UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Bone health and osteoporosis in women diagnosed with breast cancer in British Columbia Tseng, Ling-I Olivia 2017

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

Item Metadata

Download

Media
24-ubc_2017_november_tseng_lingiolivia.pdf [ 3.02MB ]
Metadata
JSON: 24-1.0355207.json
JSON-LD: 24-1.0355207-ld.json
RDF/XML (Pretty): 24-1.0355207-rdf.xml
RDF/JSON: 24-1.0355207-rdf.json
Turtle: 24-1.0355207-turtle.txt
N-Triples: 24-1.0355207-rdf-ntriples.txt
Original Record: 24-1.0355207-source.json
Full Text
24-1.0355207-fulltext.txt
Citation
24-1.0355207.ris

Full Text

  BONE HEALTH AND OSTEOPOROSIS  IN WOMEN DIAGNOSED WITH BREAST CANCER  IN BRITISH COLUMBIA by  Ling-I Olivia Tseng   M.Sc. Simon Fraser University, Canada, 2007 M.D. Chung Shan University, Taiwan, 1997  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Interdisciplinary Oncology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August 2017  © Ling-I Olivia Tseng, 2017 ii  Abstract Background: Women diagnosed with breast cancer are at higher risk of osteoporosis and osteoporotic fractures. Information is lacking on utilization of bone mineral density testing in British Columbia, and fracture risks associated with tamoxifen and aromatase inhibitors to plan care.     Methods: Three studies were conducted on women diagnosed with breast cancer. Study 1, a retrospective cross-sectional study evaluated the utilization of bone mineral density testing in 1995-2008 and identified factors associated with different testing rates using secondary data-linkage in older women aged ≥65 and diagnosed with breast cancer for ≥3 years in British Columbia, Canada. Study 2, a pilot randomized controlled trial, assessed the feasibility of a protocol designed to improve bone health management, especially bone mineral density testing rates, with educational material in older women aged ≥65 and diagnosed with breast cancer for ≥3 years. And study 3, a systematic review with meta-analysis, estimated fracture risks associated with tamoxifen and aromatase inhibitors in younger women aged ≤65.   Results: In older women aged ≥65, proportions of women with ≥1 bone mineral density test per calendar year increased from 1.0% in 1995 to 10.1% in 2008. Women with lower socio-economic status or rural residence were significantly less likely to have a bone mineral density test. The study protocol is feasible with a promising effect of educational material on bone mineral density testing rates (17%, 95% CI=6 to 33) in the 54 participants during the pilot study six-month follow-up period. In younger women aged ≤65, fracture risk did not differ between the  iii  tamoxifen and no-tamoxifen groups. Aromatase inhibitor-associated fracture risk was 17% and 35% higher than the risks in the no-aromatase inhibitor group and tamoxifen group respectively. The higher aromatase inhibitor-associated fracture risk compared with tamoxifen descreased slowly over time. The risk was significantly higher during the treatment period, but not the post-treatment period.   Conclusions: Increased risk of fractures is reported in women diagnosed with breast cancer and treated with aromatase inhibitors, while screening for osteoporosis with bone mineral density testing is sub-optimal. There is a need to improve bone health management programs which should include educational materials.      iv  Lay Summary  Women diagnosed with breast cancer are at higher risk of developing osteoporosis and breaking bones. Bone health management including bone scans, plays a key role in preventing fractures for this group of women. My studies show low rates of bone scans for women aged 65 years and over, especially women with lower income or living in rural areas. Nine of the 54 women had bone scans within six months after being provided with our educational material. An additional 35 broken bones happen to every 100 women, aged 65 years and under, taking aromatase inhibitor treatment compared with women taking tamoxifen treatment. Broken bones are more likely to happen during the treatment period. Women diagnosed with breast cancer should be encouraged to have bone scans. Bone health management programs including physical activity and calcium/vitamin D intake are important for women diagnosed with breast cancer, especially those taking aromatase inhibitors.   v  Preface This dissertation is original and independent work completed by Ling-I Olivia Tseng (me). This thesis is manuscript-based. I am the lead author of all three manuscripts included in this thesis.   A version of Chapter Two (study 1) of this thesis has been submitted to Osteoporosis International. The co-authors include Martin Dawes, John Spinelli, Carolyn Gotay, and Mary McBride. The title is “Utilization of bone mineral density testing among breast cancer survivors in British Columbia, Canada”. This study provides a better understanding of the utilization of bone mineral density (BMD) testing using secondary administrative healthcare data linkage. I developed this project using data acquired under Mary McBride’s research project “Late morbidity and health care utilization among three-year survivors of breast cancer in BC, Canada”. Ethics approval H09-01957 was obtained from the University of British Columbia (UBC) / British Columbia (BC) Cancer Agency Ethics Board. Financial support for this project was provided through Mary McBride’s Canadian Breast Cancer Foundation research grant. I designed the study, prepared the study protocol, analyzed and interpreted the data, and prepared the manuscript. The co-authors contributed to the study design and data interpretation. Mary McBride also provided administrative support for the data-linkage and John Spinelli aided in the statistical analysis. All co-authors reviewed and involved in revising the manuscript. All inferences, opinions, and conclusions drawn in this chapter are those of the authors, and do not reflect the opinions or policies of the Data Steward(s).   A version of Chapter Three (study 2) from this thesis will be submitted for journal publication. The co-authors include Martin Dawes, John Spinelli, Carolyn Gotay, Mary McBride, and Wan vi  Yu (Julia) Ho. The title is “Promoting bone health management in women diagnosed with breast Cancer: a pilot randomized controlled trial”. This project evaluated the feasibility of a randomized controlled trial protocol designed to evaluate the effect of educational material on bone health management. I developed and designed this project with guidance from Martin Dawes. I prepared the ethics application and obtained ethics approval H15-00849, from the UBC / BC Cancer Agency Ethics Board. I obtained a Janus research grant provided by the College of Family Physicians Canada to support this study. I wrote the study protocol and educational material, and selected validated questionnaires based on literature review. The first four co-authors contributed to the study design, data interpretation, educational material review, and questionnaire preparation. John Spinelli also contributed to the statistical analysis. All co-authors reviewed and were involved in revising the manuscript.  A version of Chapter Four (study 3) will be submitted for journal publication. The co-authors include Martin Dawes, John Spinelli, Carolyn Gotay, Mary McBride and W.Y. (Julia) Ho. The title is “Aromatase inhibitors are associated with a higher fracture risk than tamoxifen: a systematic review and meta-analysis. This study evaluated fracture risks associated with breast cancer treatments, tamoxifen and aromatase inhibitors, using a systematic review format with meta-analysis. No ethics approval was required and no research grant was obtained for this study. I developed this project, wrote the study protocol and selected tools to assess methodology quality with guidance from Martin Dawes. I conducted article searches with the guidance of a UBC biomedical librarian. I recruited and supervised a first-year family practice resident Nicole Redding, who screened the selected articles by title and abstract. I also recruited and supervised W.Y. (Julia) Ho, who reviewed the selected full-text articles and extracted data from the vii  included articles. I screened the selected articles by title and abstract, reviewed the selected full-text articles, prepared Excel worksheets for data extraction, conducted methodology quality assessment, and extracted data from the included studies. I conducted all data analyses. The first four co-authors contributed to the study design, data interpretation and manuscript preparation. Martin Dawes and John Spinelli also contributed to the statistical analysis. All co-authors reviewed and were involved in revising the manuscript.   viii  Table of Contents Abstract ii Lay Summary ............................................................................................................................... iv Preface v Table of Contents ....................................................................................................................... viii List of Figures ..............................................................................................................................xv List of Symbols ........................................................................................................................... xvi List of Abbreviations ................................................................................................................ xvii Acknowledgements .................................................................................................................... xix Chapter 1: Introduction ................................................................................................................1 1.1 Breast Cancer ...................................................................................................................... 2 1.1.1 Introduction .............................................................................................................. 2 1.1.2 Epidemiology ........................................................................................................... 2 1.1.3 Risk factors ............................................................................................................... 4 1.1.4 Clinical manifestations ............................................................................................. 5 1.1.5 Screening and early detection................................................................................... 6 1.1.6 Diagnosis .................................................................................................................. 7 1.1.6.1 Diagnostic evaluation ........................................................................................................... 7 1.1.6.2 Diagnostic criteria ................................................................................................................ 7 1.1.6.3 Staging, grading, and receptor status .................................................................................... 7 1.1.7 Treatment................................................................................................................ 11 1.1.7.1 Surgery ............................................................................................................................... 11 1.1.7.2 Radiation therapy................................................................................................................ 11 1.1.7.3 Chemotherapy .................................................................................................................... 12 1.1.7.4 Hormonal treatment ............................................................................................................ 12 1.1.7.5 Biological treatment ........................................................................................................... 13 1.1.8 Follow-up care after completing breast cancer treatment ...................................... 13 1.1.8.1 Increasing population of women diagnosed with breast cancer ......................................... 13 ix  1.1.8.2 Follow-up care in women completing initial breast cancer treatments............................... 13 1.2 Osteoporosis and Osteoporotic Fractures.......................................................................... 14 1.2.1 Introduction ............................................................................................................ 14 1.2.2 Epidemiology ......................................................................................................... 14 1.2.3 Risk factors for osteoporosis and osteoporotic fractures........................................ 15 1.2.4 Osteoporosis fulfills the Wilson-Jungner criteria for a screening program ........... 16 1.2.5 Screening for osteoporosis ..................................................................................... 20 1.2.5.1 Risk assessment to identify high risk individuals for bone mineral density testing ........... 20 1.2.5.2 Bone mineral density testing to diagnose osteoporosis or determine fracture risk ............. 21 1.2.5.3 Benefits and harms of osteoporosis screening .................................................................... 23 1.2.5.4 Potential barriers to access bone mineral density testing .................................................... 