway  1996  43.9  Paris, France Sweden  2000 2003  9.03 (7.4-10.6) 29.9 (23.6-36.2)  Australia  Reinhold-Keller 2000 (365) Watts 2000 (7)  Germany  1995-1999 2000-2004 1994  Avina-Zubieta 2011 (1) Cotch 1995 (295) 1996 (411) Gibson 2006 (412)  British Columbia  1997 1988-1997 2007  9.5 (7.69-11.61) 14.8 (12.5-17.4) 19.5 (15.3-23.6) 21.6 (17.3-25.9) 14.45 (11.04-18.53) 22.14 (17.92-27.02) 10.3  New York  1986-1990  2.6 (99%CI: 1.7-3.5)  New Zealand  2003  9.35 (6.6-12.1) 11.2 (8.3-14.2)  1999-2003  13.1 (9.9-16.3) 15.2 (11.7-18.6) 5.3  PM only DM only PM/DM SARDs-VD (all)  Wegener’s  GCA  UK  Haugeberg 1998 (408) Koldingsnes 2001 (413)  Norway  1996  Norway  Mahr 2004 (409) Mohammad 2007 (363) Ormerod 2008 (410)  Paris, France Sweden  1988 1993 1998 1984-1988 1989-1993 1994-1998 2000 2003  Reinhold-Keller 2000 (365) Watts 2000 (7)  Germany  Avina-Zubieta 2011 (1) Boesen 1987 (370)  British Columbia  1997 1988-1997 2007  Denmark  1982  Cotch 1995 (295)  New York  1986-1990  Australia  UK  1995-1999 2000-2004 1994  3.04 (1.66-5.10) 4.93 (3.12-7.39) 9.51 (6.91-12.90 3.01 (1.65-5.06) 6.29 (4.21-9.04) 10.90 (8.11-14.33) 2.37 (1.6-3.1) 12.9 (8.7-17.0) 16.0 (11.4-20.6) 6.43 (4.93-8.17) 9.50 (7.69-11.61) 4.2 (2.3-6.2) – 5.8 (3.6-8.0) 6.29 (4.15-9.16) 10.64 (7.73-14.28) 17.3 all ages:37.8 50+ years: 135.4 19.6 31  Disorder  Study  Location  Year  GCA  Reinhold-Keller 2000 (365)  Germany  1994  PAN  Avina-Zubieta 2011 (1) Cotch 1995 (295) Haugeberg 1998 (408) Mahr 2004 (409) Mohammad 2007 (363) Ormerod 2008 (410)  British Columbia  2007  Prevalence (95% CI) per 100,000 50+: 24.0 (16.4-31.5) 30.0 (20.7-39.3) 4.0  New York Norway  1986-1990 1996  2.7 3.3  Paris, France Sweden  2000 2003  3.07 (2.1-4.0) 3.1 (1.1-5.2)  Australia  Reinhold-Keller 2000 Haugeberg 1998 (408) Mahr 2004 (409) Mohammad 2007 (363) Ormerod 2008 (410)  Germany  1995-1999 2000-2004 1994  Norway  1996  2.05 (1.30-3.21) 2.23 (1.38-3.33) 0.2 (0-0.7) – 0.9 (0-1.8) 1.3  Paris, France Sweden  2000 2003  1.07 (0.5-1.7) 1.4 (0.03-2.7)  Australia Germany  1995-1999 2000-2004 1994  British Columbia  2007  1.17 (0.62-2.96) 2.23 (1.34-3.33) 0.2 (0.0-0.7) – 0.7 (0.0-1.4) 1.7  CSD  TA  Reinhold-Keller 2000 (365) Avina-Zubieta 2011 (1)  32  Table 1.3 Drugs Used in the Treatment of SARDs Drug Class Examples Disorder non-steroidal anti-inflammatory ibuprofen, naproxen, SLE (4), SSc (57), drugs (NSAIDs ) diclofenac SjD (20) corticosteroids prednisone, SLE (4), SSc (21), SjD methylprednisone, (20), PM (23), DM (24), prednisolone PAN (25), Wegener’s (19), CSD (18), TA (29), GCA (31) anti-malarials hydroxychloroquine SLE (4), SjD (20), DM (24) immunosuppressants cyclophosphamide, SLE, SSc (21), PM (23), methotrexate, azathioprine, DM (24), PAN (hepatitis mycophenolate, B-negative cases) (25), chlorambucil, thalidomide Wegener’s (19), CSD (18), TA (29), GCA B-cell depletors rituximab SLE (4), SjD (20), PM (23), DM (24), Wegener`s (19) B-cell-stimulator inhibitor belimumab SLE (4) tumour necrosis factor (TNF) inhibitors  etanercept, infliximab  PM (23), TA (29)  chelating agents  penicillamine  SSc (21,57)  vasodilators  sildenafil, bosentan  SSc (21,414)  proton-pump inhibitors (PPIs)  omeprazole (21)  SSc (21), PM (23), DM (24), GCA  histamine-II receptor antagonists (H2RAs)  cimetidine, ranitidine  SSc (21), PM (23), DM (24), CSD (18)  diltiazem, nifedipine (29)  SSc (21), PM, DM (24), TA (29)  calcium channel blockers  parasympathetic agonists immune globulins  pilocarpine  SjD (20) SLE (4), PM (23), DM (24), Wegener`s (19), CSD (18)  Role alleviate joint pain decrease inflammation and immune response  modulate immune response inhibit cell growth, reduce inflammation  reduce number of B cells inhibit B-cell differentiation inhibit pro-inflammatory cytokines reduce collagen metabolism and fibrosis (21,57) address pulmonary hypertension, alleviates vasospasm and fibrosis (57,414) alleviate GERD, prevent corticosteroidassociated gastrointestinal damage alleviate GERD, prevent corticosteroidassociated gastrointestinal haemorrhage (18) reduce vasospasm and fibrosis in SSc (57), calcinosis in PM/DM (24), hypertension in TA (29) alleviate oral dryness  33  Drug Class  Examples  Disorder  Role  anticoagulants  warfarin, heparin  SLE (28), PM/ DM (24), TA (29)  anti-platelet agents  aspirin, clopidogrel  SLE (28)SSc (21)TA (29), GCA (22)  anti-virals (25)  vidarabine, lamivudine, interferon (25)  PAN (hepatitis Bpositive cases) (25)  antibiotics (19)  trimethoprimsulfamethoxazole (TMPSMZ) (19)  Wegener’s (19)  bisphosphonates (19)  pamidronate (24)  SLE (28), PM (23), DM (24), Wegener’s (19), GCA  improve renal function (29), calcinosis (24), enhance blood flow in SSc (57), prevent stroke and vision loss in GCA (22), improve renal function in TA (29) clear hepatitis B (HBV) infected cells, as HBV often leads to PAN (25) prevent opportunistic pneumonia, may reduce inflammation (19) reduce calcinosis (24), prevent glucocorticoidassociated osteoporosis (19,28)  34  Table 1.4 Estimated Annual Per-Patient Costs of Standard and Emerging SARDs Therapies in Canada Standard Therapies Cost (CDN) Emerging Therapies Cost (CDN) corticosteroids $240 (32) mycophenolate $1020-$2040 (88) hydroxychloroquine  $240 (32)  methotrexate  $240-480 (32)  cyclophosphamide  $240-480 (32)  azathioprine  $600 (12)  calcium channel blockers  $360 (255)  infliximab  $13,000 (225)  rituximab  $9500 (32)  belimumab  $20,000 (32)  etanercept  $22,000 (225)  sildenafil  $1825-$5475 (415)  epoprostenol  $40,515 2003 CDN (256)  bosentan  $43,800 2003 CDN (256)  35  Table 1.5a Study Characteristics for Included Studies using Clinic-Based Data – Mean Annual Direct Medical Costs, Per-Patient Study Year Disease Country Study Number of Data Mean Total Mean Annual Mean Annual Years Subjects Collection / Annual Direct Outpatient Total Hospital Recall Medical Costs Costs Costs Period Aghdassi et al 2011 SLE Canada 2004141: 4 weeks LN=$12,122 LN=$3,578 n/a (297) 2009 79 LN, LNN=$10,186 LNN=$3,390 62 LNN Bernatsky et al(a) 2009 SSc Canada ? 457 12 months $5,038 $1,492 $1,670 (274) Callaghan et al 2007 SjD UK 2001 129 6 months $5,443 $3848 $1221 (311) Clarke et al (276) 1993 SLE Canada 1989, 1989=164 6 months 1989=$8,667 1989=$3,165 1989=$4,183 1990 1990=155 1990=$10,752 1990=$2922 1990=$6649 Clarke et al (315) 1999 SLE Canada, 1995Canada= 6 months Canada=$6210 Canada=$1902 Canada=$3238 USA, UK 1997 229 USA=$6,753 USA=$2,106 USA=$2,271 USA=268 UK=$6,084 UK=$1,677 UK=$2753 UK=211 (adjusted) (unadjusted) (unadjusted) Clarke et al (314) 2004 SLE Canada, 1995Canada= 6 months – Canada=$16,570 Canada=$4266 n/a USA, UK 2001 231 4 years USA=$22,090 USA=$5,497 USA=269 UK=$18,927 UK=$4,265 UK=215 (4-year (4-year cumulative) cumulative) Finn et al (305) 1993 SLE USA 198874 5 years $11,705 $4,155 $7,128 (same) 1992 Gironimi et al 1996 SLE USA 1990174 6 months $17,413 $8,098 $4,754 (313) 1991 Krulichova et al 2004 TA Italy 199867 12 months $6,801 $1,607 $3,700 (318) 2000  Mean Annual Inpatient Hospital Costs  Mean Annual Medication Costs  LN=$951 LNN=$1,476  LN=$4,979 LNN=$2,865  $1,448  $1,575  n/a  $241  1989=$3,385 1990=$6,242 Canada=$2946 USA=$1,176 UK=$1,649 (adjusted) Canada=$4172 USA=$6,296 UK=$6,682 (4-year cumulative) $7,128  1989=$1,452 1990=$1,181 Canada=$1296 USA=$1,562 UK=$1,478 (adjusted) Canada=$5454 USA=$6,599 UK=$6,510 (4-year cumulative) $1,150  $4367  $1713 $1,494  Minier et al (319)  2010  SSc  Hungary  2006  80  12 months  $5,177  $223  $4,322  $4,322 (same)  $395  Panopalis et al (302)  2008  SLE  USA  20032005  812  12 months  $16,368 (sensitivity analysis: $13,411$17,925)  $2,967  $8,723  $7,966  $4,200  36  Study  Year  Disease  Country  Study Years  Number of Subjects  Sutcliffe et al (312)  2001  SLE  UK  19951996  105  Zhu et al(a, b) (322,323)  2009  SLE  Hong Kong  20052007  306  Data Collection / Recall Period 2x6 months  Mean Total Annual Direct Medical Costs  Mean Annual Outpatient Costs  Mean Annual Total Hospital Costs  Mean Annual Inpatient Hospital Costs  Mean Annual Medication Costs  $7,253  $3,059  $2,784  n/a  $1,252  12 months  $9,765  $2,551  n/a  $5,116  $389  37  Table 1.5b Study Characteristics for Included Studies using Clinic-Based Data –Mean Annual Health Resource Utilization, Per-Patient Aghdassi et BernatsClarke Clarke Edwards GironiKrulichoMinier Panopal- Petri and Sutcliffe Study al (297)  ky et al (b) (274, 296)  et al (276)  et al (315)  et al (320)  mi et al (313)  va et al (318)  et al (319)  is et al (302)  Genovese (325)  et al (312)  Teh et al (321)  Yelin et al (309)  Zhu et al(a,b) (322, 323)  Zink et al (317)  Year Disease Country  2011 SLE  2009 SSc  1993 SLE  1999 SLE  2003 SLE  1996 SLE  2004 TA  2010 SSc  2008 SLE  1992 SLE  2001 SLE  2008 SLE  2007 SLE  2009 SLE  2004 SLE  Canada  Canada  Canada  Canada, USA, UK  Singapore  USA  Italy  Hungary  USA  USA  UK  Malaysia  USA  Hong Kong  Germany  Study Years  2004-2009  2000  19951996 105  20012005  20052007 306  2001  80  20032005 812  2006  1698  19982000 67  1989-1990  Data Collection/ Recall Period  4 weeks  12 months  Canada= 229 USA=268 UK=211 6 months  19901991 174  2006  141: 79 LN 62 LNN  1989, 1990 1989= 164 1990= 155 6 months  1995-1997  Number of Subjects  20042005 352  12 months  6 months  12 months  12 months  12 months  24 months  2x6 months  12 months  12 months  12 months  # Outpatient Visits  LN=3.0 LNN=2.5 (4 weeks)  7.3  11.8  n/a  22.2  17.4  30.6  17.4  n/a  7.25  21.9  Annual Hospitalizatio n Rate  LN=6.5% LNN=5.0% (4 weeks)  n/a  1989= 18% 1990= 15% (6 months)  Canada=19 USA=20 UK=19 (unadjusted) Canada= 18% USA=11% UK=12%  13.10%  n/a  98+%  21.10%  1989=37% 1990=39%  1st 6months= 13.66%; 2nd 6months= 8.78%  n/a  27%  ~25%  Annual # Hospitalizations (whole cohort)  n/a  n/a  0.2  0.5  0.3  1989=0.70 5 1990=0.66 9 Overall= 0.69  n/a  n/a  Annual # Hospitalizations (hospitalized cases)  n/a  n/a  1.56  n/a  4.9  1.5  1.58  n/a  Mean LOS per Hospitalizatioo  LN=2.8 LN=5.7 (4 weeks)  n/a  median= 4  n/a  n/a  n/a  1989= 1.13 1990= 1.09 (6 months)  261  9.6  79  median= 6  12 month s  n/a  1248  n/a  38  Study  Aghdassi et al (297)  n/a  Mean LOS per Patient  Proportion Dispensed Medication Mean # Prescriptions (recipient cases)  Bernatsky et al (b) (274, 296)  LN=99% LNN= 100%  Clarke et al (276)  1989=9. 5 1990= 18.0 (6 months)  Clarke et al (315)  Canada=12 USA=8 UK=11 (6 months)  Edwards et al (320)  Gironimi et al (313)  Krulichova et al (318)  Minier et al (319)  n/a  Panopalis et al (302)  Sutcliffe et al (312)  Teh et al (321)  n/a  1st 6months= 8 2nd 6months= 9 95%  n/a  n/a  n/a  n/a  n/a  99%  n/a  n/a  n/a  6.1  Petri and Genovese (325)  Yelin et al (309)  Zhu et al(a,b) (322, 323)  n/a  n/a  21  n/a  4.9  97%  n/a  n/a  n/a  Zink et al (317)  39  Table 1.6a Study Characteristics for Included Studies using Administrative Databases – Mean Annual Direct Medical Costs, Per-Patient Study Year Disease Country Study Number of Data Mean Total Mean Mean Years Subjects Collection Annual Direct Annual Annual Period Medical Costs Outpatient Total Costs Hospital Costs  Mean Annual Inpatient Hospital Costs  Mean Annual Medication Costs  Bernatsky et al (294)  2011  PM/DM  Canada  2003  1,102  12 months  $40,06  ~$1,042  ~$2,964  ~$,2964 (same)  n/a  Carls et al (299)  2009  SLE  USA  20002005  6,269  12 months  $23,847  $10,391  n/a  $9831  $3238  Chiu and Lai (324)  2010  SLE  Taiwan  20002007  22,182  maximum 8 years  $1,399-$1,727  $713-$908  n/a  $682-$835  n/a  Cotch et al (295) Cotch (416)  1995 2000  PAN, Wegener’s GCA, TA  USA  19861990  PAN=480 Wegener’s= 571 GCA=3519 TA=154 (New York State)  5 years  n/a  n/a  n/a  PAN: NY=$4879, USA=$4896; Wegener’s: NY=$5324, USA=$7198; GCA: NY=$2820, USA=$2817; TA: NY=$3631, USA=$3710  n/a  Li et al (304)  2009  SLE  USA  20002005  2,298  5 years  $18,206$28,312  n/a  n/a  n/a  n/a  Pelletier et al (301)  2009  SLE  USA  20072008  15,590: LN=1,068 LNN=14,522  12 months  LN=$36,507 LNN=$14,327 overall= $15,846  LN=$18,183 LNN=$7,387  n/a  LN=$11,067 LNN=$3,140  LN=$7,256 LNN=$3,799  Wilson (300)  1997  SSc  USA  1994  77  12 months  $8,685 (main) $5,699 (2o)  $1,835  $3,871 ($3,819$4,296)  n/a  $1,882  40  Table 1.6b Study Characteristics for Included Studies using Administrative Databases – Mean Annual Health Resource Utilization, Per-Patient Bernatsky et al Chiu and Chung et Cotch et al Krishnan (306) Li et al Molina Nietert et Pelletier et al Study (294) Lai (324) al (307) (295) (304) et al al (308) (301) Cotch (416) (310)  Wilson (300)  Year Disease  2011 PM/DM  2010 SLE  2007 SSc  1995, 2000 PAN, Wegener’s GCA, TA  2006 SLE  2009 SLE  2008 SLE  2001 SSc  2009 SLE  1997 SSc  Country  Canada  Taiwan  USA  USA  USA  USA  USA  USA  USA  Study Years  2003  2000-2007  1986-1990  1998-2002  2007-2008  1994  1102  22,182  PAN=480 Wegener’s=571 GCA=3519 TA=154 (New York State)  n/a  20002005 2298  1995  Number of Subjects  20022003 n/a  Puerto Rico 2003 RH=665 GP=92  n/a  15,590: LN=1068 LNN=14,522  77  Data Collection/ Recall Period  maximum 12 months  maximum 8 years  12 months  5 years  5 years  5 years  12 months  12 months  12 months  12 months  # Outpatient Visits  n/a  12  n/a  n/a  n/a  5.6-6.9  n/a  n/a  10.8  Annual Hospitalization Rate  24.20%  n/a  n/a  n/a  n/a  n/a  n/a  Annual # Hospitalizations (whole cohort)  n/a  0.4  n/a  n/a  n/a  year 1=24% years 2-5=18% year 1=0.5 years 2-4=0.3 year 5= 0.4  LN=20.49 LNN=18.93 LN=30.3% LNN=13.6%  RH=0.2 GP=0  n/a  n/a  Annual # Hospitalizations (hospitalized cases)  n/a  n/a  n/a  PAN=1.39 Wegener’s=1.71 GCA=1.28 TA=1.55 (New York State) (5 years)  n/a  n/a  n/a  LN=2.08 LNN=1.55  41  Study  Bernatsky et al (294)  Chiu and Lai (324)  Chung et al (307)  Mean LOS per Hospitalization  n/a  9.6  6.6  Mean LOS per Patient Proportion Dispensed Medication  n/a  n/a  n/a  Mean # Prescriptions (recipient cases)  n/a  Cotch et al (295) Cotch (416)  Krishnan (306)  Li et al (304)  Molina et al (310)  Nietert et al (308)  Pelletier et al (301)  Wilson (300)  PAN=20 Wegener’s=17 GCA=13 TA=11 (New York State)  6  4.4-6.0  n/a  7.5  LN=6.93 LNN=5.32  n/a  n/a  n/a  n/a  n/a  n/a  1.6  n/a  n/a  n/a  n/a  n/a  n/a  n/a  LN=16.52 LNN=9.69 n/a  n/a  n/a  n/a  n/a  54.562.9  n/a  n/a  LN=60.62 LNN=42.68  n/a  n/a  42  Table 1.7 Studies Reporting a Greater SARDs Burden in Older Individuals Disorder Study Location Year Avina-Zubieta 2011 British Columbia 2007 SARDs-CTD (all) (1) Bernatsky 2011 (5) Canada 2003  SLE  SSc  Avina-Zubieta 2011 (1) Bernatsky 2007 (381)  British Columbia  2007  Quebec  19942003  Hopkinson 1993 (336)  UK  19891990  Naleway 2005 (339)  Wisconsin  2001  Stahl-Hallengren 2000 (391)  Sweden  19811991  Arias-Nunez 2008 (348)  Spain  19882006  Avina-Zubieta 2011 (1)  British Columbia  2007  Report incidence in 45+ years=52.9 vs. <45 years=33.8 greatest prevalence in those aged 45 years and over, age a significant predictor of prevalence; rural Manitoba females: 1250 (1160-1340) for >45 years vs. 200 (170-230); urban Nova Scotia females: 1150 (920-1670) for >45 years vs. 110 (80-170) incidence in 45+ years=16.8 vs. <45 years=10.8 highest incidence and prevalence rates in those aged 45-64 years highest incidence rates in those aged 50-59 years, median age at diagnosis=55.5 for males and 47.0 for females highest incidence rate in those aged 60-79 years: 11.5 vs. 5.1 (3.6-6.6) overall; highest prevalence rate in those aged 80+ years: 252.7 vs. 78.5 (59.0-98.0) overall highest incidence rates in 65-74 years: females=14.1 and males=3.2 vs. 4.5-4.8 overall highest incidence rates in those 45 years : 3.1 (2.0-4.1) for 4564 years and 3.0 (2.1-4.1) for 65+ years vs. 2.3 (1.6-2.5) overall incidence in 45+ years=4.9 vs. <45 years=1.9 43  Disorder SSc  Study Steen 1997 (353)  Location Pennsylvania  Year 19631972  SjD  Alamanos 2006 (355)  Greece  19822003  Avina-Zubieta 2011 (1) Haugen 2008 (417)  British Columbia  2007  Norway  19971999  Pillemer 2001 (358)  Minnesota  19761992  Thomas 1998 (359)  UK  ?  Weng 2011 (360)  Taiwan  20052007  Avina-Zubieta 2011 (1) Oddis 1990 (361)  British Columbia  2007  Pennsylvania  19631982  Avina-Zubieta 2011 (1) Benbassat 1980 (418) Avina-Zubieta 2011 (1)  British Columbia  2007  Israel  19601976 2007  PM only  DM only PM/DM SARDs-VD (all)  British Columbia  Report highest incidence rate in black women aged 45-54 years (2.12 (0.40-3.83) vs. 0.19 (0.01-0.36) in white women aged 15-24 years highest incidence rate for males 46-55 years and females 56-65 years incidence in 45+ years=5.8 vs. <45 years=2.4 prevalence for 71-74 years: 1400 (95%CI: 1020-1920) 3390 (95%CI: 2770-4140); prevalence for 40-44 years: 220 (95%CI: 150-320) – 440 (95%CI: 340-570) peak incidence for females aged 55-64 years, no cases in males <75 years highest prevalence in 55 years+: 4900 (3700-6400) vs. 3100 (2100-4400) highest incidence in females aged 55-64 years (23.4 (21.6-25.1)) and males aged 65-74 years (4.0 (3.1-4.9)) incidence in 45+ years=2.2 vs. <45 years=0.5 highest incidence in those 65 years+: 1.05 (0.11-1.99) vs. 0.55 (0.03-1.07) overall incidence in 45+ years=1.3 vs. <45 years=0.7 highest incidence in those 70-79 years (0.63 vs. 0.22 overall) incidence in 45+ years=8.1 vs. <45 years=2.0  44  Disorder  Study  Location  Year  Report  SARDs-VD (all)  Gonzalez-Gay 2003 (6)  Spain  19882001  Herlyn 2008 (419)  Germany  19982005  Koldingsnes 2001 (413)  Norway  19841998  Mohammad 2009 (364)  Sweden  19972006  Reinhold-Keller 2000 (365)  Germany  1994  Reinhold-Keller 2002 (366)  Germany  1998 1999  Reinhold-Keller 2005 (367)  Germany  19882002  highest incidence in those 55-64 years: 3.49 (2.86-4.26) vs. 1.31 (0.89-1.92) overall greater incidence in those 50+ years: 6.94 (5.3-8.5) – 11.43 (9.3-13.5) vs. 1.24 (0.6-1.9) - 2.48 (1.5-3.5) for <50 years for males, highest incidence in males aged 65-74 years 2.90 (1.25-5.71) vs. 1.22 (0.84-1.70) for all adult males highest incidence in those aged 75 years+ (7.91 (5.52-10.30) vs. 2.18 (1.82-2.54) overall prevalence 5x higher in those 50+ (44.2 (34.0-54.5) – 45.0 (33.63-56.4) vs. 7.8 (4.6-11.1) - 8.7 (5.3-12.1) for <50 years greater incidence in those 50+ years: 8.18 (5.9-10.5) - 10.47 (7.8-13.1) vs. 2.32 (0.7-4.0) - 2.36 (0.7-4.1) incidence 2-5x higher in those aged 50 years: 7.40 (5.70-9.10) - 11.43 (9.30-13.50) vs. 1.24 (0.06-1.90) – 2.48 (1.50-3.50)  Watts 2000 (7), 2001 Avina-Zubieta 2011 (1)  UK  19881997 2007  Wegener’s  British Columbia  highest incidence in those aged 65-74 years : 6.01 (4.08-8.53) incidence in 45+ years=2.9 vs. <45 years=1.4  45  Disorder Wegener’s  Study O’Donnell 2007 (420)  Location New Zealand  Year 19992003  Report highest incidence in those 70-79 years, approximately 25.0 vs. 3.9 overall  PAN  Avina-Zubieta 2011 (1) Gonzalez-Gay 2003 (6) Avina-Zubieta 2011 (1) Bengtsson 1981 (421)  British Columbia  2007  Spain  19882001 2007  incidence in 45+ years=0.8 vs. <45 years=0.4 highest incidence in those 55-64 years: 0.93 (0.58-1.49) incidence in 45+ years=4.8 vs. <45 years=0.2 incidence for 50+ years: 28.6 vs. 9.3 overall  Boesen 1987 (370)  Denmark  1982  Friedman 1982 (284)  Israel  19601978  Friedman 1982 (284)  Israel  19601978  Gonzalez-Gay 2001 (371)  Spain  19811998  Gonzalez-Gay 2007 (372)  Spain  19812005  Gran 1997 (374)  Norway  19871994  Petursdottir 1999 (375)  Sweden  19761995  GCA  British Columbia Sweden  incidence for 70-79 years: 142.9 incidence for 50+ years: 76.6 overall incidence: 21.5 highest incidence of temporal arteritis in those 70-79 years: 1.16 vs. 0.49 for whole 50+ population greatest incidence in those 70+ years: 1.16 (0.69-1.63) vs. 0.49 (0.35-0.63) for all 50+ years and 0.02 (0.00-0.11) for those 50-59 years highest average annual incidence in those aged 70-79 years: 20.78 (13.33-28.23) vs. 10.24 (8.13-12.58) overall incidence rate for 70-79 years: 23.16 (19.52-27.28) vs. 10.13 (8.93-11.46) overall highest incidence in those 75-79 years: 308.6 vs. 141.7 for all 50+ years and 41.1 overall incidence for 50+ years: 22.2 vs. 7.7 overall 46  Disorder GCA  TA  Study Reinhold-Keller 2000 (365)  Location Germany  Year 1994  Report higher prevalence in those aged 50 years and older: 24.0 (16.431.5) - 30.0 (20.7-39.3) vs. 8.7 (5.9-11.4) - 9.4 (6.5-12.3) overall  Salvarani 1995 (376)  Minnesota  19501991  Salvarani 2004 (377)  Minnesota  19501999  Sonnenblick 1994 (378)  Jerusalem  19801991  Avina-Zubieta 2011 (1)  British Columbia  2007  highest incidence in those 70-79 years 48.9 (33.9-68.3) vs. 17.8 (14.7-21.0) for all 50+ years highest incidence in those aged 80+ years: 51.9 (37.6-65.3) vs. 18.8 (15.9-21.6) for all those 50+ years highest incidence in those aged 75+ years 28.4 (21.0-35.7) vs. 2.2 (0.8-5.2) in those aged 5064 incidence in 45+ years=0.5 vs. <45 years=0.2  47  Table 1.8 Studies Reporting an Increase in the Incidence of SARDs Disorder Study Location Initial Years Incidence (95% CI) per-100,000 Avina-Zubieta British Columbia 1996 16.2 SARDs-CTD 2011 (1) (all) Alamanos 2003 Greece 19821.41 SLE (330) 1986 (0.99-1.83) Avina-Zubieta British Columbia 1996 4.7 2011 (1) Laustrup 2009 Denmark 1995 0.52 (388) (0.05-1.90) Uramoto 1999 Minnesota 19501.51 (288) 1979 (0.85-2.17) Voss 1998 (344) Denmark 1980 1.1 (0.3-2.9) Arias-Nunez 2008 Spain 19881.4 (0.6-2.3) SSc (348) 1992 Avina-Zubieta British Columbia 1996 1.8 2011 (1) Steen 1997 (353) Pennsylvania 19630.97 1967 (0.75-1.18) Avina-Zubieta British Columbia 1996 0.6 SjD 2011 (1) Avina-Zubieta British Columbia 1996 0.3 PM only 2011 (1) Avina-Zubieta British Columbia 1996 0.6 DM only 2011 (1) Benbassat 1980 Israel 19600.13 PM/DM (418) 1964 Oddis 1990 (361) Pennsylvania 1963all: 0.25 1972 females: 0.32 males: 0.17 Avina-Zubieta British Columbia 1996 2.4 SARDs-VD (all) 2011 (1) Watts 2001 (368) UK 19881.55 1993 (1.06-2.19)  Subsequent Years Incidence (95% CI) per 100,000 2007 44.3 19972001 2007  2.19 (1.78-2.60) 14.1  2002  1.04 (0.27-3.69) 5.56 (3.93-7.19) 3.6 (2.0-6.1) 2.5 (1.4-4.1)  19801992 1994 20032006 2007  3.6  19781982 2007  1.82 (1.50-2.13) 4.3  2007  1.4  2007  1.0  19701974 19731982  0.26  2007  all: 0.89 females: 1.16 males: 0.58 5.3  19941998  2.10 (1.56-2.74)  48  Disorder  Study  Location Years  Wegener’s  GCA  PAN TA  Initial Incidence (95% CI) per-100,000  Subsequent Years Incidence (95% CI) per 100,000  Avina-Zubieta 2011 (1) Koldingsnes 2001 (413) Avina-Zubieta 2011 (1) Franzen 1992 (422)  British Columbia  1996  0.6  2007  2.2  Norway  19841988 1996  0.52 (0.27-0.90) 1.4  19941998 2007  1.20 (0.80-1.73) 2.7  Friedman 1982 (284) Gonzalez-Gay 1997 (373)  Israel  19841987 (44 months) 19601964 19861990  overall= 22.5 50+ years=69.8 0.16 (0.00-0.32) annual average: 8.26  19871988 (16 months) 19751978 19911995  overall= 30.4 50+ years= 94.4 0.86 (0.51-1.22) annual average: 10.49  Gonzalez-Gay 2007 (372) Petursdottir 1999 (375) Salvarani 2004 (377) Avina-Zubieta 2011 (1) Avina-Zubieta 2011 (1)  British Columbia Finland  Spain  19811985 1976  5-year average: 41.30 3.18 (1.82-5.16) 9.6  20012005 1995  5-year average: 51.99 12.92 (9.97-16.46) 30.1  British Columbia  19501954 1996  6.7 (0.0-14.3) 0.2  19951999 2007  18.5 (11.3-25.7) 0.6  British Columbia  1996  0.03  2007  0.4  Spain Sweden Minnesota  49  Table 1.9a Study Characteristics for Included Studies Reporting Incremental/Attributable Mean Annual Direct Medical Cost Data, Per-Patient Callaghan et Carls et al Huscher et Li et al Panopalis et al Pelletier et al Study al (311) (299) al (316) (304) (302) (301) Year Disease Country Study Years Number of Controls Data Collection/Recall Period Mean Total Annual Direct Medical Costs (Controls) Total Mean Annual Attributable Direct Medical Costs Mean Annual Outpatient Costs (Controls) Mean Annual Attributable Outpatient Costs Mean Annual Hospital Costs (Controls) Mean Annual Attributable Hospital Costs  2007 SjD United Kingdom 2001  2009 SLE USA  2006 SLE Germany  2009 SLE USA  2008 SLE USA  2009 SLE USA  2000-2005  2002  2003-2005  2007-2008  92  6,269  844  20002005 2,298  n/a  6 months  12 months  3 months  12 months  n/a: US national average 12 months  12 months  $2,361  $8,882  n/a  $10,920$19,181  $5,802  n/a  $3,085 (57%)  $14,964 (63%)  $5,878 (n/a)  $4,475$9,569 (15-42%)  $10,566 (66%)  LN=$8,326 (23%) LNN=$2,964 (31%)  $1,903  $5,839  n/a  n/a  n/a  n/a  $1,943 (50%)  $5,659 (54%)  $680 (n/a)  n/a  n/a  $358  n/a  n/a  n/a  n/a  LN=$2,332 (13%) LNN=$1,000 (14%) n/a  $863 (71%)  n/a  n/a  n/a  n/a  n/a  50  Study  Callaghan et al (311)  Carls et al (299)  Huscher et al (316)  Li et al (304)  Panopalis et al (302)  Pelletier et al (301)  Mean Annual Attributable Inpatient Hospital Costs Mean Annual Medication Costs (Controls) Mean Annual Attributable Medication Costs  n/a  $7,576 (77%)  $2,853 (n/a)  n/a  n/a  $67  $1,777  n/a  n/a  n/a  LN=$3,829 (35%) LNN=$1,096 (35%) n/a  $174 (72%)  $1,500 (46%)  $1,566 (n/a)  n/a  n/a  LN=$1,818 (25%) LNN=$868 (23%)  51  Table 1.9b Study Characteristics for Included Studies with Incremental/Attributable Health Resource Utilization Data, Per-Patient Clarke et al Li et al (304) Pelletier et al (301) Petri and Genovese Study (276) (325) 1993 2009 2009 1992 Year SLE SLE SLE SLE Disease Canada USA USA USA Country 1989, 1990 2000-2005 2007-2008 1989-1990 Study Years n/a: Quebec 2,298 n/a n/a Number of Controls average 6 months 12 months 12 months 24 months Data Collection/Recall Period 6.9 3.4-3.8 n/a n/a Mean # Outpatient Visits (Controls) 4.9 2.2-3.2 n/a n/a Mean Attributable Outpatient Visits n/a 10.4-12.0% LN=16.1% n/a Annual Hospitalization Rate LNN=5.9% (Controls) n/a 6.3-14.4% LN=14.2% 70% of Annual Attributable LNN=7.7% hospitalizations Hospitalization Rate n/a 0.2 n/a n/a Mean Annual Admissions (whole cohort) n/a 0.1-0.3 n/a n/a Annual Attributable Admissions n/a 4.2-5.8 n/a n/a Mean LOS per Hospitalization (Controls) n/a -2.4 n/a n/a Mean Attributable LOS 1.16 n/a n/a n/a Mean LOS per Patient (controls) 3.69 n/a n/a n/a Mean Attributable LOS n/a 37.4-47.2 n/a n/a Mean # Prescriptions (Controls) n/a 15.0-17.1 n/a n/a Mean # Attributable Prescriptions  52  Table 1.10 Mean Annual Direct Medical Costs of Other Arthritides in Canada, Per-Patient Disease Canadian Location Study Year Mean Annual PerPrevalence Patient Costs (per (2007 CDN) 100,000 adults) Rheumatoid 1,000 (12) Saskatchewan & Clarke et al 1983-1989 $4,997 Arthritis (RA) Quebec (277) 1990-1994 $6,142 Ontario Maetzel et al 2000 $7,414 (279) Quebec Fautrel et al 2002 $11,436 (278) Juvenile 1,000 perBritish Columbia Bernatsky et al mid-2000’s gross=$3,143 Idiopathic 100,000 & Quebec (424) incremental=$1,765 Arthritis (JIA) children and teenagers (423) Osteoarthritis 10,000 Ontario Maetzel et al 2000 $5,690 (OA) (425) (279) Fibromyalgia 2,000-4,000 Canada Penrod et al 2001 $5,269 (426) (427) Ontario White et al (428) 1994 $1,386 (outpatient only)  53  2 Methods In the previous chapter we presented our study objectives – to determine the health resource utilization and economic burden of SARDs – and explained how we will employ provincial administrative health data while meeting them. The many advantages of this data source were presented, including large and less-biased patient samples, comprehensive capture of health resource utilization, and more precise costing of this utilization. In this chapter we will detail our case and data sources and the methods we will use for identifying cases, costing services, and producing these estimates of mean direct per-capita medical costs.  2.1 Case and Data Source-British Columbia Linked Health Database Data on the health resource utilization and demographics of the cohort were obtained from the British Columbia Linked Health Database (BCLHD). These are administrative datasets containing the claims for health care services consumed by BC residents. BCLHD was administered by the provincial government (specifically the Ministry of Health) but this data is now made available through a separate organization, Population Data BC (PopDataBC). These claims are linked to the Consolidation File, or vital statistics datasets, through each resident’s unique Personal Health Number (PHN). This allows researchers access to health resource utilization data at the patient level. Although the PHN is used to link claims, a separate study identification number (study id) generated by PopDataBC is used to differentiate patients in the research data, rendering it anonymous. Because the data are de-identified, informed consent was not required from the selected cases and ethics approval was not needed.  In BC, health insurance is available through a single provider, the Medical Services Plan (MSP). The Consolidation File mainly contains data collected by MSP to register beneficiaries and collect their annual premiums. The datafields of interest include the birth year and month, and the dates when MSP coverage started and ended (429). Also available are data on all deaths registered in BC (429). It is through these registered deaths and MSP coverage end-dates that we determined the end-of-follow-up date for each case. If there was no record of a case’s death or termination of coverage, they were followed through the end of the study period (December 31, 2007).  54  We used the following health claims datasets: Medical Services Plan Payment Information File (MSP) for outpatient encounters, Discharge Abstract Database (DAD) for hospital separations, and PharmaNet for prescriptions. These are detailed in sections 2.1.1, 2.1.2, and 2.1.3, respectively, of this chapter.  Coverage for the DAD and MSP data extends from April 1, 1985 onwards (429). PharmaNet coverage is from January 1, 1995 onwards (430), but since this dataset is not considered to be well-populated for the first twelve months of this time, our study period began on January 1, 1996. It extended for twelve years, through December 31, 2007, the last full calendar year for which data was available at the time of request.  All BC residents are both eligible for, and required to, enroll for MSP. This entitles them to the use of all medicallynecessary outpatient and hospital services (429) without upfront charges. Neither the inability, nor the failure, to pay premiums bars access to health services or the capture of claims in the database (431). Residents cannot opt-out of PharmaNet when dispensed a prescription (430). This means the health care consumption of nearly every BC resident is captured by BCLHD.  Given the publically-operated and funded nature of BC’s health care system, where a full range of services are provided to all residents regardless of financial means, the BCLHD’s coverage is considered comprehensive and unbiased. Only a small and select group of BC residents (mainly active members of the RCMP and Canadian Armed Forces - but not their dependants - and Status Native and Inuit individuals, all of whom receive federal health benefits) are not covered by MSP (429). There are a small number of private medical and surgical facilities in the province whose services are not captured by BCLHD. However, the quantity and cost of these services are relatively insignificant in relation to provincially-funded consumption, particularly as complex procedures requiring an extended inpatient stay are not provided by these facilities.  2.1.1 Medical Services Plan Dataset MSP is strictly a BC entity – claims for services provided to BC residents are submitted to and adjudicated by MSP directly within the province. The outpatient services captured in the MSP dataset essentially refer to any mediallynecessary services provided, ordered, or interpreted by a physician or some select non-physician fee-for-service  55  (FFS) practitioners. This includes any eligible physician visits/consultations (including those occurring in a hospital), procedures or operations performed, and laboratory, radiological, and other investigations are ordered or interpreted by them (431). Also included are surgical podiatry services and those provided by a dentist or oral surgeon in a hospital setting (431). Neither routine physical or eye examinations, dental services provided outside hospital, cosmetic procedures, counselling/psychological services (431), nor any services provided through other, non-FFS arrangements are captured in the MSP data. The latter would include physicians reimbursed through a salary (429). Concerning other supplementary-benefit services (those not provided by a physician), the scope of eligible benefits was reduced significantly as-of April 1, 2002. This resulted in the MSP database offering only partial coverage of these encounters over our study period.  Each MSP claim includes datafields for the service date, practitioner number, specialty code (referring to the specialty of the physician or profession of the non-physician practitioner), fee item (referring to the exact service provided), one diagnostic code (using International Classification of Diseases, 9 th revision (ICD-9) nomenclature, up to six digits), and the amount paid to the practitioner by MSP. This amount corresponds to the fee item billed and follows the Medical Services Commission payment schedule. The sex and birthdate of the beneficiary are also listed (429). The specialty codes do change over time; for instance, many specialties previously coded under Internal Medicine are now coded more specifically as Respirology or Nephrology (432). This has implications for SARDs in that rheumatology, the primary specialty involved in SARDs care, was not coded separately until January 1, 1998.  2.1.2 Discharge Abstract Database The hospital separations in the DAD include any discharges, deaths, or transfers – as part of an inpatient or day admission - from any acute-care hospital. These include any separations from an extended or long-term care facility attached to an acute-care hospital (referred to as Alternate Level of Care, or ALC, stays) (429). While the MSP database includes the services provided by physicians during a hospitalization (such as consultations or surgeries), the resource consumption captured by the DAD covers every other medical service and cost involved in providing this care such as nursing costs, consumables, housekeeping, equipment and other overhead. However, the DAD does not capture outpatient emergency room services (429).  56  Unlike MSP, the DAD is a national entity maintained by the Canadian Institute for Health Information (CIHI). CIHI collects discharge records directly from all BC hospitals, processes and performs further data analysis, and submits this back to the BC Ministry of Health on a monthly basis (429). This centralized structure of data collection and processing is thought to increase the quality of the information contained in the DAD (429). In addition to the sex and birthdate of the beneficiary, each DAD record has fields on the hospital, level of care (inpatient, day, or ALC), admission and separation dates, diagnostic code(s) and type(s) (up to 6-digits each and up to 25 total diagnoses), length of stay (LOS), case-mix group (CMG) and resource intensity weight (RIW). The CMG is a method of grouping together similar diagnoses that should necessitate a similar level of resource consumption. The RIW is a weighted measure of the actual resource consumption was necessitated by that hospitalization (429). More detail about the RIW is provided in section 2.6.3.  2.1.3 PharmaNet PharmaNet contains the records for all prescriptions dispensed by any community pharmacy in BC. The Ministry of Health subsidizes some prescription costs through the PharmaCare program, but all prescriptions are recorded in PharmaNet regardless of the payment arrangement. PharmaNet does not capture medications provided in hospital. Each PharmaNet record has fields for the dispensing date, drug identification number (DIN), drug quantity, number of days supplied, drug cost, and professional fee charged.  2.2 Inclusion Criteria Any BC resident who, at any point during the study period (January 1, 1996 through December 31, 2007) was registered with MSP, and was at least 18 years of age, was eligible for inclusion. There was no minimum follow-up period. The MSP dataset and the DAD were searched from January 1, 1990 onwards for potential cases. These were individuals with a claim containing a diagnostic code for any SARD. The specific International Classification of Diseases (ICD) codes, for both the ninth (ICD-9) and tenth (ICD-10) revisions, that this included are listed in Appendix A.  57  To be eligible for inclusion cases had to meet one of the three definitions below: a) ≥2 codes by any non-rheumatologist physician on an outpatient encounter, >60 days apart but within a single twoyear period; b) ≥1 code by a rheumatologist on an MSP encounter; c) ≥1 hospital discharge code as either the primary or one of the 24 secondary discharge diagnoses;  Definition a) required two visits to increase diagnostic specificity. The intention was to exclude individuals who did not actually have a SARD, but had a single SARD-coded encounter to rule-out this diagnosis, or where a nonSARDs-specialist (non-rheumatologist) was uncertain of the SARDs diagnosis. The 61-day (two-month) minimum timeframe was implemented to exclude those with transient, inflammatory, undifferentiated connective tissue diseases that did not meet the criteria for a SARD.  2.2.1 Rationale Similar case definitions have been used to identify cases from US administrative databases for several SLE costing studies (299,301,304). High sensitivity (0.85(95%CI: 0.73-0.97)) and specificity (0.90(95%CI: 0.81-0.99)) values for ICD-9-CM 710.0 (SLE) have been reported for rheumatology billings in a US Medicare database (433). But of particular importance for our work, these same ICD-9 codes and nearly-identical case definitions have been applied to Canadian provincial administrative data when measuring the incidence and prevalence of SARDs-CTD (5,349,362,381). The one Canadian population-based study of SARDs costs also used a similar definition (294). The accuracy of these (and SARDs-VD) diagnoses in Canadian data was assessed in a subsequent validation study (434) which compared them to cases’ medical records. The diagnosis was confirmed for 83% of cases. Among cases with a false-positive diagnosis for one SARD, most were deemed to still have a SARD (or similar disorder), only a different one, upon chart review (5,434).  Using multiple datasets and definitions helps maximize case ascertainment and minimize bias. These Canadian studies compared the different sources and definitions and as shown in Table 2.1, they differed in the number of cases detected and their sensitivity and specificity. For instance, the prevalence of PM/DM, when calculated using all three sources (rheumatologist billings, all physician billings, and hospital separations) was nearly twice as large  58  as when only rheumatologist billings were examined (362). With only 24% of SLE cases identified in both physician and hospital data (381), there is clearly limited overlap in the coverage provided by each source. This means fewer cases will be captured if just one is used, limiting sample size and statistical power. Using multiple sources can increase diagnostic specificity, with the validation study finding this was higher when multiple sets were used (434).  Bias is also minimized since the cases captured by each source can differ in their demographic characteristics, including age (362,381), sex (5), and urban/rural residence (5,349,362,381). As detailed in Chapter 1, these factors can influence healthcare consumption. With hospital-sourced cases more likely to have a greater severity of disease and/or a greater comorbidity burden than others (349,362), using only hospital data could overestimate costs. Instead, by combining the three sources and definitions together, we should obtain the most-representative and leastbiased cohort possible.  2.3 Exclusion Criteria 2.3.1 Diagnostic Specificity After potential SARDs cases were identified, we implemented additional exclusion criteria to increase the reliability of the SARDs diagnosis and specificity of our cost estimates. First, any potential case was excluded if more than five years passed between the dates of the last encounter (MSP claim or hospital separation) coded for the SARD and end-of-follow-up. Given the high rate of SARDs-related consumption expected for true SARDs cases, such a gap in utilization reduced the certainty of the SARDs diagnosis. Potential cases only meeting definition a) were excluded if they had a rheumatologist encounter subsequent to their SARDs index date (the date of the first qualifying visit) which was not coded for a SARD. This was also to increase the reliability of the SARD diagnosis.  2.3.2 Costing Specificity Even though an individual can legitimately have both a SARD and another form of arthritis, such as RA, due to overlap in the types of health resource utilization necessitated by these two classes of disorders, we excluded these cases. It would otherwise be difficult to attribute any healthcare consumption to SARDs. This meant potential cases that were later diagnosed with another type of inflammatory arthritis (including RA, psoriatic 59  arthritis, spondyloarthropathy, systemic vasculitis and other non-SARD connective tissue diseases) at any time during the follow-up period were excluded. This was determined through the presence of any diagnostic code corresponding to these disorders (ICD-9 274, 696, 711, 712, 713, 714, or 720; ICD-10 M02, M05, M07, M10, M11, M14, M45, or M148) on a health encounter claim.  