23 1.2.5.5 Potential interventions to improve utilization of dual-energy X-ray absorptiometry ......... 24 1.2.6 Clinical manifestation – osteoporotic fractures ...................................................... 25 1.2.6.1 The impact of osteoporotic fractures – morbidity, mortality and economic burdens ......... 25 1.2.7 Diagnosis ................................................................................................................ 25 1.2.7.1 Post-menopausal women .................................................................................................... 26 1.2.7.2 Pre-menopausal women ...................................................................................................... 26 1.2.8 Treatment................................................................................................................ 26 1.2.8.1 Non-pharmacological treatment – lifestyle advice ............................................................. 27 1.2.8.2 Pharmacological treatment - medications ........................................................................... 28 1.2.8.3 Monitoring .......................................................................................................................... 28 1.3 Osteoporosis in Women Diagnosed with Breast Cancer .................................................. 29 1.3.1 Epidemiology ......................................................................................................... 29 1.3.2 Effects of breast cancer treatments on bones and fractures .................................... 29 1.3.2.1 Overall review .................................................................................................................... 29 1.3.2.2 Tamoxifen .......................................................................................................................... 30 1.3.2.3 Aromatase inhibitors .......................................................................................................... 31 1.3.2.4 Sequential treatments with tamoxifen and aromatase inhibitors ........................................ 33 1.3.3 Guidelines for bone mineral density measurements with dual-energy X-ray absorptiometry in women diagnosed with breast cancer in Canada ...................... 33 1.4 Rationale, Objectives, and Hypotheses ............................................................................. 36 1.4.1 Utilization of bone mineral density testing in women diagnosed with breast cancer in British Columbia, Canada (Chapter Two, study 1) ............................................ 36 1.4.2 Promoting bone health management in women diagnosed with breast cancer: A pilot randomized controlled trial (Chapter Three, study 2) .................................... 37 x  1.4.3 Aromatase inhibitors are associated with a higher fracture risk than tamoxifen: a systematic review and, meta-analysis (Chapter Four, study 3) .............................. 38 Chapter 2: Utilization of Bone Mineral Density Testing in Women Diagnosed with Breast Cancer in British Columbia, Canada (study 1) .....................................................40 2.1 Introduction: ...................................................................................................................... 40 2.2 Method .............................................................................................................................. 41 2.2.1 Study groups ........................................................................................................... 41 2.2.2 Data sources ........................................................................................................... 42 2.2.3 Outcome variable, bone mineral density test ......................................................... 43 2.2.4 Potential modifying factors for association analysis .............................................. 43 2.2.5 Statistical analysis .................................................................................................. 45 2.3 Results ............................................................................................................................... 46 2.3.1 Trend analysis......................................................................................................... 46 2.3.2 Association analysis ............................................................................................... 49 2.4 Discussion ......................................................................................................................... 54 2.4.1 Trend analysis......................................................................................................... 54 2.4.2 Association analysis ............................................................................................... 57 2.4.3 Limitations and future directions ........................................................................... 59 2.5 Conclusion ........................................................................................................................ 61 Chapter 3: Promoting Bone Health Management in Women Diagnosed with Breast Cancer: A Pilot Randomized Controlled Trial (study 2) ....................................................62 3.1 Introduction ....................................................................................................................... 62 3.2 Methods ............................................................................................................................. 63 3.2.1 Study protocol ........................................................................................................ 63 3.2.2 Recruitment and randomization ............................................................................. 63 3.2.3 Interventions ........................................................................................................... 67 3.2.4 Self-reported participant questionnaires ................................................................. 67 3.2.5 Procedure / intervention outcome measures ........................................................... 68 3.2.6 Statistical analysis .................................................................................................. 69 3.2.7 Ethics and clinical trial registration ........................................................................ 70 3.3 Results ............................................................................................................................... 70 xi  3.3.1 Feasibility of the study protocol ............................................................................. 70 3.3.2 Study participants ................................................................................................... 73 3.3.3 Primary outcome – bone mineral density testing rates........................................... 76 3.3.4 Secondary outcomes ............................................................................................... 76 3.4 Discussion ......................................................................................................................... 78 3.4.1 Educational material ............................................................................................... 79 3.4.2 Bone mineral density testing rate ........................................................................... 80 3.4.3 Secondary outcome measures ................................................................................ 81 3.4.4 Limitation ............................................................................................................... 81 Chapter 4: Aromatase Inhibitors are Associated with a Higher Fracture Risk than Tamoxifen: A Systematic Review and Meta-analysis (study 3) ...........................83 4.1 Introduction ....................................................................................................................... 83 4.2 Method .............................................................................................................................. 85 4.2.1 Search strategy ....................................................................................................... 85 4.2.2 Study selection ....................................................................................................... 86 4.2.3 Study quality assessment ........................................................................................ 86 4.2.4 Data extraction ....................................................................................................... 87 4.2.5 Data synthesis ......................................................................................................... 88 4.2.6 Statistical analysis .................................................................................................. 89 4.3 Results ............................................................................................................................... 89 4.3.1 Characteristics of included studies ......................................................................... 91 4.3.2 Study quality assessment ........................................................................................ 96 4.3.3 Tamoxifen .............................................................................................................. 98 4.3.4 Aromatase inhibitors .............................................................................................. 98 4.3.5 Comparison of aromatase inhibitors and tamoxifen............................................... 99 4.3.6 Comparison of aromatase inhibitors and tamoxifen – time effect ....................... 105 4.4 Discussion ....................................................................................................................... 107 4.4.1 Tamoxifen ............................................................................................................ 107 4.4.2 Aromatase inhibitors ............................................................................................ 108 4.4.4 Study methodology............................................................................................... 111 4.4.5 Limitation ............................................................................................................. 115 xii  4.5 Conclusion ...................................................................................................................... 116 Chapter 5: Discussion and Conclusion ....................................................................................117 5.1 Key Findings ................................................................................................................... 118 5.1.1 Utilization of bone, mineral density testing among women diagnosed with breast cancer in British Columbia, Canada ..................................................................... 118 5.1.2 Promoting bone health management in women diagnosed with breast cancer: a pilot randomized controlled trial .......................................................................... 119 5.1.3 Aromatase inhibitors are associated with a higher fracture risk than tamoxifen: a systematic review and meta-analysis ................................................................... 120 5.2 Strengths and Limitations ............................................................................................... 121 5.2.1 Strengths ............................................................................................................... 121 5.2.2 Limitations............................................................................................................ 124 5.3 Conclusion ...................................................................................................................... 126 5.4 Implications, Applications and Future Research ............................................................ 127 5.4.1 Older women aged 65 years and over who were diagnosed with breast cancer .. 127 5.4.2 Younger women aged 65 years and under who were diagnosed with breast cancer129 5.4.3 Women diagnosed with breast cancer .................................................................. 131 Bibliography ..............................................................................................................................132 Appendices 149 Appendix A Summary of datasets in this study ...................................................................... 149 Appendix B Summary of codes for variables ......................................................................... 150 Appendix C Educational Material .......................................................................................... 151 Appendix D List of risk factors .............................................................................................. 155    xiii  List of Tables Table 1-1 TNM staging of breast cancer ....................................................................................... 8 Table 1-2 Definition of categories of cancer relapse risk for patients with operated breast cancer. ..................................................................................................................................... 10 Table 1-3 Risk factors for osteoporosis in general female population based on the 2002 Canadian osteoporosis guideline ................................................................................. 16 Table 1-4 Summary of the Wilson-Jungner Criteria and corresponding evidence to support osteoporosis for screening programs .......................................................................... 18 Table 1-5   Identification of risk factors during osteoporosis screening ....................................... 21 Table 1-6   A comparison between tamoxifen and aromatase inhibitors for treating hormone receptor positive breast cancer without metastasis ..................................................... 34 Table 1-7   Summary of guideline recommendations on bone mineral density measurement with dual-energy X-ray absorptiometry in women with breast cancer in Canada .............. 35 Table 2-1  Trends in proportion of survivors with at least one bone mineral density test, overall and stratified by osteoporosis diagnosis from 1995 to 2008 ....................................... 47 Table 2-2   Characteristics of female breast cancer survivors for associations between factors and bone mineral density testing rates during the period 2006-2008 ................................ 50 Table 2-3   Associations between factors and bone mineral density testing rates in older female breast cancer survivors during the period 2006-2008 ................................................. 52 Table 3-1   Data sources for outcome measures ............................................................................ 69 Table 3-2   Representativeness of participants .............................................................................. 71 Table 3-3   Response patterns in questionnaires ........................................................................... 73 Table 3-4   Characteristics of study participants ........................................................................... 74 Table 4-1   Summary of studies ..................................................................................................... 92 Table 4-2   Summary of risk of bias assessment for the included randomized controlled trials ... 97 Table 4-3   Summary of Newcastle-Ottawa Scale assessment for the included cohort studies .... 98 Table 4-4   Meta-analysis including subgroup analysis of aromatase inhibitors, tamoxifen, and control groups on fractures ....................................................................................... 100 xiv  Table 4-5   Meta-analysis of aromatase inhibitors and tamoxifen on fractures at different ranges of follow-up duration and treatment phases .............................................................. 105  xv  List of Figures Figure 1-1 Age-standardized incidence rates and age-standardized mortality rates for breast cancer, females, Canada 1986-2015 ............................................................................. 3 Figure 2-1 Numbers of female breast cancer survivors and proportions of women with at least one BMD test by osteoprososis diagnosis and calendar year from 1995 to 2008 ...... 48 Figure 3-1 Consolidated Standards of Reporting Trails (CONSORT) flow diagram .................. 66 Figure 4-1 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analysis) flow diagram for systematic review of the fracture risks associated with breast cancer treatments .................................................................................................................... 90 Figure 4-2 Forest plot of comparison for fracture risk between women treated with tamoxifen and not treated with tamoxifen (control) by study design subgroups ....................... 102 Figure 4-3 Forest plot of comparison for fracture risk between women treated with AIs and not treated with AIs (control) by study design subgroups .............................................. 103 Figure 4-4 Forest plot of comparison for fracture risk between women treated with AIs and treated with tamoxifen by study design subgroups ................................................... 104 Figure 4-5 Forest plot of comparison for fracture risk between women treated with aromatase inhibitors and tamoxifen (during treatment period) .................................................. 106 Figure 4-6 Forest plot of comparison for fracture risk between women treated with aromatase inhibitors and tamoxifen (during treatment period) .................................................. 106 Figure 4-7 Consort diagram of Breast International Group 1-98 trial ........................................ 115  xvi  List of Symbols %  percent ± plus or minus ≥ greater than or equal to  ≤ less than or equal to  >  greater than <  less than  = equal                                  xvii  List of Abbreviations AAPCs Average annual percentage changes ABCSG Austrian Breast and Colorectal Cancer Study aHR  Adjusted hazard ratio  AIs Aromatase inhibitors APCs Average percent changes ARNO Arimidex-Nolvadex ATAC Arimidex, Tamoxifen, Alone or in Combination BC  British Columbia  BCOU Breast Cancer Outcome Unit BIG  Breast International Group  BMD  Bone mineral density  CAT  Calcium Assessment Tool  CI Confidence interval CONSORT  Consolidated Standards of Reporting Trails  DBCG  Danish Breast Cancer Cooperative Group  DXA Dual-energy X-ray absorptiometry  FRAX Fracture risk assessment  GLTEQ Godin Leisure-Time Exercise Questionnaire  HER2 Human epidermal growth factor receptor 2 HR  Hazard ratio IES Intergroup Exemestane Study  xviii  IRR Incidence rate ratio  IU International unit MSP  Medical Service Plan OR  Odds ratio aPR Adjusted prevalence ratio PRISMA  Preferred Reporting Items for Systematic Reviews and Meta-analysis RCT  Randomized controlled trial  RR Risk ratio SAS Statistical Analysis System SD Standard deviation SERM Selective estrogen receptor modulator SES  Socio-economic status SOFT  Suppression of Ovarian Function Trial  STROBE  Strengthening the Reporting of Observational Studies in Epidemiology  TEAM Tamoxifen, Exemestane Adjuvant Multinational  TEXT  Tamoxifen and Exemestane Trial TNM Tumor node metastasis  UBC  University of British Columbia VIDSUN Vitamin D & Sun  xix  Acknowledgements I would like to express my deepest appreciation to my two co-supervisors Mary L. McBride, the clinical professor at the School of Population and Public Health at the University of British Columbia (UBC), and Martin G. Dawes, a professor and the head of the UBC Department of Family Practice. They both provided me with incredible support for my thesis work by guiding me to explore my own research journey. My PhD study has been a unique experience allowing me to conduct research projects from concepts right through to the completion of studies.   My sincere thanks goes to my two other committee members John J. Spinelli, a professor at the UBC School of Population and Public Health and head of the Cancer Control Research of the British Columbia Cancer Agency, and Carolyn G. Gotay, a professor at the UBC School of Population and Public Health and the Canadian Cancer Society Chair in Cancer Primary Prevention. They both shared their years of research experience with me whenever I was in need.   I would like to express my sincere gratitude to all funding agencies supporting my PhD study – the UBC Clinician Scholar Program, the Canadian Breast Cancer Foundation (Telus-Canadian Breast Cancer Foundation Fellowship), the Faculty of Medicine (Graduate Award), the Interdisciplinary Oncology Program (Travel Award), Mr. Roman M. Babicki (Roman M. Babicki Fellowship in Medical Research), and the College of Family Physicians Canada (Janus Research Grant).   I would like to express my personal appreciation to volunteer student W.Y. Julia Ho, volunteer family practice resident Nicole Redding, the educational director of the Centre of Excellence in xx  Cancer Prevention Melissa Ashman, pharmacist Jun Qing Pan, all study participants, and team members consisting of my co-supervisors and committee members. This thesis wouldn’t have been completed without their support.    And special thanks goes to my husband Richard, my sisters Sophia and Suzen, and my parents. Without their love, continued encouragement, and nonstop support, this thesis wouldn’t exist.     1  Chapter 1: Introduction This thesis provides a better understanding of osteoporosis and bone health management in women diagnosed with breast cancer. Osteoporosis is a major public health issue while breast cancer is the most common female cancer worldwide. Both osteoporosis and breast cancer are strongly associated with advancing age. This thesis was developed upon two main concepts. First, women diagnosed with breast cancer are at higher risk of osteoporotic fractures compared with women without breast cancer. Second, BMD testing is recommended to high risk populations – by age (old women aged ≥65) or risk factors (younger women aged <65 with risk factors while breast cancer treatment is not consistently considered a risk factor for BMD testing eligibility). Study 1 and study 2 focus on the utilization of BMD testing in old women, a high-risk population by age. Study 3 focuses on fracture risk estimates associated with hormonal treatments in younger women, a high-risk population by risk factors. A better understanding on the effects of hormonal treatments on fracture risk in younger women may alter their eligibility for BMD testing. Older women are eligible for BMD testing regardless of breast cancer diagnosis and treatment.    Sections 1.1 and 1.2 fundamentally review both breast cancer and osteoporosis in the general population. Section 1.3 specifically reviews osteoporosis in women diagnosed with breast cancer. Study 1 evaluated the utilization of bone mineral density (BMD) testing in older women based on section 1.3.3, “Guidelines for BMD measurement with dual-energy X-ray absorptiometry (DXA) in women diagnosed with breast cancer in Canada”. Study 2 determined the feasibility of a study protocol designed to improve bone health management including BMD testing rates and lifestyle modifications in older women. This study was developed based on the 2  results from study 1, section 1.2.5.4 “Potential barriers to access BMD testing”, and section 1.2.5.5 “Potential interventions to improve utilization of DXA”. Study 3 systematically reviewed the effects of tamoxifen and aromatase inhibitors (AIs) on fracture risk in younger women based on section 1.3.2 “Effects of breast cancer treatments on bones and fractures”.   1.1 Breast Cancer  1.1.1 Introduction  Breast cancer is defined by the National Cancer Institute as “cancer that forms in tissues of the breast”. Common breast cancers include ductal carcinoma starting in the lining of the mammary ducts (80-90%) and lobular carcinoma starting in the lobules of the milk-producing glands (10%) (www.cancer.gov/publications/dictionaries/cancer-terms). Nearly 99% of all breast cancers occur in women [1].  1.1.2 Epidemiology  Breast cancer is the most common female cancer worldwide. There were approximately 1.7 million new female breast cancer cases globally in 2012. The highest incidence rates were reported in North America, Australia, New Zealand, and in western and northern Europe. The lowest incidence rates were noted in Asia and Sub-Saharan Africa [2].   Breast cancer accounts for 26% of all female cancers in Canada. One in every nine Canadian women is expected to develop breast cancer in their lifetime. The majority of female breast cancer cases are diagnosed at the age range of 50-69 years. There were an estimated 25,000 new breast cancer cases and 5,000 deaths in 2015 in Canada alone [1].  3  The historical age-standardized incidence rates in Canada increased by 15% during the period 1986-1992 (Figure 1-1). This rate increase was associated with the implementation of organized provincial screening mammogram programs in 1988. Age-standardized incidence rates had been stable over the period 1988-2015. The highest historical age-standardized mortality rate was reported in 1986. Age-standardized mortality rates had been trending down and dropped by 44% during the period 1987-2015 [1]. This was likely due to early cancer identification through screening mammogram programs and the use of more effective cancer treatment.   Figure 1-1 Age-standardized incidence rates and age-standardized mortality rates for breast cancer, females, Canada 1986-2015  Rates are age-standardized to the 1991 Canadian population. This figure was created based on data from reference article [1].  4  1.1.3 Risk factors  Many risk factors have been established for female breast cancer. The most important risk factor is “advancing age”. The probabilities of a Canadian woman developing breast cancer over her next 10-year period of life per her current age are as follows [1]:  • Age 30 – 0.4% (1 in 250 women or one in every 250 women who are currently aged 30 years, will develop breast cancer over their next 10 years.)  • Age 40 – 1.4% (1 in 71 women)  • Age 50 – 2.2% (1 in 45 women)  • Age 60 – 3.2% (1 in 31 women)  • Age 70 – 3.3% (1 in 30 women)  • Age 80 – 2.6% (1 in 38 women)  The  risk factors associated with breast cancer are categorized into three groups as follows [3]:  High risk (relative risk (RR) > 4.0)  • Old age [3] • White race [4, 5] • BRCA 1 / 2 gene positive [3] • Two first-degree relatives with breast cancer diagnosed at an early age [3] • Previous breast cancer [6] • Dense breast tissue [7] Moderate risk (2 > RR ≥ 4.0)  • One first-degree relative with breast cancer [3] • Benign breast disease [8, 9] • Therapeutic ionizing radiation exposure [10-13] 5  • No oophorectomy at younger age [3] Weak risk (2 ≥ RR) • Reproductive factors, including early menarche or late menopause [9, 14-17], nulliparity [17, 18], and older age at first pregnancy [9, 17, 18] • Short or no breast feeding [19] • Hormonal factors, including hormonal replacement therapy [20] and oral contraceptive use [21] • High socio-economic status (SES) • Obesity [22] • Tall stature [23-25] • Life style factors, including smoking [26, 27], alcohol [28-31], and night shift work [32]  1.1.4 Clinical manifestations A breast mass or lump is the most common breast cancer presentation, accounting for 55-92% of new cases. A cancerous breast mass commonly presents as a single, hard, immovable subcutaneous lesion with an irregular border. Other common manifestations of breast cancer include breast pain, nipple discharge, skin changes, and nipple changes [33-36].    Manifestations of breast cancer can change when cancer progresses. When breast cancer spreads beyond the breast(s) at the locally advanced stage, a lump or multiple lumps may develop in the armpits (axilla). These lumps are lymph nodes infiltrated by cancer cells, which could be painless, hard, and immovable. When breast cancer spreads to other organs at the metastatic stage, the manifestations that develop mainly depend on the organs involved. The common 6  organs (related symptoms) involved are bones (bone pain and pathological fracture), liver (abdominal pain, nausea, and jaundice), lungs (cough and shortness of breath), and brain (headache, nausea, vomiting, weakness, and confusion) [37].   1.1.5 Screening and early detection Screening plays a key role in the early detection of breast cancer. It can be used to identify individuals with breast cancers before symptoms occur. Mammogram, a breast imaging test using low-dose X-ray, remains the primary screening test for early detection of breast cancer. Screening mammogram has been recommended by major health authorities in North America with variations in screening intervals, and ages to initiate and discontinue screening [38, 39]. In Canada, a screening mammogram at a two- to three-year interval is recommended for women aged 50-74 years with average breast cancer risk [38].  Screening mammogram has been shown to be associated with an approximate 20% reduction in breast cancer mortality based on three meta-analyses conducted by the UK Independent Panel (relative risk (RR)=0.80, 95% confidence intervals (CI)=0.73 to 0.89; using a random-effects model) [40], the Canadian Task Force (RR=0.83, 95% CI=0.76 to 0.92; using a random-effects model) [38], and Cochrane (RR=0.81, 95% CI=0.74 to 0.87; using a fixed-effect model) [41]. However,  a more recent study reported no impact of screening mammogram on breast cancer mortality after a 25-year follow-up in Canadian women aged 40-59 years [42].   7  1.1.6 Diagnosis  1.1.6.1 Diagnostic evaluation  A diagnostic evaluation is conducted to identify the causes of mammogram-detected abnormalities or presented symptoms commonly associated with breast cancer. The diagnostic evaluation may include a full personal and family health history, a physical examination, diagnostic imaging tests, and a breast biopsy.   The individual’s personal and family health history information is used to determine the risk of breast cancer development. The physical examination by a health care provider looks for signs of breast cancer, such as a breast lump. The two most common initial diagnostic imaging tests are diagnostic mammogram (more views than a screening mammogram) and ultrasound to locate possible cancer lesions [43]. A breast biopsy is used to confirm the presence of cancer cells in the suspected lesion by primarily using needles to obtain a small sample of the lesion. The sample is then examined by a pathologist.   1.1.6.2 Diagnostic criteria A breast cancer diagnosis is confirmed by the presence of breast cancer cells with a pathological examination.   1.1.6.3 Staging, grading, and receptor status  The extent and features of the cancer should be evaluated and classified by stage, receptor status, and cancer relapse risk right after a breast cancer diagnosis is made. This information is used to determine prognosis and guide treatment.  8  Stage  The Tumor, Node, and Metastasis (TNM) staging of breast cancer is summarized in Table 1-1. The T category describes the size of the primary breast cancer. The N category describes the number and location of any regional lymph node(s) containing cancer cells. And the M category describes whether the cancer has spread beyond the breast(s). The TNM staging information is grouped into prognostic stage ranging from 0 (zero) to IV (four) with increasing severity of the cancer [44]. Stages I-II, III, and IV are also commonly referred to as early, locally advanced, and advanced/metastatic stages respectively. .  Table 1-1 TNM staging of breast cancer Overall stage Tumor (T) Nodes (N) Metastasis (M)    Stage 0  Non-invasive, cancer cells are contained in the milk duct 0  No  Early breast cancer Stage I / II  Size ≤ 5cm  ≤ 3 involved nodes No  Locally advanced  breast cancer Stage III Any size ≥ 4 involved nodes No  Any size Nodes other than in axilla No   Size > 5cm or tumor fixed to skin or chest wall Any nodes No  Metastatic breast cancer Stage IV  Any size  Any nodes Metastasis   Receptor status The status of three receptors, either positive or negative, is evaluated using immunohistochemistry staining. The three receptors evaluated are the estrogen receptor, the progesterone receptor, and the human epidermal growth factor receptor 2 (HER2). These receptors may receive signals from corresponding hormones, estrogen, progesterone, and human epidermal growth factor, to promote the growth of breast cancer cells. Women with estrogen 9  and/or progesterone receptor-positive breast cancers are likely to benefit from hormonal treatment, which are associated with better outcomes [45-47]. Women with HER2-positive breast cancers are likely to benefit from biological treatment [48].   Cancer relapse risk  Categories of cancer relapse risk are determined in women with non-metastatic stage 0-3 breast cancer based on age, tumor size, and histological features (Table 1-2) [49]. Three different histological features of tubule formation, nuclear pleomorphism, and mitotic count, are evaluated in breast cancer cells under a microscopic exam. Each feature is scored from one to three. The total score of the three features is classified into three grades: grade 1 or low grade (total score 3-5), grade 2 or intermediate grade (total score 6-7), and grade 3 or high grade (total score 8-9). Higher grade cancer cells tend to grow faster and are more likely to spread [50].    10  Table 1-2 Definition of categories of cancer relapse risk for patients with operated breast cancer. Reprinted with permission [49] Risk category   Low risk a Node negative and all of following features  pT ≤ 2cm, AND  Grade 1 b, AND  Absence of peritumoral vascular invasion c, AND  HER2/neu gene neither overexpressed nor amplified d, AND  Age ≥ 35 years  Intermediate risk e Node negative AND at least one of the following features:   pT > 2cm, OR  Grade 2-3 b, OR  Presence of peritumoral vascular invasion c, OR  HER2/neu gene overexpressed or amplified d, OR  Age < 35 years  Node positive (1-3 involved nodes) AND   HER2 / neu gene neither overexpressed nor amplified d High risk  Node positive (1-3 involved nodes) AND   HER/neu gene overexpressed or amplified d  Node positive (4 or more involved nodes).  pT pathological tumor size (i.e. size of the invasive component), HER human epidermal growth factor receptor a Some Panel members view pT1a and pT1b (i.e. pT <1 cm) tumors with node-negative disease as representing low risk even if higher grade and/or younger age b Histologic and/or nuclear grade c Peritumoral vascular invasion was considered controversial as a discriminatory feature of increased risk; its presence defined intermediate risk for node-negative disease, but did not influence risk category for node-positive disease d HER2/neu gene overexpression or amplification must be determined by quality-controlled assays using immunohistochemistry or fluorescence in situ hybridization analysis e Note that the intermediate-risk category includes both node-negative and node-positive 1–3 disease    11  1.1.7 Treatment  Breast cancer treatments are primarily determined by menopausal status, receptor status, and cancer relapse risk. Breast cancer treatments are categorized into two major groups – loco-regional treatments and adjuvant systemic treatments. The loco-regional treatments include surgery and radiation therapy. The adjuvant systematic treatments include chemotherapy, hormonal treatment, and biological therapy.   1.1.7.1 Surgery Surgery involves the removal of the cancer tissues. Lumpectomy is primarily performed as a definite treatment in women with early breast cancer. This is a surgical removal of the cancerous tissues and some surrounding normal breast tissues [51]. Mastectomy and lymph node dissection in the ipsilateral axilla are primarily performed in women with locally advanced breast cancer. This is a surgical removal of the entire breast with cancer and the axillary lymph nodes [52]. Surgery may also be an option for women with metastatic breast cancer.   1.1.7.2 Radiation therapy  Radiation therapy shrinks or kills cancer cells using high-energy radiation. Radiation therapy is primarily provided after breast surgery to reduce the local recurrence of the breast cancer in women with early or locally advanced breast cancer [53, 54]. Radiation therapy may also be offered to women with metastatic cancer to relieve symptoms, such as pain associated with bone metastasis and neurological symptoms associated with brain metastasis.   12  1.1.7.3 Chemotherapy Chemotherapy uses both oral and injectable drugs to kill cancer cells. It is primarily provided after surgery, but before radiation therapy, to reduce the recurrence of the breast cancer. Common chemotherapy drugs are cyclophosphamide, methotrexate, 5-fluorouracil, adriamycin, doxorubicin, and epirubicin [55, 56]. Chemotherapy is indicated in women at intermediate or high cancer relapse risk, regardless of menopausal status.   1.1.7.4 Hormonal treatment Hormonal treatment is primarily given to women with hormone receptor-positive (estrogen and/or progesterone receptor-positive) breast cancer, and determined based on menopausal status and cancer relapse risk [49]. Hormonal treatment stops or slows cancer growth by reducing the available estrogen to cancer cells. Hormonal treatments are selective estrogen receptor modulators (SERMs), aromatase inhibitors (AIs), and ovarian suppression treatment. SERMs, such as tamoxifen, block estrogen from binding to breast tissues. AIs, such as letrozole, reduce the production of estrogen. Ovarian suppression by radiation, surgery, or gonadotropin-releasing hormone agonists, are commonly provided after surgery [55-57]. Ovarian suppression is primarily used to treat pre-menopausal women, especially those at intermediate or high cancer relapse risk. Tamoxifen is primarily used to treat pre-menopausal women, and post-menopausal women at lower cancer relapse risk. AIs are primarily used to treat post-menopausal women at higher cancer relapse risk.    13  1.1.7.5 Biological treatment  Biological treatment stops or slows cancer cell growth by blocking the growth signals with antibodies, such as trastuzumab. Biological treatment is an option for women with HER2-positive breast cancer [58].  1.1.8 Follow-up care after completing breast cancer treatment  The impact of the care needs in women with breast cancer is escalating due to (1) an increasing population of women diagnosed with breast cancer; (2) the increasing complexity of care needs in this population; and (3) new recognition of the long-term and late effects associated with breast cancer and cancer treatment.   1.1.8.1 Increasing population of women diagnosed with breast cancer  The projected Canadian population will increase by 19% from 32 million residents in 2000 to 38 million residents over the period from 2000 to 2028 [59]. Over the same time period, the estimated new cases of female breast cancer will increase by 63% (19,200 cases in 2000 to 31,255 cases in 2028) in Canada [60, 61] and by 80% (2,600 cases in 2000 to 4,675 cases in 2028) in BC [62]. At least 87% of these women diagnosed with breast cancer will survive five years or more [60].  1.1.8.2 Follow-up care in women completing initial breast cancer treatments The care needs differ in women at different phases of their breast cancer such as at the phases of cancer diagnosis, cancer treatment, and post-cancer treatment [63, 64]. Caring for women after completing their initial breast cancer treatments is challenging due to a lack of standardized 14  guidelines. A recognized care gap in women completing their cancer treatments has been emphasized by the Institute of Medicine and the American Society of Clinical Oncology  [65]. Women completing their initial cancer treatments require care for cancer recurrence surveillance, primary or secondary cancer prevention, and monitoring and management of common long-term and late effects such as treatment related osteoporosis, heart failure, coronary artery disease, diabetes, and premature menopause [66-71].  