Individuals who had originally met definition a) for SARDs were excluded if, after their second qualifying SARD visit, they had ≥2 encounters (MSP claims or hospital separations) ≥61 days apart coded for a different form of arthritis. The two encounters, however, did not need to occur within a two-year period. Individuals who originally met either case definition b) or c) were excluded if they had either ≥1 rheumatologist encounter or ≥1 hospital separation coded for one of these arthritides.  Colleagues have used this same cohort to measure the incidence and prevalence of SARDs in BC, and their estimates were consistent with those reported in other jurisdictions (1). This supports the reliability of these case definitions, diagnostic codes, and exclusion criteria for appropriately identifying SARDs cases.  2.4 Index Dates and Subgroups 2.4.1 Index Dates For cases meeting definition a), the index date was the date of the first visit in the set of two qualifying encounters (the index visit). For cases meeting definitions b) or c), the index date was equal to the date of the single qualifying encounter (again referred to as index visit). When a case met more than one of the above criteria, the actual index date was set to the earliest of any of the possible index dates.  2.4.2 Subgroups Aside from being included in the entire SARDs cohort, cases coded with a SARDs-CTD diagnosis at any index visit were included in a SARDs-CTD subgroup. Similarly cases coded for a SARDs-VD diagnosis at any index visit were included in a SARDs-VD subgroup. Any case with both a SARDs-CTD and SARDs-VD index diagnosis was  60  included in the main SARDs cohort but not assigned to a subgroup. All subsequent analyses were performed for each group separately: the entire cohort (all-SARDs), and the SARDs-CTD and SARDs-VD subgroups.  2.5 Health Resource Utilization 2.5.1 Follow-Up Parameters All health resources consumed by cases during their follow-up period were tabulated. For some components, any consumption occurring within 30 days after the end-of-follow-up date was also included. This thirty-day window was implemented to account for errors and inconsistencies in billing dates. We assumed an encounter billed shortly after the end-of-follow-up date likely occurred during the follow-up period, particularly if follow-up ceased due to death. When the index date fell on or after January 1, 1996, follow-up began from that same date (referred to from here-on as the start date). When the index date fell before this time, follow-up began on January 1, 1996 (the start date). For cases that were less than 18 years old in the year of their index date, follow-up did not begin until January 1 of the year in which they turned 18. The last day of follow-up was the earliest date of either death, de-enrollment in MSP (usually from moving out of BC), or December 31, 2007, the last day of the study period. Any resource consumption occurring on this end-of-follow-up date (referred to from here-on as the end date) was included. A summary of the timeframes during which encounters were included is provided in Table 2.2.  We determined the number of unique cases followed each year and during the entire study period. For each year, a case was counted if their start date fell on or before January 1 of that year and their end date fell after December 31 of the previous year. Therefore, a case would be included in the year-2000 cohort if their death date was recorded as January 1, 2000, because they were alive for a portion of the year 2000. Since there was no minimum length of follow-up, many cases were followed for only part of a calendar year. To account for this, the contribution of each case was weighted by the amount of follow-up time they had, rounded up to the nearest month and expressed in patient-years or person-years (PY). For example, a case with a start date of January 1, 1996, and an end date of March 5, 1997, would have contributed a total of 1.25 PY (twelve months in 1996 and three months in 1997).  61  2.5.2 Outpatient Encounters For each component, the number of services consumed was summed both annually and for the entire study period. For outpatient encounters, each unique combination of study id, service date, and specialty-type was considered a single encounter. Therefore, only the first of any set of claims made by the same practitioner-type on the same date for the same case was counted. This still allowed an individual to have multiple encounters in one day if they were with different specialists. All claims occurring on or after the start date were considered, as were any occurring within 30 days after the end date (but before 2008). For annual tabulations, encounters were allocated to the year of the service date.  2.5.3 Prescriptions The prescription tabulations included all those dispensed on or after the start date, and those dispensed within 30 days following the end date (but before 2008). For annual tabulations, prescriptions were allocated to the year of the service/dispensing date.  2.5.4 Hospitalizations When tabulating the number of admissions, only the first admission per-case, per-day, was considered. This eliminated multiple admissions in a single day that typically resulted from inter-hospital transfers, or multiple day procedures. For annual tabulations, hospitalizations were allocated to the year of the admission date.  Considering that inpatient hospitalizations span multiple days, and therefore could span parts of two calendar years and even fall outside the follow-up period, the eligible inclusion dates were modified slightly. Hospitalizations spanning outside a case’s follow-up period were included under the following circumstances:  a) admission within 30 days before the start date and separation during follow-up b) separation within 30 days after the end date but admission during follow-up c) admission outside of follow-up, but within 30 days of the start date, AND separation outside of follow-up, but within 30 days of the end date  62  However, those hospitalizations where the admission occurred more than 30 days before the start date AND the separation occurred more than 30 days past the end date were excluded. For patients followed through December 31, 2007, any separations occurring within the first 30 days of 2008 were included (and allocated to the year 2007) as long as admission occurred before 2008. Similarly, for patients whose start date was January 1, 1996, any admissions occurring in the last 30 days of 1995 were included (and allocated to 1996) as long as separation occurred after 1995.  2.6 Cost Calculation As detailed below, we determined the unit price of each health service consumed. The total direct medical costs were equal to the sum of these prices. Totals were produced for each component (outpatient, hospital, and prescription), and for all health resources combined. Table 2.2 summarizes the methods used to cost each type of encounter.  2.6.1 Outpatient Encounters Included with each claim was the specific amount paid to the practitioner for that service (the Amount Paid field). Therefore, the cost of each encounter was equal to this figure. All submitted claims were included, even multiple claims pertaining to the same case, specialist, and date. Annual costs were allocated to the year of the service date.  2.6.2 Prescriptions Costs for the other components were calculated from the provincial government (payer) perspective, but prescription costs were calculated from a societal perspective. This meant the total cost of each prescription was considered, not just the amount subsidized by the provincial government under the PharmaCare program. This total cost was equal to the sum of the total drug cost and the total professional fee, as listed on each PharmaNet record. A secondary analysis of just the subsidized cost of each prescription was also undertaken. The subsidized cost was equal to the Pharmacare Paid field (the sum of the drug cost and professional fees that were subsidized) on the PharmaNet record. Annual costs were allocated to the year of the service/dispensing date.  63  2.6.3 Hospitalizations The costs of most hospitalizations, including all acute-care stays and about half of the day surgeries, were calculated using CIHI’s established case-mix methodology. In this, the cost of each hospitalization is equal to its RIW multiplied by the cost-per-weighted-case (CPWC) specific to the fiscal year. Both of these figures are calculated by CIHI. This standard method of pricing hospitalizations has been used in many Canadian costing studies for individual SARDs (274,297,313-315) and other disorders. The cost of each hospitalization was allocated to the year of admission whereas inflation adjustments were made according to the year of separation. As with the MSP claims, all admissions were included for costing, even multiple admissions from the same date.  Cost-Per-Weighted-Case The CPWC is the average cost of an inpatient hospitalization for a ‘standard’ patient across all hospitals in a particular jurisdiction, and is specific to a fiscal year (435). We used CPWC values specific to BC, and those corresponding to the fiscal year of the separation date.  Provincial CPWC figures for the fiscal years 2004/2005 to 2007/2008 were calculated and provided by CIHI (436) (Table 2.3), but those for the fiscal years prior to 2004/05 were unavailable. Instead, we had to estimate them, and explored several methods of doing so:  a) deflating the earliest-available CPWC figure, that for 2004/05, by the percent-change in the Canadian Consumer Price Index (CPI) between 2004 and each previous year; b) constructing a line-of-best-fit with the known CPWC values and using it to extrapolate the values for the previous years; c) using the CPWC for fiscal year 2007/08 for each study year and not making further inflation adjustments;  After some test calculations, the first method was rejected because the CPWC estimates seemed too high. The third method was rejected because it is better to use CPWC figures specific to each year. The methods used in calculating it are “not consistent across years” (437), and year-to-year changes in health services delivery can all drive-up the annual CPWC independent of any inflation increases (438). These changes could include an older or more complex  64  inpatient caseload, a tendency to treat straightforward and less-expensive cases on an outpatient basis, and the introduction of more expensive technologies and procedures (438). It has also been suggested that using only the most recent CPWC value to calculate previous years’ costs can overestimate them (439). Therefore, we decided to extrapolate the earlier values from a line-of-best-fit (method b), which are listed in Table 2.3.  Resource Intensity Weight The RIW is a measure of the relative resource consumption (or cost) of each hospitalization, taking into account the specifics of the patient and setting, including LOS (435). It is derived from the age of the patient, complexity of the case, and its assigned CMG – a method of grouping cases with similar diagnoses, resource utilization patterns, and clinical characteristics (438). The RIW is expressed as a decimal - either less than, equal to, or greater than 1.00 (depending on whether the hospitalization required the consumption of greater or fewer resources than the average inpatient case). This makes the RIW is a multiplier of the cost of the average hospitalization (the CPWC).  To ensure consistency, the same edition of RIW should be used to price all admissions. However, this was not possible since our study period spanned three editions of the RIW. The first edition, cRIW01, was available for separations through March 31, 2001. It was derived from the ICD-9 system and CMG/Plx case-mix methodology. The second edition, cRIW05, was available for separations spanning April 1, 2001 through March 31, 2007. It was also derived from ICD-9 and CMG/Plx methodology. The third edition, p_RIW07, was available for separations spanning April 1, 2001 through December 31, 2007. This was instead derived from the newer ICD-10-CA system and CMG+ case-mix methodology. Therefore, the only RIW available for the first 63 months (5.25 years) of the study period was derived from one system (ICD-9), while the only RIW available for the last 9 months (0.75 years) was derived from another system (ICD-10), with 72 months (6 years) of overlapping systems in the middle. RIWs calculated using the old methodology are not fully comparable to those from the latest edition, but if we only used separations with a p_RIW07 value, we would have lost 5.25 valuable years of observation. Instead, we stayed as consistent as possible while maximizing the quantity of data available, using the cRIW01 and cRIW05 for most of the study period, and the p_RIW07 for only the 2007 calendar year.  65  Alternate Level of Care Separations Case-mix methodology could not be used to price ALC separations since the RIW value was set to 0. Instead, these stays were priced by multiplying the LOS by a provincial per-diem rate specific to extended care beds in acute care hospitals. Such rates were not available for every year of study, so the 2007 rate of $225 CDN - quoted by several BC centres (440,441) - was used, and no further inflation adjustment was needed. Two Canadian studies of the costs of SLE have also a used per-diem rate (one provided by Statistics Canada) to price non-acute hospitalizations (314,315).  Day Procedure Separations Case-mix methodology could not be used to cost some day procedures either. The RIW calculated by CIHI for day procedures is referred to as the Day Procedure Group Weight (DPG)-RIW, or DPG-RIW. PopDataBC only provided these values through the 2000-01 fiscal year so we investigated several methods of estimating the costs after this time. We decided that using a flat, per-diem rate (as was done for ALC separations) would not reflect the actual costs of care since this would apply an average charge to a wide range of procedures. Also, a Saskatchewan study on diabetes costs found, when compared to other imputation methods, this method produced the highest hospital cost estimates (442). Instead the day procedure costs for the year 2001 and beyond were extrapolated (using the annual change in the CPI) from the actual total costs incurred for the year 2000, the latest full calendar year for which the DPG-RIW data was available. A method presented by Pohar et al (439) - a version of which was used in the Saskatchewan diabetes study above - was considered. In this, the mean DPG-RIW value for the year 2000 would be applied to each day case occurring in the subsequent years and multiplied by the corresponding CPWC. However, these authors found that cost estimates obtained this way were higher than those obtained using the cost-inflation method described above. Therefore, we chose the latter method to ensure the most conservative estimates.  2.7 Attribution Three separate utilization and cost analyses were undertaken, with the primary (gross) analysis encompassing every health resource that a case consumed. However, to estimate the net burden of SARDs we undertook a secondary analysis that only included SARDs-related encounters, which were selected by two rheumatologists (J.A. Avina-  66  Zubieta, D. Lacaille). We used two definitions for these encounters, one narrow and one broad. Table 2.4 summarizes the encounters included in each analysis.  2.7.1 Narrow Definition. For outpatient encounters, the narrow definition encompassed two mutually-exclusive types of MSP claims based on the diagnostic code and physician specialty. The first was any claim billed by a non-rheumatologist practitioner containing any SARD diagnostic code (as listed in Appendix A). The second was any claim billed by a rheumatologist (as indicated by Specialty Code ‘44’), which could contain any diagnostic code. Any hospital separation with a primary discharge diagnosis for SARDs (as listed in Appendix A) was included. A list of the classes of drugs relating to SARDs was compiled (Appendix A), and all PharmaNet records with a DIN pertaining to these drugs were included.  2.7.2 Broad Definition We suspected many SARDs-related encounters would not be captured under the narrow definition. Because only one diagnosis is recorded on each MSP claim, the morbidity necessitating the encounter (such as nephritis) may be coded instead of the underlying SARD, particularly for laboratory investigations. Therefore, our broad definition captured an additional set of encounters on the basis of physician specialty, specific SARD diagnosis, drug class, and/or fee item. These additional encounters are listed in Appendix A. The claims captured under the narrow definition were also included under the broad. For hospitalizations, the broad definition included any separation with a SARD code listed in any of the twenty-five diagnostic positions. The attributable drugs were the same as under the narrow.  2.8 Statistical Analysis The number of resources consumed, and cost of each one, was summed annually and cumulatively for the twelveyear period. Crude mean annual per-patient-year (PY) utilization and cost estimates were obtained by dividing these totals by the number of patient-years (PY) contributed each year. For some components, annual mean per-PY estimates were also calculated amongst only users (for example, the mean number of prescriptions dispensed to each  67  patient calculated only amongst those patients actually dispensed a prescription in that year). The annual mean LOS per admission and per-patient was calculated for cases that had an inpatient hospitalization.  Analyses were performed using SAS Software, Version 9.2 of the SAS System for Unix (443). All costs are reported in 2007 Canadian dollars, and were adjusted for inflation using the BC Health Care component of the Canadian Consumer Price Index (CPI) (293).  In this chapter we have described our population-based data source in detail and how we identified the SARDs cases. We detailed the types of health care encounters we included in our cost estimates, the methods employed in calculating these costs, and our rationale for selecting them. These estimates and other important findings from our analysis will be presented in the next chapter.  68  2.9 Tables Table 2.1 Comparison of Incidence, Prevalence, Sensitivity, and Specificity Calculated from Different Administrative Sources Incidence (per-100,000) (95% CI) Prevalence (per-100,000) (95% CI) Sensitivity (95% CI) Specificity (95% CI) Study Disease Hospital Separations Rheumatologist All Physician Billings Billings Bernatsky 2007 (381) SLE 2.8 (2.6-3.0) 3.0 (2.6-3.4) 31.9 32.8 42 (41-43)% – 68 (67-68)% 45 (43-47)% – 56 (56-57)% 99.99 (99.98-99.99) % – 99.99 (99.98-99.99)% – 99.99 (99.99-100)% 99.99 (99.99-100)% Bernatsky 2009 (349) SSc 20% minimum 73% maximum Bernatsky 2009 (362) PM/DM 10.5 8.1 10.2 Bernatsky 2011 (5) all SARDs -CTD 40-70% 50-90% -  All Sources 51.0 --15.6 -  69  Table 2.2 Summary of Included Encounters and Costing Procedures COMPONENT SERVICE/SEPARATION DATES ELIGIBLE FOR INCLUSION [start date] to [end date +30 days]a Outpatient [start date] to [end date +30 days]a Prescription Inpatient: fiscal years 1995/1996-2000/2001 Inpatient: fiscal years 2001/2002 – December 31, 2006 Inpatient: January 1, 2007 – December 31, 2007 Day Case: fiscal years 1995/1996-2000/2001 Day Case: fiscal years 2001/2002-2007/2008  Hospital  UNIT COST CALCULATION  [start date -30 days]b to [end date +30 days]c  = [Amount Paid] datafield = [Drug Cost Paid datafield] + [Professional Fee Paid datafield] =[cRIW01 x CPWC]  [start date -30 days]b to [end date +30 days]c  =[cRIW05 x CPWC]  [start date -30 days]b to [end date +30 days]c  =[p_RIW07 x CPWC]  [start date -30 days]b to [end date +30 days]c  =[DPG-RIW x CPWC]  [start date -30 days]b to [end date +30 days]c  unit costing n/a; total annual costs= [total costs for 2000] x [annual change in inflation from 2000] =[LOS x $225]  ALC [start date -30 days]b to [end date +30 days]c not including encounters during 2008 b where separation occurred on or after the start date c where admission occurred on or before the end date a  70  Table 2.3 Provincial (BC) CPWC Values Fiscal Year Extrapolated CPWC 1995/96 $2,737.50 1996/97 $2,925.20 1997/98 $3,112.90 1998/99 $3,300.60 1999/00 $3,488.30 2000/01 $3,676.00 2001/02 $3,863.70 2002/04 $4,051.40 2003/04 $4,239.10  Fiscal Year 2004/05 2005/06 2006/07 2007/08  Actual CPWC, Obtained from CIHI $4,325.00 $4,767.00 $4,802.00 $4,939.00  71  Table 2.4 Summary of Encounters Considered for Each Utilization and Cost Analysis COMPONENT ANALYSIS Gross Attribution - Narrow Attribution - Broad -all eligible - all eligible claims from a non- - all eligible claims from Outpatient claims rheumatologist practitioner that a non-rheumatologist contained any SARD diagnostic practitioner containing code (as listed in Appendix A) any SARD diagnostic code (as listed in - all eligible claims from a Appendix A) rheumatologist with any diagnostic code - all eligible claims from a rheumatologist with any diagnostic code; -all eligible claims from Appendix A -all eligible -all eligible separations with a -all separations with a Hospital separations primary discharge diagnosis for primary or secondary SARDs discharge diagnosis for SARDs -all eligible -all eligible prescriptions as -all eligible prescriptions Prescriptions prescriptions listed in Appendix A as listed in Appendix A  72  3 Results 3.1 Descriptive Statistics Over twelve years, 18,741 SARDs cases were identified, contributing 82,140 patient-years (PY) of total follow-up. 16,773 (89%) of SARDs cases had a SARDs-CTD diagnosis and 1,680 had a SARDs-VD diagnosis. The number of cases and patient-years observed each year are listed in Table 3.1. The SARDs-VD group had a lower proportion of females than the SARDs-CTD group (65% vs. 76%) and the mean age of SARDs-VD cases at the start date was higher (70 years) when compared to SARDs-CTD (52 years) (Table 3.2). These differences are consistent with published reports (described in Chapter 1) of SARDs-VD having an older peak age of incidence, and less of a female predominance, than SARDs-CTD.  3.2 Gross Overall Direct Medical Costs Over twelve years, the entire SARDs cohort incurred over $571,216,780 in direct medical costs. The cumulative costs for the SARDs-CTD and SARDs-VD groups were $469,854,838 and $75,697,339, respectively. With regards to all SARDs cases, the relative contributions of each cost category were $154,580,563 (27%) from outpatient services, $291,664,951 (51%) from hospitalizations, and $124,971,267 (22%) from prescription medications (Figure 3.1a). As illustrated in Figure 3.2 the hospital proportion decreased over the study period (from 63% of costs in 1996 to 46% in 2007) while that for prescriptions increased (from 13% to 27%) and actually equalled the outpatient in the final year of study. For SARDs-VD, the overall proportion of costs from hospital was higher (67%) and that from prescriptions lower (13%) than for SARDs-CTD cases (Figures 3.1b & 3.1c).  SARDs case incurred, on-average, annual per-patient-year (PY) overall direct medical costs of $6,954 per-PY. The mean annual per-PY costs for SARDs-CTD cases were $6,230/PY, while SARDs-VD cases incurred mean annual per-PY costs that were more than double SARDs-CTD at $15,892/PY (Table 3.3). Annual overall mean per-PY costs for all-SARDs decreased by 32% over twelve years, and by 27% for SARDs-CTD (Figures 3.3-3.4). The decrease was even greater (47%) amongst SARDs-VD cases (Figure 3.3c).  73  3.3 Gross Outpatient Consumption Over twelve-years, the SARDs cohort had a total of 2,445,748 outpatient claims (Appendix B). On average allSARDs, SARDs-CTD, and SARDs-VD cases had 30, 28, and 48 outpatient encounters per PY, respectively. For SARDs-CTD cases these diagnostic tests and physician visits accounted for 29% of overall mean annual per-PY costs but for SARDs-VD this was just 20%. Still, at $3,146/PY the mean annual per-PY costs for SARDs-VD were almost twice as high as those for SARDs-CTD ($1,783/PY). Annual mean per-PY costs decreased over the period, by 26%, 25%, and 32% for SARDs, SARDs-CTD, and SARDs-VD, respectively (Figure 3.5). These cost decreases were accompanied by similar decreases (19%, 18%, 21% for the three groups, respectively) in mean per-PY outpatient encounters (Appendix B).  Table 3.4 lists the five most-frequently billed medical specialties that combined for 85% of all encounters. Tests and investigations were a major contributor to outpatient utilization, as 56% of all outpatient encounters by SARDs cases were billings from the specialist physicians associated with these services (Laboratory Medicine, Medical Microbiology, and Radiology). The most frequently visited practitioners were primary care physicians and internists, whose claims combined for another 29% of outpatient encounters. The percents varied slightly, but these practitioners were visited with the same general frequency by each diagnostic group.  3.4 Gross Hospital Consumption SARDs cases had 48,055 hospital admissions over the twelve years. Nearly two-thirds (64%) were inpatient stays, while 35% were day-case (admissions for surgeries or procedures that did not involve an overnight stay) and just 0.6% were extended or alternate level of care (ALC). Amongst all cases the annual mean number of admissions averaged 0.59 per-PY. Inpatient admissions averaged 0.37 per-PY amongst all cases, but over the study period only half the cohort had such an admission. These admitted cases averaged 1.72 (STD=1.31, range=1-24) annual admissions with a mean LOS of 10.32 days for each (Table 3.5). As compared to SARDs-CTD, SARDs-VD cases consumed more hospital services: more were hospitalized (78% vs. 47%) at any point, and those hospitalized SARDs-VD cases had more annual admissions, on average (1.90 (STD=1.47, range=1-22) vs. 1.68 (STD=1.27, range=1-24) per-PY). These admissions also tended to be longer (12.0 days vs. 9.9 days for SARDs-VD and  74  SARDs-CTD, respectively), on-average. The annual and twelve-year sums for each type of admission, and all admissions combined, are available in Appendix B.  Averaging $3,551/PY for all-SARDs, hospital care contributed the most toward overall mean per-PY direct medical costs. Not surprisingly inpatient care was disproportionately costly, accounting for 64% of all-SARDs admissions but nearly all (94%) of their total hospital costs. Though SARDs-VD cases averaged more than twice as many annual admissions per-PY (1.31/PY) as SARDs-CTD (0.53/PY), their average annual per-PY costs were more than triple, at $10,700/PY vs. $2,968/PY.  Mean per-PY hospital costs amongst all SARDs cases decreased by half over twelve years (Figure 2.6). This was accompanied by not only a similar decrease (46%) the mean number of admissions per-PY (Appendix B), but also a longitudinal decrease in the intensity of hospital use. The annual proportion of cases admitted to hospital decreased from 32% in 1996 to 15% in 2007, while amongst these admitted cases the mean number of annual admissions decreased from 1.87 (STD=1.53, range=1-22) in 1996 to 1.66 (STD=1.29, range=1-15) in 2007. The mean length of these admissions also decreased – by 7% - from 10.3 to 9.6 days (Figure 3.7, Table 3.5).  3.5 Gross Prescription Medication Consumption Annual mean per-PY medication costs over twelve years were $1,521/PY, $1,479/PY, and $2,046/PY for SARDs, SARDs-CTD, and SARDs-VD, respectively. However Figure 3.9 illustrates our most dramatic finding: while mean per-PY outpatient and hospital costs decreased over the study period, mean per-PY prescription medication costs increased in this time. For all-SARDs this was a 50% increase, from $1,117/PY in 1996 to $1,670/PY in 2007 (Appendix B), while the increase was 49% for SARDs-CTD and 59% for SARDs-VD. With 86-90% of the cohort dispensed a prescription each year, the annual mean per-PY costs just amongst users were very similar to the crude annual estimates, and increased almost identically (by 54%) over twelve years (Appendix B).  Similarly, while mean per-PY utilization of outpatient and hospital resources decreased over the twelve years, perPY prescription utilization increased, and quite substantially. SARDs cases were dispensed an average of 30 prescriptions per-PY overall, and (as shown in Figure 3.10 and Appendix B) annual per-PY consumption increased  75  by 49%, nearly as much as per-PY costs did. While SARDs-VD cases had a mean per-PY medication cost increase (59%) that was similar to SARDs-CTD cases (49%), their twelve-year increase in per-PY medication consumption (96%) was much greater than SARDs-CTD (44%), and overall SARDs-VD cases averaged 25 more prescriptions per-PY than SARDs-CTD (53/PY vs. 28/PY). Consistent with what we observed for prescription costs, the annual mean number of prescriptions dispensed per-PY amongst users overall - 29 per-PY for all-SARDs - almost matched the crude estimate, as did the twelve-year increase in mean annual per-PY prescriptions (55%) (Appendix B).  3.5.1 Costs and Consumption by Drug and Drug Class Spending and consumption were examined for each drug (by active generic ingredient) and drug class (according to Anatomical Therapeutic Chemical (ATC) Classification, Second Level). The most common agents and categories within each ATC-2 class are available in Appendix A. Tables 3.6 and 3.8 and Figures 3.11-3.12 illustrate the most dispensed drug classes and medications overall, while the annual breakdowns are available in Appendix B. Even amongst all prescriptions dispensed to the cohort, the heavy burden of SARDs is clear, with many of these therapies used to treat SARDs directly, or to manage the adverse effects of SARDs treatments and comorbidities that frequently arise in these cases. With corticosteroids a mainstay therapy for all SARDs, it was fitting this class made the fifth-largest contribution to cumulative prescriptions and one such GC, prednisone, was almost always the mostfrequently prescribed drug each year. The anti-malarial hydroxychloroquine, another common SARDs therapy, was also heavily-prescribed. Considering that many SARDs medications can cause gastritis and osteoporosis, it was also fitting that the proton-pump inhibitors (PPIs) omeprazole and rabeprazole (part of the Antacid class), and, especially for SARDs-VD cases, the bisphosphonates etidronate and alendronate, were prescribed frequently to protect against these adverse effects.  Those drug classes prescribed most frequently generally accounted for the greatest costs as well (Table 3.7, Figure 3.13), though immunosuppressants were disproportionatly costly, making up 7.8% of drug costs but only 1.7% of all prescriptions. Concerning specific drugs, omeprazole and hydroxychloroquine were among both the most dispensed (Table 3.8, Figure 3.12) and most costly (Table 3.9, Figure 3.14) medications overall.  76  Emerging Vasodilators In Chapter 1 we described how interest is growing in the use of several vasodilators – that have other indications - to address digital ulcers, Raynaud’s phenomenon, and pulmonary arterial hypertension (PAH) in SSc. One is bosentan, approved in Canada in 2001 (89) to treat PAH (254), and after its release it quickly became one of the costliest drugs prescribed to this SARDs cohort (Table 3.9). In 2003 there were only 89 prescriptions filled for bosentan (and none to SARDs-VD cases), but with each costing $4,268 2007 CDN on-average, total costs for bostenan exceeded $379,000 that year. As shown in Appendix B this trend continued with bosentan actually being the most-costly drug in 2007, despite only 250 prescriptions for it. Sildenafil and epoprostenol, two other emerging and expensive vasodilators, did not influence drug costs in the same way. There were relatively few sildenafil prescriptions overall – just 547 in 2007– and its mean unit cost during our study ($153 2007 CDN) was high but nothing compared to that of bosentan. Epoprosentol prescriptions were much more expensive than sildenafil (with each averaging $1,823 2007 CDN) but this drug was rarely dispensed.  Mycophenolate Mycophenolate, a more expensive but less-toxic immunosuppressant when compared to cyclophosphamide or azathioprine (62), is officially approved for preventing the rejection of transplanted organs (63), and not the management of SARDs. Still it is available for this purpose under special access, and was one of only two new therapies to emerge for SARDs over our study period. With a mean unit cost of $535 2007 Canadian dollars, it heavily influenced the annual drug costs of our cohort. In 2002 it was the eighth-costliest drug prescribed to SARDs cases - even with only 313 prescriptions - and in subsequent years continued to move up the cost rankings (Appendix B). This immense cost contribution continued despite infrequent dispensing: in 2007 only 0.32% of all dispensed prescriptions were for mycophenolate. For context, while the average unit price amongst all drugs over twelve years was $116 2007 CDN, the average unit price of mycophenolate was almost five-times greater this, and about forty-five times greater than the cost of the average prednisone prescription ($12 2007 CDN), which was the most-frequently prescribed drug overall.  77  Biologic Therapies We also examined the costs contributed by the newer biologic therapies we detailed in Chapter 1: ritiximab (approved in 2000 (89), this was the second SARDs therapy that emerged during our study period) and etanercept and infliximab (both approved in 2001 (89)). These drugs were prescribed to 88 discrete cases over the twelve years: etanercept to 45 cases, infliximab to 26, and rituximab to 24. Table 3.10 shows these cases had a range of individual SARD diagnoses. While infliximab may actually induce SLE (17), amongst recipient cases with an SLE diagnosis, only one received this drug prior to the first record of that diagnosis (for this case, the first infliximab prescription was filled 16 months prior).  These three very expensive agents (costing approximately $10,000-$20,000 per-patient annually (32,225)) were not prescribed frequently enough to be amongst the top ten costliest drugs in any year for the cohort at large. Still their influence on costs rose in the last three years of our study, and with the exception of rituximab in 2006 total annual costs for these drugs in the final three years exceeded their proportion of total prescriptions by at least ten-times (Table 3.11). Together the three drugs accounted for 1.74% of total costs in 2005 but more than 3% in 2007, when etanercept also became the tenth-costliest drug prescribed to SARDs-CTD cases (Appendix B).  Lipid-Modifying Agents/Atorvastatin Though lipid-modifying agents - including statins - are not used in the direct treatment of SARDs, they are often dispensed to SARDs cases since they, especially those with SLE, have an elevated risk of cardiovascular disease due to underlying inflammation and exposure to GC (28). As such we were not surprised the volume of prescriptions for this class grew five-fold over the period. Consumption rose for one statin in particular, atorvastatin, which became the twelfth-most prescribed drug in 2007. With a higher-than-average unit price - at $128 2007 CDN, it was almost twice that of hydroxychloroquine ($66 2007 CDN) – and mean per-PY prescriptions rising for it, atorvastatin was another emerging cost contributor and amongst all SARDs cases was actually the third-costliest drug overall (Table 3.9).  78  3.5.2 Major Contributors Toward Twelve-Year Per-Patient-Year Consumption and Cost Increases To gain insight into the reasons for the tremendous growth in mean per-PY prescription quantities and costs, we examined which drugs and drug classes contributed most to it. As outlined by the Canadian Institute for Health Information (CIHI) (444) we did so by dividing, for each drug and class, the mean difference in its per-PY consumption and costs from 1996 to 2007, by the mean net increase for all prescriptions. This was the number of additional prescriptions dispensed to, and costs incurred by, the average SARD case in 2007, as compared to 1996. For costs the total mean net increase was $553/PY (from $1117/PY in 1996 to $1670/PY in 2007). Table 3.12a and Figure 3.14a list the five most influential drug classes, with the top-two - immunosuppressants and lipid-modifying agents - combining for one-third ($180) of the entire mean per-PY cost increase.  Looking at specific drugs, the four greatest contributors to the mean per-PY cost increase (Table 3.12b, Figure 3.14b) – bosentan, mycophenolate, atorvastatin, and rabeprazole – were also the four mostly-costly drugs in 2007 (Appendix B), though none were prescribed to the cohort (or even approved in Canada (89)) in 1996. SARDs likely influenced the contribution from these drugs, and from gabapentin: although classified as an anti-epileptic agent, it is also used to relieve neuropathic pain (445) and pain in fibromyalgia, a comorbidity of SLE (28).  It must be  emphasized that while bosentan and mycophenolate were the greatest contributors, on-average, few cases were actually prescribed them, so for most SARDs cases they would not have been responsible for the increase in drug costs. Still this is another indication of their small but growing influence on overall healthcare costs in SARDs.  Over twelve years, the mean number of prescriptions dispensed per-PY increased by 11 per-PY (from 23/PY in 1996 to 34/PY in 2007). Of interest two of the five classes that contributed most to the increase in annual per-PY prescriptions – lipid-modifying agents and anti-epileptics – and two of the five drugs – rabeprazole and atorvastatin were also major contributors to the mean increase in per-PY costs (Table 3.13, Figure 3.15). About 6% of SLE cases develop hypothyroidism (28) which may explain levothyroxine being the third-highest contributing drug, onaverage. As well, with ramipril (the second-highest contributing drug) having multiple applications in SARDs from hypertension to renal disease - and bisphosphonates (the fifth-highest contributing class) frequently prescribed as prophylaxis against GC-induced osteoporosis, much of the twelve-year increase in mean per-PY drug consumption and costs can be attributed to SARDs.  79  3.6 Net Overall Direct Medical Costs To determine the net healthcare burden of SARDs, or the additional direct medical costs imparted by these disorders, we separately tabulated the costs of just those healthcare services attributable to SARDs. As described in Chapter 2 we used both a narrow and broad definition of SARDs-related consumption. The narrow definition included all SARDs-coded and/or rheumatology-associated outpatient encounters and all hospital discharges where any SARD was the primary diagnosis. We also included these services under the broad definition while adding more SARDs-related outpatient services (Appendix A) and any hospital discharges with a SARD coded in any diagnostic position. A single list of SARDs-related medications was compiled (Appendix A) and employed for both definitions.  3.6.1 Net Overall Costs - Narrow Definition Approximately 29% of the direct medical costs incurred by the cohort were attributable to SARDs under our narrow definition. These totalled approximately $165,165,682, with $10,316,313 (6%) from outpatient, $29,969,223 (18%) from hospital, and $124,880,146 (76%) from prescriptions (Figure 3.17). The total attributable costs for each year are provided in Appendix B. As illustrated in Figure 3.18, the proportion of attributable costs increased over time. Less of the cumulative costs incurred by SARDs-VD cases were attributable (23%) as compared to SARDs-CTD (30%) though by 2007 this gap had narrowed (Figures 3.18b &c).  Hospitalizations were the largest component of costs in the gross analysis but prescription costs were by-far the largest here for all SARDs cases. This is likely because of the large proportion (99%) of gross dispensed prescriptions that were attributable to SARDs. SARDs-VD was an exception with the hospital proportion exceeding the prescription in most years up to 2004.  The hospital proportion decreased substantially over twelve years  (Figure 3.19), making up, for SARDs, SARDs-CTD, and SARDs-VD, respectively, 38%, 34%, and 55% of costs in 1996 but only 11%, 8%, and 27% of costs in 2007. The contribution from outpatient services – 6% for all SARDs and 7% for SARDs-CTD, but only 3% for SARDs-VD, changed little over the study period.  80  Per-capita, the annual mean per-PY overall direct medical costs that we attributed to SARDs averaged $2,011 perPY over twelve years. These annual mean per-PY costs were almost twice as high ($3,646/PY) for SARDs-VD cases than SARDs-CTD ($1,868/PY). While gross overall annual per-PY costs decreased by 32%, as illustrated in Table 3.15 and Figure 3.20, net annual per-PY costs remained stable for all groups over the study period.  3.6.2 Net Overall Costs - Broad Definition Under our broad definition we attributed nearly half ($263,040,026, or 46%) of our cohort’s gross direct medical costs to SARDs. This cumulative total was 59% greater than the attributable total under the narrow definition. From these costs $23,764,332 (9%) were from outpatient, $114,395,548 (43%) from hospital and $124,880,146 (48%) were from prescriptions (Figure 3.21). The total attributable costs for each year are provided in Appendix B. Unlike what we observed for the narrow, SARDs-VD cases had almost the same percent of attributable costs (47%) as SARDs-CTD (46%). For all groups these annual percents changed little over the study period (Figure 3.18).  The proportion of costs from hospital for all-SARDs (43%) was about two-and-a-half times greater here than under the narrow definition (18%). Still the proportion from hospital decreased over time - making up 68% of costs in 1996 for all-SARDs but just 27% in 2007 – while prescription costs increased almost identically, from 27% to 62% (Figure 3.22). Outpatient costs were slightly more influential using this definition (making up 9% of costs overall) when compared to the narrow (6%), and their annual contribution increased slightly, from 6% of costs to 10%, over twelve years.  With this expanded definition the net, or incremental, annual mean per-PY overall direct medical costs of SARDs averaged $3,202 per-PY, about one-and-a-half times (and $1,191) higher than under the narrow. Unlike with the narrow, these annual mean per-PY costs decreased over twelve years by 36% (Figure 3.21), slightly more than by what gross per-PY costs decreased (32%). But as with the narrow, the incremental mean per-PY costs for SARDsVD ($7,390/PY) were more than double those for SARDs-CTD ($2,866/PY) (Table 3.16).  81  3.7 Net Prescription Medication Consumption The impact of SARDs on prescription use was important: our attribution analysis captured nearly every prescription (over 99% for all-SARDs and SARDs-CTD, and 98% for SARDs-VD) dispensed to the cohort (Appendix B). This meant most prescription costs (at least 99.9% in all groups) were captured too (Appendix B). On an annual basis this percentage did decrease slightly: in 1996, only 0.03% of all prescriptions were not attributable to SARDs, but this increased to 1.00% of the prescriptions in 2007. The same prescriptions were included under the broad and narrow definitions.  This heavy SARDs-related consumption was widespread amongst the cohort, with an attributable prescription dispensed to every case who received any prescription during the period. The overall net mean per-PY estimates barely differed from the gross, with attributable annual mean per-PY prescription costs averaging $1,520/PY, $1,478/PY, and $2,044/PY for SARDs, SARDs-CTD, and SARDs-VD, respectively (Appendix B). Similar too were the twelve-year mean per-PY cost increases. The average SARDs case was still dispensed 30 prescriptions per-PY, with the same twelve-year increase (from 23 per-PY in 1996 to 34 per-PY in 2007) as under the gross (Appendix B).  3.8 Net Outpatient and Hospital Consumption - Narrow Definition 3.8.1 Net Outpatient Consumption Outpatient services had little influence on SARDs healthcare consumption under this definition, with only 7% of all outpatient encounters considered attributable. While the annual percent of attributable encounters changed little amongst SARDs-CTD cases, it increased slightly over twelve years (from 2% in 1996 to 4% in 2007) for SARDsVD cases (Appendix B). Given these low percentages it is not surprising SARDs, SARDs-CTD, and SARDs-VD cases averaged only 2.0, 1.9, and 1.7 attributable outpatient encounters per-PY. Attributable outpatient costs (Appendix B) totalled $10,316,313 and made up 6% of all incremental costs. The annual mean per-PY costs were not tremendous ($126/PY for all-SARDs) and changed little over the period (Appendix B).  82  3.8.2 Net Hospital Consumption Just seven percent of the cohort’s hospitalizations were attributable to SARDs under this definition, a proportion which decreased from 12% of 1996 hospitalizations to just 3% of those in 2007. There was a decrease in the annual attributable-proportion of hospital costs for SARDs-CTD cases (from 14% to 6%) but a slight increase (from 12% to 14%) for SARDs-VD. SARDs cases had an average of 0.04 attributable admissions per-PY, which decreased by 84% from 0.10-0.02/PY (Appendix B). Over twelve years the average annual per-PY hospital costs ($365/PY for SARDs and $266/PY for SARDs-CTD) decreased by 70% and 76%, respectively - more than gross per-PY costs did. For SARDs-VD cases these costs were considerably higher ($1,477/PY) and decreased less (by 51%) over twelve years (Appendix B).  3.9 Net Outpatient and Hospital Consumption – Broad Definition 3.9.1 Net Outpatient Consumption With our expanded definition almost all SARDs cases (98%) had an attributable encounter, including 98% of SARDs-CTD and 93% of SARDs-VD cases (as compared to 94% of SARDs-CTD cases and just 59% of SARDsVD cases under the narrow). More than twice as many (15% vs. 7%) encounters for all-SARDs were attributable here, and this increased slightly over twelve years from 11% of encounters in 1996 to 15% in 2007. The annual mean number of encounters per-PY was still small, but under this definition doubled from the narrow to 4.4 per-PY (Appendix B). The total attributable outpatient costs ($23,764,332) exceeded those under the narrow by 130% and accounted for 15% of gross outpatient costs. Annual mean per-PY costs ($289/PY for all-SARDs, $284/PY for SARDs-CTD, and $315/PY for SARDs-VD) were also more than double those under the narrow. In contrast to the twelve-year decreases we observed in mean per-PY gross outpatient costs, these costs actually increased over time, by 12% for all-SARDs, 6% for SARDs-CTD, and 93% for SARDs-VD (Appendix B).  3.9.2 Net Hospital Consumption When all twenty-five discharge positions were considered, 25% of the cohort’s admissions were attributable to SARDs - more than three-times that under the narrow – and 39% of their gross hospital costs – more than twice that under the narrow (Appendix B). These attributable-percentages did decrease over time, admissions from 42% of the gross in 1996 to 15% in 2007, and costs from 51% of the gross to just 26%. With the percent of attributable costs  83  exceeding that for attributable admissions in each year and for each diagnostic group, SARDs-related hospitalizations for our cohort may be especially costly.  SARDs cases averaged about four-times more attributable admissions (0.15/PY) annually under this definition than under the narrow (0.04/PY). Still, mean annual per-PY admissions decreased over twelve years, by 81% (from 0.37/PY-0.07/PY). SARDs-VD cases averaged more attributable admissions (0.44/PY) than SARDs-CTD (0.13/PY) but had a similar twelve-year decrease (78%) in these admissions as SARDs-CTD (81%). The average annual per-PY costs ($1,393/PY, $1,104/PY and $5,031/PY for SARDs, SARDs-CTD, and SARDs-VD, respectively) were also about four-times greater here than under the narrow, but decreased by about 74% over twelve years.  In this chapter we have presented the most relevant findings of our analysis. These will be discussed in greater depth in the next chapter, which will conclude with a final summary of our current work and the follow-up analyses we plan to undertake.  84  3.10 Tables Table 3.1 Number of Cases and Patient-Years Contributed SARDS Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Overall  # Cases 3,305 3,649 4,224 4,893 5,425 6,054 6,779 8,371 9,856 11,463 13,235 14,372 18,741  SARDs-CTD Patient-Years 2,925.58 3,237.08 3,612.50 4,217.17 4,833.50 5,387.67 6,052.75 7,288.08 8,750.50 10,373.00 12,045.33 13,416.83 82,139.99  # Cases 2,967 3,287 3,840 4,471 4,988 5,595 6,302 7,734 9,121 10,608 12,192 13,245 16,773  SARDs-VD Patient- Years 2,658.42 2,942.67 3,305.08 3,892.67 4,480.58 5,015.58 5,654.42 6,787.67 8,144.75 9,629.00 11,156.33 12,385.25 76,052.42  # Cases 287 304 315 342 342 351 350 490 572 674 848 928 1,680  Patient- Years 221.58 242.83 249.17 255.08 271.00 277.00 291.83 368.00 459.00 579.17 708.08 840.42 4,763.17  85  Table 3.2 Cohort Characteristics SARDs  SARDs-CTD  SARDs-VD  Total Cases  18,741  16,773  1,680  # Female (%)  13,948 (74.43%)  12,682 (75.61%)  1085 (64.58%)  Mean Age at Index Date (years)  53.97  52.35  69.65  Maximum Age at Index Date (years)  101  101  99  Mean Length of Follow-Up (months)  51.83  53.65  33.18  86  Table 3.3 Crude Annual Overall Mean Per-Patient-Year Costs (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $8,901.13 $7,699.89 $20,709.78 1996 $9,126.09 $7,913.16 $22,332.14 1997 $8,520.70 $7,511.71 $19,019.20 1998 $8,420.50 $7,263.20 $22,575.60 1999 $7,883.24 $6,813.42 $20,301.37 2000 $7,147.81 $6,254.67 $17,957.15 2001 $7,056.73 $6,402.34 $16,981.82 2002 $6,947.75 $6,208.18 $18,341.26 2003 $6,375.23 $5,883.99 $14,045.00 2004 $6,326.98 $5,790.73 $13,473.23 2005 $6,333.86 $5,704.59 $13,475.82 2006 $6,087.17 $5,603.42 $10,964.38 2007 Overall $6,954.19 $6,230.17 $15,892.23 -31.61% -27.23% -47.06% %-Change  87  Table 3.4 Top Five Most-Frequent Outpatient Encounters, by Billing Specialty SARDs SARDs-CTD SARDs-VD Specialty Number of % of Total Number of % of Total Number of % of Total Claims Claims Claims Claims Claims Claims Laboratory Medicine 2,861,612 47 2,453,581 46 318,743 53 General Practice 1,318,767 22 1,188,155 22 106,167 18 Internal Medicine 416,531 7 344,381 6 58,483 10 Medical Microbiology 387,257 6 342,602 6 31,791 5 Radiology 189,598 3 170,007 3 15,054 3 subtotal 5,173,765 85 4,498,726 85 530,238 89 Other Specialties 916,037 15 822,153 1 68,883 12 Total 6,089,802 100 5,320,879 100 599,121 100 -due to rounding, some percents may not sum to 100% exactly  88  Table 3.5 Annual Mean Length-of-Stay Per-Year and Per-Admission (amongst hospitalized cases) (days) Mean Annual Total LOS SARDs  SARDs-CTD  Mean LOS per Admission SARDs-VD  SARDs  SARDs-CTD  SARDs-VD  Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  Mean 19.32 18.83 17.87 19.03 19.94 19.47 18.92 17.39 15.85 16.54 17.07 15.90  STD 29.10 32.84 26.95 29.81 32.96 30.00 30.28 26.51 25.43 26.65 30.60 28.10  Mean 18.19 17.73 16.85 17.74 18.62 17.68 17.57 16.38 15.57 15.72 16.08 14.90  STD 27.69 33.23 26.44 27.49 31.27 27.65 27.29 26.27 25.25 25.65 28.40 27.12  Mean 24.52 23.82 22.21 25.71 24.92 27.21 26.01 21.47 18.43 21.32 20.74 21.38  STD 34.45 31.47 26.53 40.23 38.62 36.35 42.85 26.59 28.10 32.24 35.64 33.07  Mean 10.33 10.47 10.13 10.65 11.37 11.31 11.05 10.18 9.64 9.94 10.14 9.58  STD 15.88 19.21 17.32 19.54 21.19 20.01 18.44 16.55 16.04 17.05 18.54 15.61  Mean 9.95 10.05 9.78 10.26 11.13 10.50 10.40 9.78 9.49 9.68 9.75 9.29  STD 15.07 19.06 16.58 18.19 20.50 18.30 15.81 16.26 15.77 16.44 16.91 15.30  Mean 11.86 12.51 11.27 12.41 11.92 14.05 13.73 11.51 11.27 11.57 11.86 10.85  STD 19.36 20.59 18.06 24.97 23.89 23.85 27.65 16.89 18.73 20.35 23.75 17.29  Overall  17.72  28.99  16.66  27.67  22.74  33.80  10.32  17.88  9.93  16.94  11.97  21.31  %-Change  -17.70%  -18.08%  -12.78%  -7.30%  -6.70%  -8.55%  89  Table 3.6 Top-Ten Most Frequently-Prescribed Drug Classes SARDs Class Proportion (%) 8 12-year Analgesics Cumulative Psycholeptics 7 Psychoanaleptics 6 Antacids 6 Systemic Corticosteroids 4 Systemic Antibiotics 4 4 RAAS Agents Diuretics 4 3 Anti-inflammatory Agents Anti-protozoals 3  1996  2007  SARDs-CTD Class Proportion (%) 9 Analgesics Psycholeptics 8 Psychoanaleptics 7 Antacids 6 Systemic Antibiotics 4 RAAS Agents 4 4 Systemic Corticosteroids Diuretics 4 4 Anti-inflammatory Agents 4 Anti-protozoals subtotal 54% Analgesics 10  SARDs-VD Class  Systemic Corticosteroids Analgesics Antacids Psycholeptics Diuretics RAAS Agents Psychoanaleptics Systemic Antibiotics  Proportion (%) 7 7 6 6 6 5 5 4 4  Bisphosphonates Drugs for Obstructive Airway Diseases subtotal Systemic Corticosteroids  53% 10  3  subtotal Analgesics  49% 10  Psycholeptics Systemic Corticosteroids Systemic Antibiotics Antacids  9 6 6 6  Psycholeptics Systemic Antibiotics Antacids Systemic Corticosteroids  9 6 6 6  Analgesics Antacids Systemic Antibiotics Psycholeptics  8 6 6 6  Psychoanaleptics Anti-inflammatory Agents  5 4  Psychoanaleptics Anti-inflammatory Agents  5 4  Diuretics Cardiac therapy  5 5  Diuretics  4  Diuretics  4  4  Drugs for Obstructive Airway Diseases Calcium Channel Blockers subtotal Analgesics Psychoanaleptics  4  Drugs for Obstructive Airway Diseases Anti-protozoals subtotal Analgesics Psychoanaleptics  3  Drugs for Obstructive Airway Diseases RAAS agents  3 56% 9 8  Diabetes Therapies subtotal Antacids Systemic Corticosteroids  3 56% 6 6  3 57% 8 8  3  90  SARDs  2007  SARDs-CTD Class Proportion (%)  SARDs-VD Class  Class  Proportion (%)  Proportion (%)  Psycholeptics  8  Psycholeptics  7  Analgesics  6  Antacids  6  Antacids  6  Psycholeptics  6  RAAS Agents  5  RAAS Agents  5  Diuretics  5  Diuretics  4  Anti-epileptics  4  RAAS Agents  5  Anti-epileptics  4  Diuretics  4  Psychoanaleptics  5  Systemic Antibiotics  4  Systemic Antibiotics  4  Bisphosphonates  4  Systemic Corticosteroids  4  Anti-protozoals  3  Lipid-Modifying Agents  4  Lipid-Modifying Agents  3  Systemic Corticosteroids  3  Beta-Blockers  3  subtotal  54%  subtotal  53%  subtotal  50%  91  Table 3.7 Top-Ten Most-Costly Drug Classes (2007 Canadian dollars) SARDs SARDs-CTD Class Proportion Class Proportion (%) (%) 10 10 12-year Antacids Antacids cumulative Immunosuppressants 8 Immunosuppressants 8 Psychoanaleptics 7 Psychoanaleptics 7 7 7 Analgesics Analgesics RAAS Agents 6 RAAS Agents 6 5 5 Lipid-Modifying Agents Lipid-Modifying Agents 4 Calcium Channel 4 Calcium Channel Blockers Blockers Anti-protozoals 4 Anti-protozoals 4 Drugs for Obstructive Airway 3 Anti-inflammatory 4 Diseases Agents 3 3 Anti-inflammatory Agents Psycholeptics subtotal 57% subtotal 58% Antacids 12 Antacids 12 1996 Calcium Channel Blockers 7 Calcium Channel 7 Blockers Psychoanaleptics 6 Psychoanaleptics 7 Immunosuppressants 6 Analgesics 6  2007  Analgesics Drugs for Obstructive Airway Diseases RAAS agents Anti-inflammatory Agents  5 5  Systemic Antibiotics Anti-protozoals subtotal Immunosuppressants Antacids  4 4 59% 10 9  5 5  Immunosuppressants Anti-inflammatory Agents RAAS Agents Drugs for Obstructive Airway Diseases Anti-protozoals Systemic Antibiotics subtotal Immunosuppressants Antacids  SARDs-VD Class  Antacids Immunosuppressants RAAS Agents Drugs for Obstructive Airway Diseases Lipid-Modifying Agents Calcium Channel Blockers  Proportion (%) 10 9 8 7 6 5 5  Bisphosphonates Psychoanaleptics  5 4  Analgesics 3  6 5  Systemic Antibiotics subtotal Antacids Calcium Channel Blockers RAAS Agents Drugs for Obstructive Airway Diseases Systemic Antibiotics Immunosuppressants  5 5  Cardiac therapy Anti-neoplastic agents  4 4  5 4 62% 10 9  Psychoanaleptics Anti-thrombotic agents subtotal Immunosuppressants Antacids  4 3 58% 12% 9  62% 12 8 6 6 6 5  92  SARDs Class 2007  SARDs-CTD Class  SARDs-VD  Analgesics  Proportion (%) 7  Analgesics  Proportion (%) 7  Class  Proportion (%) 7  Psychoanaleptics  7  Psychoanaleptics  7  Drugs for Obstructive Airway Diseases Lipid-Modifying Agents  Lipid-Modifying Agents  6  Lipid-Modifying Agents  5  RAAS Agents  7  RAAS Agents Anti-epileptics Anti-hypertensives Calcium Channel Blockers  6 4 4 4  5 4 4 4  3  3  Analgesics Bisphosphonates Psychoanaleptics Calcium Channel Blockers Psycholeptics  6 5 5 5  Drugs for Obstructive Airway Diseases subtotal  RAAS Agents Anti-hypertensives Anti-epileptics Calcium Channel Blockers Psycholeptics  60%  subtotal  58%  subtotal  66%  7  3  93  Table 3.8 Top-Ten Most Frequently-Prescribed Drugs Overall SARDs SARDs-CTD Drug Quantity Drug Prednisone 99,302 Prednisone Acetaminophen76,474 Acetaminophencodeine(30mg) codeine(30mg) Hydroxychloroquine 70,095 Hydroxychloroquine Levothyroxine 62,641 Levothyroxine Zopiclone 44,698 Zopiclone Furosemide 41,408 Ramipril Lorazepam 38,607 Lorazepam Rabeprazole 35,485 Furosemide Methadone 31,556 Methadone Omeprazole 31,472 Rabeprazole  Quantity 77,415 71,594 68,963 54,393 40,081 36,064 34,780 33,440 31,292 29,421  SARDs-VD Drug Prednisone Furosemide Levothyroxine Ramipril Rabeprazole Etidronate Hydrochlorothiazide Warfarin Alendronate Zopiclone  Quantity 18,122 7,274 7,066 5,870 5,398 4,817 4,075 3,688 3,684 3,662  94  Table 3.9 Top-Ten Most Costly Drugs Overall (2007 Canadian dollars) SARDs SARDs-CTD Drug Cost Drug Cost Omeprazole $4,716,312.42 Hydroxychloroquine $4,544,282.46  SARDs-VD Drug Atorvastatin $  Cost 346,784.88  Hydroxychloroquine Atorvastatin  $4,607,507.46 $3,201,112.96  Omeprazole Bosentan  $ 4,281,761.58 $3,155,985.26  Omeprazole Ramipril  $ $  340,208.82 281,443.42  Bosentan  $3,155,985.26  Atorvastatin  $2,793,593.36  Rabeprazole  $  245,091.38  Mycophenolate mofetil  $2,891,982.68  Mycophenolate mofetil  $2,570,013.53  Alendronate  $  206,689.58  Ramipril  $2,632,366.36  Ramipril  $2,271,717.88  Cyclosporine  $  206,282.44  Azathioprine  $2,235,403.73  Azathioprine  $ 2,050,439.81  Prednisone  $  196,966.25  Rabeprazole  $2,134,508.76  Gabapentin  $1,867,080.09  Mycophenolate mofetil  $  195,008.65  Gabapentin  $2,049,041.56  Rabeprazole  $1,846,544.82  Pantoprazole  $  189,677.68  Morphine  $1,916,976.26  Morphine  $ 1,815,148.99  Etidronate  $  187,880.67  95  Table 3.10 Distribution of Specific SARD Diagnoses Amongst Cases Dispensed a Biologic Therapy Biologic Total Specific SARD Diagnoses Cases SLE SSc SjD PM DM PAN Wegener’s GCA TA 9 4 2 3 4 1 0 3 2 Etanercept 45 3 1 0 2 2 0 1 0 1 Infliximab 26 3 1 3 1 4 2 4 1 0 Rituximab 24 -sums of individual diagnoses do not equal total number of cases since some cases had multiple diagnoses and others simply one for SARDs-CTD or SARDs-VD  96  Drug  Etanercept  Infliximab  Rituximab  combined  Table 3.11 Annual Total and Mean Per-Patient-Year Prescription Quantity and Costs for the Biologic Therapies, 2005-2007 Year Rank for Prescription Proportion of Mean Rank for Total Cost Proportion Mean Cost Total Quantity Total Prescriptions Total (2007 CDN) of Total perPrescription Prescriptions per-PY Prescription Prescription Prescription Quantity Cost Costs (2007 CDN)  Mean Cost per-PY (2007 CDN)  2005  283  127  0.04%  0.01  20  $179,580.37  0.59%  $1414.02  $17.31  2006  280  159  0.04%  0.01  13  $258,183.34  0.72%  $1623.79  $21.43  2007  257  214  0.05%  0.02  11  $348,392.62  0.94%  $1628.00  $25.97  2005  488  22  0.007%  0.00  48  $91,754.47  0.32%  $ 4,170.66  $8.85  2006  471  35  0.01%  0.00  36  $138,375.57  0.38%  $ 3953.59  $11.49  2007  386  75  0.02%  0.01  17  $276,036.63  0.74%  $3680.49  $20.57  2005  552  11  0.003%  0.00  176  $15,289.29  0.05%  $1389.94  $1.47  2006  455  39  0.01%  0.00  173  $19,414.30  0.05%  $497.80  $1.61  2007 2005 2006 2007  428 -  51 160 233 340  0.01% 0.05% 0.06% 0.07%  0.00 0.01 0.01 0.03  97 -  $54,114.94 $286,624.13 $415,973.21 $678,544.19  0.15% 1.74% 2.13% 3.03%  $1061.08 -  $4.03 $27.63 $34.53 $50.57  97  Table 3.12a Drug Classes Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Prescription Costs Class Proportion (%) Twelve-Year Net Increase in Per-PY Costs (2007 CDN) Immunosuppressants 20 $109 Lipid-Modifying Agents 13 $71 Anti-hypertensives 11 $61 Analgesics 10 $55 Anti-epileptics 9 $50 subtotal 63% $346 Total Net Difference 100% $553 Table 3.12b Drugs Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Prescription Costs Generic Drug Proportion (%) Twelve-Year Net Increase in Per-PY Costs (2007 CDN) Bosentan 11 $61.31 Mycophenolate mofetil 10 $57.68 Atorvastatin 10 $54.01 Rabeprazole 9 $49.92 Gabapentin 6 $33.26 subtotal 46% $256.18 Total Net Difference 100% $553.05  98  Table 3.13a Drug Classes Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions Class Proportion (%) Twelve-Year Net Increase in Per-PY Prescriptions Psychoanaleptics 12 1.30 RAAS Agents 9 0.97 Lipid-Modifying Agents 8 0.92 Anti-epileptics 8 0.91 Bisphosphonates 6 0.64 subtotal 43% 4.74 Total Net Difference 100% 11.21 Table 3.13b Drugs Contributing Most to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions Generic Drug Proportion (%) Twelve-Year Net Increase in Per-PY Prescriptions Rabeprazole 8 0.87 Ramipril 7 0.75 Levothyroxine 5 0.54 Atorvastatin 4 0.48 Methadone 4 0.45 subtotal 28% 3.09 Total Net Difference 100% 11.21  99  Table 3.14 Twelve Year Changes in Prescription Quantity and Cost, Selected Drug Classes Drug Class % of Total %-Change in Mean %-Change in Total %-Change in Mean Prescriptions Price Prescriptions Prescriptions perPY Analgesics 8% 55% -16% 25% ($26.13-$40.55) (9.88-8.27) (2.27-2.83) 7% 7% -19% 21% Psycholeptics ($20.69-$22.16) (8.66-7.02) (1.99-2.40) Psychoanaleptics  6%  -29% ($61.94-$44.28) -25% ($95.04-$71.28) -11% ($13.47-$11.99) 6.41% ($34.65-$36.87) -35% ($90.80-$58.64) -30% ($12.28-$8.63) -25% ($54.61-$41.19)  44% (4.97-7.14) -3% (6.15-5.95) -43% (6.26-3.56) -41% (6.23-3.69) 75% (2.65-4.62) 3% (3.77-3.89) -36% (4.12-2.64)  114% (1.14-2.44) 44% (1.41-2.03) -15% (1.44-1.22) -12% (1.43-1.26) 160% (0.61-1.58) 53% (0.87-1.33) -5% (0.95-0.90)  Antacids  6%  Systemic Corticosteroids  4%  Systemic Antibiotics  4%  RAAS Agents  4%  Diuretics  4%  Anti-inflammatory Agents  3%  Anti-malarials  3%  -33% ($70.46-$47.51) -34% ($112.40-$73.90)  0.35% (2.88-2.89) -21% (3.11-2.46)  49% (0.66-0.99) 18% (0.71-0.84)  Calcium Channel Blockers  3%  Drugs for Obstructive Airway Diseases  3%  7% ($70.08-75.32)  -39% (3.52-2.15)  -9% (0.81-0.73)  Beta-Blockers  2%  Bisphosphonates  2%  Sex Hormones and Modulators Lipid-Modifying Agents  2%  Diabetes Therapies  2%  Immunosuppressants  2%  Cardiac Therapies  2%`  Anti-neoplastics  1%  -53% ($47.60-$22.40) -35% ($81.43-$53.03) -1.25% ($41.50-40.98) -43% ($155.06-$88.22) -3% ($40.52-$39.31) 89% ($167.48-$315.86) -38% ($41.86-$26.12) 9% ($68.17-$74.24)  54% (1.79-2.75) 156% (0.99-2.53) -39% (2.57-1.57) 398% (0.62-3.10) 2% (1.83-1.88) -2% (1.61-1.58) -61% (2.83-1.11) 6% (0.94-1.00)  129% (0.94-0.41) 281% (0.23-0.86) -9% (0.59-0.54) 641% (0.14-1.06) 52% (0.42-0.64) 47% (0.37-0.54) -42% (0.65-0.38) 58% (0.22-0.34)  2%  100  Table 3.15 Crude Annual Mean Net Overall Per-Patient-Year Costs – Narrow Definition (2007 Canadian dollars) SARDsYear SARDS SARDsVD CTD $1,982.16 $1,834.61 $3,285.35 1996 $1,986.95 $1,797.41 $3,960.67 1997 $2,140.39 $1,962.26 $3,692.23 1998 $1,933.33 $1,748.99 $4,265.96 1999 $1,978.37 $1,783.07 $4,219.59 2000 $2,020.59 $1,876.23 $3,477.23 2001 $2,037.15 $1,931.06 $3,381.31 2002 $2,122.98 $1,923.07 $5,018.92 2003 $1,993.88 $1,906.44 $3,142.83 2004 $1,937.44 $1,814.87 $3,241.42 2005 $2,022.44 $1,879.85 $3,725.51 2006 $2,004.39 $1,876.04 $3,293.68 2007 Overall $2,010.78 $1,867.69 $3,645.53 %1.12% 2.26% 0.25% Change  101  Table 3.16 Crude Annual Mean Net Overall Per-Patient-Year Costs – Broad Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $4,200.63 $3,610.06 $10,406.56 1996 $4,523.96 $3,933.19 $11,400.50 1997 $3,973.99 $3,505.34 $9,063.90 1998 $3,910.43 $3,363.26 $11,004.19 1999 $3,583.64 $3,189.72 $8,461.29 2000 $3,442.54 $3,088.13 $8,143.64 2001 $3,345.85 $3,014.51 $8,022.00 2002 $3,327.45 $2,932.49 $8,805.63 2003 $2,894.18 $2,671.09 $6,191.89 2004 $2,815.94 $2,566.50 $6,015.81 2005 $2,862.51 $2,537.94 $6,785.12 2006 $2,673.74 $2,471.95 $4,520.48 2007 Overall $3,202.34 $2,866.14 $7,390.01 %-Change  -36.35%  -31.53%  -56.56%  102  3.11 Figures Figure 3.1a Twelve-Year Proportion of Gross Medical Costs by Component – SARDs Outpatient  Hospital  21.88%  Prescription  27.06%  51.06%  103  Figure 3.1b Twelve-Year Proportion of Gross Medical Costs by Component – SARDs-CTD Outpatient  Hospital  23.94%  Prescription  28.86%  47.20%  104  Figure 3.1c Twelve-Year Proportion of Gross Medical Costs by Component – SARDs-VD Outpatient  Hospital  12.88%  Prescription  19.80%  67.33%  105  Figure 3.2a Annual Proportion of Gross Direct Medical Costs by Component – SARDs  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  106  Figure 3.2b Annual Proportion of Gross Direct Medical Costs by Component – SARDs-CTD  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  107  Figure 3.2c Annual Proportion of Gross Direct Medical Costs by Component – SARDs-VD  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  108  Figure 3.3a Crude Annual Mean Overall Gross Direct Medical Costs, Per-Patient-Year (2007 Canadian dollars)  SARDs  CTD  VD  $23,000 $21,000  Mean Costs Per-PY (2007 CDN)  $19,000 $17,000 $15,000 $13,000 $11,000 $9,000 $7,000 $5,000  Year  109  Figure 3.3b Crude Annual Mean Overall Gross Direct Medical Costs, Per-Patient-Year (2007 Canadian dollars) – SARDs and SARDs-CTD only  SARDs  CTD  $10,000 $9,500  Mean Costs Per-PY (2007 CDN)  $9,000 $8,500 $8,000 $7,500 $7,000 $6,500 $6,000 $5,500 $5,000  Year  110  Figure 3.4 Crude Annual Mean Gross Direct Medical Costs, Per-Patient Year – Overall and By Component, all SARDs cases (2007 Canadian dollars)  Overall  Outpatient  Hospital  Prescription  Mean Costs Per-PY (2007 CDN)  $9,100 $8,100 $7,100 $6,100 $5,100 $4,100 $3,100 $2,100 $1,100  Year 111  Figure 3.5 Crude Annual Mean Gross Outpatient Costs, Per-Patient-Year (2007 Canadian dollars)  SARDs  CTD  VD  $4,000  Mean Costs Per-PY (2007 CDN)  $3,750 $3,500 $3,250 $3,000 $2,750 $2,500 $2,250 $2,000 $1,750 $1,500  Year  112  Figure 3.6 Crude Annual Mean Gross Hospital Costs, Per-Patient-Year (2007 Canadian dollars)  SARDs  CTD  VD  $17,000 Mean Costs Per-PY (2007 CDN)  $15,500 $14,000 $12,500 $11,000 $9,500 $8,000 $6,500 $5,000 $3,500 $2,000  Year  113  Figure 3.7 Annual Mean Length of Stay Per-Admission, Amongst Hospitalized Cases (days)  Mean Length of Stay (days)  SARDs  CTD  VD  14.00 13.75 13.50 13.25 13.00 12.75 12.50 12.25 12.00 11.75 11.50 11.25 11.00 10.75 10.50 10.25 10.00 9.75 9.50 9.25 9.00  Year  114  Figure 3.8 Annual Mean Length of Stay Per-Year, Amongst Hospitalized Cases (days)  SARDs  CTD  VD  28.00 27.00  Mean Length of Stay (days)  26.00 25.00 24.00 23.00 22.00  21.00 20.00 19.00 18.00 17.00 16.00 15.00 14.00  Year  115  Figure 3.9 Crude Annual Mean Gross Prescription Costs, Per-Patient-Year (2007 Canadian dollars)  SARDs  CTD  VD  $2,300 $2,200 Mean Costs Per-PY (2007 CDN)  $2,100 $2,000 $1,900 $1,800 $1,700 $1,600 $1,500 $1,400 $1,300 $1,200 $1,100 $1,000  Year  116  Figure 3.10 Gross Crude Annual Mean Number of Prescriptions Dispensed, Per-Patient-Year  SARDs  CTD  VD  60 56 52  Quantity  48 44 40 36 32 28 24 20  Year  117  Figure 3.11a Top-Ten Most Frequently-Prescribed Drug Classes – SARDs  Analgesics 8.49%  Psycholeptics 7.39% Psychoanaleptics 6.42% Antacids 5.84% Systemic Corticosteroids 4.36% Systemic Antibiotics 4.34% RAAS Agents 4.06% Diuretics 3.92% Anti-inflammatory Agents 3.36% Anti-protozoals 3.16%  118  Figure 3.11b Top-Ten Most Frequently-Prescribed Drug Classes – SARDs-CTD  Analgesics 8.77%  Psycholeptics 7.58% Psychoanaleptics 6.70% Antacids 5.78% Systemic Antibiotics 4.36% RAAS Agents 3.96% Systemic Corticosteroids 3.93% Diuretics 3.76% Anti-inflammatory Agents 3.62% Anti-protozoals 3.52%  119  Figure 3.11c Top-Ten Most Frequently-Prescribed Drug Classes – SARDs-VD  Systemic Corticosteroids 7.39%  Analgesics 6.45% Antacids 6.25% Psycholeptics 6.06%  Diuretics 5.42% RAAS Agents 4.91% Psychoanaleptics 4.51% Systemic Antibiotics 3.90% Bisphosphonates 3.78% Drugs for Obstructive Airway Diseases 3.14%  120  Figure 3.12a Top-Ten Most Frequently-Prescribed Drugs – SARDs  Prednisone 4.03%  Acetaminophencodeine(30mg) 3.10% Hydroxychloroquine 2.84% Levothyroxine 2.54%  Zopiclone 1.81% Furosemide 1.75% Lorazepam 1.68% Rabeprazole 1.57% Methadone 1.44% Omeprazole 1.28%  121  Figure 3.12b Top-Ten Most Frequently-Prescribed Drugs – SARDs-CTD  Prednisone 3.58%  Acetaminophencodeine(30mg) 3.31% Hydroxychloroquine 2.84% Levothyroxine 2.51%  Zopiclone 1.85% Ramipril 1.67% Lorazepam 1.61% Furosemide 1.55% Methadone 1.45% Rabeprazole 1.36%  122  Figure 3.12c Top-Ten Most Frequently-Prescribed Drugs – SARDs-VD  Prednisone 7.22%  Furosemide 2.90% Hydroxychloroquine 2.84%  Ramipril 2.34% Rabeprazole 2.15% Etidronate 1.92% Hydrochlorothiazide 1.62%  Warfarin 1.47% Alendronate 1.47% Zopiclone 1.46%  123  Figure 3.13a Top-Ten Most Costly Drug Classes – SARDs  Antacids 9.94% Immunosuppressats 7.87% Psychoanaleptics 6.83% Analgesics 6.75% RAAS Agents 5.73% Lipid-Modifying Agents 4.61%  Anti-protozoals 3.83%  Anti-inflammatory Agents 3.36%  Calcium Channel Blockers 4.26%  Drugs for Obstructive AirwayDdiseases 3.42%  124  Figure 3.13b Top-Ten Most Costly Drug Classes – SARDs-CTD  Antacids 9.92% Immunosuppressats 7.87% Psychoanaleptics 7.05% Analgesics 7.03% RAAS agents 5.58% Lipid-Modifying Agents 4.52% Calcium channel blockers 4.19% Anti-protozoals 4.18% Anti-inflammatory Agents 3.58% Psycholeptics 3.31%  125  Figure 3.13c Top-Ten Most Costly Drug Classes – SARDs-VD  Antacids 10.14% Immunosuppressats 7.87% RAAS Agents 7.41%  Drugs for Obstructive Airway Diseases 7.01%  Lipid-Modifying Agents 5.73% Calcium Channel Blockers 5.06% Bisphosphonates 4.89%  Psychoanaleptics 4.72% Analgesics 4.36% Systemic Antibiotics 3.13%  126  Figure 3.14a Top-Ten Most Costly Drugs – SARDs  Omeprazole 3.77% Hydroxychloroquine 3.69%  Atorvastatin 2.56% Bosentan 2.53% Mycophenolate mofetil 2.31% Ramipril 2.11% Azathioprine 1.79% Rabeprazole 1.71% Gabapentin 1.64% Morphine 1.53%  127  Figure 3.14b Top-Ten Most Costly Drugs – SARDs-CTD  Hydroxychloroquine 4.04% Omeprazole 3.81%  Bosentan 2.81% Atorvastatin 2.48% Mycophenolate mofetil 2.28% Ramipril 2.02% Azathioprine 1.82% Gabapentin 1.66% Rabeprazole 1.64% Morphine 1.61%  128  Figure 3.14c Top-Ten Most Costly Drugs – SARDs-VD  Atorvastatin 3.56% Omeprazole 3.49%  Ramipril 2.89% Rabeprazole 2.51% Alendronate 2.12% Cyclosporine 2.12% Prednisone 2.02% Mycophenolate mofetil 2.00% Pantoprazole 1.95% Etidronate 1.93%  129  Figure 3.15a Drug Classes Contributing Most, On-Average, to the Twelve-Year Increase in Mean Per-Patient-Year Prescription Costs  Immunosuppressants 20%  other 37% Lipid-Modifying Agents 13%  Anti-epileptics 9%  Antihypertensives 11% Analgesics 10%  130  Figure 3.15b Drugs Contributing Most, On-Average, to the Twelve-Year Increase in Mean Per-Patient-Year Prescription Costs  Bosentan 11% Mycophenolate mofetil 10%  Atorvastatin 10%  other 54%  Rabeprazole 9%  Gabapentin 6%  131  Figure 3.16a Drug Classes Contributing Most, On-Average, to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions  Psychoanaleptics 11.59% RAAS Agents 8.67% Lipid-Modifying Agents 8.17% other 56.02% Anti-epileptics 8.12% Bisphosphonates 7.43%  132  Figure 3.16b Drugs Contributing Most, On-Average, to the Twelve-Year Increase in Mean Per-Patient-Year Dispensed Prescriptions  Rabeprazole 8.53% Ramipril 7.81% Levothyroxine 6.72% Atorvastatin 4.83%  other 67.80%  133  Figure 3.17a Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Narrow Definition – SARDs  Outpatient  Hospital  Prescription  6.25% 18.14%  75.61%  134  Figure 3.17b Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Narrow Definition – SARDs-CTD  Outpatient  Hospital  Prescription  6.60% 14.12%  79.28%  135  Figure 3.17c Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Narrow Definition – SARDs-VD  Outpatient  Hospital  Prescription  3.40%  40.42% 56.08%  136  Figure 3.18a Annual Proportion of Gross Overall Direct Medical Costs Attributable to SARDs – all SARDs cases  Narrow Definition 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Broad Definition  47% 50% 47% 46% 45% 48% 47% 48% 45% 45% 45% 44% 22% 25% 22%  23% 25% 28% 29% 31%  31%  31%  32%  33%  Year  137  Figure 3.18b Annual Proportion of Gross Overall Direct Medical Costs Attributable to SARDs – SARDs-CTD  Narrow Definition 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Broad Definition  47% 50% 47% 46% 47% 49% 47% 47% 45% 44% 44% 44% 24% 23% 26% 24% 26% 30% 30% 31% 32% 31% 33% 33%  Year  138  Figure 3.18c Annual Proportion of Gross Overall Direct Medical Costs Attributable to SARDs – SARDs-VD  Narrow Definition 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  50% 51%  48% 49%  42% 45% 47% 48%  Broad Definition  44% 45% 50%  41% 16% 18% 19% 19% 21% 19% 20% 27% 22% 24% 28% 30%  Year  139  Figure 3.19a Twelve-Year Proportion of Net Direct Medical Costs by Component, Narrow Definition – SARDs  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  140  Figure 3.19b Twelve-Year Proportion of Net Direct Medical Costs by Component, Narrow Definition – SARDs-CTD  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  141  Figure 3.19c Twelve-Year Proportion of Net Direct Medical Costs by Component, Narrow Definition – SARDs-VD  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  142  Figure 3.20 Crude Annual Mean Attributable Overall Direct Medical Costs, Per-Patient-Year – Narrow Definition (2007 Canadian dollars)  SARDs  CTD  VD  $4,950 $4,700  Mean Costs Per-PY (2007 CDN)  $4,450 $4,200 $3,950 $3,700 $3,450 $3,200 $2,950 $2,700 $2,450 $2,200 $1,950 $1,700  Year  143  Figure 3.21a Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Broad Definition – SARDs  Outpatient  Hospital  Prescription  9.03% 47.48% 43.49%  144  Figure 3.21b Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Broad Definition – SARDs-CTD  Outpatient  Hospital  Prescription  9.90%  51.58%  38.52%  145  Figure 3.21c Twelve-Year Proportion of Overall Net Direct Medical Costs by Component, Broad Definition – SARDs-VD Outpatient  Hospital  Prescription  4.26% 27.66%  68.08%  146  Figure 3.22a Twelve-Year Proportion of Net Direct Medical Costs by Component, Broad Definition – SARDs  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  147  Figure 3.22b Twelve-Year Proportion of Net Direct Medical Costs by Component, Broad Definition – SARDs-CTD  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  148  Figure 3.22c Twelve-Year Proportion of Net Direct Medical Costs by Component, Broad Definition – SARDsVD  100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%  Year Outpatient  Hospital  Prescription  149  Figure 3.23 Crude Annual Mean Attributable Overall Direct Medical Costs, Per-Patient-Year – Broad Definition (2007 Canadian dollars)  SARDs  CTD  VD  $11,200  Mean Costs Per-PY (2007 CDN)  $10,200 $9,200 $8,200 $7,200 $6,200 $5,200 $4,200 $3,200 $2,200  Year  150  4 Discussion and Conclusion  4.1 Summary of Key Findings This research project was a longitudinal analysis of the health resource utilization and direct medical costs of SARDs at the population-level. It was one of few either longitudinal or population-based studies in Canada, or even the world, and the first to study all the SARDs together. We found the cumulative direct medical costs over a period of twelve years to be $571,216,780, $469,854,837, and $75,697,339, for SARDs, SARDs-CTD, and SARDs-VD cases, respectively. From the cumulative total for all-SARDs, $154,580,563 (27%) was for outpatient encounters, $291,664,951(51%) was from hospitalizations, and $124,971,267 (22%) was for prescription medications.  Prescription medications have been implicated in this unprecedented analysis as an important driver of the healthcare costs of SARDs at the population-level, both now and in the future. Here we have not only calculated the medical costs for all SARDs collectively, which has never been done before, but our twelve-year study period was the longest of any investigation like it. These factors, when combined with our population-based Canadian data source, made our estimates of the health resource utilization and direct medical costs of SARDs the most representative and comprehensive to-date. Our innovative efforts certainly paid off: the longitudinal trends we observed in these annual per-PY costs were most intriguing, but have some potentially alarming implications as well.  We identified 18,741 SARDs cases and found the healthcare burden of SARDs to be substantial, with annual direct medical costs averaging $6,954/PY, $6,230/PY, and $15,892/PY for all-SARDs, SARDs-CTD, and SARDs-VD, respectively. From the $6,954/PY for all-SARDs, $1,882/PY (27%) was for outpatient encounters, $3,551/PY (51%) for hospitalizations, and $1,521/PY (22%) for prescription medications. As compared to SARDs-CTD, SARDs-VD cases had a higher proportion of their costs from hospital (67% vs. 47%) and a lower proportion from medications (13% vs. 24%). Utilization of health resources is substantial too, with SARDs cases averaging 30 outpatient encounters, 0.59 hospitalizations, and 30 prescriptions per-PY. Those SARDs, SARDs-CTD, and SARDs-VD cases that had an inpatient hospitalization averaged 1.72 (STD=1.31, range=1-24), 1.68 (STD=1.27, range=1-24), and 1.90 (STD=1.47, range=1-22) annual admissions, respectively. Amazingly, although total annual  151  healthcare costs grew over the twelve-year period, overall mean per-PY costs decreased by 32% in this time, from $8,901/PY in 1996 to $6,087/PY in 2007 for all-SARDs. Mean per-PY outpatient costs decreased by 26%, from $2,205 to $1,641 per-PY, as did outpatient utilization, by 19% from 34 to 27 encounters per-PY. Hospitalizations showed an even larger decrease, with per-PY costs decreasing by half. Amongst the whole cohort mean annual admissions decreased by nearly as much (46%), with mean annual admissions amongst hospitalized SARDs cases decreasing too (by 11%, from 1.87-1.66). However, despite most healthcare costs decreasing over twelve years, the per-capita prescription costs for SARDs cases increased substantially in this time, by 50% from $1,117/PY to $1,670/PY. This was accompanied by an almost identical increase in prescription quantity, with the mean number dispensed annually to each case increasing by 49% in all-SARDs from 23 per-PY in 1996 to 34 per-PY in 2007.  We also compared the costs of SARDs-CTD cases with those of SARDs-VD cases, and SARDs-VD imparted a much heavier strain on the healthcare system. Their overall mean per-PY costs ($15,892/PY) were more than double those for SARDs-CTD cases ($6,230/PY), and their mean per-PY hospital costs ($10,700/PY) were more than triple those of SARDs-CTD ($2,968/PY). Correspondingly these cases had, on-average, twice as many hospital admissions (1.31/PY vs. 0.53/PY amongst all cases, 1.90 vs. 1.68 for hospitalized cases), 20 more outpatient encounters (48/PY vs. 28/PY), and 25 more prescriptions (53/PY vs. 28/PY) per-PY than did SARDs-CTD cases. However each subgroup demonstrated the same longitudinal trends, including the dramatic increase in per-PY medication costs.  Many aspects combined to make our analysis unique: it was population-based, longitudinal, used Canadian administrative data, and examined all the SARDs together. Its uniqueness, however, restricts comparisons with those studies conducted previously. The only other population-based, Canadian study (of PM/DM) reported (in 2007 Canadian dollars) similar annual per-patient outpatient ($1,042) and inpatient ($2,964) (294) costs to our perPY estimates ($1,881/PY for outpatient and $3,146/PY for inpatient hospitalizations), thus supporting the validity of our findings. Medication costs were not included in that study. Although no investigators have calculated the healthcare costs of SARDs as a group, some clinic-based estimates do exist for some individuals SARDs. Our overall annual per-PY costs ($6,954/PY) were close to those reported (and inflated to 2007 dollars) for one Canadian SLE cohort ($6,210 per-patient) (315) but exceeded those for a Canadian SSc cohort ($5,038 per-patient).  152  In that study, the outpatient costs ($1,492 per-patient) were similar to ours ($1,882/PY) and mean annual medication costs ($1,575 per-patient) nearly identical to ours ($1,521/PY), but the hospital costs ($1,670 per-patient) (274) were approximately half of ours ($3,551/PY). Two additional Canadian SLE investigations produced much higher annual estimates than ours ($6,954/PY), but the first set, $8,667-$10,752 2007 CDN per-patient (276) were calculated for 1989 and 1990 so are unlikely to reflect current patterns of health resource delivery and utilization. The other investigation was more recent, but the per-patient estimates ($10,186-$12,122 2007 CDN (297)) may be less accurate since they were derived from just four weeks of utilization data. Also, the healthcare needs of clinic-based populations tend to be higher, on-average, than those in population-based cohorts like ours.  With published estimates of the direct healthcare costs of SARDs-VD severely lacking, comparisons to previous reports are difficult. The overall mean per-PY costs that we calculated for all SARDs-VD cases ($15,892/PY) were more than double those reported amongst just 67 TA cases ($6,801 2007 CDN per-patient) in Italy (318). The annual mean inpatient costs reported for TA ($3,631 2007 CDN per-patient), PAN ($4,879 per-patient) and Wegener’s ($5,324 per-patient) in New York State (295) were also lower than our overall annual per-PY estimate for all SARDs-VD ($10,700/PY). However, this could be explained by US healthcare costs tending to be higher, on-average, than Canadian. We do emphasize our corresponding annual estimate for 2007 ($6,179/PY) was closer to these published estimates.  