1.2 Osteoporosis and Osteoporotic Fractures  1.2.1 Introduction  Osteoporosis was first defined as “a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration of bone tissue with a subsequent increase in fragility and susceptibility of fracture” at the 1993 consensus development conference [72]. This definition has recently been modified to be “a skeletal disorder characterized by compromised bone strength predisposing a person to an increased risk of fracture” by the National Institutes of Health in 2001 [73]. The term “osteopenia” describes a lower loss of bone mass than osteoporosis. Osteoporotic fractures or fragility fractures are fractures associated with low bone mass or osteoporosis, which commonly occur on the spine, wrist, or hip [74].    1.2.2 Epidemiology  Osteoporosis is a major public health issue strongly associated with advancing age. The prevalence and incidence of osteoporosis are continuing to increase due to progressively aging populations [75]. Osteoporosis affects an estimated 200 million women worldwide. When stratified by age, osteoporosis affects approximately one-tenth, one-fifth, two-fifths, and two-15  thirds of women aged 60, 70, 80, and 90 respectively [76]. Approximately one in four Canadian women have osteoporosis [77]. An estimated 15.8% and 45.9% of Canadian women aged over 50 suffer from osteoporosis and osteopenia respectively, per the osteoporosis diagnostic criteria defined by the World Health Organization [78].    There were an estimated nine million osteoporotic fractures in 2000 worldwide. One in every three women over age 50 years will develop osteoporotic fractures [79, 80]. Projected new hip fracture cases in women will increase by 240% from 1990 to 2050 [81]. Osteoporotic fractures account for 80% of all fractures in post-menopausal Canadian women over age 50 [82]. Projected annual new hip fracture cases among women aged 65 and over, will increase by threefold over the period from 1993 to 2041 [83].  1.2.3 Risk factors for osteoporosis and osteoporotic fractures Many risk factors for osteoporosis and osteoporotic fractures have been identified. The major risk factors for osteoporosis listed in the 2002 Canadian Osteoporosis guideline are summarized in Table 1-3. The four major risk factors for osteoporotic fractures are advanced age, osteoporotic fracture, family history of osteoporotic fracture, and low bone mineral density (BMD) [84]. The first three risk factors for osteoporotic fracture are also risk factors for osteoporosis.     16  Table 1-3 Risk factors for osteoporosis in general female population based on the 2002 Canadian osteoporosis guideline Major risk factors of osteoporosis Minor risk factors of osteoporotic Age >65 Rheumatoid arthritis Vertebral compression fracture Past history of clinical hyperthyroidism Osteoporotic fractures over age 40  Chronic anticonvulsant therapy Family history of osteoporotic fracture  Low dietary calcium intake Systemic glucocorticoid therapy >3 month duration  Smoker Malabsorption syndrome Excessive alcohol intake Primary hyperparathyroidism Excessive caffeine intake Propensity to fall Weight <57 kg Osteopenia apparent on x-ray film Weight loss of  >10% of weight at age 25 Hypogonadism Chronic heparin therapy Early menopause (before age 45)    1.2.4 Osteoporosis fulfills the Wilson-Jungner criteria for a screening program  The Wilson-Jungner criteria guide the selection of diseases that would benefit from and are suitable for screening. These criteria were defined for the World Health Organisation in 1968 [85]. Osteoporosis was first considered for screening programs by the World Health Organization in 1994. Osteoporosis meets the criteria for a screening program except the criterion “there should be a recognizable latent or early symptomatic stage”. While osteoporosis remains silent without symptoms before fractures occur, the target population for osteoporosis screening can be identified using validated risk factors, such as age and gender instead. The criteria and rationales supporting osteoporosis screening with BMD testing are summarized in Error! Reference source not found.   The target population for osteoporosis screening with BMD testing are the high-risk individuals (high-risk screening), but not everyone in the population (population screening). The efficacy of 17  population screening has been lacking whereas benefits of high-risk screening from cost-effective fracture prevention strategy of treating high-risk individuals with screen-detected osteoporosis, have been demonstrated [86, 87]. However, the effect of high-risk screening on mortality remains unclear.    The goal of osteoporosis screening has been shifting from identifying “individuals at high risk of osteoporosis” to “individuals at high risk of osteoporotic fractures” [88]. This is because (1) osteoporotic fractures have a higher impact than osteoporosis diagnosis on the individual’s life quality and healthcare systems; (2) not every individual diagnosed with osteoporosis will develop osteoporotic fractures in their lifetime [89]; and (3) individuals without osteoporosis diagnosis could develop osteoporotic fractures [90, 91].  18  Table 1-4 Summary of the Wilson-Jungner Criteria and corresponding evidence to support osteoporosis for screening programs Criteria Evidence supporting osteoporosis screening The condition sought should be an important health problem  - A globally significant health issue  - An estimated 200 million women with osteoporosis worldwide [75]  - Increasing prevalence and incidence of osteoporosis due to aging populations [75]  - One fracture in every three women over age 50 years [79, 80]  The natural history of the condition, including development from latent to declared disease, should be adequately understood - Well-understood pathophysiology of osteoporosis [92] There should be a recognizable latent or early symptomatic stage * - Osteoporosis develops slowly over years [93]  - Lacking symptoms till fractures occur [93] - Target populations could be identified using the validated risk factors (Table 1-3) There should be a suitable test or examination  - BMD testing is a suitable test  - BMD testing could be done using different technologies. Of them, DXA scan is the most effective and widely used test for osteoporosis screening [94-96] The test should be acceptable to the population - BMD testing is widely accepted Case finding should be a continuing process and not a “once and for all” project - A BMD test at a one- to three-year interval is recommended by major guidelines [97] Facilities for diagnosis and treatment should be available - BMD testing with DXA is readily available in hospitals and imaging clinics in Canada - Osteoporosis could be treated by family doctors or specialists in the community There should be an accepted treatment for patients with recognized disease - Osteoporosis treatment is associated with a 11% reduction in mortality (pooled risk ratio (RR)=0.89, 95% CI=0.80 to 0.99, p=0.036) in post-menopausal women [98] - The number needed to treat  for over two year treatment in post-menopausal women at higher osteoporosis risk is 24 (95% CI=19 to 37) for alendronate and 43 (95% CI=30 to 89) for residronate [99] - The number needed to treat would be lower over longer time periods and in individuals at higher osteoporosis risk [99]   19  Continued  Criteria Evidence supporting osteoporosis screening There should be an agreed policy on whom to treat as patients  - Diagnostic criteria of osteoporosis were initially defined by the World Health Organization in 1994 [95].  - Osteoporosis treatment has been recommended for anyone with osteoporosis, high risk of osteoporotic fracture or history of osteoporotic fracture [88, 97] The cost of case finding (including diagnosis and treatment of patients diagnosed) should be economically balanced in relation to possible expenditure on medical care as a whole - Treating women with screening-detected osteoporosis is cost-effective for fracture prevention in older women  - The cost per quality-adjusted life year is US$43,000 for women aged 65 and US$5,600 for women aged 75 [86, 87] DXA dual-energy X-ray absorptiometry, RR risk ratio, CI confidence interval *Did not match criteria 20  1.2.5 Screening for osteoporosis  Osteoporosis screening primarily involves the two stages of identifying high-risk individuals for BMD testing, and diagnosing osteoporosis and determining fracture risk based on BMD measurements. The goal of osteoporosis screening is to identify individuals at high risk who would benefit from treatment before fractures occur [97].   1.2.5.1 Risk assessment to identify high risk individuals for bone mineral density testing Any individuals aged 50 years and over should be assessed for risk. Risk assessment involves either (1) identifying specific risk factors [97], or (2) determining fracture risk using risk assessment instruments (www2.gov.bc.ca/gov/content/health/practitioner-professional-resources/bc-guidelines).   Risk factor identification  Identifying risk factors involves appropriate history taking, physical examinations, biochemical testing, and possibly radiographic examinations. Important risk factors that can be identified during this process are summarized in Table 1-5 [97].         21  Table 1-5   Identification of risk factors during osteoporosis screening Risk assessment  Risk factors  History taking Family history of osteoporotic fractures Personal history of osteoporotic fractures High-risk medications (e.g. aromatase inhibitors, glucocorticoids) Smoking Excessive drinking Diseases (e.g. rheumatoid arthritis) Post-menopausal status  Physical examination Low body mass index  Biochemical tests 25-hydrocyvitamin D (detecting vitamin D deficiency) Thyroid-stimulating hormone (detecting thyroid disease) Radiological examination  Vertebral compression fractures  Risk assessment instruments  Multiple risk assessment instruments have been developed using different combinations of risk factors. In British Columbia (BC), the FRAX tool without BMD is suggested. FRAX is an international tool developed by the World Health Organization and calibrated for each country [100]. The Canadian version of FRAX was released in 2008. Any individuals with moderate fracture risk (10-20%) should have BMD testing for further fracture risk stratification (www2.gov.bc.ca/gov/content/health/practitioner-professional-resources/bc-guidelines).   1.2.5.2 Bone mineral density testing to diagnose osteoporosis or determine fracture risk BMD is used to quantify bone mass as the bone mineral content per unit in g/cm2 [101]. BMD can be measured using a variety of imaging technologies, such as dual-energy X-ray absorptiometry (DXA), calcaneal quantitative ultrasound, and quantitative computer tomography. The DXA scan is favored over the other technologies due to lower radiation exposure, reasonable cost, greater precision, and better sensitivity to detect osteoporosis [84, 102]. A DXA scan of the spine and hip (total hip, femoral neck or trochanter) is the gold 22  standard test and widely accepted for BMD measurements in Canada. BMD testing with DXA remains a better single quantifiable risk predictor while its sensitivity increases when combined with FRAX [103].   Indications of BMD testing vary significantly among international and local guidelines. In Canada, one BMD test is recommended for individuals aged 65 and over, or individuals aged under 65 with pre-selected risk factors at one- to five-year intervals based on each individual’s risk level [97].   Dual-energy X-ray absorptiometry  A DXA scan involves an individual lying on a flat surface table. Two distinct X-ray energy beams, one high and one low, pass through the bones being examined. High- and low-energy X-rays are absorbed by both soft tissues and bone tissues at different ratios. BMD is then calculated using a mathematical formula based on absorption ratios. A DXA scan of the spine or hip is the only diagnostic tool for osteoporosis before fractures occur [94-96]. DXA can detect osteoporosis with 88.2% sensitivity and 62.5% specificity [104].   DXA is non-invasive with low radiation exposure equivalent to one-tenth the amount of a chest X-ray. A DXA scan only takes 10-15 minutes to complete. Having a DXA scan increases an individual’s willingness to promote lifestyle modifications and initiate osteoporosis treatment [105-109]. A major disadvantage of DXA is that DXA machines are not easily portable, which may lead to poor access to DXA machines in remote areas. DXA reading can be influenced by factors, such as inappropriate patient positioning, osteoarthritis, and vertebral compression 23  fracture [110]. Another limitation of DXA is that DXA only estimates relative fracture risk. DXA is unable to identify every individual who will develop fractures in the future [89-91, 111]. False-negative results from DXA tests may lead to future missed treatment opportunities [91, 111]. False-positive results from DXA tests may also lead to unnecessary treatment and potential adverse events associated with treatment.  