The first-ever Canadian estimates of the net health resource and economic burden of any SARD at the population level were also produced by us. This was done according to two definitions of SARDs-attributable healthcare consumption, though the same prescription medications were included in both definitions. The narrow definition included all outpatient claims from a rheumatologist and/or coded with a SARDs diagnosis, and all hospitalizations with a primary diagnosis of a SARD. Even under this narrow, most conservative definition of SARDs-related health services, SARDs-related annual costs accounted for 29% of gross healthcare costs and averaged $2,011 per-PY. Prescription medications were especially influential here, with 76% of net costs ($1,520/PY) from medications, and SARDs cases averaging 30 attributable prescriptions per-PY. Hospital services contributed to 18% ($365 per-PY) of costs and outpatient just 6% ($126 per-PY), with SARDs responsible for an average of 0.04 hospital admissions and two outpatient encounters per-PY in these cases. While SARDs-VD cases incurred higher gross per-PY  153  healthcare costs than SARDs-CTD, somewhat less (only 23%) of these were actually attributable to SARDs under this definition.  All services from the narrow were included under the broad definition too, as were other SARDs-related outpatient services and hospitalizations with a SARD as any discharge diagnosis. The immense burden of SARDs was particularly apparent under this definition, with these disorders responsible for nearly half (46%) of all the direct medical costs incurred by our cases. This was consistent for the SARDs-CTD and SARDs-VD subgroups, with the overall annual burden of SARDs averaging $3,202/PY, $2,866/PY, and $7,390/PY, respectively for each group. Attributable hospital costs were almost four-times greater under this definition at $1,393/PY (44%) for all-SARDs, though prescription costs remained the largest contributor (47% of net costs). Mean per-PY net outpatient costs and encounters doubled under this broad definition but - averaging $289 and four encounters per-PY for all-SARDs were still relatively low. This attribution analysis suggests SARDs-VD-specific care may actually be more expensive than care specific to SARDs-CTD. Nearly the same percentage of gross healthcare costs for SARDs-VD and SARDs-CTD cases (47% and 46%, respectively) were attributable to these disorders under the broad definition, but the actual annual mean per-PY attributable cost estimate was still two-and-a-half-times greater for SARDs-VD ($7,390/PY) than SARDs-CTD ($2,866/PY).  Longitudinally, annual mean per-PY costs were relatively stable for each group under the narrow definition but decreased under the broad. The twelve-year decrease in mean per-PY net costs (36%) was slightly greater than that for gross costs (32%). Over twelve years these net mean per-PY costs decreased by twice as much in SARDs-VD cases (by 57%) than SARDs-CTD (28%).  As with gross costs, the innovative design of our study - our grouping of the SARDs, population-based analysis, twelve-year study period, and use of Canadian administrative data - also limit comparisons of our attribution estimates to previous work. This problem is compounded by the paucity of estimates available as to the incremental costs of any SARD. Table 1.9a lists the findings (all standardized to 2007 Canadian dollars) of the six previous investigations that produced estimates of incremental mean per-patient costs. Our net overall per-PY estimate under our broad definition was close to annual per-patient estimates for SjD in the UK ($3,085) (311) and SLE (in the  154  absence of nephritis) in the US ($2,964) (301). Although the four other per-patient cost estimates were much higher than ours (299,302,304,316), the gross per-patient estimates were much higher in these investigations too . We found 29-46% of gross healthcare costs could be attributed to SARDs, and this percent varied amongst the prior studies. Three reported higher percents (57% of total costs for SjD cases (311), and 63% (299), and 66% (302) of total costs for SLE cases) while the percent was similar or lower (15-42% (304), 23-31% (301)) than ours in two other SLE investigations.  4.1.1 Burden of SARDs Even when viewed in isolation the per-capita healthcare costs of SARDs cases are clearly substantial. When compared to average per-capita provincial healthcare spending in BC, our most current annual estimate for allSARDs cases ($6,087/PY in 2007) exceeded the BC estimate for that year ($3,231 2007 Canadian dollars (290)) nearly two-times. The magnitude of these costs is made more apparent when they are compared to the costs of other types of arthritis. The annual mean per-patient costs, in 2007 Canadian dollars, incurred by Canadians with RA, OA, juvenile idiopathic arthritis (JIA), and fibromyalgia are tabulated in Table 1.10. As described in Chapter 1 some previous cost estimates for some individual SARDs exceeded multiple estimates for RA. The estimates we calculated reinforced this, being comparable to, and actually exceeding, many of the estimates for RA and other arthritides, all of which are more prevalent than SARDs (Table 1.10). As well, these studies and others have suggested the intensity of health resource consumption, particularly with regards to hospital use, is lower in these other forms of arthritis (278,279,319). This may mean SARDs not only impart a high cost burden, but also a health services burden that is even larger than these other arthritides.  While population-based estimates for arthritis in Canada are lacking, our approach has been used to estimate the costs of other chronic conditions at the population-level in Canada. With two such investigations (for the costs of diabetes cases in Saskatchewan) producing overall per-patient estimates for 1996 ($4,296 2007 CDN)(442) and 2001 ($3,777 2007 CDN)(439) that were about half of the mean per-PY costs incurred by SARDs cases in those years ($8,901/PY and $7,148/PY, respectively) we do not believe we have overestimated the direct medical costs of SARDs with these methods.  155  4.1.2 Longitudinal Trends Over our twelve-year study we observed that the overall mean per-PY direct medical costs of SARDs cases have decreased (with specific decreases for the outpatient and hospital components), while mean per-PY medication costs have increased. We described in Chapter 1 how no longitudinal Canadian studies have produced annual per-patient cost estimates. However in the approximately eight years between two cross-sectional studies on the direct medical costs of SLE cases - by the same investigators using a similar patient population - the average annual per-patient costs decreased by 44%, from $11,124 2007 CDN (for 1989-1990) (276) to $6,210 2007 CDN (for 1995-1997) (315). Between two similarly-conducted Canadian investigations of RA, a 67% decrease in the per-patient hospital costs of RA cases (from $2,767(277) to $907(278) 2007 CDN) was observed over approximately eight years. All of these decreases add to the plausibility of our findings, and as we will describe further in section 4.2 of this chapter, may stem from increased healthcare efficiencies.  Only two other longitudinal individual SARD studies exist for comparison (304,324), both of which were also population-based. The first followed a US cohort of “newly-active” SLE cases in a Medicaid database for five years: their mean annual per-patient costs decreased in the second year but increased by an average of 16% in each subsequent year (304). These authors attribute their findings to a pattern where the acute illness that first led to diagnosis or flare stabilizes in the second year, thus decreasing per-patient costs. They believe the annual costs subsequently increased from continued “disease progression” (304). But the cohort in this study was restricted to “newly-active” cases – either incident cases or prevalent cases beginning a new “episode of care” (defined as those who did not have an SLE-coded health encounter in the six-months preceding their index SLE encounter) (304). Since we did not employ the same restrictions one may expect the SLE cases in this US study to have different perPY cost and utilization patterns, on average, than our cohort, which may explain why we instead observed a decrease in mean per-PY costs. The second study (by Chiu and Lau) followed a cohort of SLE cases in Taiwan for up to eight years, from 2000 to 2007. The data source was the Taiwanese National Health Insurance Research Database, covering at least 96% of the population. Diagnostic accuracy was enhanced in this study since all SLE cases had their diagnosis verified in order to qualify for free medical insurance (324). Mean annual per-patient costs, which included outpatient and  156  hospital components, increased during this time but only by 15% (from $1,421 to $1,628), or $207 (324) 2007 CDN. Two factors could account for the disparity between these observations and ours: healthcare costs and delivery practices may differ between Canada and Taiwan, and the per-patient Taiwanese costs could have increased if Taiwan implemented more expensive SLE treatments during this time that had already been established in Canada. Second, although Chiu and Lau did not intend to exclusively tabulate the net costs of SLE, only those claims coded for SLE were included in their analysis, which limits its comparability to our gross analysis.  In addition to these disease-specific estimates, our findings are consistent with some longitudinal healthcare spending trends observed in the general Canadian population. From 1990 to 2010, mean per-capita drug costs (including non-prescription drugs) increased by 265% (from $250 to $912 1997 CDN) (444). Our observations as to the types of drugs consumed longitudinally also mirrored national trends. In our analysis the two classes contributing most to the growth in mean per-PY costs over twelve years were immunosuppresants (accounting for 20% of the difference in mean per-PY costs) and lipid-modifying agents (accounting for 13%). These same classes were also the top-two contributors (at 12% and 10%, respectively) towards wholesale drug spending increases from 2004/05 to 2009/10 amongst the general Canadian population (444).  4.2 Interpretation of Findings  4.2.1 Decreases in Mean Per-Patient-Year Outpatient Utilization and Costs In longitudinal analyses it is important to be consistent regarding the types and quantities of encounters captured each year. But this was impossible for outpatient encounters since billings from supplementary benefit practitioners (including physiotherapists, chiropractors, optometrists, massage therapists, naturopaths, acupuncturists, and podiatrists) were only available through the 2001/02 fiscal year. MSP stopped covering most of their services after this point (429,431). These billings only contributed to 8% of the outpatient claims we captured through 2001, but the average annual decrease in per-PY outpatient costs was more than twice as high in the years after 2002 (3.9%), when this change took place, than the years before (1.7%). Therefore, this longitudinal inconsistency may explain some of the corresponding decrease in per-PY outpatient costs, and, to a lesser extent, per-PY overall costs. To examine this, we compared the magnitude of the five-year decreases in outpatient costs and encounters before and  157  after this change took place. The decreases in the five years following it (2002-2007) were more than twice as much as those in the five years previous to it (1997-2001): 16.9% vs. 8.3%, for costs, and 8.43% vs. 3.64% for encounters. The partial absence of these encounters may also explain why our twelve-year average outpatient and overall per-PY estimates ($1,882/PY and $6,954/PY, respectively) were lower than per-patient estimates from some other Canadian reports ($2,922-$3,165 2007 CDN for outpatient, $8,667-$10,752 overall) (276) ($3,390-$3,578 2007 CDN for outpatient, $10,186-$12,122 overall) (297), which did include these encounters.  4.2.2 Increases in Mean Per-Patient-Year Prescription Costs Our most remarkable finding was the twelve-year increase in annual mean per-PY prescription costs, which was in contrast to the twelve-year decrease in overall annual mean per-PY direct medical costs. There are likely multiple reasons for such a large (50%) increase. But we must first emphasize this mean per-capita increase was observed after all annual costs were standardized to 2007 dollars, so inflationary increases are not the explanation.  Proportion of Cases Dispensed a Prescription The Canadian Institute for Health Information (CIHI) has proposed many other explanations for longitudinal increases in drug costs (444). Since our annual per-PY estimates were crude (calculated amongst all cases followed each year, and not just those receiving prescriptions), total prescription costs, and therefore mean per-PY costs as well, may increase if there were annual increases in the proportion of cases dispensed a prescription. But we did not observe a large increase: this annual proportion only grew by 4% over the twelve years, from 86% of cases in 1996 to 90% in 2007. Therefore this factor would only have made a minor contribution to the nearly 600% increase in total annual drug costs we observed. As well, the annual mean per-PY estimates calculated amongst users (or only those cases dispensed a prescription) (Appendix B) were very similar to the crude annual per-PY estimates (Appendix B) and these costs demonstrated nearly the same twelve-year increase (54%) as crude mean per-PY costs. This shows, instead of being driven by the rising costs of small number of cases, the 50% increase in mean per-PY costs reflected an increase in the costs incurred by most cases.  158  Mean Number of Prescriptions Dispensed Per-Patient-Year A rise in the annual number of prescriptions dispensed to each case, on-average, could also drive up both total (444), and mean per-PY drug costs. Such a rise (11 prescriptions per-PY, on-average or 49%) did occur within our cohort: cases were dispensed an average of 23 prescriptions per-PY in 1996 and 34 per-PY in 2007 (Appendix B). Of interest, the cost of each prescription over twelve years averaged $50.71 2007 CDN, which, when multiplied by 11 (the mean difference in per-PY prescriptions over twelve years), equals $557.81. As this amount is almost equal to the actual dollars by which mean per-PY drug costs increased from 1996 to 2007 ($553.05, on-average), this rise in annual mean prescriptions dispensed per-PY can be considered a major factor in the twelve-year drug cost increase.  Unit Drug Prices CIHI cites two other potential contributors to rising drug costs, unit price increases (particularly for the same drugs), and a changing and more-expensive drug mix (444). These factors are difficult to separate because the drug mix for SARDs cases was not stagnant over the twelve years. But unit price increases are an unlikely explanation since we observed nearly-identical increases in total drug costs and prescription quantities (586% and 582%, respectively), and mean per-PY costs and prescriptions, over the twelve years. If it was just a matter of the prices increasing, prescription costs would have increased much more than the quantity did. CIHI also discounts the influence of price increases. They reported that while nationally total drug costs increased from 1998 to 2007, drug prices overall actually decreased in this time, by an average of 2.7% each year (444). Our observations were similar: the average cost of each prescription changed little from 1996 to 2007, increasing by only 0.48% over the twelve years (from $48.61 to $48.84 2007 CDN), and by 0.10% annually, on-average.  More importantly, average prices decreased within many drug classes. Table 3.14 shows - for the major classes of drugs dispensed to the cohort - the twelve-year change in their annual mean prices, and in thirteen of these twenty classes the average annual price decreased. Of particular note are the many classes in which the annual mean unit price decreased while mean per-PY consumption actually increased. The drugs in many of these classes - RAAS agents, calcium channel blockers, bisphosphonates, and lipid-modifying agents – are commonly used in the management of SARDs and their associated comorbidities. Therefore with the mean annual unit price decreasing in most classes where mean per-PY consumption increased, unit price increases should not account for our findings.  159  Emergence of Expensive Prescriptions Although unit price increases for the same drugs would explain little of the per-PY cost increase, a more-expensive drug mix would.  As illustrated in Chapter 3 and Appendix B, some of the most frequently-prescribed drugs each  year were acetaminophen-codeine, prednisone, and levothyroxine. But given their low mean unit costs - $13, $12, and $12, respectively - increased dispensing of just these types of prescriptions could not lead to such a massive (50% per-PY) cost increase. In fact, over twelve years mean per-PY consumption actually decreased for two of these oft-prescribed-but-inexpensive drugs: prednisone by 16% and acetaminophen-codeine(30mg) by 38%.  Instead, some more expensive medications were prescribed to the cohort over the course of the study period, and in greater quantities. These drugs (and their average unit prices) included mycophenolate, which was only prescribed to our cohort after 1997 ($535 2007 CDN), and bosentan ($3,877), infliximab ($3,956), and etanercept ($1,609), all three of which were approved in 2001 (89). Given their high prices, the emergence of these drugs certainly had an impact, albeit moderate, on annual prescription costs. While they made no contribution to drug costs in 1996 or 1997, these five medications accounted for 10% of mean per-PY prescription costs in 2007. This occurred despite these being rarely dispensed (only accounting for 0.45% of total prescriptions in 2007). And their growing contribution to annual drug expenditures was not from their unit costs increasing, since the mean unit price for three of these drugs decreased between 2005 and 2007. Table 3.12b shows how bosentan and mycophenolate were the top contributors, on-average, to the twelve-year increase in per-PY drug costs. With so few prescriptions dispensed for these drugs, they would not have been responsible for the cost increase for most individuals. Still with these drugs costing so much and accounting for a rising share of total annual prescriptions (from 0.29% in 2003 to 0.37% in 2005 and 0.45% in 2007) the impact of this changing and more-expensive drug mix should be noted.  Higher Outpatient Drug Consumption Rises in per-PY prescription consumption and the emergence of more expensive drugs are likely the primary explanations for the increases in prescription medication quantity and costs; however, there is an additional, methodological one. Nationally, there has been a shift in drug spending from the hospital to outpatient setting. In 1996, about 40% of drug spending occurred in an outpatient setting and 60% in hospital, but this share has steadily  160  increased and in 2007 actually exceeded the inpatient share(446). Since PharmaNet captures all prescriptions dispensed on an outpatient basis, but not inpatient medications, this shift would lead to more prescriptions appearing in PharmaNet over time, followed by a higher mean number of prescriptions dispensed per-PY.  4.2.3 Potential for Improved Healthcare Efficiencies Our study has many positive findings. From the patient perspective, spending less time in hospital and making fewer physician visits may provide them with a better quality of life. From the government (payer) perspective our findings are suggestive of improvements in healthcare delivery, with care being provided more efficiently and at a lower cost. One example is the proportion of hospital admissions that were inpatient, which decreased from 75% of admissions in 1996 to 56% of admissions in 2007. In terms of spending, the crude average cost of each outpatient encounter decreased by 9% over twelve years, and the cost of each hospital admission by 8% (from $6,285$6,069/PY). Also, the twelve-year decreases we observed in mean per-PY outpatient and hospital utilization were less (19% and 46%, respectively) than the corresponding decreases in mean per-PY costs (26% and 50%, respectively). If these decreases were equal, it would suggest the mean per-PY cost decreases resulted only from decreases in consumption.  These longitudinal decreases in spending and resource consumption may be concerning to some but simply spending more money on healthcare and/or providing more services does not always lead to better outcomes. SARDs are no exception and this point was emphasized by Clarke et al while reporting on the healthcare costs and outcomes for Canadian, American, and British SLE cases in two studies. In their first study, the overall direct per-patient costs of the three cohorts did not differ significantly, but the Canadian cohort experienced significantly better outcomes (315). When this analysis was expanded, the Canadian and British cohorts incurred lower (20% and 13%, respectively) average per-patient costs but had similar outcomes to the American (314). The scope of our analysis prevented us from examining which outpatient encounters or types of admissions showed the largest decreases. With this information of particular value to healthcare administrators, we may detail this in the future.  161  4.2.4 Potential for Lower Costs and Better Outcomes from Greater Prescription Use The increases that occurred in per-PY drug use and costs can be interpreted positively or negatively. The positive interpretation is the increased drug use per-PY (and subsequent drug costs) has contributed to the decrease in overall per-PY costs by helping to improve the health status of these cases and thus reduce the need to consume many health services. This would be in addition to the system-wide improvements in health care delivery that have certainly reduced many costs. We could not study clinical outcomes directly, and as described above, the quantity of health resources consumed does not always correlate with disease state or outcomes. However, one indicator of SARDs outcomes may be the mean number of all hospital admissions combined (inpatient and day case) per-PY. Even if healthcare efficiencies shifted certain inpatient procedures to the day care setting, the total number of admissions should not be affected by this. Since we observed a 46% decrease in total admissions per-PY over twelve years, outcomes may actually be improving for the average SARD case.  The emerging SARDs therapies are themselves expensive, but concerns have been raised about the additional healthcare costs associated with their use. For instance, rituximab needs to be given by IV infusion (90) and monthly lab tests are required during the first year of mycophenolate treatment (63). But much of this additional consumption (including any additional laboratory tests and physician visits) would have been billed to MSP and therefore captured in our data. This means per-PY overall and outpatient costs still decreased, on-average, in spite of any additional outpatient services that were required. In fact, in actual dollars the twelve-year ($564/PY for allSARDs) decrease in mean annual per-PY outpatient costs actually exceeded the same mean per-PY increase in prescription costs ($553/PY for all-SARDs). With the mean per-PY increases in drug spending offset by decreases in outpatient and hospital spending, prescription drugs could be a good healthcare investment. From this perspective access to new drug therapies, even the more expensive ones, should be expanded if outcomes could continue to improve without an increase in overall average per-capita healthcare costs. Greater drug consumption may be the trade-off to reduce hospitalization and improve the management of these chronic disorders, Given that many of the drugs of interest were only available during the last half of our study period (or less), further investigation of their impact on long-term outcomes is required.  162  4.2.5 Potential for Rising Comorbidity Burdens and Complication Rates Unfortunately a much more alarming explanation exists for the increasing mean per-PY drug costs. In it, more efficient health care delivery is the primary reason for the decreases in overall per-PY costs. This means - instead of greater medication consumption contributing to better outcomes in SARDs – it may be an indication of patients requiring additional drug therapies after developing more comorbidities and/or complications.  This possibility is supported by many of our findings. Although new and expensive drugs contributed somewhat to the longitudinal increase, the five main emerging drugs still only combined for 10% of drug costs in 2007. This means the majority of costs were from ‘ordinary’, less-expensive drugs. And while two of these expensive drugs bosentan and mycophenolate - combined to account for 21% of the twelve-year increase in mean per-PY drug costs (Table 3.12b), this contribution was averaged amongst the whole cohort. Although the increase was experienced by most cases - with at least 86% of cases dispensed a prescription each year, and the twelve-year increase in mean prescriptions dispensed to users per-PY (55%) almost matching the number dispensed to all cases (49%) - most cases were not dispensed these therapies so they would not have affected their annual drug costs. Finally, unlike for the per-PY outpatient and hospital decreases, the twelve-year increases we observed in per-capita costs (50%) and prescriptions (49%) were nearly identical. If the increase in drug costs resulted mainly from funding these expensive but infrequently-used drugs, the number of prescriptions dispensed per-PY would not have increased so much alongside it.  We were not able to investigate which secondary conditions developed in SARDs cases, and at what rates, in this analysis but as described in section 4.7, we intend to explore this in our future work. We suspect, however, they may include diabetes, nephritis, and cardiovascular disease. Risk for cardiovascular disease is elevated in many SARDs including SLE (447,448) and Wegener’s (449), and this may stem from the inflammatory nature of these diseases and exposure to GC (28,450,451). GC exposure may also increase the risk of diabetes (17,26), and other chronic kidney diseases may result from SARDs-VD, SSc, and SjD (17,452,453), in addition to lupus nephritis. With diabetes managed primarily medically and on an outpatient basis, an increase in mean per-PY admissions from these conditions would not be expected, and hospital admissions for our cohort actually decreased by 44%.  163  However with multiple medications required to manage these conditions (4,27,454,455), a rise in mean per-PY prescriptions would be expected.  The longitudinal per-PY increases we observed for specific drugs and drug classes support these possibilities even more. Along with GC and immunosuppressants, ACE-II inhibitors and/or ACE-II receptor antagonists are often prescribed for lupus nephritis (27,455) and other chronic kidney diseases (17,453), and in our cohort, mean annual per-PY prescriptions for this class (RAAS Agents) rose by 160%. In addition to RAAS Agents, mean per-PY prescriptions increased for other anti-hypertensive classes including calcium channel blockers (by 18%), diuretics (by 53%), and beta-blockers (by 129%) (Table 3.14). This suggests the prevalence of hypertension (one cardiovascular morbidity that is particularly common in TA (29) and may be present in half of lupus cases (28)) may have increased amongst our cohort. Similarly an increase in the development of atherosclerosis (which may be accelerated in Wegener’s (456), and is the “most common manifestation of cardiovascular disease” in lupus cases, according to the Johns Hopkins Lupus Centre (28)) could explain the 641% increase in mean per-PY prescriptions for lipid-modifying agents we observed. The annual mean number of prescriptions for diabetes medications also increased, by 52% per-PY (Table 3.14), with mean per-PY costs for these drugs rising almost equally (by 48%).  If comorbidity burdens are increasing, new SARDs therapies like belimumab may be useful (particularly for lupus nephritis (4,27)), and could actually reduce the long-term risk of comorbidities. Their potential long-term benefits warrant further investigation, and as detailed more in section 4.7, such an investigation is in our plans.  4.2.6 Implications for Patients The decreases in mean per-PY health resource utilization that we observed, such as less-frequent hospitalization, are a positive development for patients. But mean per-PY medication consumption is rising, and the implications of this for patients could be quite negative. The side effects of these drugs may reduce quality-of-life, while the possible development of secondary illnesses could further decrease physical health and well-being. The serious financial implications of these medication increases should not be ignored either. In BC, all residents pay an annual premium for medical insurance, which entitles them to receive all medically-necessary hospital and outpatient services (431) without additional charge or co-payment. However they are personally responsible for most prescription costs, with  164  partial subsidy available from the province under the income-tested PharmaCare program (431). Although our main prescription medication estimates were based on the full cost of each prescription, we also explored those prescription costs paid by PharmaCare (using data available on each PharmaNet record), and our findings were disturbing. As total mean per-PY prescription costs increased (by 50%), those costs that PharmaCare covered decreased by almost the same magnitude, 46% on-average, over twelve years. In 2007, the mean per-PY prescription costs for all-SARDs were $1,670/PY, but on average, only $509 was subsidized by the province, leaving patients responsible for an average of $1,161.  When costs increase for other healthcare components, on an individual-level responsibility for these extra costs is shared amongst all taxpayers. Instead as prescription costs rise SARDs cases will face additional personal costs. Many BC residents have employer-sponsored insurance to reduce prescription costs but, as noted by Aghdassi et al with regards to SLE, complications can limit patients’ ability to work and thus their access to prescriptions. This manifests in two ways: their income may be insufficient to pay for their prescriptions, and their access to this employer-sponsored insurance is reduced (297). Therefore in addition to research aimed at reducing comorbidity risks, these increasing drug costs call for a re-examination of public drug policies and programs. For instance, Health Canada’s approval of rituximab for Wegener’s granulomatosus in late 2011 has increased the likelihood that, for Wegener’s cases, its cost will be covered. But CanVasc, a Canadian SARDs-VD advisory group, has noted such funding and approval decisions are made by each province separately (116). With these decisions tending to be inconsistent amongst provinces, some patients may still be forced to pay these extremely high costs themselves or go without. Further research is needed to determine the effectiveness of these high-cost medications in the longterm. If they could actually improve health status and decrease health resource utilization, they would become costeffective and this would be of great interest to policy makers and patients. However if per-capita drug costs continue to rise but public drug policies remain stagnant, adherence to many medications (and not just the most expensive) could very well decrease, potentially leading to further health problems and higher healthcare costs for patients and government in the long term.  165  4.3 Limitations The array of data available for each claim, and systematic way in which it is collected and recorded, are but some advantages of administrative data. However we still faced some limitations using this data source with regards to the exclusion of some health encounters, costing precision, and the accuracy of our cases’ SARDs diagnoses. These must be kept in mind when interpreting our estimates.  4.3.1 Omitted Health Care Costs Our tabulation of healthcare costs was restricted to provincially-funded health services. This meant services funded by other public agencies and levels of government (federal and municipal) were not included in our estimates. However with 93% of public healthcare spending in BC funded by the province in each of 1996 and 2007 (290), we still captured most costs incurred by the public payer. With the 2007 estimate of average per-capita public sector healthcare spending in BC ($3,458 2007 Canadian dollars (290)) differing little from the baseline (non-SARDs) mean per-PY overall healthcare costs we calculated for the same year ($3,414/PY), we have further assurance of the completeness of our cost capture.  Within the scope of provincially-paid health resources, we attempted to include all funded health care services in our cost estimates. But since we could only include consumption captured by MSP and the hospital dataset (Discharge Abstract Database, or DAD), some costs were omitted. The provincial government pays a large portion of ambulance costs (431), approximately $450 per-trip in 2007 (457), but data on the use of ambulance services are not available from PopData BC. The MSP database includes amounts paid to all fee-for-service (FFS) practitioners but not to those compensated under different schemes like contracts or salaries. With these arrangements employed more and more in rural areas of the province (429), we may have omitted many services provided there.  Many ER physicians are compensated under a non-FFS arrangement, and at the same time, outpatient ER visits (those not resulting in an inpatient admission) are not included in the DAD either (429). A portion of physician costs from these visits were captured, as some ER physicians (91% in 1998/99 (458) ) do bill MSP for each encounter. But most other associated costs, including staff, equipment, supplies, and medications (which are not recorded in PharmaNet either) were not included. To give an idea of these costs, a recent Canadian SLE study  166  estimated the average cost per ER visit at $173 (297), but without ER utilization data we could not reliably approximate these costs for our cohort. Since a BC report found ER users were more likely to be female (458) – and the majority of SARDs cases are female – and had poorer health and greater health resource utilization than the general population (458), a not-insignificant, and likely growing quantity (as nationally, the cost of an average ER visit increased by 47% from 2003-2008 (446)) portion of health care costs may have been missed. This omission must be considered when interpreting our annual cost estimates.  Since the consumption of other health care components is not recorded by the province, we were unable to tabulate their costs. These items include non-prescription medications, vitamins, natural health products, assistive devices (including eyeglasses and hearing aids), and most dental services (431). But with the provincial government not funding these items – patients do – their omission is justified. Still, caution must be taken when comparing our estimates to others, particularly clinic-based ones, which did include these additional items.  In the end, the vast quantity of health care expenses that we did include is a reliable proxy of provincial healthcare spending. In fact while keeping in mind SARDs cases consume more health resources than the average individual, our overall mean per-PY cost estimate for 2007 ($6,087 2007 Canadian dollars per-PY) far exceeded average provincial per-capita healthcare spending in BC ($3,231 2007 Canadian dollars (290)) that same year. With the exception of non-physician outpatient encounters (whose contribution should not be great), any omissions were consistent amongst study years and disease groups so would not impact our subgroup comparisons or longitudinal analyses. If anything, we are presenting a conservative, not an exaggerated, estimate of the net burden economic healthcare burden of SARDs, and this more ideal.  4.3.2 Underestimated Day Surgery Costs Day surgeries performed from 2001 and onwards comprised about 27% of all our cohort’s admissions but we very likely underestimated the costs of these procedures. Instead of using case-mix methodology, these costs were calculated by adjusting the year-2000 sum by the subsequent annual increases in inflation. But with care continuously shifting from the inpatient to the day-case setting, the average annual increase in inflation (2.42%) was actually much less than the average annual increase in day surgery costs from 1996 to 2000 (20%) and the annual  167  increase in number of day-case hospitalizations (21%, on-average, from 2001 to 2007). We had little choice in this matter though. PopDataBC did not provide the Resource Intensity Weight (RIW) for each day surgery so we could not use case-mix methodology. As detailed in Chapter 2, several approaches for estimating these costs were explored, and we selected the most conservative one. This may explain why we calculated lower per-capita day surgery costs than some Canadian investigations did (274,315). It may also explain why our overall per-PY costs were similar to those reported for SLE in 1989 and 1990 (276), when day surgeries were less frequent. Still with day surgeries making up a relatively small proportion of overall costs in our analysis (at most, just 3.7% of annual hospital costs from 1996-2000), any underestimation should have minimal effect on our overall per-PY estimates or longitudinal observations.  4.3.3 Minor Costing Uncertainties Other uncertainties exist in our hospital cost estimates. Since case-mix methodology could not be used for extended or Alternate Level of Care (ALC) separations either, their costs were determined by multiplying a widely-used BC government per-diem rate by the LOS. But any imprecision here would only have a minor impact on our annual costs with ALC separations accounting for just 0.6% of all separations, and 11.6% of all hospital days. The annual number of ALC separations also decreased greatly over the period, from a total of 53 in 1996 to just two in 2007.  We also faced some uncertainties even when calculating the cost of each inpatient hospitalization with case-mix methodology. This involves multiplying the annual average cost for a hospitalization in BC (called the Cost-PerWeighted-Case, or CPWC) by the Resource Intensity Weight (RIW) of each separation, a measure of the resource consumption required for the specific stay. Since CIHI did not provide annual cost-per-weighted-case (CPWC) values previous to 2004/05, these were estimated from the later years’ values. But each year’s value reflects specific year-to-year changes in health service delivery, including expensive technological innovations and increasinglycomplex caseloads (438). This means, even with our conservative approach (as detailed in the Methods, Chapter 2), the older CPWC values, and the corresponding-years’ hospital costs, may have been slightly overestimated.  Our study period spanned several editions of the RIW, and the one we had to use for the year-2007 separations was derived differently than the others. The magnitude and direction of any impact from these different derivations is 168  currently unknown.  In theory it reduces the comparability of our 2007 costs to previous years’, but the real impact  is likely minimal. For both hospital and overall costs, the decrease in per-PY costs from 2006 to 2007 (7% and 4%, respectively) differed little from the average annual cost decreases observed over twelve years (6% and 3%, respectively).  4.3.4 Capture of Attributable Outpatient Encounters The limitations of administrative pertaining to the coding of diagnoses and specialist codes may have affected our attribution analysis. While we tried to capture SARDs-related encounters by including all (non-rheumatologist) billings coded for SARDs, many would have been coded for the morbidity necessitating that encounter, such as nephritis or pneumonia. Since MSP only allows one diagnosis per claim, these diagnoses would have taken the place of the underlying SARD, thus reducing the number of SARDs-related encounters captured under the narrow definition. Given the high consumption of health resources mandated by SARDs treatment and follow-up, it is difficult to believe only seven percent of all outpatient encounters by SARDs cases, or, on-average, only two per-PY (as were attributable under our narrow definition), were related to SARDs. By failing to capture all attributable visits under this definition, we may have underestimated our annual per-PY incremental cost estimates. With this method of cost attribution dependant on the logistics of diagnostic coding, it may lack precision and estimates as to the burden of disease should be interpreted with caution.  Some logistics with coding also limited the capture of rheumatologist billings. Although these were identified by the specialty code on the claim, Rheumatology was not coded as a separate specialty by MSP until January 1, 1998. In theory this would also have underestimated our SARDs-attributable cost estimates. However the actual implications of this appear mixed. For instance, the percent of the total gross costs that were attributable to SARDs under the narrow definition was lower in 1996 and 1997 (22%) when compared to other years (Appendix B). But the proportion of attributable outpatient costs for all-SARDs changed minimally (from 5.1% to 5.7%) from 1997 (when Rheumatology was not coded) to 1998 (the first year that it was). The proportion of outpatient encounters that were deemed attributable in 1996 and 1997 (6%) was also not much different in later years. The effect may have been greater for SARDs-VD cases than SARDs-CTD, with percent of attributable costs and encounters rising more from 1997 to 1998 in these cases (SARDs-VD costs and encounters each by about 1% vs. about 0.5% for costs 169  and 0.15% for encounters in SARDs-CTD). We attempted to make up for these potential omissions by including additional encounters in our broad definition. But while the sensitivity for attributable encounters increased in this definition, we may also have included encounters not actually related to SARDs and thus decreased the specificity of the attribution estimate.  4.3.5 Accuracy of SARDs Diagnoses An odd pattern became apparent while examining the number of SARD cases we captured each year. As shown in Table 3.1, we had 3,305 cases in 1996 and 14,372 in 2007, making for an enormous twelve-year increase of about 335%. In comparison, the general population of BC only increased by 11% in this period (273). Avina-Zubieta et al calculated the annual prevalence and incidence of SARDS in BC residents aged 20 years and older with the same data, and some of their findings are listed in Table 4.1. According to their estimates (which accounted for growth in the general population), the prevalence of SARDs-CTD increased by 302% over twelve years and SARDs-VD by 252%, while the incidence of each increased by 172% and 125%, respectively (1). We described in Chapter 1 how the incidence of many SARDs appears to be increasing, and given the chronic nature of these disorders, this would also contribute to an increase in prevalence. The increased incidence may stem from increased clinical recognition, the aging population, or a true rise in their development, but for such massive increases to occur over just twelve years is not clinically-plausible.  We examined the external validity of these case numbers by comparing the annual prevalence rates for SARDs-CTD in BC (237 per-100,000 in 2003 and 389 per-100,000 in 2007 (1)) as calculated from our data, to those also calculated with administrative data by Bernatsky et al from three other Canadian provinces. Since these prevalence estimates were similar - ranging from 260 to 410 per-100,000 in 2003 (5)) (Table 1.2) – they support the validity of our data and prevalence estimates, at least for the later years of our study. Many other explanations are possible for these apparent increases. The introduction of a new or more sensitive diagnostic test could have suddenly driven up case numbers by identifying cases that would otherwise have done undetected.  It is suspected the incidence of  some SARDs-VD in the UK increased in the 1980s after the introduction of the ANCA test (7,459), but no major diagnostic test for any SARD was introduced during this period. A large rise in the number of practicing rheumatologists in BC could have led to more SARDs diagnoses, but between the 2001/02 and 2007/08 fiscal years  170  the number of rheumatologists only increased by 10 (from 43 to 53) (460). Since the BC population also increased in this time (by 6% (273)), it meant the average number of patients per rheumatologist only increased by 5% (from 954 to 1005 (460)). With access to rheumatology services (and potentially, in turn, SARDs diagnoses) increasing very little in this period, this is unlikely to have accounted for the 137% increase we observed from 2001 to 2007. Finally the BC population on the whole is growing older and with some SARDs, particularly SARDs-VD, mainly affecting those 50 years of age and older, somewhat of an increase in case numbers would be expected. Still the proportion of the BC population that was aged 50 years and older only increased by 28% (from 27% of the population in 1996 to 34% in 2007) in this period (273), meaning this age shift should only have contributed to onetwelfth of the twelve-year (335%) increase.  Instead we believe the primary explanation lies with the exclusion criteria applied to potential cases after initial screening. Potential cases with eligible SARD-coded encounters from 1990 onwards were identified, but to optimize diagnostic accuracy these individuals continued to be screened for exclusionary encounters through either 2007 or end-of-follow-up (from either death or de-registration from MSP). These encounters are detailed in Chapter 2 but included (for cases first diagnosed by a non-rheumatologist) a rheumatologist visit where the SARD diagnosis was not confirmed or, for any case, a subsequent diagnosis of a different type of inflammatory arthritis, such as RA or psoriatic arthritis. Each additional year of follow-up that was available for a potential case equated to another year in which that case could be excluded. The unbalanced case numbers arose from those identified late in the study period having fewer years in which to be excluded. There was no minimum follow-up required of any case, or maximum timeframe for the exclusionary encounter to occur. This meant, when compared with those identified in the 1990’s, many more of the potentially ineligible cases identified in later years remained in the final cohort. In fact, when our colleagues went back and did not apply these exclusion criteria to any potential cases, the annual case numbers balanced out.  Instead of eliminating our exclusion criteria, diagnostic accuracy could be increased by implementing a minimum follow-up period. This would allow all potential cases to be thoroughly screened for exclusionary encounters, but there is a trade-off. Requiring cases to have, for instance, five years of follow-up - as in Li et al (304) - may introduce a survival bias since only cases that could be followed for that length of time would be eligible. If the  171  mean annual healthcare costs of these cases were systematically different from those with less follow-up time (which is possible if more complex cases tended to incur high annual costs for 1-2 years but then died from these complications), biased estimates would be produced.  With eighteen years of potential follow-up time (1990-2007), some potential cases may have had an exclusionary encounter many years after the index visit. However we suspect most exclusionary encounters occurred within five years of the initial SARD diagnosis, and in our cohort the highest annual increases in the prevalence of SARDs-CTD did occur between approximately 2002 and 2005 (two-to-five years before our follow-up period ended) (Table 3.1). A similar pattern was with the annual prevalence of SARDs-VD.  While this injects some uncertainty into the SARDs-status of our cohort, we at least have assurance our case definitions were quite sensitive. Still, the effect it had on our annual and longitudinal cost estimates is unclear. In later years the inclusion of milder cases with less-certain diagnoses may have brought down our annual per-PY cost estimates and partially accounted for the 32% decrease in overall per-PY costs. Looking at Figures 3.3 and 3.4, overall mean per-PY costs decreased the most from 1997 until 2001, and it is after this point the ‘excess’ cases really became apparent. In fact, the annual average decrease in mean per-PY costs from 1997 to 2001 was 4.2%, but dropped to 2.6% from 2002 to 2007. Using consistent methods of case ascertainment throughout a longitudinal analysis is vital for making valid comparisons of annual estimates. However it may be more correct to include these uncertain cases in our cohort. While still maintaining some diagnostic specificity (which we did with our rigorous inclusion criteria), capturing the costs of all potential SARDs cases (even questionable ones) may actually produce a more accurate estimate of the total healthcare burden of these diseases. Further research on the internal validity of this diagnostic algorithm is required, and we intend to undertake this.  4.4 Strengths Despite these limitations, our study has several strengths. By using population-based data our analysis incorporated a large patient sample, multiple years of health claims, and precise unit costs. Given this country’s single-payer healthcare system, using Canadian administrative data allowed for a comprehensive tabulation of health resource consumption and increased the generalizability of our findings by minimizing selection bias. Not only is ours the  172  first analysis of the SARDs together, but these factors also make ours the largest and longest study of the healthcare costs of SARDs to date.  4.4.1 Maximum Case Ascertainment and Sample Size By grouping SARDs diagnoses and identifying potential cases over eighteen years (1990-2007) we obtained 18,741 cases - the second-largest sample known for any SARDs costing study. This maximized statistical power and the external validity of our findings. Bernatsky et al have proposed that administrative data may fail to capture cases with milder disease who don’t seek care during the study period (5,381,434), but it is unlikely a true case would go for a period as long as ours without needing SARDs-related care. SLE, like many individual SARDs, often relapses and remits but the period of disease quiescence (during which cases often don’t need treatment specific to SLE (340)) tends to span only a few years. In a study of the Hopkins Lupus Cohort, this phase accounted for the smallest proportion of follow-up time, with an average of only 2.3 years and a maximum of 5.7 (461). With the exception of new arrivals to BC, most potential cases would have had at least this much screening time. Finally, Bernatsky et al have also reported high sensitivities for the ICD-9 codes pertaining to SARDs (exceeding 88% for all but SLE) (434) that we used to identify cases, meaning we should have captured most cases with these codes. It must be emphasized that clinic-based data holds no advantage over population-based in this respect, with such mild or earlyonset cases unlikely to attend at a tertiary rheumatology clinic.  4.4.2 Minimally-Biased and Comprehensive Estimates Even population-based cohorts may over-represent one socioeconomic class or health state, but our study population was minimally biased in these respects. Most adult SARD cases in the province could have been included, regardless of care provider or area of residence, two factors identified in Chapter 1 as potentially influencing health resource utilization. Our estimates therefore incorporated the health resource consumption of a wide spectrum of SARDs cases, and should accurately reflect the per-PY costs of the average case. Enhancing this further was the systematic and detailed records held by PopDataBC and single-payer nature of BC’s health system. This allowed us to capture nearly every publically-funded health service consumed, including every community-dispensed prescription, records unavailable in most other Canadian databases. Since the exact cost of each outpatient and  173  prescription claim was available, these estimates should highly reflect the amounts paid by the provincial government, maximizing the accuracy of our cost estimates.  4.4.3 Longitudinal Design We not only estimated the annual healthcare consumption of SARDs cases, but produced these estimates for each of twelve years. When a cross-sectional investigation produces a solitary cost estimate, there is no context available for interpretation, and without knowing if healthcare costs are rising or falling, it is difficult to predict future research and health care needs. Since multiple cross-sectional estimates may not be comparable if determined with different methods and study populations, any longitudinal inferences made from these must be taken with caution. We know of only two other individual SARDs studies producing such longitudinal estimates (304,324), and these had much shorter follow-up periods. Our longitudinal data also let us evaluate the potential impact of new therapies and changes in funding and service delivery on annual mean per-PY costs. For instance, our average annual decreases in per-PY outpatient costs were greater after paramedical services (including physiotherapists, chiropractors, optometrists, massage therapists, naturopaths, acupuncturists, and podiatrists) ceased to be reimbursed. Of particular interest to health policymakers is how the approval of bosentan, infliximab, and etanercept in 2001 may have driven up per-PY medication costs in subsequent years.  4.5 Contribution  4.5.1 Knowledge Gaps In quantifying the previously-unknown burden of SARDs on the Canadian health care system at the populationlevel, we filled many gaps in this research that were identified in Chapter 1. Despite their shared pathogenesis, manifestations, and treatments, this is, to our knowledge, the first time health resource utilization and direct medical costs have been quantified for the SARDs collectively. Grouping the SARDs made great biological sense and greatly increased the size of our cohort: at 18,741 cases, it was nearly unmatched and dwarfed those seen in clinicbased studies. In quantifying the healthcare burden of these rare and relatively-unknown diseases, we made clear their public health impact. This should make research to improve outcomes and care for SARDs (and in doing so,  174  reduce this health resource burden) a priority in Canada, and inspire more groups to conduct clinical and health services research relating to SARDs.  Prior to our work Canadian estimates existed for some individual SARDs, but most were determined with a tertiary clinic population. The small samples and short follow-up periods inherent with these studies, and greater severity of disease within their cohorts, limited the external validity of these estimates. Our estimates were instead produced using population-based data. This was only the second Canadian study to do so, and the first to encompass all three healthcare components, making ours the most generalizable Canadian estimates to-date. Little was unknown about the trajectory of these costs over time, with few longitudinal estimates produced from any country (and none from Canada), so we have made a significant contribution in studying these costs over twelve years (the longest of any SARDs study). This information could be valuable when planning for future needs: the positive effects of any past efficiencies implemented in the delivery of outpatient and hospital services are supported by the decreases in mean per-PY outpatient and hospital costs we observed. At the same time the contribution to overall costs from prescriptions came to equal the outpatient contribution in the last year of our study, and in subsequent years will likely exceed it. The large increases we observed in per-PY medication consumption and costs should make it clear this healthcare component will be a main driver of future healthcare costs in SARDs, and we would not have been able to determine this in a cross-sectional analysis.  On top of this work, we also estimated the incremental costs of SARDs, which though minimally studied and not ever reported for Canada, are especially important for health policy and justifying increased research support. When previously faced with estimates of the healthcare costs of individuals SARDs, it was difficult for Canadian decisionmakers to know how much of these costs were actually due to the SARD, but this information is now available. These factors combine to make our analysis truly unique and capable of producing the most reliable and comprehensive estimates to date.  4.5.2 Future Cost-Effectiveness Analysis This study was not a formal cost-effectiveness analysis but the estimates we produced could make a valuable contribution towards one in the future, particularly for the new-but-costly drugs for SARDs. Similar breakdowns of  175  healthcare costs were useful in earlier discussions of care for RA. It was reported that much of medical costs in RA stemmed from hospitalizations, particularly operations to correct joint destruction and deformity caused by this disease. Having this information supported the widespread use of the some of the same biologic therapies now being introduced for SARDs (namely infliximab and etanercept), if they could reduce this joint damage and subsequent need for costly surgery in RA (462).  4.6 Applications and Knowledge Translation With our work spanning the clinical pharmacy, rheumatology, and population health and economics disciplines, these findings will be of interest to many audiences. Estimates of the current costs of SARDs could be used by patients to lobby for better treatments to improve their health and quality of life. Our detailed breakdowns by component, year, and disease subgroup will be especially useful to health care administrators, policy makers, and health economics researchers to help further improve health care efficiencies and allocate future resources. It is imperative clinicians are made aware of the rises we observed in drug use and costs, particularly as they may be tied to rising comorbidity burdens. By working with patients to reduce their occurrence, and detect and treat these conditions earlier, clinicians could put a significant dent in these burdens.  With this in mind, and the innovative nature of our assembled cohort and study design, we will be proactive in disseminating our findings. Preliminary results were presented in poster format to the North American arthritis community at the Canadian Arthritis Network (CAN), and Canadian Network for Improved Outcomes in Systemic Lupus Erythematosus (CaNIOS) annual meetings in 2011. With these organizations valuing consumer partnerships, we were able to educate patients at these venues too. We presented a poster at the American Association for the Advancement of Science (AAAS) annual meeting in February 2012, which allowed us to reach a broader audience interested in scientific and public policy, and did so too at the meeting of the Canadian Rheumatology Association (CRA) in March 2012. Since our CRA abstract will be published in the international, peer-reviewed Journal of Rheumatology, our findings will reach beyond the physical conference. Estimates pertaining to the net burden of SARDs will be communicated to those who regularly measure the value and impact of drugs and technologies in a poster at the Canadian Agency for Drug and Technologies in Health (CADTH) Symposium in April 2012. We will  176  also present a poster at the 2012 conference of the Canadian Association for Population Therapeutics (CAPT), with the abstract appearing later in the Journal of Population Therapeutics and Clinical Pharmacology. CAPT focuses on pharmacoepidemiology, so there we will emphasize this possibility that rising comorbidity burdens are driving drug costs in SARDs. We also intend to publish this work, with a focus on the attributable costs of SARDs, in a relevant, peer-reviewed rheumatology journal. Aside from these formal settings, our findings and their implications will be discussed with colleagues associated with the Arthritis Research Centre of Canada (ARC), and other institutions across the country.  4.7 Future Work Our findings here were very exciting, but only preliminary. With this vast quantity of health utilization data we look forward to investigating these findings more deeply and gaining a more comprehensive understanding of the health resource burden of SARDs.  4.7.1 Discrete Longitudinal Analysis Although our analysis was longitudinal, the entire cohort was not followed from a single point in time. Cases could enter at any time during the period, with no minimum follow-up period. This served to minimize survival bias but also restricted our ability to interpret the longitudinal trends. For instance the 32% decrease in overall mean per-PY costs could be from our longstanding (1996-entry) cases requiring fewer health services as their disease stabilizes. In an alternative scenario the healthcare needs of such cases do not decrease over time, but annual mean per-PY costs decreased on the whole because many of these costly cases died and thus ceased to incur these high costs. Stratifying our cohort by year of entry would let us compare the consumption patterns of cases followed for different periods, including the types and intensities of inpatient hospitalizations. We performed a trial analysis with only those SLE cases followed from 1996, and the decreases in annual mean per-PY outpatient and overall costs were half those observed for the whole SLE cohort. At the same time their mean per-PY prescription increases were more than double. With this further indication of disease progression and complications necessitating greater prescription use, a complete stratified analysis of the whole cohort would help us understand the role of comorbidities in the long-term healthcare costs of SARDs.  177  4.7.2 Updated Health Utilization Data Our dataset was large in terms of the size of the cohort (over 18,000 cases), number of encounters captured (over six million outpatient claims), and study period (twelve years). However this will be surpassed when PopDataBC updates our data through December 31, 2011. Our request for this has already been approved. Not only will our sample size increase but we will be able to see if the same cost and utilization trends continued. The contribution made by some expensive drugs – including the biologics, mycophenolate, and atorvastatin – was growing in our final years of study, and we are interested as to how much they are impacting drug costs now. What was most intriguing though, was in 2007 – the final year of study - mean per-PY outpatient and prescription costs were nearly the same, especially after differing by over $1000 per-PY, on-average, in 1996. With prescription costs increasing so much over twelve years and actually exceeding the outpatient (by an average of $29 per-PY) in the last year, we anticipate the annual per-PY prescription costs of the average SARDs case to now exceed the outpatient. This could have huge implications on health policy, further justifying a re-examination of prescription drug subsidies.  The precision of our hospital cost estimates will also be enhanced. Not only will the RIWs for all ALC and day separations be included, but we will have at least ten years with which to calculate costs using most current, ICD-10derived, RIW values.  This will provide insight into how this missing hospital data impacted our current estimates,  particularly the extent to which our day surgery costs were underestimated.  4.7.3 Control Data One of our objectives was to quantify the net health resource and economic burden that SARDs impart on the provincial healthcare system. We did so here by tabulating separately the costs for healthcare services deemed attributable to SARDs, but this approach may have lacked precision. Instead several studies on individual SARDs (299,304,311) have used the costs incurred by a matched control group to calculate this burden. Unforeseen delays in obtaining PharmaNet data for a control group prevented us from doing so here, but our request has now been approved. By comparing the costs of SARDs cases to those of an age- and sex-matched random sample of BC residents who do not have a SARD diagnosis, we will soon produce an even more reliable estimate of the incremental costs of SARDs.  178  4.7.4 Impact of Complications and Comorbidities These preliminary findings suggest comorbidity burdens and complication rates may be growing amongst SARDs cases, which may explain the growing drug costs. This possibility needs to be investigated in more detail, and with these conditions arising after many years, our updated dataset will make us even better equipped to do so. We would identify these secondary conditions through the appearance of their corresponding diagnostic codes on the outpatient claims and hospital discharges of SARDs cases. With our patient-level data, we could then access the prescription records of these cases to see how many prescriptions they were dispensed, and which ones, and evaluate any association between these conditions and per-PY prescriptions quantities and costs. Stratifying cases by year of cohort entry would be involved here too, since we would expect the incidence of these comorbidities to be greater in the more longstanding cases. Finally our analysis was motivated, in-part, by the expense of some emerging SARDs therapies. To assess their impact on long-term complication risk and thus their cost-effectiveness, we may also perform a pharmacoepidemiologic analysis where the complication rates of cases exposed to these new drugs would be compared to those not exposed.  4.7.5 Significant Cost Predictors Identifying any clinical or demographic predictors of costs in SARDs - which may include female sex, age, and disease severity - could also help reduce complications and healthcare costs. With our updated dataset, we plan to do this using multivariate linear regression. If high-risk patients could be better-identified and treated aggressively, long-term outcomes may be improved.  4.8 Conclusion This population-based study of the collective healthcare burden of SARDs over twelve years was the first of its kind in the world. We have shown the direct medical costs of SARDs cases at the population-level are substantial (totalling $571,216,780 2007 CDN over twelve years) - with mean per-PY costs averaging $6,087/PY CDN in 2007. With the net healthcare costs of SARDs averaging $2,674/PY in 2007 (and accounting for 44% of total healthcare costs for these cases), and total net costs exceeding $38 million in BC that year, SARDs themselves are clearly responsible for much of this burden. 179  The long-term reductions in health care costs we observed are encouraging and suggestive of more efficient health service delivery. But in contrast to the overall cost decreases, per-PY prescription quantities and costs are increasing greatly, with many patients personally financing a growing proportion of these costs. The use of immunosuppressants and bisphosphonates, and some specific ACE inhibitors, PPIs, thyroid therapies, and statins accounted for much of these increases. In addition, the contribution from some very expensive vasodilators and biologic therapies toward these drug costs is small but growing. The proportion of overall costs from prescriptions has more than doubled over twelve years and came to exceed the outpatient proportion in 2007. If this trend continues for SARDs cases, medication costs will become more important than the costs of outpatient care.  With per-PY prescription numbers increasing nearly as much as costs, comorbidity burdens and complication rates may be rising amongst longstanding cases, and this warrants further investigation. The connection between SARDs and these secondary conditions (such as cardiovascular disease) is currently underappreciated, but must be communicated, since preventive therapies and other measures could slow their development and reduce healthcare costs. Some expensive ($20,000 per-patient annually), but potentially-better, SARDs therapies are emerging, and further research to assess their impact on long-term comorbidity risk, and not just short-term outcomes, will be needed.  180  4.9 Tables Table 4.1 Changes in the Incidence and Prevalence Rates for SARDs in BC, 1996-2007 (1) SARDs-CTD SARDs-VD Incidence Prevalence Incidence Prevalence (per-100,000) (per-100,000) (per-100,000) (per-100,000) 1996 2007 12-year increase  16.25 96.62 172%  44.26 388.55 302%  2.37 9.06 125%  5.34 31.91 252%  181  References 1. Avina-Zubieta JA Sayre EC Bernatsky S Lehman AJ Shojana K Esdaile JM Lacaille D. Adult Prevalence of Systemic Autoimmune Rheumatic Diseases (SARDs) in British Columbia, Canada. 2011 American College of Rheumatology ACR/AHRP Annual Scientific Meeting, Chicago, IL, USA, November 5-9, 2011 2. Gota CE. Vasculitis [Online]. 2008 May [cited 12/18/2011]; Available from: URL: http://www.merckmanuals.com/professional/musculoskeletal_and_connective_tissue_disorders/vasculitis/overview_ of_vasculitis.html?qt=systemic vasculitis&alt=sh. 3. Vasculitis – Johns Hopkins Vasculitis Center [Online]. 2011 [cited 3/20/2012]; Available from: URL: http://www.hopkinsvasculitis.org/. 4. Bartels CM Krause RS Lakdawala VS Leber MJ Muller D. Systemic Lupus Erythematosus (SLE) [Online]. 2011 November 15 [cited 12/17/2011]; Available from: URL: http://emedicine.medscape.com/article/332244overview. 5. Bernatsky S Lix L Hanly J Hudson M Badley E Peschken C, et al. Surveillance of systemic autoimmune rheumatic diseases using administrative data. Rheumatol Int 2011;31(4):549-554. 6. Gonzalez-Gay MA Garcia-Porrua C Guerrero J Rodriguez-Ledo P Llorca J. The epidemiology of the primary systemic vasculitides in northwest Spain: implications of the Chapel Hill Consensus Conference definitions Arthritis Rheum 2003; 49(3):388-393. 7. Watts RA Lane SE Bentham G Scott DG. Epidemiology of systemic vasculitis: a ten-year study in the United Kingdom Arthritis Rheum 2000; 43(2):414-419. 8. Consensus Conference on SARD [Online]. 2007 December 20 [cited 12/7/2011]; Available from: URL: http://www.arthritis.ca/look at research/sard/default.asp?s=1&province=bc. 9. Barnabe C Joseph L Belisle P Labrecque J Edworthy S Barr SG, et al. Prevalence of systemiclupus erythematosus and systemic sclerosis in the First Nations population of Alberta, Canada. Arthritis Care and Research 2012; 64(1):138-143. 10. Peschken CA Esdaile JM. Systemic lupus erythematosus in North American Indians: a population based study. J Rheumatol 2000; 27(8):1884-1891. 11. Canadian Organization for Rare Disorders [Online]. 2012 [cited 1/19/2012]; Available from: URL: http://www.raredisorders.ca/. 12. Hazlewood G Bykerk VP. e-Therapeutics+ : Therapeutics : Musculoskeletal Disorders: Rheumatoid Arthritis [Online]. 2011 May [cited 12/11/2011]; Available from: URL: http://www.etherapeutics.ca/tc.showChapter.action?documentId=c0060. 13. Kopec JA Rahman MM Berthelot JM Le Petit C Aghajanian J Sayre EC, et al. Descriptive epidemiology of osteoarthritis in British Columbia, Canada J Rheumatol 2007; 34(2):386-393. 14. Kopec JA Sayre EC Flanagan WM Fines P Cibere J Rahman MM, et al. Development of a population-based microsimulation model of osteoarthritis in Canada Osteoarthritis Cartilage 2010; 18(3):303-311.  182  15. Gelber A, Levine S, Rosen A. Chapter 24. Inflammatory Rheumatic Diseases. In: McPhee S, Hammer G, editors. Pathophysiology of Disease. 6th ed. New York: McGraw-Hill; 2010. 16. Delves PJ. Acquired Immunity: Biology of the Immune System: Merck Manual Home Edition [Online]. 2008 September [cited 12/18/2011]; Available from: URL: http://www.merckmanuals.com/home/immune_disorders/biology_of_the_immune_system/acquired_immunity.html. 17. DiPiro JT, Talbert RL, Yee GC. Pharmacotherapy : A Pathophysiologic Approach. New York, NY, USA: McGraw-Hill Professional Publishing; 2008. 18. Rust RS. Churg-Strauss Disease [Online]. 2010 January 20 [cited 12/16/2011]; Available from: URL: http://emedicine.medscape.com/article/1178795-overview. 19. Papadopoulos PJ. Wegener Granulomatosis [Online]. 2011 December 2 [cited 12/16/2011]; Available from: URL: http://emedicine.medscape.com/article/332622-overview. 20. Miller AV Ranatunga SKM Francis ML Pema K. Sjogren Syndrome [Online]. 2011 October 31 [cited 12/15/2011]; Available from: URL: http://emedicine.medscape.com/article/332125-overview#showall. 21. Schwartz RA Dziankowska-Bartkowiak B Zalewska A Sysa-Jedrzejowska A. Systemic Sclerosis [Online]. 2011 April 25 [cited 12/15]; Available from: URL: http://emedicine.medscape.com/article/1066280overview#showall. 22. Seetharaman M Paget SA Leibowitz E. Giant Cell Arteritis Medication [Online]. 2011 October 12 [cited 12/17/2011]; Available from: URL: http://emedicine.medscape.com/article/332483-medication#showall. 23. Pappu R Seetharaman M. Polymyositis [Online]. 2011 September 30 [cited 12/15/2011]; Available from: URL: http://emedicine.medscape.com/article/335925-overview. 24. Callen JP. Dermatomyositis [Online]. 2011 October 12 [cited 12/15/2011]; Available from: URL: http://emedicine.medscape.com/article/332783-overview. 25. Jacobs-Kosmin D Jackson JM. Polyarteritis Nodosa [Online]. 2011 September 28 [cited 12/16/2011]; Available from: URL: http://emedicine.medscape.com/article/330717-overview. 26. Clarke AE Panopalis P. Systemic lupus erythematosus: clinical manifestations, treatment and economics. Expert Review of Pharmacoeconomics & Outcomes Research 2011; 6:563. 27. Brent L Karhadkar A Bloom E. Lupus Nephritis [Online]. 2011 November 29 [cited 2/26/2012]; Available from: URL: http://emedicine.medscape.com/article/330369-overview. 28. Lupus Information: Symptoms, Diagnosis, Treatment [Online]. 2012 [cited 4/3/2012]; Available from: URL: http://www.hopkinslupus.org/. 29. Roberts JR Rossman MG Ahmed MM Wolf RE. Takayasu Arteritis [Online]. 2011 October 12 [cited 12/16/2011]; Available from: URL: http://emedicine.medscape.com/article/332378-overview. 30. Hanly JG. e-Therapeutics+ : Therapeutics : Musculoskeletal Disorders: Polymyalgia Rheumatica and Giant-Cell Arteritis [Online]. 2011 May [cited 12/11/2011]; Available from: URL: http://www.etherapeutics.ca/tc.showChapter.action?documentId=c0057. 31. Albertini JG Marks VJ Cronin H. Temporal (Giant Cell) Arteritis [Online]. 2009 May 29 [cited 12/17/2011]; Available from: URL: http://emedicine.medscape.com/article/1084911-overview.  183  32. Smith CD. e-Therapeutics+ : Therapeutics : Musculoskeletal Disorders: Systemic Lupus Erythematosus [Online]. 2011 September [cited 12/11/2011]; Available from: URL: http://www.etherapeutics.ca/tc.showChapter.action?documentId=c0128#c0128n01061. 33. Chan TM Li FK Tang CS Wong RW Fang GX Ji YL, et al. Efficacy of mycophenolate mofetil in patients with diffuse proliferative lupus nephritis. Hong Kong-Guangzhou Nephrology Study Group N Engl J Med 2000; 343(16):1156-1162. 34. Ong LM Hooi LS Lim TO Goh BL Ahmad G Ghazalli R, et al. Randomized controlled trial of pulse intravenous cyclophosphamide versus mycophenolate mofetil in the induction therapy of proliferative lupus nephritis Nephrology (Carlton) 2005; 10(5):504-510. 35. El-Shafey EM, Abdou SH, Shareef MM. Is mycophenolate mofetil superior to pulse intravenous cyclophosphamide for induction therapy of proliferative lupus nephritis in Egyptian patients? Clin Exp Nephrol ; 2011(12/13/2011). 36. Wang J Hu W Xie H Zhang H Chen H Zeng C, et al. Induction therapies for class IV lupus nephritis with non-inflammatory necrotizing vasculopathy: mycophenolate mofetil or intravenous cyclophosphamide Lupus 2007; 16(9):707-712. 37. Ginzler EM Wofsy D Isenberg D Gordon C Lisk L Dooley MA, et al. Nonrenal disease activity following mycophenolate mofetil or intravenous cyclophosphamide as induction treatment for lupus nephritis: findings in a multicenter, prospective, randomized, open-label, parallel-group clinical trial Arthritis Rheum 2010; 62(1):211-221. 38. Dooley M Jayne D Ginzler E Isenberg D Olsen N Wofsy D, et al. Mycophenolate versus azathioprine as maintenance therapy for lupus nephritis N Eng J Med 2011; 2012(20):1886-95. 39. Hu W Liu Z Chen H Tang Z Wang Q Shen K, et al. Mycophenolate mofetil vs cyclophosphamide therapy for patients with diffuse proliferative lupus nephritis Chin Med J (Engl) 2002; 115(5):705-709. 40. Chan TM Tse KC Tang CS Mok MY Li FK Hong Kong Nephrology Study Group. Long-term study of mycophenolate mofetil as continuous induction and maintenance treatment for diffuse proliferative lupus nephritis J Am Soc Nephrol 2005; 16(4):1076-1084. 41. Contreras G Pardo V Leclercq B Lenz O Tozman E O'Nan P, et al. Sequential therapies for proliferative lupus nephritis N Engl J Med 2004; 350(10):971-980. 42. Fujinaga S Ohtomo Y Hara S Umino D Someya T Shimizu T, et al. Maintenance therapy with mycophenolate mofetil for children with severe lupus nephritis after low-dose intravenous cyclophosphamide regimen Pediatr Nephrol 2008; 23(10):1877-1882. 43. Cross J Dwomoa A Andrews P Burns A Gordon C Main J, et al. Mycophenolate mofetil for remission induction in severe lupus nephritis Nephron Clin Pract 2005; 100(3):c92-100. 44. Laskari K Mavragani CP Tzioufas AG Moutsopoulos HM. Mycophenolate mofetil as maintenance therapy for proliferative lupus nephritis: a long-term observational prospective study Arthritis Res Ther 2010; 12(6):R208. 45. Dooley MA Cosio FG Nachman PH Falkenhain ME Hogan SL Falk RJ, et al. Mycophenolate mofetil therapy in lupus nephritis: clinical observations J Am Soc Nephrol 1999; 10(4):833-839.  184  46. Sahin GM Sahin S Kiziltas S Masatlioglu S Oguz F Ergin H. Mycophenolate mofetil versus azathioprine in the maintenance therapy of lupus nephritis Ren Fail 2008; 30(9):865-869. 47. Rabrenović V, Poskurica M, Kovacević Z, Nesić V, Savin M, Mitić B, Dimković N, Cucković C, Vujić D, Pljesa S, Perunicić-Peković G, Curić S, Mitić I, Ratković M, Marinković J, Jovanović D. Treatment of lupus nephritis by mycophenolate mofetil. Kidney Blood Press Res ; 2011(12/13/2011). 48. Kingdon EJ McLean AG Psimenou E Davenport A Powis SH Sweny P, et al. The safety and efficacy of MMF in lupus nephritis: a pilot study Lupus 2001; 10(9):606-611. 49. Kazderova M, Jancova E, Rysava R, Merta M, Tesar V. Mycophenolate mofetil in low doses stabilizes and improves antineutrophil cytoplasmic antibody-associated vasculitis and lupus nephritis. Arch Med Res ; 2011(12/13/2011). 50. Weng MY, Weng CT, Liu MF. The efficacy of low-dose mycophenolate mofetil for treatment of lupus nephritis in Taiwanese patients with systemic lupus erythematosus. Clinical rheumatology 2010; 29(7):771-5. 51. Pisoni CN Sanchez FJ Karim Y Cuadrado MJ D'Cruz DP Abbs IC, et al. Mycophenolate mofetil in systemic lupus erythematosus: efficacy and tolerability in 86 patients J Rheumatol 2005; 32(6):1047-1052. 52. Kapitsinou PP Boletis JN Skopouli FN Boki KA Moutsopoulos HM. Lupus nephritis: treatment with mycophenolate mofetil Rheumatology (Oxford) 2004; 43(3):377-380. 53. Ginzler EM Dooley MA Aranow C Kim MY Buyon J Merrill JT, et al. Mycophenolate mofetil or intravenous cyclophosphamide for lupus nephritis N Engl J Med 2005; 353(21):2219-2228. 54. Ortega LM Schultz DR Lenz O Pardo V Contreras GN. Review: Lupus nephritis: pathologic features, epidemiology and a guide to therapeutic decisions Lupus 2010; 19(5):557-574. 55. Riskalla MM Somers EC Fatica RA McCune WJ. Tolerability of mycophenolate mofetil in patients with systemic lupus erythematosus J Rheumatol 2003; 30(7):1508-1512. 56. Dall'Era M. Mycophenolate mofetil in the treatment of systemic lupus erythematosus. Current Opinion in Rheumatology 2011; 23(5):454-8. 57. Scleroderma Treatment Options | Johns Hopkins Scleroderma Center [Online]. 2012 [cited 1/17/2012]; Available from: URL: http://www.hopkinsscleroderma.org/patients/scleroderma-treatment-options/. 58. Bernatsky S Boivin JF Joseph L Manzi S Ginzler E Gladman DD, et al. Mortality in systemic lupus erythematosus Arthritis Rheum 2006; 54(8):2550-2557. 59. Urowitz MB Gladman DD Abu-Shakra M Farewell VT. Mortality studies in systemic lupus erythematosus. Results from a single center. III. Improved survival over 24 years J Rheumatol 1997; 24(6):1061-1065. 60. Mak A Mok CC. Mycophenolate mofetil for refractory haemolytic anemia in systemic lupus erythematosus Lupus 2005; 14(10):856-858. 61. Pisoni CN Karim Y Cuadrado MJ. Mycophenolate mofetil and systemic lupus erythematosus: an overview Lupus 2005; 14 Suppl 1:s9-11. 62. Jones R Walsh M Smith K. What is the value of mycophenolate mofetil as induction and maintenance therapy in lupus nephritis? Curr Opin Rheumatol ; 2012(3/5/2012):61.  185  63. e-Therapeutics+ : Therapeutics: Mycophenolate [Online]. 2010 February 23 [cited 1/19/2012]; Available from: URL: http://www.e-therapeutics.ca/cps.showPopupMonograph.action?monographId=m114200. 64. Alba P Karim MY Hunt BJ. Mycophenolate mofetil as a treatment for autoimmune haemolytic anaemia in patients with systemic lupus erythematosus and antiphospholipid syndrome Lupus 2003; 12(8):633-635. 65. Chang HK. Successful treatment of refractory thrombocytopenia with mycophenolate mofetil in a patient with systemic lupus erythematosus J Korean Med Sci 2005; 20(5):883-885. 66. Vasoo S Thumboo J Fong KY. Refractory immune thrombocytopenia in systemic lupus erythematosus: response to mycophenolate mofetil Lupus 2003; 12(8):630-632. 67. Moradinejad M. Treatment of intractable pulmonary hemorrhage in two patients with childhood systemic lupus erythematosus. Rheumatol Int 2009; 29(9):1113-5. 68. Dittrich K Ross S Benz K Amann K Dotsch J. Experience with mycophenolate mofetil as maintenance therapy in five pediatric patients with severe systemic lupus erythematosus Klin Padiatr 2009; 221(7):425-429. 69. Willeke P Schluter B Becker H Schotte H Domschke W Gaubitz M. Mycophenolate sodium treatment in patients with primary Sjogren syndrome: a pilot trial Arthritis Res Ther 2007; 9(6):R115. 