1.2.5.3 Benefits and harms of osteoporosis screening The benefits of high-risk screening come from cost-effective fracture prevention when treating high-risk individuals with screen-detected osteoporosis [86, 87]. Several potential harmful effects associated with osteoporosis screening have been reported. Osteoporosis diagnosis may cause anxiety perceived from vulnerability to fracture [112].   1.2.5.4 Potential barriers to access bone mineral density testing   Potential barriers to access osteoporosis screening can be categorized into three groups – patient, physician, and healthcare system.   Many patient factors associated with lower utilization of BMD testing have been reported. These reported factors include advancing age [113], lower socio-economic status (SES) [114], comorbidities [113], cognitive impairment [115], nursing home residency [116], recent osteoporotic fracture [117], underestimated perceived fracture risk [118-120], and poor access to DXA machines due to limited mobility [121] or long travel distance [122]. Personal perceived fracture risk is underestimated among 52% of women with moderate fracture risk and 70-80% of women with high fracture risk, when compared with calculated fracture risk based on each 24  individual’s risk factors [118, 119]. This is consistent with the findings that osteoporosis is of less concern to women, compared with cancers, cardiovascular diseases, and neurological diseases [119]. Underestimated personal perceived fracture risk can reduce the personal willingness to have BMD testing.  The potential physician barriers for patients accessing BMD testing include patients’ frailty [123], time restraint for preventive care [124], and inconsistent guideline recommendations for BMD testing [125, 126]. The potential barriers associated with healthcare system include a lack of consistent guidelines for BMD testing, standardized reporting, and incentive for preventive care, and limited availability of BMD testing machines [127].   1.2.5.5 Potential interventions to improve utilization of dual-energy X-ray absorptiometry  Patient intervention combined with or without physician intervention has shown positive effects on BMD testing rates in different populations. Interventions improve BMD testing rates by 22-51% in high-risk populations with recent fractures [128], and by 18% in patients aged 65 and over in primary care settings [129]. The most common patient intervention is educational material, followed by notification and counseling. The most common physician intervention is reminders [128, 129]. Two interventions are directed at family doctors through patient education; five questions to take to their family doctors regarding investigation, diagnosis, and management of osteoporosis [130], and advice to visit their family doctors for further investigation [131]. 25  1.2.6 Clinical manifestation – osteoporotic fractures Osteoporosis presents with no clinical manifestation until fractures occur. The most common manifestations of osteoporosis are associated with osteoporotic vertebral fractures, such as kyphosis, height loss, back pain, and reduced rib-pelvis distance.   1.2.6.1 The impact of osteoporotic fractures – morbidity, mortality and economic burdens Excess mortality during the first year after hip fractures ranges from 8.4 to 36% [132]. Worldwide, osteoporotic fractures caused a loss of 5.8 million disability-adjusted life-years in the year 2002. Of the total, more than 50% of the loss was associated with fracture events in America and Europe [133]. Physical disability and nursing home admissions are reported in 10% and 19% of patients with hip fractures respectively [134]. The estimated annual cost of hip fracture care in Canada will increase from CA$650 million in 1997 to CA$2.4 billion in 2041 [135].   1.2.7 Diagnosis  Osteoporosis is diagnosed by T and Z scores which are calculated based on BMD measurements with DXA. A T score compares a woman’s BMD to an average BMD of a reference young female adult group. A Z score compares a woman’s BMD to an average BMD of a healthy female population at the same age. Both T and Z scores are measurement units expressed in standard deviations (SD). Formulas for calculating T and Z scores are as follows [136]:  T score = invididual′s BMD−population peak BMDstandard deviation of population peak BMD Z score = invididual′s BMD−population age related BMDstandard deviation  of population age−related BMD  26  1.2.7.1 Post-menopausal women A post-menopausal woman’s T score is categorized into four groups which were established by the World Health Organization in 1994. Osteoporosis diagnosis is made when a woman’s T score is -2.5 or below. The four categories include:   • Normal: a woman’s T score ≥ -1    • Low bone mass (osteopenia): a woman’s T score between -1 and -2.49  • Osteoporosis: a woman’s T score ≤ -2.5  • Severe osteoporosis: osteoporosis with the presence of one or more osteoporotic fractures  1.2.7.2 Pre-menopausal women  Two diagnostic categories in pre-menopausal women were suggested by the International Society for Clinical Densitometry in 2003 [137] and adopted in Canada in 2005 [110]. The Z score is preferred over the T score. The two categories include:    • Normal: a woman’s Z score > -2.5 • Reduced bone density: a women’s Z score ≤ - 2.5   1.2.8 Treatment  The goal of treatment is to prevent future fractures. Treatments could be categorized into two groups – non-pharmacological and pharmacological treatments.   27  1.2.8.1 Non-pharmacological treatment – lifestyle advice  Non-pharmacological treatment is recommended for individuals with osteoporosis and is used to promote bone health in any individuals over age 50. The most current Canadian osteoporosis guidelines recommend adequate vitamin D and calcium intake, exercise, fall prevention, smoking cessation, and avoiding excessive alcohol consumption [97].   Calcium and vitamin D intake  The recommended daily intake of elemental calcium through diet and supplements is 1,200 mg. The recommended daily intake of vitamin D3 in individuals is based on their risk of vitamin D deficiency: 400-800 international units (IU) for low risk; 800-1,000 IU for moderate risk; and 1000 IU for high risk. Calcium, combined with vitamin D or not, may reduce fracture risk by 3% to 23% [138].  Exercise and fall prevention  Resistance, balance, and core strengthening exercises are suggested to strengthen muscles, compensate for posture abnormalities, and improve balance [139]. This could reduce risk of falls, improve physical functions, and reduce pain in individuals with osteoporotic fractures. There seems to be a potentially small effect of exercise on reducing fracture risk in elders [140]. Fall prevention, including improved home safety such as better lighting and walking aids, might reduce the risk of fall-induced osteoporotic fractures.   28  1.2.8.2 Pharmacological treatment - medications Medications are indicated in any individuals with high fracture risk (> 20% of 10-year fracture risk), osteoporotic hip fracture, or more than one osteoporotic fracture [97]. An individual’s 10-year fracture risk is determined using the Canadian Association of Radiologists and Osteoporosis Canada tool [110]. Available medications in Canada include bisphosphonates, receptor activator for nuclear factor kappa-B ligand inhibitor, selective estrogen receptor modulators (SERMs), hormone replacement therapy, calcitonin, and teriparatide. Most medications have shown to reduce fracture risk at varied degrees, especially in post-menopausal women [99, 141]. Raloxifene, but not tamoxifen, is the only SERM approved for prevention and treatment of osteoporosis.  1.2.8.3 Monitoring  Repeated BMD testing with DXA could be a useful clinical tool to monitor BMD changes and identify individuals with poor compliance to treatment or lacking response to treatment. However, recommendations for testing intervals between successive DXA tests have not gained consensus. A short testing interval may lead to mistaking random fluctuation or artifact as true BMD changes [142]. Testing intervals increased from a one- to two-year interval in the 2002 Canadian osteoporosis guideline to one- to three-year interval in the 2010 guideline. The testing interval may be prolonged beyond a three-year interval once treatment effectiveness is shown.   29  1.3 Osteoporosis in Women Diagnosed with Breast Cancer 1.3.1 Epidemiology Breast cancer diagnosis is associated with a 32% higher prevalence of osteoporosis diagnosis, compared with women without breast cancer in the US [143]. The prevalence of osteoporosis diagnosis in women with breast cancer is 33.4% in the US [143] and remains unclear in Canada. The majority (77%) of post-menopausal women with osteoporosis remain undiagnosed in the US [144]. Higher fracture risk in women diagnosed with breast cancer was reported in the Tsa et. al. and Chen et.al. studies with more than 80,000 participants [145, 146]. The estimated fracture rate in post-menopausal women diagnosed with breast cancer is 20 per 1,000 women-years [147].   1.3.2 Effects of breast cancer treatments on bones and fractures 1.3.2.1 Overall review  Estrogen, a type of hormone, plays a key role in both breast cancer treatment and bone health. Estrogen preserves bone mass by reducing bone reabsorption [148]. Estrogen deficiency, which could be caused by most adjuvant systemic breast cancer treatments, increases the risk of osteoporosis and osteoporotic fractures. Adjuvant systemic treatments have been widely used to reduce the risk of cancer recurrence and improve survival. The adjuvant systemic treatments can be categorized into hormonal treatment, chemotherapy, and ovarian suppression. Hormonal treatment is primarily provided to women with hormone receptor-positive breast cancer, which accounts for two-third of all breast cancers [149, 150]. Chemotherapy is offered to any women at higher cancer relapse risk, which is especially true for women with hormone receptor-negative breast cancer [151]. Ovarian suppression by radiation, surgery, or gonadotropin-releasing hormone agonists is only offered to a small percentage of pre-menopausal women with breast 30  cancer and at higher cancer relapse risk [55, 57]. A comparison between tamoxifen and AIs for treating hormone receptor positive breast cancer is summarized in Table 1-6.   1.3.2.2 Tamoxifen Tamoxifen is a selective estrogen receptor modulator (SERM). A SERM acts as either an estrogen agonist (stimulation) or antagonist (inhibition) depending on the target tissues that tamoxifen binds to. Tamoxifen acts as an antagonist in breast tissues that competitively inhibits the binding of estrogen to estrogen receptors which will reduce available estrogen to breast cancer cells.   The indications for which tamoxifen is used for treating breast cancer have been extending over time. Tamoxifen was only initially approved to treat advanced breast cancer in the late 1970s and then added to treat early stage breast cancer in the early 1990s by the US Food & Drug Administration (www.fda.gov). The use of tamoxifen was extended again to women with stage 0 / ductal carcinoma in situ in 2003 [152]. Tamoxifen is currently recommended for pre-menopausal women, and post-menopausal women with lower cancer relapse risk. Tamoxifen is also an optional treatment in women with stage 0 / ductal carcinoma in situ breast cancer with higher cancer relapse risk [151]. Tamoxifen is primarily given for two to five years, alone or as part of sequential treatments with AIs. Tamoxifen should be avoided in women with personal or family history of deep vein thrombosis, pulmonary embolism, severe depression, or newly diagnosed endometrial cancer (www.bccancer.bc.ca).   31  The common side effects of tamoxifen are associated with its estrogenic activity on the tissues that tamoxifen binds to. The common important side effects include hot flashes, amenorrhea, and mood changes [153]. The effect of tamoxifen on bone tissue is inconsistent across studies. In animal models, estrogen stimulates bone formation as an agonist, which leads to higher bone mass [154]. In clinical studies, tamoxifen cause a BMD decrease in healthy pre-menopausal women but a BMD increase in healthy post-menopausal women [155]. Tamoxifen may slightly increase or decrease BMD by up to 2% in both pre- and post-menopausal women diagnosed with breast cancer [156-161]. Tamoxifen is associated with a 9% lower fracture risk in post-menopausal women diagnosed with breast cancer when compared with healthy post-menopausal women [147]. Tamoxifen, compared with no hormonal treatment, is not associated with an increased fracture risk in post-menopausal women with breast cancer [162, 163]. The effect of tamoxifen on prevention of osteoporotic fractures is neutral in the general population [164]. There might be a positive effect of tamoxifen on fracture risk. However, tamoxifen has not been approved for the prevention or the treatment of osteoporosis by the US Food & Drug Administration.    