70. Edge JC Outland JD Dempsey JR Callen JP. Mycophenolate mofetil as an effective corticosteroid-sparing therapy for recalcitrant dermatomyositis Arch Dermatol 2006; 142(1):65-69. 71. Rouster-Stevens KA Morgan GA Wang D Pachman LM. Mycophenolate mofetil: a possible therapeutic agent for children with juvenile dermatomyositis Arthritis Care Res (Hoboken) 2010; 62(10):1446-1451. 72. Morganroth PA Kreider ME Werth VP. Mycophenolate mofetil for interstitial lung disease in dermatomyositis Arthritis Care Res (Hoboken) 2010; 62(10):1496-1501. 73. Gelber AC Nousari HC Wigley FM. Mycophenolate mofetil in the treatment of severe skin manifestations of dermatomyositis: a series of 4 cases J Rheumatol 2000; 27(6):1542-1545. 74. Stassen PM Tervaert JW Stegeman CA. Induction of remission in active anti-neutrophil cytoplasmic antibody-associated vasculitis with mycophenolate mofetil in patients who cannot be treated with cyclophosphamide Ann Rheum Dis 2007; 66(6):798-802. 75. Hu W Liu C Xie H Chen H Liu Z Li L. Mycophenolate mofetil versus cyclophosphamide for inducing remission of ANCA vasculitis with moderate renal involvement Nephrol Dial Transplant 2008; 23(4):1307-1312. 76. Iatrou C, Zerbala S, Revela I, Spanou E, Marinaki S, Nakopoulou L, Boletis J. Mycophenolate mofetil as maintenance therapy in patients with vasculitis and renal involvement. Clin Nephrol ; 2011(12/13/2011). 77. Langford CA Talar-Williams C Sneller MC. Mycophenolate mofetil for remission maintenance in the treatment of Wegener's granulomatosis Arthritis Rheum 2004; 51(2):278-283. 78. Osuna A GJ. Cyclophosphamide-intolerant Wegener's granulomatosis successfully treated with mycophenolate mofetil. Acta Reumatol Port ; 2011(12/13/2011). 79. Joy MS Hogan SL Jennette JC Falk RJ Nachman PH. A pilot study using mycophenolate mofetil in relapsing or resistant ANCA small vessel vasculitis Nephrol Dial Transplant 2005; 20(12):2725-2732.  186  80. Nowack R Gobel U Klooker P Hergesell O Andrassy K van der Woude FJ. Mycophenolate mofetil for maintenance therapy of Wegener's granulomatosis and microscopic polyangiitis: a pilot study in 11 patients with renal involvement J Am Soc Nephrol 1999; 10(9):1965-1971. 81. Silva F Specks U Kalra S Hogan MC Leung N Sethi S, et al. Mycophenolate mofetil for induction and maintenance of remission in microscopic polyangiitis with mild to moderate renal involvement--a prospective, openlabel pilot trial Clin J Am Soc Nephrol 2010; 5(3):445-453. 82. Han F, Liu G, Zhang X, Li X, He Q, He X, Li Q, Wang S, Wang H, Chen J. Effects of mycophenolate mofetil combined with corticosteroids for induction therapy of microscopic polyangiitis. Am J Nephrol ; 2011(12/13/2011). 83. Lin S, Shan L, Mingcai Q. A case of Wegener's granulomatosis treated with mycophenolate mofetil. Nephron ; 2011(12/13/2011). 84. Osuna A GJ. Cyclophosphamide-intolerant Wegener's granulomatosis successfully treated with mycophenolate mofetil. Acta Reumatol Port ; 2011(12/13/2011). 85. Waiser J Budde K Braasch E Neumayer HH. Treatment of acute c-ANCA-positive vasculitis with mycophenolate mofetil Am J Kidney Dis 1999; 34(3):e9. 86. Koukoulaki M Jayne DR. Mycophenolate mofetil in anti-neutrophil cytoplasm antibodies-associated systemic vasculitis Nephron Clin Pract 2006; 102(3-4):c100-7. 87. Yalcindag FN Amer R Forrester JV. Mycophenolate mofetil in the treatment of ocular inflammation in ANCA-associated vasculitis J Ocul Pharmacol Ther 2008; 24(2):249-254. 88. Swain MG. e-Therapeutics+ : Therapeutics : Gastrointestinal Disorders: Chronic Liver Diseases [Online]. 2011 May [cited 12/13/2011]; Available from: URL: http://www.etherapeutics.ca/tc.showChapter.action?documentId=c0043. 89. Health Canada Drug Product Database (DPD) [Online]. 2011 July 21 [cited 1/19/2012]; Available from: URL: http://www.hc-sc.gc.ca/dhp-mps/prodpharma/databasdon/index-eng.php. 90. AccessMedicine | RiTUXimab [Online]. [cited 12/13/2011]; Available from: URL: http://www.accessmedicine.com/drugContent.aspx?mid=6919&section=10. 91. Gottenberg JE Guillevin L Lambotte O Combe B Allanore Y Cantagrel A, et al. Tolerance and short term efficacy of rituximab in 43 patients with systemic autoimmune diseases Ann Rheum Dis 2005; 64(6):913-920. 92. Smith KG Jones RB Burns SM Jayne DR. Long-term comparison of rituximab treatment for refractory systemic lupus erythematosus and vasculitis: Remission, relapse, and re-treatment Arthritis Rheum 2006; 54(9):2970-2982. 93. Tony HP Burmester G Schulze-Koops H Grunke M Henes J Kotter I, et al. Safety and clinical outcomes of rituximab therapy in patients with different autoimmune diseases: experience from a national registry (GRAID) Arthritis Res Ther 2011; 13(3):R75. 94. Keogh KA Ytterberg SR Fervenza FC Carlson KA Schroeder DR Specks U. Rituximab for refractory Wegener's granulomatosis: report of a prospective, open-label pilot trial Am J Respir Crit Care Med 2006; 173(2):180-187.  187  95. Martinez Del Pero M Chaudhry A Jones RB Sivasothy P Jani P Jayne D. B-cell depletion with rituximab for refractory head and neck Wegener's granulomatosis: a cohort study Clin Otolaryngol 2009; 34(4):328-335. 96. Eriksson P. Nine patients with anti-neutrophil cytoplasmic antibody-positive vasculitis successfully treated with rituximab J Intern Med 2005; 257(6):540-548. 97. Khan A Lawson CA Quinn MA Isdale AH Green MJ. Successful Treatment of ANCA-Negative Wegener's Granulomatosis with Rituximab Int J Rheumatol 2010; 2010:846063. 98. Ooka S, Maeda A, Ito H, Omata M, Yamada H, Ozaki S. Treatment of refractory retrobulbar granuloma with rituximab in a patient with ANCA-negative Wegener's granulomatosis: a case report. ; 2011(12/12/2011):3. 99. Minami R Miyamura T Watanabe H Takahama S Yamamoto M Suematsu E. Successful treatment of a patient with refractory Wegener's granulomatosis by rituximab Nihon Rinsho Meneki Gakkai Kaishi 2007; 30(2):133-138. 100. Sharma A, Kumar S, Wanchu A, Lal V, Singh R, Gupta V, Singh S, Gupta A. Successful treatment of hypertrophic pachymeningitis in refractory Wegener's granulomatosis with rituximab. Clinical rheumatology ; 2011(12/12/2011):10. 101. Oristrell J Bejarano G Jordana R Monteagudo M Mari B Casanovas A, et al. Effectiveness of rituximab in severe Wegener's granulomatosis: report of two cases and review of the literature Open Respir Med J 2009; 3:94-99. 102. Kowalewska B, Szechiński J, Roszkowska E. Wegener's granulomatosis effectively treatreated with rituximab: a case study Pol Arch Wewn ; 2011(12/12/2011):5-5. 103. Brihaye B Aouba A Pagnoux C Cohen P Lacassin F Guillevin L. Adjunction of rituximab to steroids and immunosuppressants for refractory/relapsing Wegener's granulomatosis: a study on 8 patients Clin Exp Rheumatol 2007; 25(1 Suppl 44):S23-7. 104. Seo P Specks U Keogh KA. Efficacy of rituximab in limited Wegener's granulomatosis with refractory granulomatous manifestations J Rheumatol 2008; 35(10):2017-2023. 105. Freidlin J Wong IG Acharya N. Rituximab treatment for peripheral ulcerative keratitis associated with Wegener's granulomatosis Br J Ophthalmol 2007; 91(10):1414. 106. Cheung CM Murray PI Savage CO. Successful treatment of Wegener's granulomatosis associated scleritis with rituximab Br J Ophthalmol 2005; 89(11):1542. 107. Huerva V Sanchez MC Traveset A Jurjo C Ruiz A. Rituximab for peripheral ulcerative keratitis with wegener granulomatosis Cornea 2010; 29(6):708-710. 108. Onal S, Kazokoglu H, Koc A, Yavuz S. Rituximab for remission induction in apatient with relapsing necrotizing scleritis associated with limited Wegener's granulomatosis Ocul Immunol Inflamm ; 2011(12/12/2011):2. 109. Stone JH Merkel PA Spiera R Seo P Langford CA Hoffman GS, et al. Rituximab versus cyclophosphamide for ANCA-associated vasculitis N Engl J Med 2010; 363(3):221-232. 110. Specks U Fervenza FC McDonald TJ Hogan MC. Response of Wegener's granulomatosis to anti-CD20 chimeric monoclonal antibody therapy Arthritis Rheum 2001; 44(12):2836-2840.  188  111. Omdal R Wildhagen K Hansen T Gunnarsson R Kristoffersen G. Anti-CD20 therapy of treatment-resistant Wegener's granulomatosis: favourable but temporary response Scand J Rheumatol 2005; 34(3):229-232. 112. Stasi R Stipa E Del Poeta G Amadori S Newland AC Provan D. Long-term observation of patients with anti-neutrophil cytoplasmic antibody-associated vasculitis treated with rituximab Rheumatology (Oxford) 2006; 45(11):1432-1436. 113. Tamura N Matsudaira R Hirashima M Ikeda M Tajima M Nawata M, et al. Two cases of refractory Wegener's granulomatosis successfully treated with rituximab Intern Med 2007; 46(7):409-414. 114. Tzaribachev N Koetter I Kuemmerle-Deschner JB Schedel J. Rituximab for the treatment of refractory pediatric autoimmune diseases: a case series Cases J 2009; 2:6609. 115. Garcia Hernandez FJ Ocana Medina C Gonzalez Leon R Garrido Rasco R Colorado Bonilla R Castillo Palma MJ, et al. Rituximab for treatment of patients with systemic autoimmune diseases Med Clin (Barc) 2007; 128(12):458-462. 116. Pagnoux C Pagnoux C. Vasculitis [Online]. 2012 January 31 [cited 1/19/2012]; Available from: URL: http://canvasc.com/vasculitis.htm. 117. Scheinberg M Hamerschlak N Kutner JM Ribeiro AA Ferreira E Goldenberg J, et al. Rituximab in refractory autoimmune diseases: Brazilian experience with 29 patients (2002-2004) Clin Exp Rheumatol 2006; 24(1):65-69. 118. Vigna-Perez M Hernandez-Castro B Paredes-Saharopulos O Portales-Perez D Baranda L Abud-Mendoza C, et al. Clinical and immunological effects of Rituximab in patients with lupus nephritis refractory to conventional therapy: a pilot study Arthritis Res Ther 2006; 8(3):R83. 119. Gunnarsson I Sundelin B Jonsdottir T Jacobson SH Henriksson EW van Vollenhoven RF. Histopathologic and clinical outcome of rituximab treatment in patients with cyclophosphamide-resistant proliferative lupus nephritis Arthritis Rheum 2007; 56(4):1263-1272. 120. van Vollenhoven RF Gunnarsson I Welin-Henriksson E Sundelin B Osterborg A Jacobson SH, et al. Biopsy-verified response of severe lupus nephritis to treatment with rituximab (anti-CD20 monoclonal antibody) plus cyclophosphamide after biopsy-documented failure to respond to cyclophosphamide alone Scand J Rheumatol 2004; 33(6):423-427. 121. Garcia-Carrasco M Mendoza-Pinto C Sandoval-Cruz M Soto-Vega E Beltran-Castillo A JimenezHernandez M, et al. Anti-CD20 therapy in patients with refractory systemic lupus erythematosus: a longitudinal analysis of 52 Hispanic patients Lupus 2010; 19(2):213-219. 122. Nwobi O Abitbol CL Chandar J Seeherunvong W Zilleruelo G. Rituximab therapy for juvenile-onset systemic lupus erythematosus Pediatr Nephrol 2008; 23(3):413-419. 123. Lindholm C Borjesson-Asp K Zendjanchi K Sundqvist AC Tarkowski A Bokarewa M. Longterm clinical and immunological effects of anti-CD20 treatment in patients with refractory systemic lupus erythematosus J Rheumatol 2008; 35(5):826-833. 124. Sousa E ID. Treating lupus: from serendipity to sense, the rise of the new biologicals and other emerging therapies. Best Pract Res Clin Rheumatol ; 2011(12/18/2011).  189  125. Coca A Sanz I. B cell depletion in lupus and Sjogren's syndrome: an update Curr Opin Rheumatol 2009; 21(5):483-488. 126. Chehab G Sander O Fischer-Betz R Schneider M. Anti-CD20 therapy for inducing and maintaining remission in refractory systemic lupus erythematosus Z Rheumatol 2007; 66(4):328, 330-6. 127. Lu TY Ng KP Cambridge G Leandro MJ Edwards JC Ehrenstein M, et al. A retrospective seven-year analysis of the use of B cell depletion therapy in systemic lupus erythematosus at University College London Hospital: the first fifty patients Arthritis Rheum 2009; 61(4):482-487. 128. Willems M Haddad E Niaudet P Kone-Paut I Bensman A Cochat P, et al. Rituximab therapy for childhoodonset systemic lupus erythematosus J Pediatr 2006; 148(5):623-627. 129. Kumar S Benseler SM Kirby-Allen M Silverman ED. B-cell depletion for autoimmune thrombocytopenia and autoimmune hemolytic anemia in pediatric systemic lupus erythematosus Pediatrics 2009; 123(1):e159-63. 130. Tokunaga M Fujii K Saito K Nakayamada S Tsujimura S Nawata M, et al. Down-regulation of CD40 and CD80 on B cells in patients with life-threatening systemic lupus erythematosus after successful treatment with rituximab Rheumatology (Oxford) 2005; 44(2):176-182. 131. Sfikakis PP Boletis JN Lionaki S Vigklis V Fragiadaki KG Iniotaki A, et al. Remission of proliferative lupus nephritis following B cell depletion therapy is preceded by down-regulation of the T cell costimulatory molecule CD40 ligand: an open-label trial Arthritis Rheum 2005; 52(2):501-513. 132. Boletis JN Marinaki S Skalioti C Lionaki SS Iniotaki A Sfikakis PP. Rituximab and mycophenolate mofetil for relapsing proliferative lupus nephritis: a long-term prospective study Nephrol Dial Transplant 2009; 24(7):2157-2160. 133. Camous L Melander C Vallet M Squalli T Knebelmann B Noel LH, et al. Complete remission of lupus nephritis with rituximab and steroids for induction and rituximab alone for maintenance therapy Am J Kidney Dis 2008; 52(2):346-352. 134. Pepper R Griffith M Kirwan C Levy J Taube D Pusey C, et al. Rituximab is an effective treatment for lupus nephritis and allows a reduction in maintenance steroids Nephrol Dial Transplant 2009; 24(12):3717-3723. 135. Carroll RP Brown F Kerr PG. Anti-CD20 antibody treatment in refractory Class IV lupus nephritis Nephrol Dial Transplant 2007; 22(1):291-293. 136. Galarza C, Valencia D, Tobón GJ, Zurita L, Mantilla RD, Pineda-Tamayo R, Rojas-Villarraga A, Rueda JC, Anaya JM. Should rituximab be considered as the first-choice treatment for severe autoimmune rheumatic diseases? Clin Rev Allergy Immunol ; 2011(12/12/2011):8. 137. Van den Bergh B, Selleslag D, Boelaert JR, Matthys EG, Schurgers M, Vandecasteele S, De Vriese A. Management of therapy-resistant systemic lupus erythematosus with rituximab: report of a case and review of the literature. ; 2011(12/12/2011):5. 138. Menon S, Hari P, Bagga A. Beneficial effects of rituximab therapy for systemic lupus erythematosus. ; 2011(12/12/2011):82.  190  139. Leandro MJ Cambridge G Edwards JC Ehrenstein MR Isenberg DA. B-cell depletion in the treatment of patients with systemic lupus erythematosus: a longitudinal analysis of 24 patients Rheumatology (Oxford) 2005; 44(12):1542-1545. 140. Cambridge G Isenberg DA Edwards JC Leandro MJ Migone TS Teodorescu M, et al. B cell depletion therapy in systemic lupus erythematosus: relationships among serum B lymphocyte stimulator levels, autoantibody profile and clinical response Ann Rheum Dis 2008; 67(7):1011-1016. 141. Ng KP Leandro MJ Edwards JC Ehrenstein MR Cambridge G Isenberg DA. Repeated B cell depletion in treatment of refractory systemic lupus erythematosus Ann Rheum Dis 2006; 65(7):942-945. 142. Marks SD Patey S Brogan PA Hasson N Pilkington C Woo P, et al. B lymphocyte depletion therapy in children with refractory systemic lupus erythematosus Arthritis Rheum 2005; 52(10):3168-3174. 143. Leandro MJ Edwards JC Cambridge G Ehrenstein MR Isenberg DA. An open study of B lymphocyte depletion in systemic lupus erythematosus Arthritis Rheum 2002; 46(10):2673-2677. 144. Podolskaya A Stadermann M Pilkington C Marks SD Tullus K. B cell depletion therapy for 19 patients with refractory systemic lupus erythematosus Arch Dis Child 2008; 93(5):401-406. 145. Kneitz C Wilhelm M Tony HP. Effective B cell depletion with rituximab in the treatment of autoimmune diseases Immunobiology 2002; 206(5):519-527. 146. Tokunaga M Saito K Kawabata D Imura Y Fujii T Nakayamada S, et al. Efficacy of rituximab (anti-CD20) for refractory systemic lupus erythematosus involving the central nervous system Ann Rheum Dis 2007; 66(4):470475. 147. Perrotta S Locatelli F La Manna A Cennamo L De Stefano P Nobili B. Anti-CD20 monoclonal antibody (Rituximab) for life-threatening autoimmune haemolytic anaemia in a patient with systemic lupus erythematosus Br J Haematol 2002; 116(2):465-467. 148. Kotani T Takeuchi T Kawasaki Y Hirano S Tabushi Y Kagitani M, et al. Successful treatment of cold agglutinin disease with anti-CD20 antibody (rituximab) in a patient with systemic lupus erythematosus Lupus 2006; 15(10):683-685. 149. Sabugo F, Llanos C, Soto L, Gutiérrez J, Cuchacovich M.ui. [Rituximab (anti-CD20 monoclonal antibody) for refractory systemic lupus erythematosus: report of one case]. Rev Med Chil ; 2011(12/12/2011):4. 150. Abdwani R Mani R. Anti-CD20 monoclonal antibody in acute life threatening haemolytic anaemia complicating childhood onset SLE Lupus 2009; 18(5):460-464. 151. Albert D Dunham J Khan S Stansberry J Kolasinski S Tsai D, et al. Variability in the biological response to anti-CD20 B cell depletion in systemic lupus erythaematosus Ann Rheum Dis 2008; 67(12):1724-1731. 152. Looney RJ Anolik JH Campbell D Felgar RE Young F Arend LJ, et al. B cell depletion as a novel treatment for systemic lupus erythematosus: a phase I/II dose-escalation trial of rituximab Arthritis Rheum 2004; 50(8):25802589. 153. Terrier B Amoura Z Ravaud P Hachulla E Jouenne R Combe B, et al. Safety and efficacy of rituximab in systemic lupus erythematosus: results from 136 patients from the French AutoImmunity and Rituximab registry Arthritis Rheum 2010; 62(8):2458-2466.  191  154. Tanaka Y Yamamoto K Takeuchi T Nishimoto N Miyasaka N Sumida T, et al. A multicenter phase I/II trial of rituximab for refractory systemic lupus erythematosus Mod Rheumatol 2007; 17(3):191-197. 155. Melander C Sallee M Trolliet P Candon S Belenfant X Daugas E, et al. Rituximab in severe lupus nephritis: early B-cell depletion affects long-term renal outcome Clin J Am Soc Nephrol 2009; 4(3):579-587. 156. Li EK Tam LS Zhu TY Li M Kwok CL Li TK, et al. Is combination rituximab with cyclophosphamide better than rituximab alone in the treatment of lupus nephritis? Rheumatology (Oxford) 2009; 48(8):892-898. 157. Weide R Heymanns J Pandorf A Koppler H. Successful long-term treatment of systemic lupus erythematosus with rituximab maintenance therapy Lupus 2003; 12(10):779-782. 158. Rech J Hueber AJ Kallert S Requadt C Kalden JR Schulze-Koops H. Immunoadsorption and CD20 antibody treatment in a patient with treatment resistant systemic lupus erythematosus and preterminal renal insufficiency Ann Rheum Dis 2006; 65(4):552-553. 159. Roccatello D Sciascia S Rossi D Alpa M Naretto C Baldovino S, et al. Intensive short-term treatment with rituximab, cyclophosphamide and methylprednisolone pulses induces remission in severe cases of SLE with nephritis and avoids further immunosuppressive maintenance therapy Nephrol Dial Transplant 2011; 26(12):39873992. 160. Sabugo F, Llanos C, Soto L, Gutiérrez J, Cuchacovich M. [Rituximab (anti-CD20 monoclonal antibody) for refractory systemic lupus erythematosus: report of one case]. Rev Med Chil ; 2011(12/12/2011). 161. ten Cate R Smiers FJ Bredius RG Lankester AC van Suijlekom-Smit LW Huizinga TW, et al. Anti-CD20 monoclonal antibody (rituximab) for refractory autoimmune thrombocytopenia in a girl with systemic lupus erythematosus Rheumatology (Oxford) 2004; 43(2):244. 162. Hundae A Peskoe S Grimsley E Patel S. Rituximab therapy for refractory thrombotic thrombocytopenic purpura and autoimmune-mediated thrombocytopenia in systemic lupus erythematosus South Med J 2008; 101(9):943-944. 163. Lee JW Kim HA Sung JM Suh CH. Successful treatment of refractory immune thrombocytopenia with antiCD20 antibody in a patient with systemic lupus erythematosus Lupus 2010; 19(2):227-228. 164. Ng KP Cambridge G Leandro MJ Edwards JC Ehrenstein M Isenberg DA. B cell depletion therapy in systemic lupus erythematosus: long-term follow-up and predictors of response Ann Rheum Dis 2007; 66(9):12591262. 165. Bosello S De Santis M Lama G Spano C Angelucci C Tolusso B, et al. B cell depletion in diffuse progressive systemic sclerosis: safety, skin score modification and IL-6 modulation in an up to thirty-six months follow-up open-label trial Arthritis Res Ther 2010; 12(2):R54. 166. Daoussis D Liossis SN Tsamandas AC Kalogeropoulou C Kazantzi A Sirinian C, et al. Experience with rituximab in scleroderma: results from a 1-year, proof-of-principle study Rheumatology (Oxford) 2010; 49(2):271280. 167. Vasil'ev VI, Logvinenko OA, Kokosadze NV, Gaĭduk IV, Varlamova EIu, Kovrigina AM, Gorodetskiĭ VR, Nasonov EL. [First experience with the application of rituximab for the treatment of patients with Sjogren's syndrome and disease]. Vestn Ross Akad Med Nauk ; 2(12/12/2011).  192  168. Devauchelle-Pensec V Pennec Y Morvan J Pers JO Daridon C Jousse-Joulin S, et al. Improvement of Sjogren's syndrome after two infusions of rituximab (anti-CD20) Arthritis Rheum 2007; 57(2):310-317. 169. Meijer JM Meiners PM Vissink A Spijkervet FK Abdulahad W Kamminga N, et al. Effectiveness of rituximab treatment in primary Sjogren's syndrome: a randomized, double-blind, placebo-controlled trial Arthritis Rheum 2010; 62(4):960-968. 170. Seve P Gachon E Petiot P Stankovic K Charhon A Broussolle C. Successful treatment with rituximab in a patient with mental nerve neuropathy in primary Sjogren's syndrome Rheumatol Int 2007; 28(2):175-177. 171. Ring T Kallenbach M Praetorius J Nielsen S Melgaard B. Successful treatment of a patient with primary Sjogren's syndrome with Rituximab Clin Rheumatol 2006; 25(6):891-894. 172. Pijpe J Meijer JM Bootsma H van der Wal JE Spijkervet FK Kallenberg CG, et al. Clinical and histologic evidence of salivary gland restoration supports the efficacy of rituximab treatment in Sjogren's syndrome Arthritis Rheum 2009; 60(11):3251-3256. 173. Pijpe J van Imhoff GW Spijkervet FK Roodenburg JL Wolbink GJ Mansour K, et al. Rituximab treatment in patients with primary Sjogren's syndrome: an open-label phase II study Arthritis Rheum 2005; 52(9):2740-2750. 174. Pijpe J van Imhoff GW Vissink A van der Wal JE Kluin PM Spijkervet FK, et al. Changes in salivary gland immunohistology and function after rituximab monotherapy in a patient with Sjogren's syndrome and associated MALT lymphoma Ann Rheum Dis 2005; 64(6):958-960. 175. Seror R Sordet C Guillevin L Hachulla E Masson C Ittah M, et al. Tolerance and efficacy of rituximab and changes in serum B cell biomarkers in patients with systemic complications of primary Sjogren's syndrome Ann Rheum Dis 2007; 66(3):351-357. 176. Mok CC Ho LY To CH. Rituximab for refractory polymyositis: an open-label prospective study J Rheumatol 2007; 34(9):1864-1868. 177. Levine TD. Rituximab in the treatment of dermatomyositis: an open-label pilot study Arthritis Rheum 2005; 52(2):601-607. 178. Chung L Genovese MC Fiorentino DF. A pilot trial of rituximab in the treatment of patients with dermatomyositis Arch Dermatol 2007; 143(6):763-767. 179. Noss EH Hausner-Sypek DL Weinblatt ME. Rituximab as therapy for refractory polymyositis and dermatomyositis J Rheumatol 2006; 33(5):1021-1026. 180. Chiappetta N Steier J Gruber B. Rituximab in the treatment of refractory dermatomyositis J Clin Rheumatol 2005; 11(5):264-266. 181. Yáñez V J, Cisternas M M, Saldías H V, Saldías P F. [Refractory dermatomyositis associated with chronic organizing pneumonia treated with rituximab: report of one case]. ; 2011(12/12/2011):93. 182. Touma Z Arayssi T Kibbi L Masri AF. Successful treatment of cardiac involvement in dermatomyositis with rituximab Joint Bone Spine 2008; 75(3):334-337. 183. Cooper MA Willingham DL Brown DE French AR Shih FF White AJ. Rituximab for the treatment of juvenile dermatomyositis: a report of four pediatric patients Arthritis Rheum 2007; 56(9):3107-3111.  193  184. Dinh HV McCormack C Hall S Prince HM. Rituximab for the treatment of the skin manifestations of dermatomyositis: a report of 3 cases J Am Acad Dermatol 2007; 56(1):148-153. 185. Jones RB Ferraro AJ Chaudhry AN Brogan P Salama AD Smith KG, et al. A multicenter survey of rituximab therapy for refractory antineutrophil cytoplasmic antibody-associated vasculitis Arthritis Rheum 2009; 60(7):2156-2168. 186. Rhee EP Laliberte KA Niles JL. Rituximab as maintenance therapy for anti-neutrophil cytoplasmic antibodyassociated vasculitis Clin J Am Soc Nephrol 2010; 5(8):1394-1400. 187. José RJ, Chrysochou C, Shurrab AE, New D, Wood GN. Rituximab for rescue and maintenance therapy in rapidly progressive life-threatening antineutrophil cytoplasmic autoantibody-associated systemic vasculitis. Scand J Urol Nephrol ; 2011(12/12/2011):462. 188. Asamiya Y, Moriyama T, Takano M, Iwasaki C, Kimura K, Ando Y, Aoki A, Kikuchi K, Takei T, Uchida K, Nitta K. Successful treatment with rituximab in a patient with TTP secondary to severe ANCA-associated vasculitis. Intern Med ; 2011(12/12/2011). 189. Baird EM Lehman TJ Worgall S. Combination therapy with rituximab and cyclophosphamide in the treatment of anti-neutrophil cytoplasmic antibodies (ANCA) positive pulmonary hemorrhage: case report Pediatr Rheumatol Online J 2011; 9(1):33. 190. Wendt M, Gunnarsson I, Bratt J, Bruchfeld A. Rituximab in relapsing or refractory ANCA-associated vasculitis: a case series of 16 patients. Scand J Rheumatol 2011 Nov 28. 191. Lovric S Erdbruegger U Kumpers P Woywodt A Koenecke C Wedemeyer H, et al. Rituximab as rescue therapy in anti-neutrophil cytoplasmic antibody-associated vasculitis: a single-centre experience with 15 patients Nephrol Dial Transplant 2009; 24(1):179-185. 192. Roccatello D Baldovino S Alpa M Rossi D Napoli F Naretto C, et al. Effects of anti-CD20 monoclonal antibody as a rescue treatment for ANCA-associated idiopathic systemic vasculitis with or without overt renal involvement Clin Exp Rheumatol 2008; 26(3 Suppl 49):S67-71. 193. Roccatello D, Sciascia S, Rossi D, Alpa M, Naretto C, Russo A, Menegatti E, Baldovino S. Long-term effects of rituximab added to cyclophosphamide in refractory patients with vasculitis. Am J Nephrol ; 2011(12/12/2011). 194. Kaushik VV Reddy HV Bucknall RC. Successful use of rituximab in a patient with recalcitrant ChurgStrauss syndrome Ann Rheum Dis 2006; 65(8):1116-1117. 195. Saech J Owczarczyk K Rosgen S Petereit H Hallek M Rubbert-Roth A. Successful use of rituximab in a patient with Churg-Strauss syndrome and refractory central nervous system involvement Ann Rheum Dis 2010; 69(6):1254-1255. 196. Pepper RJ Fabre MA Pavesio C Gaskin G Jones RB Jayne D, et al. Rituximab is effective in the treatment of refractory Churg-Strauss syndrome and is associated with diminished T-cell interleukin-5 production Rheumatology (Oxford) 2008; 47(7):1104-1105. 197. Koukoulaki M Smith KG Jayne DR. Rituximab in Churg-Strauss syndrome Ann Rheum Dis 2006; 65(4):557-559.  194  198. Ribeiro E Cressend T Duffau P Grenouillet-Delacre M Rouanet-Lariviere M Vital A, et al. Rituximab Efficacy during a Refractory Polyarteritis Nodosa Flare Case Report Med 2009; 2009:738293. 199. Tan AL Holdsworth J Pease C Emery P McGonagle D. Successful treatment of resistant giant cell arteritis with etanercept Ann Rheum Dis 2003; 62(4):373-374. 200. Martinez-Taboada VM Rodriguez-Valverde V Carreno L Lopez-Longo J Figueroa M Belzunegui J, et al. A double-blind placebo controlled trial of etanercept in patients with giant cell arteritis and corticosteroid side effects Ann Rheum Dis 2008; 67(5):625-630. 201. Molloy ES Langford CA Clark TM Gota CE Hoffman GS. Anti-tumour necrosis factor therapy in patients with refractory Takayasu arteritis: long-term follow-up Ann Rheum Dis 2008; 67(11):1567-1569. 202. Hoffman GS Merkel PA Brasington RD Lenschow DJ Liang P. Anti-tumor necrosis factor therapy in patients with difficult to treat Takayasu arteritis Arthritis Rheum 2004; 50(7):2296-2304. 203. Efthimiou P Schwartzman S Kagen LJ. Possible role for tumour necrosis factor inhibitors in the treatment of resistant dermatomyositis and polymyositis: a retrospective study of eight patients Ann Rheum Dis 2006; 65(9):1233-1236. 204. Hengstman GJ van den Hoogen FH Barrera P Netea MG Pieterse A van de Putte LB, et al. Successful treatment of dermatomyositis and polymyositis with anti-tumor-necrosis-factor-alpha: preliminary observations Eur Neurol 2003; 50(1):10-15. 205. Hengstman GJ van den Hoogen FH van Engelen BG. Treatment of dermatomyositis and polymyositis with anti-tumor necrosis factor-alpha: long-term follow-up Eur Neurol 2004; 52(1):61-63. 206. Hengstman GJ De Bleecker JL Feist E Vissing J Denton CP Manoussakis MN, et al. Open-label trial of anti-TNF-alpha in dermato- and polymyositis treated concomitantly with methotrexate Eur Neurol 2008; 59(34):159-163. 207. Steinfeld SD Demols P Salmon I Kiss R Appelboom T. Infliximab in patients with primary Sjogren's syndrome: a pilot study Arthritis Rheum 2001; 44(10):2371-2375. 208. Steinfeld SD Demols P Appelboom T. Infliximab in primary Sjogren's syndrome: one-year followup Arthritis Rheum 2002; 46(12):3301-3303. 209. Booth AD Jayne DR Kharbanda RK McEniery CM Mackenzie IS Brown J, et al. Infliximab improves endothelial dysfunction in systemic vasculitis: a model of vascular inflammation Circulation 2004; 109(14):17181723. 210. Uthman I Kanj N Atweh S. Infliximab as monotherapy in giant cell arteritis Clin Rheumatol 2006; 25(1):109-110. 211. Cantini F Niccoli L Salvarani C Padula A Olivieri I. Treatment of longstanding active giant cell arteritis with infliximab: report of four cases Arthritis Rheum 2001; 44(12):2933-2935. 212. Airo P Antonioli CM Vianelli M Toniati P. Anti-tumour necrosis factor treatment with infliximab in a case of giant cell arteritis resistant to steroid and immunosuppressive drugs Rheumatology (Oxford) 2002; 41(3):347349.  195  213. Andonopoulos AP Meimaris N Daoussis D Bounas A Giannopoulos G. Experience with infliximab (antiTNF alpha monoclonal antibody) as monotherapy for giant cell arteritis Ann Rheum Dis 2003; 62(11):1116. 214. Kontkanen M Paimela L Kaarniranta K. Regression of necrotizing scleritis in Wegener's granulomatosis after infliximab treatment Acta Ophthalmol 2010; 88(3):96-7. 215. Wilkinson NM Erendzhinova E Zeft A Cabral DA. Infliximab as rescue therapy in three cases of paediatric Wegener's granulomatosis Rheumatology (Oxford) 2006; 45(8):1047-1048. 216. Kleinert J Lorenz M Kostler W Horl W Sunder-Plassmann G Soleiman A. Refractory Wegener's granulomatosis responds to tumor necrosis factor blockade Wien Klin Wochenschr 2004; 116(9-10):334-338. 217. Svozílková P, Ríhová E, Brichová M, Diblík P, Kuthan P, Poch T. [Infliximab in the treatment of Wegener's granulomatosis: case report]. Cesk Slov Oftalmol ; 2011(12/12/2011). 218. Lamprecht P Voswinkel J Lilienthal T Nolle B Heller M Gross WL, et al. Effectiveness of TNF-alpha blockade with infliximab in refractory Wegener's granulomatosis Rheumatology (Oxford) 2002; 41(11):1303-1307. 219. Lamprecht P, Arbach O, Voswinkel J, Lilienthal T, Nölle B, Heller M, Gause A, Gross WL. [Induction of remission with infliximab in therapy-refractory Wegener's granulomatosis - Follow-up of six patients]. Dtsch Med Wochenschr ; 2011(12/12/2011). 220. Hermann J Reittner P Scarpatetti M Graninger W. Successful treatment of meningeal involvement in Wegener's granulomatosis with infliximab Ann Rheum Dis 2006; 65(5):691-692. 221. Bartolucci P Ramanoelina J Cohen P Mahr A Godmer P Le Hello C, et al. Efficacy of the anti-TNF-alpha antibody infliximab against refractory systemic vasculitides: an open pilot study on 10 patients Rheumatology (Oxford) 2002; 41(10):1126-1132. 222. Josselin L Mahr A Cohen P Pagnoux C Guaydier-Souquieres G Hayem G, et al. Infliximab efficacy and safety against refractory systemic necrotising vasculitides: long-term follow-up of 15 patients Ann Rheum Dis 2008; 67(9):1343-1346. 223. El-Shabrawi Y Hermann J. Anti-TNF alpha therapy in chronic necrotizing scleritis resistant to standard immunomodulatory therapy in a patient with Wegener's granulomatosis Eye (Lond) 2005; 19(9):1017-1018. 224. Keogh KA Wylam ME Stone JH Specks U. Induction of remission by B lymphocyte depletion in eleven patients with refractory antineutrophil cytoplasmic antibody-associated vasculitis Arthritis Rheum 2005; 52(1):262268. 225. e-Therapeutics+ : Therapeutics : Skin Disorders: Psoriasis [Online]. 2011 May [cited 1/20/2012]; Available from: URL: http://www.e-therapeutics.ca/tc.showChapter.action?documentId=c0068. 226. Lateef A Petri M. Biologics in the treatment of systemic lupus erythematosus Curr Opin Rheumatol 2010; 22(5):504-509. 227. Aringer M Smolen JS. The role of tumor necrosis factor-alpha in systemic lupus erythematosus Arthritis Res Ther 2008; 10(1):202. 228. Postal M Appenzeller S. The role of Tumor Necrosis Factor-alpha (TNF-α) in the pathogenesis of systemic lupus erythematosus. ; 2012(3/20/2012):43.  196  229. Enbrel [Online]. 2011 May 30 [cited 2/19/2012]; Available from: URL: http://www.etherapeutics.ca/cps.showMonograph.action?newSearch=true&simpleIndex=BrandGeneric&simpleQuery=Enbrel#. 230. Soforo E Baumgartner M Francis L Allam F Phillips PE Perl A. Induction of systemic lupus erythematosus with tumor necrosis factor blockers J Rheumatol 2010; 37(1):204-205. 231. Chogle AR Shah CV Murthy AK. Role of anti-tumor necrosis factor-alpha blockers in inducing lupus erythematosus tumidus in "rhupus syndrome" J Rheumatol 2011; 38(6):1218-1219. 232. Schneider SW Staender S Schluter B Luger TA Bonsmann G. Infliximab-induced lupus erythematosus tumidus in a patient with rheumatoid arthritis Arch Dermatol 2006; 142(1):115-116. 233. Charles PJ Smeenk RJ De Jong J Feldmann M Maini RN. Assessment of antibodies to double-stranded DNA induced in rheumatoid arthritis patients following treatment with infliximab, a monoclonal antibody to tumor necrosis factor alpha: findings in open-label and randomized placebo-controlled trials Arthritis Rheum 2000; 43(11):2383-2390. 234. Puertas-Abreu E Polanco ER Azocar M Mundarain LA Nunez-Sotelo CM Montano R, et al. Onset of lupus like syndrome in patients with spondyloarthritis treated with anti-TNF-alpha Int Arch Med 2012; 5(1):7. 235. Diri E Tello W Ratnoff WD Nugent K. Infliximab-induced SLE-like syndrome involving the lung and pleura Lupus 2007; 16(9):764-766. 236. Shakoor N Michalska M Harris CA Block JA. Drug-induced systemic lupus erythematosus associated with etanercept therapy Lancet 2002; 359(9306):579-580. 237. Debandt M Vittecoq O Descamps V Le Loet X Meyer O. Anti-TNF-alpha-induced systemic lupus syndrome Clin Rheumatol 2003; 22(1):56-61. 238. Kang MJ Lee YH Lee J. Etanercept-induced systemic lupus erythematosus in a patient with rheumatoid arthritis J Korean Med Sci 2006; 21(5):946-949. 239. Mounach A Ghazi M Nouijai A Ghozlani I Achemlal L Bezza A, et al. Drug-induced lupus-like syndrome in ankylosing spondylitis treated with infliximab Clin Exp Rheumatol 2008; 26(6):1116-1118. 240. Bodur H Eser F Konca S Arikan S. Infliximab-induced lupus-like syndrome in a patient with ankylosing spondylitis Rheumatol Int 2009; 29(4):451-454. 241. Perez-Garcia C Maymo J Lisbona Perez MP Almirall Bernabe M Carbonell Abello J. Drug-induced systemic lupus erythematosus in ankylosing spondylitis associated with infliximab Rheumatology (Oxford) 2006; 45(1):114-116. 242. Benucci M Nenci G Cappelletti C Manfredi M. [Lupus like syndrome induced by treatment with anti TNFalpha (infliximab): report of three cases]. ; 2012(3/20/2012):6. 243. De Bandt M Sibilia J Le Loet X Prouzeau S Fautrel B Marcelli C, et al. Systemic lupus erythematosus induced by anti-tumour necrosis factor alpha therapy: a French national survey Arthritis Res Ther 2005; 7(3):R54551. 244. Health Canada Approves BENLYSTA™The First New Treatment for Systemic Lupus Erythematosus in Almost 50 Years [Online]. 2011 July 12 [cited 12/12/2011]; Available from: URL: http://www.gsk.ca/english/docspdf/Benlysta_News_Release_FINAL_for_web_20110711.pdf.  197  245. Benlysta [Online]. 2011 July 6 [cited 12/12/2011]; Available from: URL: http://www.gsk.ca/english/docspdf/product-monographs/Benlysta.pdf. 246. FDA approves Benlysta to treat lupus [Online]. 2011 March 9 [cited 1/19/2012]; Available from: URL: http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm246489.htm. 247. Navarra SV Guzman RM Gallacher AE Hall S Levy RA Jimenez RE, et al. Efficacy and safety of belimumab in patients with active systemic lupus erythematosus: a randomised, placebo-controlled, phase 3 trial Lancet 2011; 377(9767):721-731. 248. Furie R Stohl W Ginzler EM Becker M Mishra N Chatham W, et al. Biologic activity and safety of belimumab, a neutralizing anti-B-lymphocyte stimulator (BLyS) monoclonal antibody: a phase I trial in patients with systemic lupus erythematosus Arthritis Res Ther 2008; 10(5):R109. 249. Furie R, Petri M, Zamani O, Cervera R, Wallace DJ, Tegzová D, Sanchez-Guerrero J, Schwarting A, Merrill JT, Chatham WW, Stohl W, Ginzler EM, Hough DR, Zhong ZJ, Freimuth W, van Vollenhoven RF BLISS-76 Study Group. A phase III, randomized, placebo-controlled study of belimumab, a monoclonal antibody that inhibits B lymphocyte stimulator, in patients with systemic lupus erythematosus. Arthritis and Rheumatism ; 2012(1/19/2012). 250. Dubey AK Handu SS Dubey S Sharma P Sharma KK Ahmed QM. Belimumab: First targeted biological treatment for systemic lupus erythematosus J Pharmacol Pharmacother 2011; 2(4):317-319. 251. Mok CC. Update on emerging drug therapies for systemic lupus erythematosus. Expert Opin Emerg Drugs ; 2012(1):53-70. 252. Arefiev K Fiorentino DF Chung L. Endothelin Receptor Antagonists for the Treatment of Raynaud's Phenomenon and Digital Ulcers in Systemic Sclerosis Int J Rheumatol 2011; 2011. 253. Kowal-Bielecka O Landewe R Avouac J Chwiesko S Miniati I Czirjak L, et al. EULAR recommendations for the treatment of systemic sclerosis: a report from the EULAR Scleroderma Trials and Research group (EUSTAR) Ann Rheum Dis 2009; 68(5):620-628. 254. Tracleer [Online]. 2011 June 27 [cited 3/7/2012]; Available from: URL: http://www.etherapeutics.ca/cps.select.preliminaryFilter.action?simplePreliminaryFilter=bosentan+monohydrate#. 255. Roussin A. e-Therapeutics+ : Therapeutics : Cardiovascular Disorders: Raynaud’s Phenomenon [Online]. 2011 May [cited 1/5/2012]; Available from: URL: http://www.etherapeutics.ca/tc.showChapter.action?documentId=c0034. 256. Patented Medicine Prices Review Board. Report of New Patended Trugs - Tracleer [Online]. 2011 April 7 [cited 3/8/2012]; Available from: URL: http://www.pmprb-cepmb.gc.ca/english/view.asp?x=551&mp=572. 257. Launay D Diot E Pasquier E Mouthon L Boullanger N Fain O, et al. Bosentan for treatment of active digital ulcers in patients with systemic sclerosis Presse Med 2006; 35(4 Pt 1):587-592. 258. Riccardi M Chaila A Lannone F Grattagliano V Covelli M Lapadula G. [Treatment of digital ulcers in systemtic sclerosis with endothelin-1 receptor antagonist (bosentan)] Reumatismo 2007; 59(2):135-9.  198  259. Garcia de la Pena-Lefebvre P Rodriguez Rubio S Valero Exposito M Carmona L Gamir Gamir ML Beltran Gutierrez J, et al. Long-term experience of bosentan for treating ulcers and healed ulcers in systemic sclerosis patients Rheumatology (Oxford) 2008; 47(4):464-466. 260. Korn JH Mayes M Matucci Cerinic M Rainisio M Pope J Hachulla E, et al. Digital ulcers in systemic sclerosis: prevention by treatment with bosentan, an oral endothelin receptor antagonist Arthritis Rheum 2004; 50(12):3985-3993. 261. Matucci-Cerinic M Denton CP Furst DE Mayes MD Hsu VM Carpentier P, et al. Bosentan treatment of digital ulcers related to systemic sclerosis: results from the RAPIDS-2 randomised, double-blind, placebo-controlled trial Ann Rheum Dis 2011; 70(1):32-38. 262. Viagra [Online]. 2011 April 12 [cited 3/7/2012]; Available from: URL: http://www.etherapeutics.ca/cps.select.preliminaryFilter.action?simplePreliminaryFilter=sildenafil+citrate#m577300n00065. 263. Levien TL. Phosphodiesterase inhibitors in Raynaud's phenomenon Ann Pharmacother 2006; 40(7-8):13881393. 264. Caglayan E Huntgeburth M Karasch T Weihrauch J Hunzelmann N Krieg T, et al. Phosphodiesterase type 5 inhibition is a novel therapeutic option in Raynaud disease Arch Intern Med 2006; 166(2):231-233. 265. Colglazier CL Sutej PG O'Rourke KS. Severe refractory fingertip ulcerations in a patient with scleroderma: successful treatment with sildenafil J Rheumatol 2005; 32(12):2440-2442. 266. Brueckner CS Becker MO Kroencke T Huscher D Scherer HU Worm M, et al. Effect of sildenafil on digital ulcers in systemic sclerosis: analysis from a single centre pilot study Ann Rheum Dis 2010; 69(8):1475-1478. 267. Rosenkranz S Caglayan E Diet F Karasch T Weihrauch J Wassermann K, et al. Long-term effects of sildenafil in a patient with scleroderma-associated pulmonary hypertension and Raynaud's syndrome Dtsch Med Wochenschr 2004; 129(33):1736-1740. 268. Merrill JT Neuwelt CM Wallace DJ Shanahan JC Latinis KM Oates JC, et al. Efficacy and safety of rituximab in moderately-to-severely active systemic lupus erythematosus: the randomized, double-blind, phase II/III systemic lupus erythematosus evaluation of rituximab trial Arthritis Rheum 2010; 62(1):222-233. 269. Rovin B Furie R Latinis K Looney R Fervenza F Sanchez-Guerrero J, et al. Efficacy and safety of rituximab in patients with active proliferative lupus nephritis: The Lupus Nephritis Assessment with RItuximab study. Arthritis and Rheumatism ; 2012(3/20/2012):26. 270. How BENLYSTA® (belimumab) Works [Online]. 2011 [cited 12/12/2011]; Available from: URL: http://www.benlysta.com/benlysta-information/how-benlysta-works.html. 271. Pollack A. F.D.A. Panel Backs Drug for Lupus. Reuters . 272. Flolan [Online]. 2011 November 11 [cited 3/7/2012]; Available from: URL: http://www.etherapeutics.ca/cps.select.preliminaryFilter.action?simplePreliminaryFilter=epoprostenol+sodium#. 273. BC STATS: British Columbia Total Population Estimates [Online]. 2010 October [cited 12/14/2011]; Available from: URL: http://www.bcstats.gov.bc.ca/data/pop/pop/BCPop.asp. 274. Bernatsky ,Sasha Hudson ,Marie Panopalis ,Pantelis Clarke A,E. Pope ,Janet Leclercq ,Sharon, et al. The cost of systemic sclerosis. Arthritis and Rheumatism 2009; 61(1):119-123.  199  275. Panopalis P Petri M Manzi S Isenberg DA Gordon C Senecal JL, et al. The systemic lupus erythematosus Tri-Nation study: cumulative indirect costs Arthritis Rheum 2007; 57(1):64-70. 276. Clarke A,E. Esdaile J,M. Bloch D,A. Lacaille ,Diane Danoff D,S. Fries J,F. A canadian study of the total medical costs for patients with systemic lupus erythematosus and the predictors of costs. Arthritis & Rheumatism 1993; 36(11):1548-1559. 277. Clarke AE Zowall H Levinton C Assimakopoulos H Sibley JT Haga M, et al. Direct and indirect medical costs incurred by Canadian patients with rheumatoid arthritis: a 12 year study J Rheumatol 1997; 24(6):1051-1060. 278. Fautrel B Clarke AE Guillemin F Adam V St-Pierre Y Panaritis T, et al. Costs of rheumatoid arthritis: new estimates from the human capital method and comparison to the willingness-to-pay method Med Decis Making 2007; 27(2):138-150. 279. Maetzel A Li LC Pencharz J Tomlinson G Bombardier C Community Hypertension and Arthritis Project Study Team. The economic burden associated with osteoarthritis, rheumatoid arthritis, and hypertension: a comparative study Ann Rheum Dis 2004; 63(4):395-401. 280. Bernatsky S Ramsey-Goldman R Isenberg D Rahman A Dooley MA Sibley J, et al. Hodgkin's lymphoma in systemic lupus erythematosus Rheumatology (Oxford) 2007; 46(5):830-832. 281. Bernatsky S Joseph L Boivin JF Gordon C Urowitz M Gladman D, et al. The relationship between cancer and medication exposures in systemic lupus erythaematosus: a case-cohort study Ann Rheum Dis 2008; 67(1):7479. 282. Bernatsky S Boivin JF Joseph L Rajan R Zoma A Manzi S, et al. An international cohort study of cancer in systemic lupus erythematosus Arthritis Rheum 2005; 52(5):1481-1490. 283. Scott DG Watts RA. Systemic vasculitis: epidemiology, classification and environmental factors Ann Rheum Dis 2000; 59(3):161-163. 284. Friedman G Friedman B Benbassat J. Epidemiology of temporal arteritis in Israel Isr J Med Sci 1982; 18(2):241-244. 285. Andrews M Edmunds M Campbell A Walls J Feehally J. Systemic vasculitis in the 1980s--is there an increasing incidence of Wegener's granulomatosis and microscopic polyarteritis? J R Coll Physicians Lond 1990; 24(4):284-288. 286. Scott DG Watts RA. Systemic vasculitis: epidemiology, classification and environmental factors Ann Rheum Dis 2000; 59(3):161-163. 287. Overview of the BC and Regional Population Projections 2011 to 2036 [Online]. 2011 September [cited 1/17/2012]; Available from: URL: http://www.bcstats.gov.bc.ca/data/pop/pop/Project/P36BCIntro.pdf. 288. Uramoto KM Michet CJ,Jr Thumboo J Sunku J O'Fallon WM Gabriel SE. Trends in the incidence and mortality of systemic lupus erythematosus, 1950-1992 Arthritis Rheum 1999; 42(1):46-50. 289. Urowitz MB Gladman DD Tom BD Ibanez D Farewell VT. Changing patterns in mortality and disease outcomes for patients with systemic lupus erythematosus J Rheumatol 2008; 35(11):2152-2158. 290. National Health Expenditure Trends, 1975 to 2011. ; 2011 November. 1-181p.  200  291. Clarke AE Panopalis P Petri M Manzi S Isenberg DA Gordon C, et al. SLE patients with renal damage incur higher health care costs. Rheumatology 2008; 47(3):329-333. 292. Historical Exchange Rates | OANDA [Online]. [cited 12/8/2011]; Available from: URL: http://www.oanda.com/currency/historical-rates/. 293. CANSIM at CHASS - University of Toronto [Online]. 2012 February 13 [cited 12/8/2011]; Available from: URL: http://dc1.chass.utoronto.ca/chasscansim/. 294. Bernatsky S Panopalis P Pineau CA Hudson M St. Pierre Y Clarke AE. Healthcare Costs of Inflammatory Myopathies. J Rheumatol 2011; 38(5):885-888. 295. Cotch MF Hoffman GS. The prevalence, epidemiology and cost of hospitalizations for vasculitis in New York State: 1986 to 1990. Arthritis and Rheumatism 1995; 38(9 Supp):S225. 296. Bernatsky S Panopolis P Hudson M Pope J Leclercq S Robinson D, et al. Demographic and Clinical Factors Associated with Physician Service Use in Systemic Sclerosis. J Rheumatol 2009; 36(1):96-98. 297. Aghdassi E Zhang W St-Pierre Y Clarke AE Morrison S Peeva V, et al. Healthcare Cost and Loss of Productivity in a Canadian Population of Patients with and without Lupus Nephritis. J Rheumatol 2011; 38(4):658666. 298. Katz JN Barrett J Liang MH Kaplan H Roberts WN Baron JA. Utilization of rheumatology physician services by the elderly. Am J Med 1998; 105(4):312-318. 299. Carls G Li T Panopalis P Wang S Mell AG Gibson TB, et al. Direct and Indirect Costs to Employers of Patients With Systemic Lupus Erythematosus With and Without Nephritis. Journal of Occupational and Environmental Medicine 2009; 51(1):66-79. 300. Wilson L. Cost-of-illness of scleroderma: The case for rare diseases. Semin Arthritis Rheum 1997; 27(2):7384. 301. Pelletier EM Ogale S Yu E Brunetta P Garg J. Economic outcomes in patients diagnosed with systemic lupus erythematosus with versus without nephritis: Results from an analysis of data from a US claims database. Clin Ther 2009; 31(11):2653-2664. 302. Panopalis ,Pantelis Yazdany ,Jinoos Gillis J,Zell Julian ,Laura Trupin ,Laura Hersh A,O., et al. Health care costs and costs associated with changes in work productivity among persons with systemic lupus erythematosus. Arthritis and Rheumatism 2008; 59(12):1788-1795. 303. Stewart M PM. Lupus nephritis outcomes : Health maintenance organizations compared to non-health maintenance organizations. Journal of Rheumatology 2000; 27(4):900-902. 304. Li T Carls GS Panopalis P Wang S Gibson TB Goetzel RZ. Long-term medical costs and resource utilization in systemic lupus erythematosus and lupus nephritis: A five-year analysis of a large medicaid population. Arthritis & Rheumatism 2009; 61(6):755-763. 305. Finn D, Allen P, Hoffstein P, Goldman D, Petri M. Economic Costs of Systemic Lupus Erythmatosus (SLE): A Five Year Prospective Study. Arthritis and Rheumatism 1993; 36(Supp 9):S192. 306. Krishnan E. Hospitalization and mortality of patients with systemic lupus erythematosus. J Rheumatol 2006; 33(9):1770-1774.  201  307. Chung L Krishnan E Chakravarty EF. Hospitalizations and mortality in systemic sclerosis: results from the Nationwide Inpatient Sample. Rheumatology 2007; 46(12):1808-1813. 308. Nietert PJ Silverstein MD Silver RM. Hospital admissions, length of stay, charges, and in-hospital death among patients with systemic sclerosis. J Rheumatol 2001; 28(9):2031-2037. 309. Yelin E Trupin L Katz P Criswell LA Yazdany J Gillis J, et al. Impact of health maintenance organizations and fee-for-service on health care utilization among people with systemic lupus erythematosus. Arthritis & Rheumatism 2007; 57(3):508-515. 310. Molina MJ Mayor AM Franco AE Morell CA Lopez MA Vila LM. Utilization of Health Services and Prescription Patterns among Lupus Patients Followed by Primary Care Physicians and Rheumatologists in Puerto Rico. Ethn Dis 2008; 18(2 Supp 2):S2-205-10. 311. Callaghan R Prabu A Allan RB Clarke AE Sutcliffe N Pierre YS, et al. Direct healthcare costs and predictors of costs in patients with primary Sjögren's syndrome. Rheumatology 2007; 46(1):105-111. 312. Sutcliffe N Clarke AE Taylor R Frost C Isenberg DA. Total costs and predictors of costs in patients with systemic lupus erythematosus. Rheumatology 2001; 40(1):37-47. 313. Gironimi G Clarke A,E. Hamilton V,H. Danoff D,S. Bloch D,A. Fries J,F., et al. Why health care costs more in the US: Comparing health care expenditures between systemic lupus erythematosus patients in Stanford and Montreal. Arthritis and Rheumatism 1996; 39(6):979-987. 314. Clarke AE Petri M Manzi S Isenberg DA Gordon C Senécal J-, et al. The systemic lupus erythematosus Tri-nation Study: absence of a link between health resource use and health outcome. Rheumatology 2004; 43(8):1016-1024. 315. Clarke A Petri MA Manzi S Isenberg DA Gordon C Senecal J, et al. An international perspective on the well being and health care costs for patients with systemic lupus erythematosus. Journal of Rheumatology 1999; 26(7):1500-1511. 316. Huscher D Merkesdal S Thiele K Zeidler H Schneider M Zink A. Cost of illness in rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis and systemic lupus erythematosus in Germany. Annals of the Rheumatic Diseases 2006; 65(9):1175-1183. 317. Zink A Fischer-Betz R Thiele K Listing J Huscher D Gromnica-Ihle E, et al. Health care and burden of illness in systemic lupus erythematosus compared to rheumatoid arthritis: results from the national database of the German Collaborative Arthritis Centres. Lupus 2004; 13(7):529-536. 318. Krulichova I Gamba S Ricci E Garattini L. Direct Medical Costs of Monitoring and Treating Patients with Takayasu Arteritis in Italy. The European Journal of Health Economics 2004; 5(4):330; 330-334. 319. Minier T Péntek M Brodszky V Ecseki A Kárpáti K Polgár A, et al. Cost-of-illness of patients with systemic sclerosis in a tertiary care centre. Rheumatology 2010; 49(10):1920-1928. 320. Edwards CJ Lian TY Badsha H Teh CL Arden N Chng HH. Hospitalization of individuals with systemic lupus erythematosus: characteristics and predictors of outcome. Lupus 2003; 12(9):672-676. 321. Teh CL Chan GYL Lee J. Systemic lupus erythematosus in a tertiary, east Malaysian hospital: admission, readmission and death. International Journal of Rheumatic Diseases 2008; 11(1):24-29.  202  322. Zhu TY Tam L Lee VWY Lee KK Li EK. Systemic lupus erythematosus with neuropsychiatric manifestation incurs high disease costs: a cost-of-illness study in Hong Kong. Rheumatology 2009; 48(5):564-568. 323. Zhu TY Tam L Lee VW- Lee KK- Li EK. The impact of flare on disease costs of patients with systemic lupus erythematosus. Arthritis & Rheumatism 2009; 61(9):1159-1167. 324. Chiu Y Lai C. Nationwide population-based epidemiologic study of systemic lupus erythematosus in Taiwan. Lupus 2010; 19(10):1250-1255. 325. Petri M GM. Incidence of and Risk Factors for Hospitalizations in Systemic Lupus Erythmatosus: A Prospective Study of the Hopkins Lupus Cohort. Journal of Rheumatology 1992; 19(10):1559-1565. 326. Yazdany J Gillis JZ Trupin L Katz P Panopalis P Criswell LA, et al. Association of socioeconomic and demographic factors with utilization of rheumatology subspecialty care in systemic lupus erythematosus. Arthritis & Rheumatism 2007; 57(4):593-600. 327. Gillis JZ Yazdany J Trupin L Julian L Panopalis P Criswell LA, et al. Medicaid and access to care among persons with systemic lupus erythematosus. Arthritis & Rheumatism 2007; 57(4):601-607. 328. Lacaille D, Clarke AE, Bloch DA, Danoff D, Esdaile JM. The Impact of Disease Activity, Treatment, and Disease Severity on Short Term Costs of Systemic Lupus Erythmatosus. Journal of Rheumatology 1994; 21(3):448449-453. 329. Slawsky KA Fernandes AW Fusfeld L Manzi S. A structured literature review of the direct costs of adult systemic lupus erythematosus in the US. Arthritis Care and Research ; 2011; 63(11/23/2011; 9):1224-1232. 330. Alamanos Y Voulgari PV Siozos C Katsimpri P Tsintzos S Dimou G, et al. Epidemiology of systemic lupus erythematosus in northwest Greece 1982-2001 J Rheumatol 2003; 30(4):731-735. 331. Alonso MD Llorca J Martinez-Vazquez F Miranda-Filloy JA Diaz de Teran T Dierssen T, et al. Systemic lupus erythematosus in northwestern Spain: a 20-year epidemiologic study. Medicine 2011; 90(5):350-358. 332. Chakravarty EF Bush TM Manzi S Clarke AE Ward MM. Prevalence of adult systemic lupus erythematosus in California and Pennsylvania in 2000: estimates obtained using hospitalization data Arthritis Rheum 2007; 56(6):2092-2094. 333. Govoni M Castellino G Bosi S Napoli N Trotta F. Incidence and prevalence of systemic lupus erythematosus in a district of north Italy Lupus 2006; 15(2):110-113. 334. Gudmundsson S Steinsson K. Systemic lupus erythematosus in Iceland 1975 through 1984. A nationwide epidemiological study in an unselected population J Rheumatol 1990; 17(9):1162-1167. 335. Hochberg MC. Prevalence of systemic lupus erythematosus in England and Wales, 1981-2 Ann Rheum Dis 1987; 46(9):664-666. 336. Hopkinson ND Doherty M Powell RJ. The prevalence and incidence of systemic lupus erythematosus in Nottingham, UK, 1989-1990 Br J Rheumatol 1993; 32(2):110-115. 337. Lopez P Mozo L Gutierrez C Suarez A. Epidemiology of systemic lupus erythematosus in a northern Spanish population: gender and age influence on immunological features Lupus 2003; 12(11):860-865. 338. McCarty DJ Manzi S Medsger TA,Jr Ramsey-Goldman R LaPorte RE Kwoh CK. Incidence of systemic lupus erythematosus. Race and gender differences Arthritis Rheum 1995; 38(9):1260-1270.  203  339. Naleway AL Davis ME Greenlee RT Wilson DA McCarty DJ. Epidemiology of systemic lupus erythematosus in rural Wisconsin Lupus 2005; 14(10):862-866. 340. Nightingale AL Farmer RD de Vries CS. Incidence of clinically diagnosed systemic lupus erythematosus 1992-1998 using the UK General Practice Research Database Pharmacoepidemiol Drug Saf 2006; 15(9):656-661. 341. Nived O Sturfelt G Wollheim F. Systemic lupus erythematosus in an adult population in southern Sweden: incidence, prevalence and validity of ARA revised classification criteria Br J Rheumatol 1985; 24(2):147-154. 342. Nossent JC. Systemic lupus erythematosus on the Caribbean island of Curacao: an epidemiological investigation Ann Rheum Dis 1992; 51(11):1197-1201. 343. Nossent HC. Systemic lupus erythematosus in the Arctic region of Norway J Rheumatol 2001; 28(3):539-546. 344. Voss A Green A Junker P. Systemic lupus erythematosus in Denmark: clinical and epidemiological characterization of a county-based cohort Scand J Rheumatol 1998; 27(2):98-105. 345. Ward MM. Prevalence of physician-diagnosed systemic lupus erythematosus in the United States: results from the third national health and nutrition examination survey J Womens Health (Larchmt) 2004; 13(6):713-718. 346. Alamanos Y Tsifetaki N Voulgari PV Siozos C Tsamandouraki K Alexiou GA, et al. Epidemiology of systemic sclerosis in northwest Greece 1981 to 2002 Semin Arthritis Rheum 2005; 34(5):714-720. 347. Allcock RJ Forrest I Corris PA Crook PR Griffiths ID. A study of the prevalence of systemic sclerosis in northeast England Rheumatology (Oxford) 2004; 43(5):596-602. 348. Arias-Nunez MC Llorca J Vazquez-Rodriguez TR Gomez-Acebo I Miranda-Filloy JA Martin J, et al. Systemic sclerosis in northwestern Spain: a 19-year epidemiologic study Medicine (Baltimore) 2008; 87(5):272-280. 349. Bernatsky S Joseph L Pineau CA Belisle P Hudson M Clarke AE. Scleroderma prevalence: demographic variations in a population-based sample Arthritis Rheum 2009; 61(3):400-404. 350. Geirsson AJ Steinsson K Guthmundsson S Sigurthsson V. Systemic sclerosis in Iceland. A nationwide epidemiological study Ann Rheum Dis 1994; 53(8):502-505. 351. Mayes MD Lacey JV,Jr Beebe-Dimmer J Gillespie BW Cooper B Laing TJ, et al. Prevalence, incidence, survival, and disease characteristics of systemic sclerosis in a large US population Arthritis Rheum 2003; 48(8):2246-2255. 352. Rosa JE Soriano ER Narvaez-Ponce L del Cid CC Imamura PM Catoggio LJ. Incidence and prevalence of systemic sclerosis in a healthcare plan in Buenos Aires. Journal of Clinical Rheumatology 2011; 17(2):59-63. 353. Steen VD Oddis CV Conte CG Janoski J Casterline GZ Medsger TA,Jr. Incidence of systemic sclerosis in Allegheny County, Pennsylvania. A twenty-year study of hospital-diagnosed cases, 1963-1982 Arthritis Rheum 1997; 40(3):441-445. 354. Valter I, Saretok S, Maricq HR. Prevalence of scleroderma spectrum disorders in the general population of Estonia.. Scand J Rheumatol ; 2011(12/9/2011):25. 355. Alamanos Y Tsifetaki N Voulgari PV Venetsanopoulou AI Siozos C Drosos AA. Epidemiology of primary Sjogren's syndrome in north-west Greece, 1982-2003 Rheumatology (Oxford) 2006; 45(2):187-191.  204  356. Birlik M, Akar S, Gurler O, Sari I, Birlik B, Sarioglu S, Oktem MA, Saglam F, Can G, Kayahan H, Akkoc N, Onen F. Prevalence of primary Sjogren's syndrome in Turkey: a population-based epidemiological study. Int J Clin Pract ; 2011(12/14/2011). 357. Miyasaka N. [Epidemiology and pathogenesis of Sjögren's syndrome]. Nihon Rinsho ; 2011(12/14/2011). 358. Pillemer SR Matteson EL Jacobsson LT Martens PB Melton LJ,3rd O'Fallon WM, et al. Incidence of physician-diagnosed primary Sjogren syndrome in residents of Olmsted County, Minnesota Mayo Clin Proc 2001; 76(6):593-599. 359. Thomas E Hay EM Hajeer A Silman AJ. Sjogren's syndrome: a community-based study of prevalence and impact Br J Rheumatol 1998; 37(10):1069-1076. 360. Weng MY Huang YT Liu MF Lu TH. Incidence and mortality of treated primary Sjogren's syndrome in Taiwan: a population-based study J Rheumatol 2011; 38(4):706-708. 361. Oddis CV Conte CG Steen VD Medsger TA,Jr. Incidence of polymyositis-dermatomyositis: a 20-year study of hospital diagnosed cases in Allegheny County, PA 1963-1982 J Rheumatol 1990; 17(10):1329-1334. 362. Bernatsky S Joseph L Pineau CA Belisle P Boivin JF Banerjee D, et al. Estimating the prevalence of polymyositis and dermatomyositis from administrative data: age, sex and regional differences Ann Rheum Dis 2009; 68(7):1192-1196. 363. Mohammad AJ Jacobsson LT Mahr AD Sturfelt G Segelmark M. Prevalence of Wegener's granulomatosis, microscopic polyangiitis, polyarteritis nodosa and Churg-Strauss syndrome within a defined population in southern Sweden Rheumatology (Oxford) 2007; 46(8):1329-1337. 364. Mohammad AJ Jacobsson LT Westman KW Sturfelt G Segelmark M. Incidence and survival rates in Wegener's granulomatosis, microscopic polyangiitis, Churg-Strauss syndrome and polyarteritis nodosa Rheumatology (Oxford) 2009; 48(12):1560-1565. 365. Reinhold-Keller E Zeidler A Gutfleisch J Peter HH Raspe HH Gross WL. Giant cell arteritis is more prevalent in urban than in rural populations: results of an epidemiological study of primary systemic vasculitides in Germany Rheumatology (Oxford) 2000; 39(12):1396-1402. 366. Reinhold-Keller E Herlyn K Wagner-Bastmeyer R Gutfleisch J Peter HH Raspe HH, et al. No difference in the incidences of vasculitides between north and south Germany: first results of the German vasculitis register Rheumatology (Oxford) 2002; 41(5):540-549. 367. Reinhold-Keller E Herlyn K Wagner-Bastmeyer R Gross WL. Stable incidence of primary systemic vasculitides over five years: results from the German vasculitis register Arthritis Rheum 2005; 53(1):93-99. 368. Watts RA Gonzalez-Gay MA Lane SE Garcia-Porrua C Bentham G Scott DG. Geoepidemiology of systemic vasculitis: comparison of the incidence in two regions of Europe Ann Rheum Dis 2001; 60(2):170-172. 369. Baldursson O Steinsson K Bjornsson J Lie JT. Giant cell arteritis in Iceland. An epidemiologic and histopathologic analysis Arthritis Rheum 1994; 37(7):1007-1012. 370. Boesen P Sorensen SF. Giant cell arteritis, temporal arteritis, and polymyalgia rheumatica in a Danish county. A prospective investigation, 1982-1985 Arthritis Rheum 1987; 30(3):294-299.  205  371. Gonzalez-Gay MA Garcia-Porrua C Rivas MJ Rodriguez-Ledo P Llorca J. Epidemiology of biopsy proven giant cell arteritis in northwestern Spain: trend over an 18 year period Ann Rheum Dis 2001; 60(4):367-371. 372. Gonzalez-Gay MA, Miranda-Filloy JA, Lopez-Diaz MJ, Perez-Alvarez R, Gonzalez-Juanatey C, SanchezAndrade A, Martin J, Llorca J. Giant cell arteritis in northwestern Spain: a 25-year epidemiologic study. . Medicine (Baltimore) ; 2011(12/10/2011):8. 373. Gonzalez-Gay MA Blanco R Sanchez-Andrade A Vazquez-Caruncho M. Giant cell arteritis in Lugo, Spain: a more frequent disease with fewer classic features J Rheumatol 1997; 24(11):2166-2170. 374. Gran JT Myklebust G. The incidence of polymyalgia rheumatica and temporal arteritis in the county of Aust Agder, south Norway: a prospective study 1987-94 J Rheumatol 1997; 24(9):1739-1743. 375. Petursdottir V Johansson H Nordborg E Nordborg C. The epidemiology of biopsy-positive giant cell arteritis: special reference to cyclic fluctuations Rheumatology (Oxford) 1999; 38(12):1208-1212. 376. Salvarani C Gabriel SE O'Fallon WM Hunder GG. The incidence of giant cell arteritis in Olmsted County, Minnesota: apparent fluctuations in a cyclic pattern Ann Intern Med 1995; 123(3):192-194. 377. Salvarani C Crowson CS O'Fallon WM Hunder GG Gabriel SE. Reappraisal of the epidemiology of giant cell arteritis in Olmsted County, Minnesota, over a fifty-year period Arthritis Rheum 2004; 51(2):264-268. 378. Sonnenblick M Nesher G Friedlander Y Rubinow A. Giant cell arteritis in Jerusalem: a 12-year epidemiological study Br J Rheumatol 1994; 33(10):938-941. 379. Al-Arfaj AS Al-Balla SR Al-Dalaan AN Al-Saleh SS Bahabri SA Mousa MM, et al. Prevalence of systemic lupus erythematosus in central Saudi Arabia Saudi Med J 2002; 23(1):87-89. 380. Anagnostopoulos I Zinzaras E Alexiou I Papathanasiou AA Davas E Koutroumpas A, et al. The prevalence of rheumatic diseases in central Greece: a population survey BMC Musculoskelet Disord 2010; 11:98. 381. Bernatsky S Joseph L Pineau CA Tamblyn R Feldman DE Clarke AE. A population-based assessment of systemic lupus erythematosus incidence and prevalence—results and implications of using administrative data for epidemiological studies. Rheumatology 2007; 46(12):1814-1818. 382. Bossingham D. Systemic lupus erythematosus in the far north of Queensland Lupus 2003; 12(4):327-331. 383. Gourley IS Patterson CC Bell AL. The prevalence of systemic lupus erythematosus in Northern Ireland Lupus 1997; 6(4):399-403. 384. Hart HH Grigor RR Caughey DE. Ethnic difference in the prevalence of systemic lupus erythematosus Ann Rheum Dis 1983; 42(5):529-532. 385. Helve T. Prevalence and mortality rates of systemic lupus erythematosus and causes of death in SLE patients in Finland Scand J Rheumatol 1985; 14(1):43-46. 386. Hopkinson ND Doherty M Powell RJ. Clinical features and race-specific incidence/prevalence rates of systemic lupus erythematosus in a geographically complete cohort of patients Ann Rheum Dis 1994; 53(10):675680. 387. Johnson AE Gordon C Palmer RG Bacon PA. The prevalence and incidence of systemic lupus erythematosus in Birmingham, England. Relationship to ethnicity and country of birth Arthritis Rheum 1995; 38(4):551-558.  206  388. Laustrup H, Voss A, Green A, Junker P. Occurrence of systemic lupus erythematosus in a Danish community: an 8-year prospective study. Scand J Rheumatol ; 2011(12/15/2011). 389. Maskarinec G Katz AR. Prevalence of systemic lupus erythematosus in Hawaii: is there a difference between ethnic groups? Hawaii Med J 1995; 54(2):406-409. 390. Serdula MK Rhoads GG. Frequency of systemic lupus erythematosus in different ethnic groups in Hawaii Arthritis Rheum 1979; 22(4):328-333. 391. Stahl-Hallengren C Jonsen A Nived O Sturfelt G. Incidence studies of systemic lupus erythematosus in Southern Sweden: increasing age, decreasing frequency of renal manifestations and good prognosis J Rheumatol 2000; 27(3):685-691. 392. Airo P Tabaglio E Frassi M Scarsi M Danieli E Rossi M. Prevalence of systemic sclerosis in Valtrompia in northern Italy. A collaborative study of rheumatologists and general practitioners Clin Exp Rheumatol 2007; 25(6):878-880. 393. Arnett FC Howard RF Tan F Moulds JM Bias WB Durban E, et al. Increased prevalence of systemic sclerosis in a Native American tribe in Oklahoma. Association with an Amerindian HLA haplotype Arthritis Rheum 1996; 39(8):1362-1370. 394. Kaliterna DM Radić M Pavić A. [Incidence, prevalence and disease characteristics of systemic sclerosis in Split-Dalmatia County]. Reumatizm 2010; 57(2):94-98. 395. Le Guern V Mahr A Mouthon L Jeanneret D Carzon M Guillevin L. Prevalence of systemic sclerosis in a French multi-ethnic county Rheumatology (Oxford) 2004; 43(9):1129-1137. 396. Lo Monaco A Bruschi M La Corte R Volpinari S Trotta F. Epidemiology of systemic sclerosis in a district of northern Italy. Clinical and Experimental Rheumatology 2011; 29(2 Suppl 65):S10-S14. 397. Maricq HR Weinrich MC Keil JE Smith EA Harper FE Nussbaum AI, et al. Prevalence of scleroderma spectrum disorders in the general population of South Carolina Arthritis Rheum 1989; 32(8):998-1006. 398. Robinson D,Jr Eisenberg D Nietert PJ Doyle M Bala M Paramore C, et al. Systemic sclerosis prevalence and comorbidities in the US, 2001-2002 Curr Med Res Opin 2008; 24(4):1157-1166. 399. Thompson AE Pope JE. Increased prevalence of scleroderma in southwestern Ontario: a cluster analysis J Rheumatol 2002; 29(9):1867-1873. 400. Bowman SJ Ibrahim GH Holmes G Hamburger J Ainsworth JR. Estimating the prevalence among Caucasian women of primary Sjogren's syndrome in two general practices in Birmingham, UK Scand J Rheumatol 2004; 33(1):39-43. 401. Dafni UG Tzioufas AG Staikos P Skopouli FN Moutsopoulos HM. Prevalence of Sjogren's syndrome in a closed rural community Ann Rheum Dis 1997; 56(9):521-525. 402. Gøransson LG, Haldorsen K, Brun JG, Harboe E, Jonsson MV, Skarstein K, Time K, Omdal R. The point prevalence of clinically relevant primary Sjögren's syndrome in two Norwegian counties. Scand J Rheumatol ; 2011(12/14/2011).  207  403. Jacobsson LT Axell TE Hansen BU Henricsson VJ Larsson A Lieberkind K, et al. Dry eyes or mouth--an epidemiological study in Swedish adults, with special reference to primary Sjogren's syndrome J Autoimmun 1989; 2(4):521-527. 404. Kabasakal Y Kitapcioglu G Turk T Oder G Durusoy R Mete N, et al. The prevalence of Sjogren's syndrome in adult women Scand J Rheumatol 2006; 35(5):379-383. 405. Tomsic M Logar D Grmek M Perkovic T Kveder T. Prevalence of Sjogren's syndrome in Slovenia Rheumatology (Oxford) 1999; 38(2):164-170. 406. Whaley K Williamson J Wilson T McGavin DD Hughes GR Hughes H, et al. Sjogren's syndrome and autoimmunity in a geriatric population Age Ageing 1972; 1(4):197-206. 407. Zhang NZ Shi CS Yao QP Pan GX Wang LL Wen ZX, et al. Prevalence of primary Sjogren's syndrome in China J Rheumatol 1995; 22(4):659-661. 408. Haugeberg G Bie R Bendvold A Larsen AS Johnsen V. Primary vasculitis in a Norwegian community hospital: a retrospective study Clin Rheumatol 1998; 17(5):364-368. 409. Mahr A Guillevin L Poissonnet M Ayme S. Prevalences of polyarteritis nodosa, microscopic polyangiitis, Wegener's granulomatosis, and Churg-Strauss syndrome in a French urban multiethnic population in 2000: a capture-recapture estimate Arthritis Rheum 2004; 51(1):92-99. 410. Ormerod AS Cook MC. Epidemiology of primary systemic vasculitis in the Australian Capital Territory and south-eastern New South Wales Intern Med J 2008; 38(11):816-823. 411. Cotch MF, Hoffman GS, Yerg DE, Kaufman GI, Targonski P, Kaslow RA. The epidemiology of Wegener's granulomatosis. Estimates of the five-year period prevalence, annual mortality, and geographic disease distribution from population-based data sources. Arthritis and Rheumatism 1996; 39(1):87-92. 412. Gibson A Stamp LK Chapman PT O'Donnell JL. The epidemiology of Wegener's granulomatosis and microscopic polyangiitis in a Southern Hemisphere region Rheumatology (Oxford) 2006; 45(5):624-628. 413. Koldingsnes W NH. Epidemiology of Wegener's granulomatosis in northern Norway. Arthritis and Rheumatism ; 2011(12/1/2011). 414. Jiminez SA Cronin PM Koenig AS O'Brien MS. Scleroderma [Online]. 2011 July 22 [cited 1/17/2012]; Available from: URL: http://emedicine.medscape.com/article/331864-treatment#showall. 415. Basson R. e-Therapeutics+ : Therapeutics : Sexual Health: Male Sexual Dysfunction [Online]. 2011 May [cited 3/7/2012]; Available from: URL: http://www.e-therapeutics.ca/tc.showChapter.action?documentId=c0078. 416. Cotch MF. The socioeconomic impact of vasculitis Curr Opin Rheumatol 2000; 12(1):20-23. 417. Haugen AJ Peen E Hulten B Johannessen AC Brun JG Halse AK, et al. Estimation of the prevalence of primary Sjogren's syndrome in two age-different community-based populations using two sets of classification criteria: the Hordaland Health Study Scand J Rheumatol 2008; 37(1):30-34. 418. Benbassat J, Geffel D, Zlotnick A. Epidemiology of polymyositis-dermatomyositis in Isreal,1960-1976. ; 2011(12/10/2011):200. 419. Herlyn K Hellmich B Gross WL Reinhold-Keller E. Stable incidence of systemic vasculitides in schleswigholstein, Germany Dtsch Arztebl Int 2008; 105(19):355-361.  208  420. O'Donnell JL Stevanovic VR Frampton C Stamp LK Chapman PT. Wegener's granulomatosis in New Zealand: evidence for a latitude-dependent incidence gradient Intern Med J 2007; 37(4):242-246. 421. Bengtsson BA Malmvall BE. The epidemiology of giant cell arteritis including temporal arteritis and polymyalgia rheumatica. Incidences of different clinical presentations and eye complications Arthritis Rheum 1981; 24(7):899-904. 422. Franzen P Sutinen S von Knorring J. Giant cell arteritis and polymyalgia rheumatica in a region of Finland: an epidemiologic, clinical and pathologic study, 1984-1988 J Rheumatol 1992; 19(2):273-276. 423. What is Juvenile Idiopathic Arthritis? [Online]. 2009 December 20 [cited 3/17/2012]; Available from: URL: http://www.aboutkidshealth.ca/tas/en/aboutjia/Pages/WhatIsJIA.aspx. 424. Bernatsky S Duffy C Malleson P Feldman DE St Pierre Y Clarke AE. Economic impact of juvenile idiopathic arthritis Arthritis Rheum 2007; 57(1):44-48. 425. Osteoarthritis [Online]. 2011 November 30 [cited 12/15/2011]; Available from: URL: http://www.arthritis.ca/types of arthritis/osteoarthritis/default.asp?s=1. 426. Finestone H. e-Therapeutics+ : Therapeutics : Musculoskeletal Disorders: Fibromyalgia [Online]. 2011 May [cited 3/17/2012]; Available from: URL: http://www.e-therapeutics.ca/tc.showChapter.action?documentId=c0053. 427. Penrod JR Bernatsky S Adam V Baron M Dayan N Dobkin PL. Health services costs and their determinants in women with fibromyalgia J Rheumatol 2004; 31(7):1391-1398. 428. White KP Speechley M Harth M Ostbye T. The London Fibromyalgia Epidemiology Study: direct health care costs of fibromyalgia syndrome in London, Canada J Rheumatol 1999; 26(4):885-889. 429. Pop Data BC [Online]. 2011 [cited 12/20/2011]; Available from: URL: http://www.popdata.bc.ca/. 430. PharmaNet [Online]. [cited 12/20/2011]; Available from: URL: http://www.health.gov.bc.ca/pharmacare/pharmanet/netindex.html. 431. Medical Services Plan [Online]. 2011 March [cited 12/19/2011]; Available from: URL: http://www.health.gov.bc.ca/msp/infoben/pdf/msp-brochure.pdf. 432. MSP Information Resource Manual 2010/2011. Victoria, BC: Medical Services Plan 112p. 433. Katz JN Barrett J Liang MH Bacon AM Kaplan H Kieval RI, et al. Sensitivity and positive predictive value of Medicare Part B physician claims for rheumatologic diagnoses and procedures. Arthritis Rheum 1997; 40(9):1594-1600. 434. Bernatsky S Linehan T Hanly JG. The accuracy of administrative data diagnoses of systemic autoimmune rheumatic diseases J Rheumatol 2011; 38(8):1612-1616. 435. Finlayson GS Reimer J Dahl M Stargardter M McGowain KL. The Direct Cost of Hospitalizations in Manitoba, 2005/06. Winnipeg, MB: Manitoba Centre for Health Policy ; 2009 March. 436. Canadian MIS Database (CMDB), Hospital Financial Performance Indicators, 2000-2001 to 2008-2009: Methodological Notes. Ottawa, Ontario: Canadian Institute for Health Information . 437. Canadian MIS Database - Hospital Financial Performance Indicators, 1999-2000 to 2008-2009: Methodological Notes. Ottawa, Ontario: Canadian Institute for Health Information ; 2010 July.  209  438. Hicks V Zhang J. Hospital Price Index Feasibility Study. Ottawa, Ontario: Canadian Institute for Health Information ; 2003 June. 439. Pohar SL Simpson SH Majunder SR, et al. Epidemiologic and Cost Trends in Diabetes in Saskatchewan, 1991 to 2001: Working Paper 05-06. Institute of Health Economics ; 2005 October. 440. WorkSafeBC Fee Items and Rates [Online]. 2007 December 5 [cited 12/21/2011]; Available from: URL: http://www.worksafebc.com/health_care_providers/Assets/PDF/hospittal_fee_codes.pdf. 441. Fraser Health Authority. Hospital Rates 1995-2010. 21 July 2011. 442. Johnson JA Simpsom SH Jacobs P Downey W Beck P Osei W. Cost of Health Care for People with Diabetes Mellitus in Saskatchewan, 1991 to 1996: Working Paper 02-02. Institute of Health Economics . 443. SAS Institute Inc. SAS System for Unix. ;9.2. 444. Health Care Cost Drivers: Drug Expenditure Trends Perspective. Ottawa, Ontario: Canadian Institute for Health Information ; 2011 November 3. 1-38p. 445. Watson C. e-Therapeutics+ : Therapeutics : Neurologic Disorders: Neuropathic Pain [Online]. 2011 July [cited 3/4/2012]; Available from: URL: http://www.e-therapeutics.ca/tc.showChapter.action?documentId=c0012. 446. Health Care Cost Drivers: Hospital Cost Drivers. Ottawa, Ontario: Canadian Institute for Health Information ; 2011 April 28. 1-80p. 447. Hak AE Karlson EW Feskanich D Stampfer MJ Costenbader KH. Systemic lupus erythematosus and the risk of cardiovascular disease: results from the nurses' health study Arthritis Rheum 2009; 61(10):1396-1402. 448. Ward MM. Premature morbidity from cardiovascular and cerebrovascular diseases in women with systemic lupus erythematosus Arthritis Rheum 1999; 42(2):338-346. 449. Faurschou M Mellemkjaer L Sorensen IJ Svalgaard Thomsen B Dreyer L Baslund B. Increased morbidity from ischemic heart disease in patients with Wegener's granulomatosis Arthritis Rheum 2009; 60(4):1187-1192. 450. Karp I Abrahamowicz M Fortin PR Pilote L Neville C Pineau CA, et al. Recent corticosteroid use and recent disease activity: independent determinants of coronary heart disease risk factors in systemic lupus erythematosus? Arthritis Rheum 2008; 59(2):169-175. 451. Petri M Perez-Gutthann S Spence D Hochberg MC. Risk factors for coronary artery disease in patients with systemic lupus erythematosus Am J Med 1992; 93(5):513-519. 452. Arora P. Chronic Kidney Disease [Online]. 2012 March 28 [cited 4/7/2012]; Available from: URL: http://emedicine.medscape.com/article/238798-overview#aw2aab6b2b3. 453. Wazny L Moist L. e-Therapeutics+ : Therapeutics : Renal Disorders: Chronic Kidney Disease [Online]. 2012 March [cited 4/7/2012]; Available from: URL: http://www.etherapeutics.ca/tc.showChapter.action?documentId=c0126. 454. Khardori R. Type 2 Diabetes Mellitus [Online]. 2012 February 23 [cited 2/26/2012]; Available from: URL: http://emedicine.medscape.com/article/117853-overview. 455. McMillan J. Nephritic and Nephrotic Syndromes: Glomerular Disorders: Merck Manual Professional [Online]. 2010 January [cited 2/26/2012]; Available from: URL:  210  http://www.merckmanuals.com/professional/genitourinary_disorders/glomerular_disorders/nephritic_and_nephrotic _syndromes.html#v1056547. 456. de Leeuw K Sanders JS Stegeman C Smit A Kallenberg CG Bijl M. Accelerated atherosclerosis in patients with Wegener's granulomatosis Ann Rheum Dis 2005; 64(5):753-759. 457. British Columbia Ambulance Service [Online]. 2012 [cited 1/1/2012]; Available from: URL: http://www.bcas.ca/EN/main/about/fees.html. 458. McKendry R Reid RJ McGrail KM Kerluke KJ. Emergency Rooms in British Columbia: A pilot project to validate current data and describe users. Vancouver, BC: The Centre for Health Services and Policy Research . 459. Carruthers D Watts R Symmons D Scott D. Wegener's Granulomatosis - Increased Incidence or Increased Recognition? Rheumatology 1996; 2(2):5. 460. MSP Physician Resource Report 2001/2002-2010/2011 [Online]. 2011 [cited 3/4/2012]; Available from: URL: http://www.health.gov.bc.ca/msp/paystats/pdf/physician-resource-report.pdf. 461. Barr SG Zonana-Nacach A Magder LS Petri M. Patterns of disease activity in systemic lupus erythematosus. Arthritis Rheum 1999; 42(12):2682-2688. 462. Russell A, Haraoui B, Keystone E, Klinkhoff A. Current and emerging therapies for rheumatoid arthritis, with a focus on infliximab: clinical impact on joint damage and cost of care in canada. Clin Ther ; 2012(1/10/2012).  