1.3.2.3 Aromatase inhibitors  Aromatase is an enzyme responsible for the synthesis of estrogen in the ovaries, normal breast tissues, breast-cancer tissues, etc. Aromatase inhibitors (AIs) reduce circulating estrogen levels in post-menopausal women by inhibiting or inactivating the aromatase enzymes in non-ovary tissues [165]. AIs act as an estrogen antagonist (inhibitors) in all tissues including bone tissues while tamoxifen act as either an estrogen antagonist or agonist depending on the target tissues that tamoxifen binds to. 32  Aminoglutethimide was the first AI used to treat post-menopausal advanced breast cancer in the 1980s, but was withdrawn from the market due to severe adverse events associated with its non-selective binding properties later [166]. Second and third generation AIs with more selective inhibition properties and fewer adverse events were then developed (www.fda.gov). Third generation AIs include non-steroidal AIs (letrozole and anastrozole) with reversible binding properties and steroidal AI (exemestane) with irreversible binding properties. AIs were approved for treating post-menopausal advanced breast cancer in the 1990s and for treating post-menopausal early breast cancer in the early 2000s. AIs were recommended for post-menopausal early breast cancer in 2005 guidelines [167]. AIs are primarily given for two to five years alone or as part of sequential treatments with tamoxifen. AIs have recently been suggested to be combined with ovarian suppression in pre-menopausal women at higher cancer relapse risk [151]. AIs should be avoided in women with severe osteopenia or osteoporosis, moderate to severe joint pain, or moderate to severe dyslipidemia (www.bccancer.bc.ca).   The common and important side effects of AIs include hot flushes, vaginal dryness, bone toxicity (osteoporosis, bone fracture, and arthralgia), and high cholesterol [168]. AI treatment reduces BMD significantly [158, 159, 169] by reducing circulating estrogen levels. AI treatment has shown to be associated with higher fracture rates in clinical trials while its degree of effect on that remains unclear [170-172].    33  1.3.2.4 Sequential treatments with tamoxifen and aromatase inhibitors  Tamoxifen and AIs can be given alone or in sequence. A sequential tamoxifen-AI treatment has tamoxifen treatment given first for two to three years and then switched to AIs for a total treatment duration of five years. A sequential AI-tamoxifen treatment has AIs given first, followed by tamoxifen [151, 152, 167]. In some special circumstances, prolonged hormonal treatment of up to 10 years would be considered. For example, AIs might be given for another five years after an initial five years of tamoxifen in women at high cancer relapse risk [151]. Sequential treatments, compared with either tamoxifen or AIs alone treatment, reduce the exposure time to both tamoxifen and AIs, which may theoretically reduce the long-term side effects associated with either tamoxifen or AIs alone, such as fracture risk.  1.3.3 Guidelines for bone mineral density measurements with dual-energy X-ray absorptiometry in women diagnosed with breast cancer in Canada  Guidelines for BMD measurements with DXA vary significantly in intervals and indications among countries and women diagnosed with breast cancer (Table 1-7). Non-Canadian guidelines recommend a BMD test with DXA at one- to two-year intervals for women 65 and over, women initiating AIs, pre-menopausal women with ovarian suppression treatment, or women aged 60-64 at high osteoporosis risk [173, 174]. In Canada, the indications and intervals of BMD testing with DXA for women diagnosed with breast cancer have been changing over time, and vary among national and provincial guidelines. The most current national guideline, the 2010 Canadian osteoporosis guidelines, recommend a BMD test with DXA at a one- to three-year interval for women 65 and over, and women under 65 with listed risk factors, such as AI treatment and premature menopause.  34  Table 1-6   A comparison between tamoxifen and aromatase inhibitors for treating hormone receptor positive breast cancer without metastasis  Tamoxifen Exemestane, anastrozole and letrozole Class  - Selective estrogen receptor modulator (SERM)  - Aromatase inhibitors (AIs)  - Non-steroidal AIs (letrozole and anastrozole); reversible binding - Steroidal AI (exemestane); irreversible binding Effect on breast cancer tissues - Inhibition  - Prevent estrogen from binding to breast tissues - Inhibition  - Reduce production of estrogen Treatment indication by menopausal status [151, 152, 167] - Pre-menopausal women  - Post-menopausal women at lower cancer relapse risk   - Pre-menopausal women at higher recurrent risk when AIs are combined with ovarian suppression treatment  - Post-menopausal women at higher cancer relapse risk  Treatment indication by stages [151] - Ductal carcinoma in situ (stage 0, optional)  - Early breast cancer  - Advanced breast cancer  - Early breast cancer  - Advanced breast cancer  Treatment duration [151] - Tamoxifen alone for 5 years  - Sequential tamoxifen (2-3 years) – AI (2-3 years) - Sequential tamoxifen (5 years) – AI (2-3 years) - AI alone for 5 years  - Sequential tamoxifen (2-3 years) – AI (2-3 years) - Sequential tamoxifen (2-3 years) – AI (5 years)  Contraindications  - Newly diagnosed endometrial cancer  - Personal or family history of deep vein thrombosis / pulmonary embolism - Severe depression - Pre-menopausal - Severe osteopenia or osteoporosis - Moderate to severe joint pain  - Moderate to severe dyslipidemia Effect on bone mineral density (BMD) - Stable with a small increase or decrease in BMD [156-161] - Significant loss [158, 159, 169] Effect on fracture risk  - Limited information  - No additional fracture risk when compared with women without hormonal treatments  [162, 163] - Increase when compared with tamoxifen in major randomized controlled trials [170-172]. - Uncertain degree of additional fracture risk  35  Table 1-7   Summary of guideline recommendations on bone mineral density measurement with dual-energy X-ray absorptiometry in women with breast cancer in Canada Year Organization  Target population  Indication of BMD measurement with DXA  Interval  National level   2002  Osteoporosis Society of Canada [84] General population - Women ≥65  - Post-menopausal women aged <65 with one major or two minor risk factors (Table 1-3) - Every 2-3 years in women with normal BMD  - Every 1-2 years in women with osteopenia or osteoporosis  2004   Canadian Task Force on Preventive Health Care [175]  General Population - Women ≥ 65  - Post-menopausal women with osteoporotic fracture(s) - Post-menopausal women with weight < 60kg - Post-menopausal women at high osteoporosis risk  - Every 2 years in women with normal BMD  - Every 1-2 years in women with osteopenia or osteoporosis  2005  Health Canada  [176] Breast cancer  - Post-menopausal women - Pre-menopausal women at high risk of osteoporosis  - Any women taking aromatase inhibitors  - Not specified 2010  Osteoporosis Canada [97] General population - Women ≥ 65  - Women aged 50-64 with listed risk factors  - Women aged <50 with listed risk factors - Every 5 years in women at low (<10%) 10-year fracture risk - Every 1-3 years in women at moderate  (10-20%) or high (>20%) 10-year fracture risk  Provincial level – British Columbia (BC)  2005  Medical Services Commission of British Columbia [177] General population  - Women≥65  - Women aged ≥50 with one major or two minor risk factors (Table 1-3) - Every 2 years  2011  Medical Services Commission of British Columbia [178] General population  - Women aged ≥65 at moderate (10-20%) or high (>20%) 10-year fracture risk using FRAX without BMD  - Women aged <65 with significant clinical risk factors  - Every 3-10 years based on a women’s risk profile  DXA dual-energy X-ray absorptiometry, BC British Columbia, BMD bone mineral density, FRAX fracture risk assessment  36 1.4 Rationale, Objectives, and Hypotheses 1.4.1 Utilization of bone mineral density testing in women diagnosed with breast cancer in British Columbia, Canada (Chapter Two, study 1) Rationale  Women diagnosed with breast cancer are at higher risk of osteoporosis [143] and osteoporotic fractures [179]. Bone Mineral Testing (BMD) testing with dual-energy X-ray absorptiometry (DXA) is the primary tool used for osteoporosis screening and treatment monitoring in Canada. BMD testing at one- to three-year intervals is recommended for women aged 65 years and over, regardless of breast cancer diagnosis. Utilization of BMD testing is unknown in women aged 65 and over, and diagnosed with breast cancer for three or more years in British Columbia (BC), Canada. Only women diagnosed with breast cancer for three or more years were included for this study as the focus of care after three years is likely to have shifted from acute active cancer treatment, to screening and management of long-term and late effects, such as osteoporosis.  Objectives  1. To evaluate trends in proportion of women, aged ≥65 and diagnosed with breast cancer for three or more years, with at least one BMD test per calendar year from 1995 to 2008 in BC, Canada. 2. To identify clinical and socio-demographic factors associated with different BMD testing rates in the three-year period 2006-2008.     37 Hypotheses  HA-1: Trends in proportion of women with at least one BMD test per calendar year from 1995 to 2008 will be positive as the population has become aware of osteoporosis.  H0-1: Trends will be stable.    HA-2: Factors, such as socio-economic status, remote residency, history of osteoporosis diagnosis, history of previous BMD testing, will be associated with different BMD testing rates.  H0-2: No factors associated with different utilization of BMD testing will be identified.   1.4.2 Promoting bone health management in women diagnosed with breast cancer: A pilot randomized controlled trial (Chapter Three, study 2) Rationale  Results from study 1 (Chapter Two) showed that less than 15% of women, aged ≥65 and diagnosed with breast cancer for three or more years, had at least one BMD test in any calendar year from 1995 to 2008. BMD testing plays a key role to identify women at higher risk of osteoporosis and osteoporotic fractures. It is important to identify an intervention to improve BMD testing rates. One way of promoting BMD is with educational material in the high-risk populations such as patients with recent fractures, and women 65 and over. However, it remains unclear whether educational material could improve BMD testing rates in another high-risk population – women diagnosed with breast cancer. In BC, prevention information sent to patients is primarily delivered by postal mail in primary care settings. With advances in communication technologies, there is an increasing interest in conveying information by email or text messaging. 38 A pilot randomized controlled trail (RCT) was designed to answer the questions - ”Does educational material improve BMD testing rates in older female breast cancer survivors” and “Do the different delivery methods of postal mail vs. patient choice of mail, email or smartphone text messaging for educational material, affect BMD testing rate differently?  Objective  To determine the feasibility of the RCT protocol by evaluating the response rate, recruitment rate and participation rate, and collecting information about effectiveness of the intervention and loss to follow-up, to inform design of a future large scale study.   Hypothesis  HA: This study protocol is feasible for a future large-scale study.   H0: This study protocol is not feasible for a future large-scale study.   1.4.3 Aromatase inhibitors are associated with a higher fracture risk than tamoxifen: a systematic review and, meta-analysis (Chapter Four, study 3) Rationale  Women diagnosed with breast cancer are at higher risk of osteoporosis and osteoporotic fractures, primarily due to adjuvant systemic breast cancer treatments, such as hormonal treatment. Aromatase inhibitors (AIs) and tamoxifen are the most common hormonal treatments in women diagnosed with breast cancer. Tamoxifen may slightly increase or decrease BMD by up to 2% in both pre- and post-menopausal women diagnosed with breast cancer. AIs significantly decrease BMD and have been associated with increased fracture risk in clinical 39 trials. The extents of effect of AIs and tamoxifen on fracture risk remain unclear. This study focuses on younger women aged 65 and under, and diagnosed with non-metastatic breast cancer. A better understanding on the effects of hormonal treatment on fracture risk may alter the future eligibility for BMD testing in younger women. Older women aged 65 and above are already eligible for BMD testing regardless of their breast cancer diagnosis and treatment. Women diagnosed with metastatic-breast cancer were excluded due to a high likelihood of pathological fractures associated with breast cancer.   Objective  To estimate fracture risk (risk ratios) in younger women aged 65 years and under, diagnosed with breast cancer, and treated with tamoxifen or AIs.   Hypotheses  HA-1: Tamoxifen increases fracture risk in younger women diagnosed with non-metastatic breast cancer.  H0-1: Tamoxifen does not increase fracture risk.  HA-2: AIs increases fracture risk in younger women diagnosed with non-metastatic breast cancer.  H0-2: AIs do not increase fracture risk.  HA-3: AIs increase fracture risk more than tamoxifen in younger women diagnosed with non-metastatic breast cancer.   H0-3: AIs do not increase fracture risk more than tamoxifen. 