211  Appendices Appendix A: Diagnostic Codes, Fee Items, and Drug Classes International Classification of Diseases (ICD) Codes for SARDs, 9th and 10th Revisions Diagnosis ICD-9 ICD-10 All connective 710, 710.X M32.1, M32.8, M32.9 tissue disorders M34.X ( SARDs-CTD) M35.0 , M35.1 M33.0, M33.1, M33.9 M33.2 Vasculitic 446.0 M30.0 disorders 446.4 M31.3 ( SARDs-VD) 446.5 M31.5 446.7 M31.4 Additional SARDs-Attributable MSP Claims Included in the Broad Definition, by Specialty, Diagnosis, and/or Drug Therapy Physician Specialty Included Claims Ophthalmology -all encounters for all cases with a SjD diagnosis at an index visit -all encounters for all cases with an SLE diagnosis at an index visit -specialty code ‘06’ -all encounters for all cases prescribed an anti-malarial at any time Respiratory Medicine -all encounters for all cases with a SSc diagnosis at an index visit -specialty code ‘49’ -beginning January 1, 2006  212  Additional SARDs-Attributable MSP Claims Included in the Broad Definition, by Fee Item Test/ Fee Item Procedure X-ray  chest  08551-08557  hand  08525  foot  08535  hip  08530  spine  08540-08543 08549  CT scan  08690-08696  Bone scan  09834  Sialogram  00723 008510  Ultrasounds  08638 08644 08648-08650 08658 08662 08670  Pulmonary function tests  00928-00946  Endoscopy  Renal biopsy  00907-00909 07780-07783 10740-10744 02357 10735 00742 08112 10912  Muscle biopsy  03211  Skin biopsy  13600  Lung biopsy  00739  Nerve biopsy  07028  Electromyography  00900-00902 00923  Dialysis  10903 33708 33723 33750-33759 33761 77380  213  Test/ Procedure Laboratory investigations  Fee Item haematology  immunology  virology  90020 90025 90027 90029 90030 90035 90039 90040 90046 90047 90050 90055 90060 90063 90065 90068 90070 90072 90073 90080 90085 90090 90095 90100 90105 90110 90115 90120 90130 90135  90140 90145 90150 90155 90160 90165 90170 90175 90180 90185 90190 90200 90205 90210 90215 90220 90225 90235 90240 90245 90265 90280 90285 90290 90295 90300 90305 90310 90315 90320  90038-90040 90045-90047 91840 91811 91845 91812 91800 91813 91801 91814 91802 91815 91803 91821 91805 91822 91810 91825 90675 90831 90690 00030 90700 90815 91765 90820  90325 90330 90335 90340 90345 90350 90355 90357 90360 90365 90370 90375 90377 90380 90385 90390 90400 90405 90410 90415 90420 90425 90427 90430 90435 90440 90445 90450 90455 90460  90465 90475 90480 90485 90490 90495 90500 90505 90510 90512 90515 90520 90525 90530 90535 90540 90545 90550 90555 90560 90565  91830 91831 91835 91850 91855 91856 91857 91858 90825 90833 90830 90831 90832  214  Test/ Procedure  Fee Item urinalysis, urine microscopy, urine culture  92367 92369 92375 92378 92382 92385 92390 92391 92395 92396 92400 92405 92406  general chemistry  91000 91005 91005 91010 91020 91021 91023 91025 91027 91030 91031 91035 91036 91037 91040 91042 91050 91055 91060 91061 91065 91070 91075 91130 91145 91146 91160 91170 91175 91180 91185 91190 91191 91195 91196 91200 91201 91205  91206 91210 91215 91216 91220 91221 91225 91226 91227 91228 91230 91231 91232 91235 91236 91240 91245 91246 91250 91260 91260 91265 91265 91270 91270 91275 91275 91280 91295 91300 91305 91310 91315 91320 91325 91326 91327 91328  91330 91335 91340 91345 91350 91351 91352 91353 91355 91356 91360 91365 91366 91367 91368 91369 91370 91375 91386 91387 91388 91400 91401 91402 91405 91406 91410 91415 91420 91421 91422 91425 91430 91434 91435 91440 91445 91450  91455 91460 91465 91470 91475 91482 91484 91486 91488 91490 91492 91494 91496 91498 91500 91502 91504 91506 91508 91510 91512 91514 91516 91518 91520 91522 91523 91524 91526 91528 91529 91530 91532 91536  215  Test/ Procedure  Fee Item 91538 91540 91542 91544 91546 91548 91550 91551 91552 91554 91556 91558 91559 91695 91705 91706 91707 91708 91709 91710 91715 91715 91716 91717 91720 91725 91936 91940 91941 91945 91946 91950 91955  91560 91561 91562 91564 91565 91566 91568 91570 91572 91573 91574 91575 91576 91730 91735 91740 91745 91750 91760 91761 91762 91770 91775 91777 91780 91785 91956 91957 91958 91959 91960 91965 91970  91599 91600 91601 91602 91603 91605 91610 91615 91620 91630 91631 91635 91636 91790 91795 91796 91860 91865 91870 91880 91881 91882 91885 91890 91895 91896 91975 91985 91990 91992 91995 91997 92000  91640 91645 91650 91660 91665 91666 91670 91675 91680 91681 91682 91685 91690 91900 91901 91902 91905 91910 91911 91912 91915 91920 91925 91930 91935  216  Test/ Procedure  Fee Item 92015 92016 92020 92025 92026 92030 92031 92035 92040 92045 92050 92055 92056 92060 92065 92070 92071 92072 92075 92080 92085 92090 92091 92092 92095  serum proteins  92100 92101 92102 92103 92105 92108 92110 92115 92120 92125 92130 92131 92135 92145 92146 92147 92148 92149 92150 92152 92155 92156 92157 92160 92165  92170 92180 92185 92190 92195 92197 92200 92201 92202 92203 92204 92205 92210 92215 92220 92225 92227 92230 92231 92232 92233 92235 92236 92240 92250  92251 92255 92260 92263 92266 92267 92270 92275 92280 92280 92285 92290 92305 92311 92315 92320 92325 92330 92332 92335 92340 92345 92346 92350 92351  92500-92515 92520-92531 92535-92550 91285 91290 91390 91395  thrombophilia investigations  90123 90125 90127  muscle enzymes  93115  synovial fluid analysis  92377 93010  217  Classes of Medications Attributable to SARDs Alpha-adrenergic blocking agents  Diuretics  Alpha-glucosidase inhibitors  Estrogens  Angiotensin II receptor antagonists  Fibric acid derivatives  Angiotensin-II converting enzyme (ACE-II) inhibitors Anti-arrhythmic agents Anti-coagulants  Gold compounds  Anti-depressants Anti-inflammatory agents Anti-malarials Anti-neoplastic agents  Insulins Meglitinides Nitrates and nitrites Non-steroidal anti-inflammatory agents  Beta-adrenergic blocking agents  Pilocarpine tablets and drops  Biguanides Calcium-channel blocking agents  Progestins Sulfonamides (systemic)  Cardiotonic agents Central alpha-agonists Contraceptives Corticosteroids (excluding intra-articular injections)  Sulfonylureas Tetracyclines Thiazolidinediones Vasodilating agents, miscellaneous  Drug Class  Heavy metal antagonists HMG-CoA reductase inhibitors  Most Prescribed Drug Categories and Generic Drugs within each Anatomical Therapeutic Chemical (ATC) Class (Second Level) Drug  RAAS Agents  ACE-II inhibitors; angiotension-II receptor antagonists  Analgesics  Morphine; ASA; paracetamol; hydromorphone; oxycodone; fentanyl; pethidine; sumatriptan; clonidine; codeine; butorphanol; tramadol; zolmitriptan Zopiclone; lorazepam; oxazepam; temazepam; quetiapine; diazepam; risperdone; alprazolam; olanzapine Amitriptyline; citalopram; venlafaxine; paroxetine; trazadone; sertraline; fluoxetine Proton pump inhibitors (PPIs); histamine-II receptor antagonists; prostaglandins; calcium carbonate Furosemide; hydrochlorothiazide; spironolactone Hydrocychloroquine; chloroquine; quinine Clonazepam; gabapentin; valproic acid; carbamazepine; phenytoin; topiramate Beta-blockers, anticholinergics, inhaled corticosteroids, leukotriene receptor antagoinists; xanthines Statins, fibrates, ezetimbe, colestyramine, nicotinic acid Azathioprine; ciclosporin; mycophenolate; tacrolimus; leflunomide; etanercept; anakinra; infliximab; sirolimus; adalimumab Digoxin, nitroglycerin, nitrates, anti-arrhythmics Methotrexate, cyclophosphamide, chlorambucil Clonidine; hydralazine; bosentan; prazosin; methyldopa Insulin, metformin, sulphonamides, thiazolidinediones  Psycholeptics Psychoanaleptics Antacids Diuretics Anti-malarials Anti-epileptics Drugs for Obstructive Airway Diseases Lipid-Modifying Agents Immunosuppressants Cardiac Therapies Anti-neoplastics Anti-hypertensives Diabetes Therapies  218  Appendix B: Additional Results Total Direct Medical Costs (2007 Canadian dollars) Year  SARDs  SARDs-CTD  SARDs-VD  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  $26,040,962.03 $29,541,868.27 $30,781,014.36 $35,510,696.81 $38,103,648.72 $38,510,053.79 $42,712,647.25 $50,635,782.60 $55,786,478.03 $65,629,712.25 $76,293,422.85 $81,670,492.92  $20,469,549.41 $23,285,820.58 $24,826,817.95 $28,273,242.91 $30,528,055.05 $31,370,786.30 $36,201,500.95 $42,139,060.40 $47,923,607.97 $55,758,978.35 $63,642,323.57 $69,399,798.03  $4,588,941.75 $5,422,988.34 $4,738,950.32 $5,758,660.04 $5,501,672.33 $4,974,130.71 $4,955,860.13 $6,749,584.07 $6,446,657.22 $7,803,246.05 $9,542,004.26 $9,214,644.21  Overall  $571,216,779.87  $469,854,837.66  $75,697,339.44  %-Change  213.62%  239.04%  100.80%  219  Total Direct Medical Costs, by Component (2007 Canadian dollars) Outpatient  Hospital  Prescription  Year  SARDs  SARDsCTD  SARDsVD  SARDs  SARDs-CTD  SARDs-VD  SARDs  SARDs-CTD  SARDs-VD  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  $6,451,603.32 $7,125,132.47 $7,632,512.63 $8,795,961.10 $9,564,219.98 $10,873,871.75 $12,139,328.11 $14,387,060.94 $16,223,482.58 $18,536,401.66 $20,827,482.73 $22,023,505.40  $5,515,450.59 $6,068,940.07 $6,593,233.24 $7,731,162.72 $8,407,042.29 $9,608,873.33 $10,802,112.48 $12,597,665.04 $14,252,443.27 $16,346,434.18 $18,325,103.99 $19,333,832.18  $817,031.51 $927,131.14 $880,521.36 $889,860.74 $912,609.82 $954,813.12 $973,693.41 $1,355,988.28 $1,540,976.48 $1,698,140.71 $1,923,135.96 $2,111,833.94  $16,322,932.25 $18,655,277.38 $18,647,434.71 $21,259,913.60 $21,842,019.69 $19,815,170.08 $21,397,645.29 $24,625,160.84 $25,264,012.18 $30,646,005.72 $35,942,543.40 $37,246,835.43  $12,070,511.68 $13,889,644.66 $14,203,903.42 $15,607,894.68 $16,019,030.79 $14,632,480.16 $17,022,732.76 $19,021,409.37 $20,706,851.57 $24,656,138.26 $27,868,079.25 $30,033,005.68  $3,455,823.63 $4,144,643.48 $3,472,855.04 $4,441,836.79 $4,116,983.60 $3,481,723.26 $3,389,377.59 $4,547,957.59 $3,879,815.58 $4,802,828.01 $6,037,175.22 $5,193,208.71  $3,266,426.46 $3,761,458.42 $4,501,067.01 $5,454,822.12 $6,697,409.04 $7,821,011.96 $9,175,673.85 $11,623,560.81 $14,298,983.28 $16,447,304.87 $19,523,396.72 $22,400,152.09  $2,883,587.15 $3,327,235.85 $4,029,681.29 $4,934,185.51 $6,101,981.98 $7,129,432.81 $8,376,655.71 $10,519,985.99 $12,964,313.14 $14,756,405.91 $17,449,140.33 $20,032,960.17  $316,086.61 $351,213.72 $385,573.92 $426,962.51 $472,078.91 $537,594.32 $592,789.13 $845,638.20 $1,025,865.17 $1,302,277.34 $1,581,693.08 $1,909,601.56  Overall %Change  $154,580,562.68 $135,582,293.36 $14,985,736.47 $291,664,950.57 $221,766,978.46 $50,964,228.50 $124,971,266.62 $112,505,565.84 $9,747,374.47 241.36%  250.54%  158.48%  128.19%  148.81%  50.27%  585.77%  594.72%  504.14%  220  Crude Annual Mean Per-Patient-Year Outpatient Costs (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $2,205.24 $2,074.71 $3,687.24 1996 $2,201.10 $2,062.39 $3,817.97 1997 $2,112.81 $1,994.88 $3,533.86 1998 $2,085.75 $1,986.08 $3,488.51 1999 $1,978.74 $1,876.33 $3,367.56 2000 $2,018.29 $1,915.81 $3,446.98 2001 $2,005.59 $1,910.38 $3,336.47 2002 $1,974.05 $1,855.96 $3,684.75 2003 $1,854.01 $1,749.89 $3,357.25 2004 $1,786.99 $1,697.63 $2,932.04 2005 $1,729.09 $1,642.57 $2,715.97 2006 $1,641.48 $1,561.04 $2,512.84 2007 Overall $1,881.92 $1,782.75 $3,146.17 -25.56% -24.76% -31.85% %-Change Total Outpatient Encounters Year  SARDs  SARDs-CTD  SARDs-VD  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  98,319 109,690 121,668 143,031 160,944 175,926 178,918 217,620 249,884 290,978 331,931 366,839  84,789 94,467 106,339 126,631 142,916 157,478 160,210 192,473 220,632 256,349 292,333 323,214  11,721 13,152 13,079 13,625 14,357 14,118 14,117 19,207 23,049 27,429 31,343 35,089  Overall  2,445,748  2,157,831  230,286  %-Change  273.11%  281.20%  199.37%  Crude Annual Mean Number of Outpatient Encounters, Per Patient-Year Year SARDs SARDs-CTD SARDs-VD 33.61 31.89 52.90 1996 33.89 32.10 54.16 1997 33.68 32.17 52.49 1998 33.92 32.53 53.41 1999 33.30 31.90 52.98 2000 32.65 31.40 50.97 2001 29.56 28.33 48.37 2002 29.86 28.36 52.19 2003 28.56 27.09 50.22 2004 28.05 26.62 47.36 2005 27.56 26.20 44.26 2006 27.34 26.10 41.75 2007 Overall 29.78 28.37 48.35 -18.64% -18.18% -21.07% %-Change  221  Total Hospital Admissions Year  SARDs  SARDs-CTD  SARDs-VD  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  2,597 2,705 2,916 3,240 3,273 3,273 3,448 4,196 4,670 5,312 5,975 6,450  2,133 2,193 2,420 2,694 2,727 2,764 3,001 3,473 3,957 4,496 5,050 5,533  409 458 434 476 442 416 350 595 567 645 733 719  Overall  48,055  40,441  6,244  %-Change  148.36%  159.40%  75.79%  Total Hospital Admissions by Type SARDs Year Inpatient Day ALC Case 1,947 597 53 1996 1,974 655 76 1997 2,070 772 74 1998 2,280 903 57 1999 2,211 1,032 30 2000 2,209 1,059 5 2001 2,207 1,241 0 2002 2,654 1,541 1 2003 2,854 1,816 0 2004 3,169 2,142 1 2005 3,558 2,417 0 2006 3,630 2,818 2 2007 Overall 30,763 16,993 299 %Change 86.44% 372.03% -96.23%  SARDs-CTD Inpatient Day ALC Case 1,563 537 33 1,575 567 51 1,682 683 55 1,839 815 40 1,766 942 19 1,805 954 5 1,857 1,144 0 2,123 1,349 1 2,367 1,590 0 2,609 1,886 1 2,899 2,151 0 3,024 2,507 2 25,109 15,125 207 93.47%  366.85%  -93.94%  341 356 341 381 368 337 288 459 394 444 521 475 4,705  SARDs-VD Day Case 50 79 74 79 66 79 62 136 173 201 212 244 1,455  39.30%  388.00%  Inpatient  ALC 18 23 19 16 8 0 0 0 0 0 0 0 84 -100.00%  222  Crude Annual Mean Number of Hospital Admissions, Per-Patient-Year Year SARDs SARDs-CTD SARDs-VD 0.89 0.80 1.85 1996 0.84 0.75 1.89 1997 0.81 0.73 1.74 1998 0.77 0.69 1.87 1999 0.68 0.61 1.63 2000 0.61 0.55 1.50 2001 0.57 0.53 1.20 2002 0.58 0.51 1.62 2003 0.53 0.49 1.24 2004 0.51 0.47 1.11 2005 0.50 0.45 1.04 2006 0.48 0.45 0.86 2007 Overall 0.59 0.53 1.31 -45.84% -44.32% -53.65% %-Change Crude Annual Mean Number of Inpatient Admissions, Per-Patient-Year Year SARDs SARDs-CTD SARDs-VD 0.67 0.59 1.54 1996 0.61 0.54 1.47 1997 0.57 0.51 1.37 1998 0.54 0.47 1.49 1999 0.46 0.39 1.36 2000 0.41 0.36 1.22 2001 0.36 0.33 0.99 2002 0.36 0.31 1.25 2003 0.33 0.29 0.86 2004 0.31 0.27 0.77 2005 0.30 0.26 0.74 2006 0.27 0.24 0.57 2007 Overall 0.37 0.33 0.99 -59.35% -58.47% -63.27% %-Change Crude Annual Mean Number of Day Admissions, Per-Patient-Year Year SARDs SARDsSARDs-VD CTD 1996 0.20 0.20 0.23 1997 0.20 0.19 0.33 1998 0.21 0.21 0.30 1999 0.21 0.21 0.31 2000 0.21 0.21 0.24 2001 0.20 0.19 0.29 2002 0.21 0.20 0.21 2003 0.21 0.20 0.37 2004 0.21 0.20 0.38 2005 0.21 0.20 0.35 2006 0.20 0.19 0.30 0.21 0.20 0.29 2007 Overall 0.21 0.20 0.31 %-Change 2.93% 0.21% 24.44%  223  Crude Annual Mean Per-Patient-Year Hospital Costs (2007 Canadian dollars) Year SARDS SARDsSARDs-VD CTD $5,579.38 $4,540.48 $15,596.04 1996 $5,763.00 $4,720.08 $17,067.85 1997 1998 $5,161.92 $4,297.60 $13,937.88 1999 $5,041.27 $4,009.56 $17,413.28 2000 $4,518.88 $3,575.21 $15,191.82 2001 $3,677.87 $2,917.41 $12,569.40 2002 $3,535.19 $3,010.52 $11,614.09 2003 $3,378.83 $2,802.35 $12,358.58 2004 $2,887.15 $2,542.36 $8,452.76 2005 $2,954.40 $2,560.61 $8,292.65 2006 $2,983.94 $2,497.96 $8,526.08 2007 $2,776.13 $2,424.90 $6,179.33 Overall $3,550.83 $2,968.11 $10,699.65 %Change -50.24% -46.59% -60.38% Annual Proportion of Admissions as Inpatient and Day-Case SARDs Year  SARDs-CTD  SARDs-VD  Inpatient Day-Case Inpatient Day-Case Inpatient Day-Case  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  74.97% 72.98% 70.99% 70.37% 67.55% 67.49% 64.01% 63.25% 61.11% 59.66% 59.55% 56.28%  22.99% 24.21% 26.47% 27.87% 31.53% 32.36% 35.99% 36.73% 38.89% 40.32% 40.45% 43.69%  73.28% 71.82% 69.50% 68.26% 64.76% 65.30% 61.88% 61.13% 59.82% 58.03% 57.41% 54.65%  25.18% 25.85% 28.22% 30.25% 34.54% 34.52% 38.12% 38.84% 40.18% 41.95% 42.59% 45.31%  83.37% 77.73% 78.57% 80.04% 83.26% 81.01% 82.29% 77.14% 69.49% 68.84% 71.08% 66.06%  12.22% 17.25% 17.05% 16.60% 14.93% 18.99% 17.71% 22.86% 30.51% 31.16% 28.92% 33.94%  Overall  64.02%  35.36%  62.09%  37.40%  75.35%  23.30%  %Change  -24.93%  90.05%  -25.41%  79.97%  -20.76%  177.60%  -these proportions may not sum to 100% because a third type of admission (ALC) made a minor contribution to the total  224  Total Number of Prescriptions Year  SARDs  SARDs-CTD  SARDs-VD  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  67,202 79,455 91,371 111,652 132,112 157,427 176,069 215,341 263,245 321,856 390,293 458,636  59,250 70,276 81,156 99,038 117,298 138,596 155,282 189,188 231,934 282,068 340,592 398,854  6,755 7,794 8,727 11,010 12,663 15,990 17,228 21,533 26,000 32,327 40,808 50,119  Overall  2,464,659  2,163,532  250,954  %-Change  582.47%  573.17%  641.95%  Crude Annual Mean Number of Prescriptions, Per-Patient-Year Year SARDs SARDs-CTD SARDs-VD 22.97 22.29 30.49 1996 24.55 23.88 32.10 1997 25.29 24.55 35.02 1998 26.48 25.44 43.16 1999 27.33 26.18 46.73 2000 29.22 27.63 57.73 2001 29.09 27.46 59.03 2002 29.55 27.87 58.51 2003 30.08 28.48 56.64 2004 31.03 29.29 55.82 2005 32.40 30.53 57.63 2006 34.18 32.20 59.64 2007 Overall 30.01 28.45 52.69 48.82% 44.49% 95.62% %-Change  225  Annual Mean Number of Prescriptions, Per-Patient-Year (amongst users) Year SARDs SARDs-CTD Mean STD Mean STD 22.38 29.67 22.14 30.09 1996 23.79 34.08 23.41 34.60 1997 23.32 34.74 22.78 34.76 1998 24.39 36.01 23.76 35.65 1999 26.14 41.46 25.34 40.46 2000 27.82 48.57 26.56 44.64 2001 28.11 52.95 26.73 49.52 2002 27.66 54.08 26.51 50.41 2003 28.76 57.19 27.63 54.68 2004 30.20 65.62 28.81 63.94 2005 31.89 74.62 30.40 72.55 2006 34.61 78.73 32.78 75.27 2007 Overall 29.06 60.24 27.85 57.91 48.03% %-Change 54.70%  SARDs-VD Mean STD 24.52 26.33 27.61 29.57 30.26 35.45 33.73 41.93 39.02 56.27 48.51 91.61 54.11 96.88 43.92 94.35 44.10 85.02 48.27 87.05 48.93 92.37 56.34 109.23 44.84 84.75 129.80%  Crude Annual Mean Per-Patient-Year Prescription Costs (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $1,116.51 $1,084.70 $1,426.49 1996 $1,161.99 $1,130.69 $1,446.32 1997 $1,245.97 $1,219.24 $1,547.45 1998 $1,293.48 $1,267.56 $1,673.82 1999 $1,385.62 $1,361.87 $1,741.99 2000 $1,451.65 $1,421.46 $1,940.77 2001 $1,515.95 $1,481.44 $2,031.26 2002 $1,594.87 $1,549.87 $2,297.93 2003 $1,634.08 $1,591.74 $2,235.00 2004 $1,585.59 $1,532.50 $2,248.54 2005 $1,620.83 $1,564.06 $2,233.77 2006 $1,669.56 $1,617.49 $2,272.21 2007 Overall $1,521.44 $1,479.32 $2,046.41 49.53% 49.12% 59.29% %-Change Annual Mean Per-Patient-Year Prescription Costs (amongst users) (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD Mean STD Mean STD Mean STD $1,097.49 1,538.50 $ 1,084.87 1,542.42 $1,179.65 1,466.15 1996 $1,130.65 1,486.48 $ 1,112.85 1,487.11 $1,249.32 1,453.90 1997 $1,151.32 1,655.07 $ 1,132.91 1,671.07 $1,343.32 1,457.38 1998 $1,200.02 1,793.27 $ 1,190.41 1,819.90 $1,316.20 1,467.37 1999 $1,333.52 1,895.72 $ 1,322.80 1,921.55 $1,493.53 1,608.28 2000 $1,398.23 1,969.33 $ 1,381.74 1,978.57 $1,633.28 1,871.81 2001 $1,474.37 2,215.64 $ 1,451.57 2,236.46 $1,847.23 1,914.96 2002 $1,506.73 2,674.41 $ 1,485.36 2,681.69 $1,740.47 2,755.12 2003 $1,576.35 3,017.10 $ 1,554.30 3,053.26 $1,787.82 2,545.91 2004 $1,549.70 2,689.77 $ 1,511.13 2,685.12 $1,950.07 2,795.29 2005 $1,606.94 2,991.50 $ 1,565.64 2,944.59 $1,938.84 3,336.52 2006 $1,694.23 3,334.38 $ 1,649.17 3,329.24 $2,157.78 3,478.33 2007 Overall $1482.62 2632.61 $1455.16 2633.49 $1759.80 2643.31 54.37% 52.02% 82.92% %-Change  226  Top-Five Most Frequently-Prescribed Drugs, by Year SARDs SARDs-CTD Year Drug Quantity Drug Acetaminophen1996 Prednisone 3916 codeine(30mg) Acetaminophencodeine(30mg) 3599 Prednisone Hydroxychloroquine 1703 Hydroxychloroquine Omeprazole 1579 Omeprazole Lorazepam 1396 Lorazepam Acetaminophen1997 Prednisone 4268 codeine(30mg) Acetaminophencodeine(30mg) 3988 Prednisone  2002  2003  Quantity 677  3111 1667 1423 1296  Prednisone Acetaminophencodeine(30mg) Furosemide Digoxin Warfarin  3737  Prednisone  700  3433  288  287 222 159 144  2135 1650 1581  4626  Hydroxychloroquine Omeprazole Lorazepam Acetaminophencodeine(30mg)  3925  Prednisone  791  4187 2766 2093  Prednisone Hydroxychloroquine Omeprazole  3716 2731 1865  363 212 199  Levothyroxine  1891  1725  Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Omeprazole Levothyroxine Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Omeprazole Levothyroxine Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Levothyroxine Omeprazole Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Levothyroxine Omeprazole Prednisone  5604  Lorazepam Acetaminophencodeine(30mg)  Furosemide Digoxin Salbutamol Acetaminophencodeine(30mg)  4605  Prednisone  973  Prednisone Hydroxychloroquine Omeprazole Levothyroxine Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Omeprazole Levothyroxine Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Levothyroxine Omeprazole Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Levothyroxine Methadone Prednisone  4505 3471 2313 2169 5038  Furosemide Omeprazole Etidronate Digoxin Prednisone  447 273 269 249 1015  4902 4272 2646 2610 5525  Furosemide Omeprazole Ramipril Hydrochlorothiazide Prednisone  457 314 308 294 1184  5203 4925 3358 3105 6138  Furosemide Ramipril Etidronate Levothyroxine Prednisone  505 441 367 349 1148  5699 5416 3775 3025 7177  Furosemide Levothyroxine Ramipril Etidronate Prednisone  518 505 421 416 1608  Hydroxychloroquine Omeprazole Lorazepam  2175 1865 1717  Prednisone Acetaminophencodeine(30mg) Hydroxychloroquine Omeprazole  1999  2001  3284  SARDs-VD Drug  Furosemide Acetaminophencodeine(30mg) Digoxin Omeprazole  1998  2000  Quantity  4840 3527 2626 2404 6213 5199 4351 3000 2892 6916 5532 5007 3770 3495 7553 5990 5495 4345 3359 9208  218 192 180  198  227  SARDs Year  Drug  Acetaminophencodeine(30mg)  6720  SARDs-CTD Quantity  Drug  SARDs-VD Quantity  Drug  Hydroxychloroquine Acetaminophencodeine(30mg) Levothyroxine Ramipril Prednisone  6629  Furosemide  736  6312 4763 3417 8056  Levothyroxine Ramipril Etidronate Prednisone  584 518 422 2004  Hydroxychloroquine Acetaminophencodeine(30mg) Levothyroxine Rabeprazole Prednisone  7433  Rabeprazole  938  7392 5993 4540 9247  Levothyroxine Furosemide Etidronate Prednisone  876 588 506 2220  Hydroxychloroquine Acetaminophencodeine(30mg) Levothyroxine Rabeprazole  8646  Rabeprazole  1172  8459 7258  Levothyroxine Ramipril Furosemide  1088 771  2003  2004  2005  2006  2007  Hydroxychloroquine Levothyroxine Ramipril Prednisone Acetaminophencodeine(30mg)  6714 5415 4010 10,476  Hydroxychloroquine Levothyroxine Rabeprazole Prednisone Acetaminophencodeine(30mg)  7549 6956 5602 12041  Hydroxychloroquine Levothyroxine  8790 8550  Rabeprazole Prednisone Levothyroxine Hydroxychloroquine Acetaminophencodeine(30mg) Rabeprazole Prednisone Levothyroxine Hydroxychloroquine  6959 13,432 10,711 10,217  7828  9004  9397 9156 15,049 12,971 11,801 11,739  Rabeprazole Ramipril  10,535  Prednisone Hydroxychloroquine Levothyroxine Acetaminophencodeine(30mg) Rabeprazole Hydroxychloroquine Prednisone Levothyroxine Rabeprazole Acetaminophencodeine(30mg)  5646 10,102 10,046 9148  Prednisone Levothyroxine Rabeprazole  759 2723 1309 1250  8674 7763 11,592 11,367 11,185  Ramipril Furosemide Prednisone Rabeprazole Levothyroxine  1127 1107 3079 1749 1514  9797  Ramipril  1496  9402  Furosemide  1284  228  Year  SARDs Drug  Total Annual Cost  Top-Ten Most Costly Drugs, by Year (2007 Canadian Dollars) SARDs-CTD Total Annual Drug Total Total Annual Quantity Annual Cost Quantity  SARDs-VD Drug  Total Annual Cost  Total Annual Quantity  1996 Omeprazole  $266,829.51  1579  Omeprazole  $ 239,729.02  1423  Omeprazole  $ 23,062.27  143  Azathioprine  $ 142,453.56  1027  Azathioprine  $ 132,167.03  945  Enalapril  $ 10,651.83  111  Hydroxychloroquine  $ 131,872.84  1703  Hydroxychloroquine  $ 128,755.45  1667  Diltiazem  $  9,495.63  62  Enalapril  $ 87,412.46  868  Nifedipine  $ 79,113.47  759  Cyclosporine  $  9,418.54  7  Nifedipine  $ 86,304.60  842  Enalapril  $ 76,504.65  753  Cyclophosphamide  $  8,942.85  119  Cisapride  $ 78,083.97  662  Cisapride  $ 70,895.80  593  Calcitonin  $  7,992.55  30  Diltiazem  $ 74,087.17  485  Diltiazem  $ 63,197.52  417  Prednisone  $  7,469.05  677  Ranitidine  $ 53,493.69  956  Ipratropium  $ 45,372.21  380  Ranitidine  $  7,235.77  124  Ipratropium  $ 50,801.18  455  Ranitidine  $ 45,227.55  818  Nitroglycerine  $  6,647.90  118  Prednisone  $ 46,861.19  3916  Morphine  $ 40,356.68  419  Warfarin  $  6,608.12  144  Omeprazole  $ 315,730.99  1865  Omeprazole  $ 280,270.93  1650  Omeprazole  $ 27,926.50  180  Hydroxychloroquine  $ 167,077.95  2175  Hydroxychloroquine  $ 163,694.16  2135  Enalapril  $ 14,528.97  158  Azathioprine  $ 153,395.47  1190  Azathioprine  $ 141,575.28  1097  Cyclosporine  $ 10,668.68  11  Enalapril  $ 86,725.80  884  Cisapride  $ 72,388.34  672  Budesonide  $  9,087.80  55  Cisapride  $ 82,480.26  781  Nifedipine  $ 71,798.26  732  Ranitidine  $  8,956.23  147  Nifedipine Cyclosporine Diltiazem  $ 78,542.37 $ 78,305.95 $ 65,242.59  804 182 571  Enalapril Cyclosporine Paroxetine  $ 71,523.21 $ 67,637.27 $ 58,655.66  717 171 573  $ $ $  7,971.11 7,950.80 7,448.55  103 94 700  Ranitidine Paroxetine Omeprazole  $ 63,008.40 $ 61,285.76 $ 377,606.98  1111 600 2093  Diltiazem Ranitidine Omeprazole  $ 57,632.98 $ 51,934.81 $ 337,425.03  508 938 1865  Cyclophosphamide Cisapride Prednisone Hematopoietic agents Nitroglycerine Omeprazole  $ 6,670.08 $ 6,665.60 $ 32,388.14  7 130 187  1997  1998  229  Year  1998  1999  2000  2001  SARDs Drug  Hydroxychloroquine Azathioprine Morphine Cisapride Nifedipine Paroxetine Enalapril Cyclosporine Sertraline Omeprazole Hydroxychloroquine Azathioprine Morphine Paroxetine Cisapride Etidronate Sertraline Cyclosporine Nifedipine Omeprazole Hydroxychloroquine Azathioprine Morphine Paroxetine Cyclosporine Ramipril Etidronate Celecoxib Pantoprazole Omeprazole Hydroxychloroquine Ramipril Morphine Atorvastatin Pantoprazole Azathioprine Paroxetine Cyclosporine  Total Annual Cost $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $  221,072.62 151,717.12 124,320.62 100,493.02 83,543.38 82,309.71 81,635.51 79,456.25 74,761.39 447,576.00 296,164.91 139,631.48 111,080.26 104,056.83 97,929.67 91,814.27 85,737.65 83,255.51 83,138.72 505,440.69 347,812.00 140,281.15 137,256.36 128,974.63 114,232.91 111,751.18 111,385.65 103,776.72 103,352.07 550,499.83 397,116.88 164,936.96 164,187.63 159,575.06 151,288.65 151,213.28 145,809.27 125,127.89  Total Annual Quantity 2766 1446 974 908 847 733 788 276 630 2626 3527 1489 1212 991 875 1603 883 429 908 3000 4351 1571 1388 1238 460 1465 2019 1334 803 3495 5007 2492 2110 1042 1269 1821 1420 381  SARDs-CTD Drug  Hydroxychloroquine Azathioprine Morphine Cisapride Nifedipine Paroxetine Cyclosporine Sertraline Enalapril Omeprazole Hydroxychloroquine Azathioprine Morphine Paroxetine Cisapride Sertraline Cyclosporine Nifedipine Etidronate Omeprazole Hydroxychloroquine Morphine Azathioprine Paroxetine Cyclosporine Celecoxib Ramipril Pantoprazole Nifedipine Omeprazole Hydroxychloroquine Morphine Azathioprine Ramipril Atorvastatin Pantoprazole Paroxetine Cyclosporine  Total Annual Cost $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $  218,476.02 143,432.36 117,386.62 87,671.96 78,789.41 75,636.48 72,521.02 66,831.27 65,839.13 404,180.78 292,300.54 131,736.00 105,581.40 95,695.71 86,105.32 79,032.96 78,999.55 78,345.80 75,124.18 459,896.81 342,483.56 132,454.88 132,079.57 117,396.43 104,929.82 98,837.10 94,100.11 93,131.90 92,938.22 508,064.80 391,547.89 156,587.54 141,246.73 140,603.91 140,393.85 136,636.37 131,777.00 115,385.58  Total Annual Quantity 2731 1356 873 776 798 667 257 553 622 2313 3471 1382 1133 889 750 813 359 849 1286 2646 4272 1324 1463 1059 398 1218 1128 718 984 3105 4925 1968 1656 2003 906 1135 1189 314  SARDs-VD Drug  Enalapril Cisapride Budesonide Etidronate Alendronate Amlodipine Prednisone Nitroglycerine Ranitidine Omeprazole Etidronate Budesonide Amlodipine Cyclophosphamide Enalapril Cisapride Prednisone Salbutamol Ipratropium Omeprazole Ramipril Etidronate Enalapril Paroxetine Budesonide Prednisone Alendronate Nitroglycerine Atorvastatin Omeprazole Ramipril Etidronate Atorvastatin Paroxetine Alendronate Prednisone Budesonide Pantoprazole  Total Annual Cost $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $  15,133.81 10,341.80 8,972.45 8,913.37 8,581.19 8,288.34 7,776.48 7,757.49 7,529.79 36,640.78 13,820.19 9,939.63 9,841.43 9,678.68 9,516.23 9,463.86 8,406.98 8,133.40 7,944.25 37,053.78 15,247.14 14,359.60 12,225.11 11,236.07 9,964.50 9,596.49 9,101.81 8,969.89 8,220.34 34,321.05 19,947.59 15,684.08 14,158.25 12,373.19 12,117.64 11,693.90 10,971.61 10,828.58  Total Annual Quantity 160 116 38 165 67 55 791 162 143 273 269 52 78 156 167 110 973 233 148 314 308 275 194 176 93 1015 99 228 64 339 441 367 108 221 171 1184 81 108 230  Year  SARDs Drug  Total Annual Cost  Total Annual Quantity  SARDs-CTD Drug  Total Annual Cost  Total Annual Quantity  2002  2003  2004  2005  Total Annual Cost  Total Annual Quantity  Enalapril  $  9,899.31  127  $ $ $ $ $ $ $ $  108,525.69 485,826.31 444,914.54 192,001.71 179,556.71 177,455.22 159,873.22 153,999.02  1179 2897 5416 1338 2564 1525 1977 1821  Omeprazole Ramipril Atorvastatin Etidronate Pantoprazole Prednisone Alendronate  $ $ $ $ $ $ $  37,234.57 22,806.10 18,891.68 16,424.41 15,571.74 13,480.24 13,198.81  402 421 170 416 189 1148 233  $ $ $ $ $ $ $  144,125.74 140,235.29 131,983.51 476,852.67 438,845.17 379,886.39 267,584.75  281 1311 29 6629 2696 89 1979  Cyclosporine Paroxetine Nifedipine Omeprazole Cyclosporine Ramipril Atorvastatin  $ $ $ $ $ $ $  10,636.49 10,133.11 9,411.06 37,120.24 35,425.20 28,281.56 26,626.81  70 239 96 377 111 518 228  $ $ $ $ $ $ $ $ $  248,754.93 237,860.23 212,391.48 179,902.70 177,566.27 173,454.45 590,581.21 480,489.67 377,895.59  452 3417 1794 618 1908 2038 138 7433 2689  21,628.27 19,269.61 18,860.94 18,677.81 18,037.31 17,494.14 43,247.80 42,293.67 41,257.87  76 1608 228 4 11 422 107 325 938  $ $ $ $ $ $ $ $  354,178.99 314,541.12 301,513.43 298,486.33 232,293.86 228,747.47 223,917.72 497,718.91  580 2224 4173 4540 2317 692 2103 125  Fentanyl Prednisone Pantoprazole Infliximab Anakinra Etidronate Cyclosporine Atorvastatin Rabeprazole Mycophenolate mofetil Ramipril Alendronate Prednisone Pantoprazole Omeprazole Interferon beta Atorvastatin  $ $ $ $ $ $ $ $ $  671 5602 4814 2461 854 2655 2886 4263  Nifedipine Omeprazole Hydroxychloroquine Atorvastatin Ramipril Pantoprazole Morphine Azathioprine Mycophenolate mofetil Paroxetine Bosentan Hydroxychloroquine Omeprazole Bosentan Atorvastatin Mycophenolate mofetil Ramipril Pantoprazole Cyclosporine Alendronate Azathioprine Bosentan Hydroxychloroquine Atorvastatin Mycophenolate mofetil Omeprazole Ramipril Rabeprazole Gabapentin Cyclosporine Venlafaxine Bosentan  $ $ $ $ $ $ $ $  35,720.54 29,528.79 26,013.90 23,675.09 23,300.47 22,575.38 22,342.09 61,412.79  65 486 437 2004 243 184 4 599  898  Hydroxychloroquine  $ 481,828.33  8646  Rabeprazole  $ 52,356.58  1172  2001 Etidronate Omeprazole Hydroxychloroquine Atorvastatin Ramipril Pantoprazole Morphine Azathioprine Mycophenolate mofetil Paroxetine Cyclosporine Omeprazole Hydroxychloroquine Bosentan Atorvastatin Mycophenolate mofetil Ramipril Pantoprazole Cyclosporine Alendronate Azathioprine Bosentan Hydroxychloroquine Atorvastatin Mycophenolate mofetil Rabeprazole Ramipril Omeprazole Cyclosporine Gabapentin Alendronate Atorvastatin Mycophenolate mofetil  SARDs-VD Drug  $ $ $ $ $ $ $ $  121,699.02 532,836.19 450,626.47 216,920.89 209,119.46 196,514.17 169,792.63 164,975.79  2448 3359 5495 1536 3085 1758 2185 1972  $ $ $ $ $ $ $  154,702.14 152,834.39 150,552.39 487,453.57 483,046.66 379,886.39 301,951.54  313 1570 489 3154 6714 89 2253  $ $ $ $ $ $ $ $ $  275,160.41 275,082.54 234,116.59 222,405.01 199,865.87 185,354.25 590,581.21 486,998.12 427,752.93  493 4010 2049 749 2290 2203 138 7549 3046  $ $ $ $ $ $ $ $  404,825.50 349,077.79 343,939.52 343,412.02 284,878.52 256,176.09 248,588.38 538,375.38  $ 499,804.68  231  Year  SARDs Drug  2005  2006  2007  Total Annual Cost  Total Annual Quantity  Bosentan  $ 497,718.91  125  Hydroxychloroquine  $ 488,313.95  8790  Rabeprazole Ramipril Gabapentin Omeprazole Venlafaxine Alendronate Bosentan Mycophenolate mofetil Atorvastatin Rabeprazole  $ $ $ $ $ $ $  440,418.57 425,639.47 329,759.15 311,260.59 293,681.86 279,380.51 733,224.07  6959 6386 3657 2664 3016 3584 183  $ 657,059.20 $ 643,066.91 $ 546,776.07  1189 5266 9156  Hydroxychloroquine Ramipril Gabapentin Venlafaxine Pantoprazole Omeprazole Bosentan Mycophenolate mofetil Atorvastatin  $ $ $ $ $ $ $  541,807.54 520,176.79 397,213.55 393,588.50 324,815.64 300,061.87 822,591.17  10217 8776 4848 4317 2989 2656 250  $ 773,937.04 $ 724,704.13  Rabeprazole Hydroxychloroquine Gabapentin Ramipril Oxycodone Pantoprazole Venlafaxine  $ $ $ $ $ $ $  669,745.82 595,597.51 483,041.48 462,368.63 373,931.39 350,602.06 349,246.53  SARDs-CTD Drug  Total Annual Cost  Total Annual Quantity  Atorvastatin  $ 468,517.57  3606  Mycophenolate mofetil  $ 451,109.02  789  Rabeprazole Ramipril Gabapentin Omeprazole Venlafaxine Oxycodone Bosentan Mycophenolate mofetil Atorvastatin Hydroxychloroquine  $ $ $ $ $ $ $  378,967.35 370,468.35 298,262.69 284,350.37 277,153.00 246,119.77 733,224.07  5646 5389 3201 2392 2831 3534 183  $ 584,665.64 $ 552,205.19 $ 534,805.32  1023 4382 10046  $ $ $ $ $ $ $  477,548.57 448,527.10 373,244.27 359,847.07 285,927.65 278,808.78 822,591.17  7763 7374 4028 4201 4121 2421 250  1475 6473  Rabeprazole Ramipril Venlafaxine Gabapentin Oxycodone Pantoprazole Bosentan Mycophenolate mofetil Atorvastatin  $ 667,532.84 $ 625,553.36  1199 5324  11739 11801 6135 10535 5690 3380 5266  Hydroxychloroquine Rabeprazole Gabapentin Ramipril Oxycodone Venlafaxine Etanercept  $ $ $ $ $ $ $  11592 9797 5174 8761 5456 4741 197  588,134.32 580,439.23 433,902.66 396,990.53 364,154.37 327,751.28 314,540.58  SARDs-VD Drug  Total Annual Cost  Total Annual Quantity  Etanercept  $ 47,809.42  31  Ramipril Mycophenolate mofetil Alendronate Cyclosporine Pantoprazole Omalizumab Prednisone Atorvastatin  $ 40,320.10  771  $ $ $ $ $ $ $  37,264.17 33,788.82 31,461.79 31,045.56 26,327.39 26,203.86 78,756.03  78 552 71 306 11 2220 804  Rabeprazole Ramipril Omalizumab Mycophenolate mofetil Etanercept Pantoprazole Alendronate Prednisone Amlodipine Atorvastatin  $ 59,977.64 $ 55,361.79 $ 51,367.87  1250 1127 26  $ $ $ $ $ $ $  42,822.78 38,469.55 37,332.30 33,755.66 30,339.74 29,103.73 89,179.82  88 21 421 655 2723 416 1032  Rabeprazole Hydromorphone Mycophenolate mofetil Ramipril Omalizumab Alendronate Clopidogrel Amlodipine Pantoprazole  $ 77,963.92 $ 57,747.91  1749 323  $ $ $ $ $ $ $  132 1496 21 964 521 547 442  57,245.95 53,984.19 43,575.60 37,174.12 36,179.45 36,159.89 35,807.57  232  Total Attributable Overall Direct Medical Costs - Narrow Definition (2007 Canadian dollars) SARDs SARDs-CTD SARDs-VD Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Overall %-Change  $5,798,955.67 $6,431,909.74 $7,732,149.20 $8,153,189.54 $9,562,438.79 $10,886,271.22 $12,330,359.85 $15,472,428.75 $17,447,450.21 $20,097,048.53 $24,360,944.79 $26,892,535.98 $165,165,682.27 363.75%  $4,877,152.89 $5,289,189.37 $6,485,422.41 $6,808,247.61 $7,989,187.31 $9,410,376.92 $10,919,025.07 $13,053,182.51 $15,527,454.49 $17,475,367.92 $20,972,226.81 $23,235,260.82 $141,817,033.17 376.41%  $727,978.01 $961,781.57 $919,981.38 $1,088,175.80 $1,143,509.55 $963,193.82 $986,779.62 $1,846,963.47 $1,442,560.52 $1,877,319.93 $2,637,971.01 $2,768,065.17 $17,364,279.84 280.24%  Total Attributable Overall Direct Medical Costs - Broad Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $12,289,281.93 $9,597,051.39 $2,305,919.75 1996 $14,644,404.25 $11,574,082.60 $2,768,422.44 1997 $14,356,046.43 $11,585,414.07 $2,258,421.75 1998 $16,490,927.41 $13,092,065.99 $2,806,986.45 1999 $17,321,546.90 $14,291,784.87 $2,293,009.04 2000 $18,547,268.34 $15,488,738.44 $2,255,787.72 2001 $20,251,613.93 $17,045,289.67 $2,341,088.43 2002 $24,250,718.08 $19,904,779.52 $3,240,471.85 2003 $25,325,486.47 $21,755,374.54 $2,842,079.53 2004 $29,209,786.96 $24,712,844.19 $3,484,157.40 2005 $34,479,836.71 $28,314,073.58 $4,804,427.39 2006 $35,873,108.94 $30,615,699.00 $3,799,084.30 2007 Overall $263,040,026.37 $217,977,197.84 $35,199,856.04 %191.91% 219.01% 64.75% Change  233  Total Attributable Dispensed Prescriptions Year SARDs SARDs-CTD 67,182 59,238 1996 79,434 70,263 1997 91,311 81,111 1998 111,512 98,927 1999 131,918 117,135 2000 157,013 138,259 2001 175,380 154,711 2002 214,301 188,505 2003 261,288 230,577 2004 318,951 279,960 2005 386,231 337,501 2006 454,041 395,524 2007 Overall 2,448,562 2,151,711 %575.84% 567.69% Change  SARDs-VD 6,747 7,786 8,712 10,981 12,633 15,918 17,117 21,209 25,478 31,620 39,897 48,897 246,995 624.72%  Total Attributable Prescription Costs (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $3,266,189.05 $2,883,381.14 $316,055.22 1996 $3,761,281.34 $3,327,086.58 $351,185.92 1997 $4,500,362.10 $4,029,046.42 $385,503.89 1998 $5,453,068.30 $4,932,494.79 $426,899.42 1999 $6,693,042.01 $6,097,891.67 $471,910.08 2000 $7,815,860.31 $7,124,775.55 $537,135.36 2001 $9,167,617.09 $8,369,378.83 $592,283.16 2002 $11,613,748.48 $10,511,419.38 $844,432.39 2003 $14,287,537.20 $12,954,657.15 $1,024,329.88 2004 $16,432,327.70 $14,743,558.74 $1,300,504.78 2005 $19,507,039.41 $17,435,050.03 $1,579,580.27 2006 $22,382,072.98 $20,017,273.67 $1,907,545.24 2007 Overall $124,880,145.99 $112,426,013.93 $9,737,365.60 %585.27% 594.23% 503.55% Change  Crude Annual Mean Per-Patient-Year Attributable Prescriptions Year SARDs SARDs-CTD SARDs-VD 22.96 22.28 30.45 1996 24.54 23.88 32.06 1997 25.28 24.54 34.96 1998 26.44 25.41 43.05 1999 27.29 26.14 46.62 2000 29.14 27.57 57.47 2001 28.98 27.36 58.65 2002 29.40 27.77 57.63 2003 29.86 28.31 55.51 2004 30.75 29.07 54.60 2005 32.06 30.25 56.35 2006 33.84 31.94 58.18 2007 Overall 29.81 28.29 51.86 %47.37% 43.31% 91.08% Change 234  Crude Annual Mean Per-Patient-Year Attributable Prescription Costs (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $1,116.42 $1,084.62 $1,426.35 1996 $1,161.94 $1,130.64 $1,446.20 1997 $1,245.77 $1,219.05 $1,547.17 1998 $1,293.06 $1,267.12 $1,673.57 1999 $1,384.72 $1,360.96 $1,741.37 2000 $1,450.69 $1,420.53 $1,939.12 2001 $1,514.62 $1,480.15 $2,029.53 2002 $1,593.53 $1,548.60 $2,294.65 2003 $1,632.77 $1,590.55 $2,231.66 2004 $1,584.14 $1,531.16 $2,245.48 2005 $1,619.47 $1,562.79 $2,230.78 2006 $1,668.21 $1,616.22 $2,269.76 2007 Overall $1,520.33 $1,478.27 $2,044.31 %49.42% 49.01% $1,426.35 Change Total Attributable Outpatient Encounters - Narrow Definition Year SARDs SARDs-CTD SARDs-VD 6,250 5,907 194 1996 6,469 6,042 264 1997 7,697 7,071 380 1998 10,138 9,309 507 1999 10,828 9,870 616 2000 11,592 10,690 455 2001 12,991 11,994 544 2002 16,934 15,328 864 2003 18,393 16,629 1,066 2004 19,623 17,811 1,156 2005 21,426 19,158 1,467 2006 22,075 19,688 1,628 2007 Overall 164,416 149,497 9,141 %253.20% 233.30% 739.18% Change Total Attributable Outpatient Encounters - Broad Definition Year SARDs SARDs-CTD SARDs-VD 11,065 10,185 625 1996 13,240 12,026 914 1997 15,997 14,567 1,029 1998 20,103 18,397 1,198 1999 22,579 20,649 1,332 2000 24,901 22,938 1,208 2001 28,299 26,058 1,381 2002 35,445 32,116 2,060 2003 38,822 35,206 2,395 2004 43,546 39,617 2,707 2005 48,923 44,386 3,163 2006 56,806 50,387 4,754 2007 Overall 359,726 326,532 22,766 %413.38% 394.72% 660.64% Change 235  Total Attributable Outpatient Costs - Narrow Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $343,184.52 $324,657.28 $11,424.98 1996 $361,451.70 $336,609.99 $16,009.68 1997 $434,286.58 $395,698.20 $24,646.63 1998 $584,602.10 $533,359.00 $32,619.64 1999 $619,502.20 $559,560.37 $39,074.01 2000 $727,971.60 $668,203.17 $32,333.87 2001 $834,513.59 $770,306.79 $34,539.50 2002 $1,092,989.26 $990,986.95 $55,797.74 2003 $1,144,774.30 $1,039,521.79 $63,465.46 2004 $1,228,712.73 $1,114,624.60 $73,188.94 2005 $1,430,693.98 $1,280,776.42 $97,445.03 2006 $1,513,630.92 $1,350,415.71 $110,584.18 2007 Overall $10,316,313.48 $9,364,720.27 $591,129.66 %341.05% 315.95% 867.92% Change Total Attributable Outpatient Costs- Broad Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $716,436.33 $662,028.79 $40,699.29 1996 $833,568.84 $755,284.60 $58,865.36 1997 $997,639.47 $902,402.83 $68,222.93 1998 $1,265,175.78 $1,153,398.95 $80,473.06 1999 $1,419,331.88 $1,291,248.41 $90,073.28 2000 $1,654,128.56 $1,519,431.81 $85,342.87 2001 $1,923,188.28 $1,768,972.22 $91,274.35 2002 $2,411,392.71 $2,192,173.02 $137,046.00 2003 $2,583,884.72 $2,350,157.00 $156,122.54 2004 $2,912,543.95 $2,649,349.87 $182,028.50 2005 $3,356,567.23 $3,052,051.39 $210,881.98 2006 $3,690,474.48 $3,284,301.15 $297,208.57 2007 Overall $23,764,332.23 $21,580,800.04 $1,498,238.73 %415.12% 396.10% 630.25% Change Crude Annual Mean Per-Patient-Year Attributable Outpatient Encounters – Narrow Definition Year SARDs SARDs-CTD SARDs-VD 2.14 2.22 0.88 1996 2.00 2.01 1.09 1997 2.13 1.83 1.53 1998 2.40 1.82 1.99 1999 2.24 2.08 2.27 2000 2.15 1.97 1.64 2001 2.15 1.89 1.86 2002 2.32 1.77 2.35 2003 2.10 1.88 2.32 2004 1.89 1.73 2.00 2005 1.78 1.60 2.07 2006 1.65 1.55 1.94 2007 Overall 2.00 1.86 1.72 %-22.98% -30.39% 121.26% Change 236  Crude Annual Mean Per-Patient-Year Attributable Outpatient Costs – Narrow Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $117.30 $122.12 $51.56 1996 $111.66 $114.39 $65.93 1997 $120.22 $119.72 $98.92 1998 $138.62 $137.02 $127.88 1999 $128.17 $124.89 $144.18 2000 $135.12 $133.23 $116.73 2001 $137.87 $136.23 $118.35 2002 $149.97 $146.00 $151.62 2003 $130.82 $127.63 $138.27 2004 $118.45 $115.76 $126.37 2005 $118.78 $114.80 $137.62 2006 $112.82 $109.03 $131.58 2007 Overall $125.59 $123.14 $124.10 %-3.83% -10.72% 155.20% Change Crude Annual Mean Per-Patient-Year Attributable Outpatient Encounters – Broad Definition Year SARDs SARDs-CTD SARDs-VD 3.78 3.83 2.82 1996 4.09 4.09 3.76 1997 4.43 4.41 4.13 1998 4.77 4.73 4.70 1999 4.67 4.61 4.92 2000 4.62 4.57 4.36 2001 4.68 4.61 4.73 2002 4.86 4.73 5.60 2003 4.44 4.32 5.22 2004 4.20 4.11 4.67 2005 4.06 3.98 4.47 2006 4.23 4.07 5.66 2007 Overall 4.38 4.29 4.78 %11.95% 6.19% 100.55% Change Crude Annual Mean Per-Patient-Year Attributable Outpatient Costs – Broad Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $244.89 $249.03 $183.67 1996 $257.51 $256.67 $242.41 1997 $276.16 $273.04 $273.80 1998 $300.01 $296.30 $315.48 1999 $293.64 $288.19 $332.37 2000 $307.02 $302.94 $308.10 2001 $317.74 $312.85 $312.76 2002 $330.87 $322.96 $372.41 2003 $295.28 $288.55 $340.14 2004 $280.78 $275.14 $314.29 2005 $278.66 $273.57 $297.82 2006 $275.06 $265.18 $353.64 2007 Overall $289.32 $283.76 $314.55 %12.32% 6.48% 92.54% Change 237  Total Attributable Hospital Admissions - Narrow Definition Year SARDs SARDs-CTD SARDs-VD 300 224 59 1996 267 209 44 1997 290 230 52 1998 263 197 50 1999 262 205 44 2000 230 170 48 2001 252 192 48 2002 354 217 109 2003 256 162 73 2004 297 190 87 2005 359 245 91 2006 219 151 48 2007 Overall 3,349 2,392 753 %-27.00% -32.59% -18.64% Change Total Attributable Hospital Admissions - Broad Definition Year SARDs SARDs-CTD SARDs-VD 1,083 882 168 1996 1,111 895 178 1997 1,107 922 157 1998 1,122 911 167 1999 1,008 812 155 2000 968 787 139 2001 937 763 137 2002 1,105 801 250 2003 978 744 180 2004 955 715 194 2005 1,088 796 237 2006 958 773 139 2007 Overall 12,162 9,801 2,101 %-11.54% -12.36% -17.26% Change Total Attributable Hospital Costs - Narrow Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $2,189,582.10 $1,669,114.47 $400,497.81 1996 $2,309,176.70 $1,625,492.80 $594,585.97 1997 $2,797,500.52 $2,060,677.79 $509,830.86 1998 $2,115,519.14 $1,342,393.82 $628,656.74 1999 $2,249,894.58 $1,331,735.27 $632,525.46 2000 $2,342,439.31 $1,617,398.20 $393,724.59 2001 $2,328,229.17 $1,779,339.45 $359,956.96 2002 $2,765,691.01 $1,550,776.18 $946,733.34 2003 $2,015,138.71 $1,533,275.55 $354,765.18 2004 $2,436,008.10 $1,617,184.58 $503,626.21 2005 $3,423,211.40 $2,256,400.36 $960,945.71 2006 $2,996,832.08 $1,867,571.44 $749,935.75 2007 Overall $29,969,222.82 $20,026,298.97 $7,035,784.58 %36.87% 11.89% 87.25% Change 238  Total Attributable Hospital Costs- Broad Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $8,306,656.55 $6,051,641.46 $1,949,165.24 1996 $10,049,554.07 $7,491,711.42 $2,358,371.16 1997 $8,858,044.86 $6,653,964.82 $1,804,694.93 1998 $9,772,683.33 $7,006,172.25 $2,299,613.97 1999 $9,209,173.01 $6,902,644.79 $1,731,025.68 2000 $9,077,279.47 $6,844,531.08 $1,633,309.49 2001 $9,160,808.56 $6,906,938.62 $1,657,530.92 2002 $10,225,576.89 $7,201,187.12 $2,258,993.46 2003 $8,454,064.55 $6,450,560.39 $1,661,627.11 2004 $9,864,915.31 $7,319,935.58 $2,001,624.12 2005 $11,616,230.07 $7,826,972.16 $3,013,965.14 2006 $9,800,561.48 $7,314,124.18 $1,594,330.49 2007 Overall $114,395,548.15 $83,970,383.87 $23,964,251.71 %17.98% 20.86% -18.20% Change Crude Annual Mean Per-Patient-Year Attributable Hospital Admissions – Narrow Definition Year SARDs SARDs-CTD SARDs-VD 0.10 0.08 0.06 1996 0.08 0.07 0.18 1997 0.08 0.07 0.21 1998 0.06 0.05 0.20 1999 0.05 0.05 0.16 2000 0.04 0.03 0.17 2001 0.04 0.03 0.16 2002 0.05 0.03 0.30 2003 0.03 0.02 0.16 2004 0.03 0.02 0.15 2005 0.03 0.02 0.13 2006 0.02 0.01 0.06 2007 Overall 0.04 0.03 0.15 %-84.08% 0.08 2.13% Change Crude Annual Mean Per-Patient-Year Attributable Hospital Costs – Narrow Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $748.43 $627.86 $1,807.44 1996 $713.35 $552.39 $2,448.54 1997 $774.39 $623.49 $2,046.14 1998 $501.64 $344.85 $2,464.52 1999 $465.48 $297.22 $2,334.04 2000 $434.78 $322.47 $1,421.39 2001 $384.66 $314.68 $1,233.43 2002 $379.48 $228.47 $2,572.64 2003 $230.29 $188.25 $772.91 2004 $234.84 $167.95 $869.57 2005 $284.19 $202.25 $1,357.11 2006 $223.36 $150.79 $892.34 2007 Overall $364.86 $266.28 $1,477.12 %-70.16% -75.98% -50.63% Change 239  Crude Annual Mean Per-Patient-Year Attributable Hospital Admissions – Broad Definition Year SARDs SARDs-CTD SARDs-VD 0.37 0.33 0.76 1996 0.34 0.30 0.73 1997 0.31 0.28 0.63 1998 0.27 0.23 0.65 1999 0.21 0.18 0.57 2000 0.18 0.16 0.50 2001 0.15 0.13 0.47 2002 0.15 0.12 0.68 2003 0.11 0.09 0.39 2004 0.09 0.07 0.33 2005 0.09 0.07 0.33 2006 0.07 0.06 0.17 2007 Overall 0.15 0.13 0.44 %-80.71% 0.33 -78.19% Change Crude Annual Mean Per-Patient-Year Attributable Hospital Costs – Broad Definition (2007 Canadian dollars) Year SARDs SARDs-CTD SARDs-VD $2,839.32 $2,276.41 $8,796.53 1996 $3,104.51 $2,545.89 $9,711.89 1997 $2,452.05 $2,013.25 $7,242.92 1998 $2,317.36 $1,799.84 $9,015.15 1999 $1,905.28 $1,540.57 $6,387.55 2000 $1,684.82 $1,364.65 $5,896.42 2001 $1,513.50 $1,221.51 $5,679.72 2002 $1,403.05 $1,060.92 $6,138.57 2003 $966.12 $791.99 $3,620.10 2004 $951.02 $760.20 $3,456.04 2005 $964.38 $701.57 $4,256.51 2006 $730.47 $590.55 $1,897.07 2007 Overall $1,392.69 $1,104.11 $5,031.16 %-74.27% -74.06% -78.43% Change  240  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0105195/manifest

Comment

Related Items