40 Chapter 2: Utilization of Bone Mineral Density Testing in Women Diagnosed with Breast Cancer in British Columbia, Canada (study 1) 2.1 Introduction: One in nine Canadian women will develop breast cancer in her lifetimes [180]. Almost 90% of these women will complete their initial cancer treatments. Most adjuvant systemic breast cancer treatments, including aromatase inhibitors (AIs), promote bone loss [66, 174]. Women with a history of breast cancer have a 32% higher prevalence of osteoporosis diagnosis than women without breast cancer history [143]. This leads to higher fracture rates compared with the general population [179].   Osteoporosis is a global medical issue with a high economic burden regardless of cancer history [181]. Osteoporotic fractures are associated with excessive mortality, physical function impairment, and more long-term care facility admissions [181]. Bone mineral density (BMD) testing using the dual-energy X-ray absorptiometry (DXA) technique plays a key role in osteoporosis screening and management. BMD testing can be used to screen individuals for osteoporosis before fractures occur. It is cost-effective to treat screen-detected osteoporosis in post-menopausal women to prevent fractures [87]. This strategy is associated with an 11% reduction in mortality associated with osteoporotic fractures [98]. For individuals with osteoporosis diagnosis, repeated BMD testing can be used to monitor treatment effectiveness by identifying individuals with persistent bone loss despite treatment.    Utilization of BMD testing remains unclear in women diagnosed with breast cancer in British Columbia (BC), Canada. This study was to provide an overall utilization picture of BMD testing 41 during the period from 1995 to 2008 in older female breast cancer survivors women aged 65 years and over, and diagnosed with breast cancer for three or more years.  2.2 Method  We conducted an observational study with two independent analyses using a provincial cancer registry and secondary administrative healthcare data linkage: (1) trend analysis to evaluate trends in the proportion of survivors with ≥1 BMD test by calendar year and osteoporosis diagnosis from 1995 to 2008 using a descriptive, serial cross-sectional study design, and (2) association analysis to evaluate associations between factors and BMD testing rates during the three-year period 2006-2008 using a cross-sectional study design. We reported study results using criteria from the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [182].   2.2.1 Study groups Women who had completed their initial breast cancer treatments with the exception of hormonal treatment were considered for this study. The care focus for this group has already shifted from active cancer treatment to surveillance and management of common long-term and late effects associated with the cancer or cancer treatment. The required duration of initial breast cancer treatment was analyzed in approximately 37,000 women diagnosed with breast cancer in BC using secondary data-linkage. The durations for different breast cancer treatments were as follows:  • Surgery alone, radiation alone or a combination of both surgery and radiation: 0.3±0.25 (mean ± standard deviation (SD)) year  42 • Chemotherapy: 0.25-0.5 year • Biological therapy: 1 year Nearly 99% of the women diagnosed with breast cancer completed their initial cancer treatments with the exception of hormonal treatment by the end of the third year after their initial breast cancer diagnosis. Hence, a three year survival time was chosen for this study.   For each observation period, we identified BC female residents from the provincial BC Cancer Agency (BCCA) registry, who were aged 65 and over; were diagnosed with breast cancer from 1989 (start of available data) to 2005; were diagnosed with breast cancer at least three years prior to the observation start date  and were not in the last year of their lives based on data from the provincial BC Cancer registry [183]. The BC Cancer Registry records virtually all new breast cancer cases among BC residents, includes treatment information, and is routinely linked with death registrations.  2.2.2 Data sources  We linked the BC Cancer Registry and BC Cancer Agency Breast Cancer Outcome Unit (BCOU) databases to the provincial healthcare administrative datasets using person-specific lifetime Personal Health Numbers. Personal Health Numbers are assigned to over 90% of BC residents who are eligible and covered under provincial health insurance plan for all “medically necessary” health services. All datasets used are summarized in Appendix A.  43 2.2.3 Outcome variable, bone mineral density test BMD tests were identified from the Medical Service Plan (MSP) Payment Information File [184] by fee codes 08688/08681/09810 (DXA, whole body), 08689/08682/09811 (DXA, single area) and 08696/08683/09812 (DXA, second area). For the trend analysis, BMD tests were identified for each survivor for each calendar year. Proportions of survivors with ≥1 BMD test were calculated for each calendar year as “number of survivors with ≥1BMD test” divided by “number of total survivors” for that year. The proportions were then stratified by survivors’ osteoporosis diagnosis which was made prior to 1st January of each calendar year and identified using diagnostic codes of International Classification of Diseases, Ninth Revision (ICD-9,  www.cdc.gov/nchs/icd/icd9.htm) and Tenth Revision (ICD-10, www.cdc.gov/nchs/icd/icd10.htm) from the MSP Payment Information File [184] and Discharge Abstracts Database [185] (Appendix B). For the association analysis, BMD tests were identified for each survivor for the three-year period 2006-2008. BMD testing rate during the period 2006-2008 was calculated as proportion of survivors with at ≥1 BMD test. A three-year period was selected to maximize capture of survivors who had BMD testing as recommended (one BMD test every 1-3 years) by the 2002 Canadian Osteoporosis guideline in effect during the study period in BC [84].   2.2.4 Potential modifying factors for association analysis We selected socio-demographic and clinical factors that could potentially influence utilization of BMD testing based on literature review and discussions with clinicians. Attained age was determined using birth date and observation period start date. Socio-economic status (SES) quintile, based on the average income per person in the survivor’s 2006 census enumeration dissemination area (identified using postal codes from the Consolidation File [186]), was created 44 using the Statistics Canada Postal Code Conversion File, Version 5J [187]. Categorization of urban/rural residential status in year 2006 was based on population size and socio-economic homogeneity using census data and a methodology developed by Statistic Canada [188]. The health service regions were categorized by the five regional health administrative areas in BC as of January 1st, 2006. Survivors (1.2%) with unknown urban/rural residential status, SES status or health service regions in 2006, were assigned with their latest available information prior to 2006. The five major non-cancer chronic diseases (coronary heart disease, cerebrovascular disease, chronic pulmonary disease, diabetes mellitus, and dementia) [189], as well as osteoporosis diagnosis and recent osteoporotic fractures, were identified separately using corresponding diagnostic ICD-9 and ICD-10 codes from the MSP Payment Information File [184] and Discharge Abstracts Database [185] (Appendix B). These five major chronic diseases account for 35% of mortality among Canadians (www.statcan.gc.ca) [190]. Each survivor’s chronic disease count was based on the number of chronic disease(s) identified prior to January 1st, 2006. Survivors with three or more chronic disease counts were grouped together due to their small sample sizes. Osteoporosis diagnosis was classified as “yes” for survivors with any osteoporosis diagnosis prior to January 1st, 2006. Osteoporotic fractures were defined as spine, hip or wrist fracture(s). Recent osteoporotic fracture was classified as “yes” for survivors who had at least one fracture within the six months prior to their BMD test, or survivors who had fractures but no BMD test, and “no” for remaining survivors. Previous BMD tests were identified with the same fee codes for the outcome variable - BMD test. Nursing home residence status on January 1st, 2006 was determined using nursing home services associated fee codes from the MSP Payment Information File (Appendix B) [184]. Age at initial cancer diagnosis was calculated using birth date and date of initial cancer diagnosis. Time since initial breast cancer 45 diagnosis was calculated by the interval between date of initial cancer diagnosis and January 1st, 2006. Stage of initial breast cancer diagnosis was obtained from the BCOU dataset. Detailed initial breast cancer treatment information was retrieved from the BCOU dataset and categorized either by type of treatments (surgery, radiation, systematic treatment, and combinations of two or three treatments) and type of initial hormonal treatment (none, tamoxifen only, and aromatase inhibitors (AIs)/ovary suppression) respectively.   2.2.5 Statistical analysis We assessed characteristics of the study group by examining the distribution of frequency counts and percentages of key variables. Trends in proportion of survivors with ≥1 BMD per calendar year by survivors’ osteoporosis diagnosis from 1995 to 2008 were evaluated using log-linear models. Up to two join points per model were fit using the Joinpoint Trend Program, Version 4.3.1.0 [191]. The statistical significance of each join point was tested using a Monte Carlo permutation procedure. Trend segments were created between join points. Average percent changes (APCs) were estimated for each trend segment. Average annual percentage changes (AAPCs) were estimated for the entire observation period from 1995 to 2008. Each AAPC was calculated as a weighted average of APCs with weights equal to the length of each corresponding trend segment. Associations between factors and BMD testing rates during the three-year period 2006-2008 were evaluated using log-binomial models. We did not use a traditional logistic model approach to avoid overestimated associations in a common outcome situation [192]. All prevalence ratios and 95% confidence intervals (CIs) were adjusted for socio-demographic factors, including attained age, SES, health service region, and urban/rural status using log-46 binomial models. Log-binomial models were fit using Statistical Analysis System version 9.3 (SAS Institute Inc., Cary, NC).  2.3 Results  2.3.1  Trend analysis The eligible survivor group nearly doubled from 4,974 in 1995 to 9,662 in 2008 period (Table 2-1, Figure 2-1). The prevalence of osteoporosis diagnosis increased from 295 (6% of all survivors) to 2,475 (25.6%) over the same period.   The proportions of survivors with ≥1 BMD were under 20% for any calendar year from 1995 to 2008. For survivors with osteoporosis diagnosis, the proportions with ≥1 BMD test during a calendar year increased from 4.4 % in 1995 to 16.8 in 2006 and then decreased slightly to 15.5% in 2008. On average, the proportions increased by 19.4% annually (95% CI=11.5 to 28.0) from 1995 to 2002 and remained relatively stable around 16% from 2002 to 2008. For the survivors without osteoporosis diagnosis, the proportions with ≥1BMD test during a calendar year increased from 0.8% in 1995 to 9.3% in 2005 and then decreased slightly to 8.2% in 2008. On average, the proportions with ≥1 BMD test during a calendar year increased annually by 33.4% (95% CI=24.6 to 42.9) from 1995 to 2001 and by 12.4% (95% CI=0.9 to 25.2) from 2001 to 2005. The proportions remained relatively stable around 8.5% in 2005-2008.     47 Table 2-1  Trends in proportion of survivors with at least one bone mineral density test, overall and stratified by osteoporosis diagnosis from 1995 to 2008  Overall Osteoporosis No osteoporosis     Year Total  BMD a (%) Total BMD a (%) Total BMD a (%) 1995 4974 1.0 295 4.4 4679 0.8 1996 5217 2.1 365 6.6 4852 1.7 1997 5657 1.8 455 5.5 5202 1.5 1998 6086 3.2 546 8.1 5540 2.7 1999 6413 3.5 676 9.3 5737 2.8 2000 6747 4.6 799 8.8 5948 4.1 2001 7119 7.0 937 14.2 6182 5.9 2002 7510 7.8 1087 14.9 6423 6.6 2003 7931 8.5 1299 14.3 6632 7.4 2004 8266 9.8 1498 16.4 6768 8.3 2005 8590 10.5 1683 15.5 6907 9.3 2006 8909 10.7 1956 16.8 6953 8.9 2007 9278 10.6 2187 16.5 7091 8.8 2008 9662 10.1 2475 15.5 7187 8.2 Trend 1- Period   1995-2001  1995-2002  1995-2001 APC b 95% CI   32.6 25.7, 39.9  19.4 11.5, 28.0  33.4 24.6, 42.9 Trend 2- Period   2001-2005  2002-2008  2001-2005 APC b 95% CI   10.9 4.6, 17.7  1.2 -1.7, 4.1  12.4 0.9, 25.2 Trend 3 -Period   2005-2008    2005-2008 APC b 95% CI  -2.8 -10.2, 5.3    -4.0 -13.3, 6.4 AAPC c  95% CI   17.7 14.2, 21.3  10.4 6.7, 14.2  17.6 12.7, 22.8 BMD bone mineral density, APC annual percent change,  AAPC average annual percent change, CI confidence interval a  Proportions of survivors who had at least one BMD test(s)  b  Calculated as (ebeta-1)*100 using log-linear models. The beta equals to the coefficient of the regression model. c  Calculated as weighted average of APCs  48  Figure 2-1 Numbers of female breast cancer survivors and proportions of women with at least one BMD test by osteoprososis diagnosis and calendar year from 1995 to 2008 WHO World Health Organization, ATAC Arimidex, Tamoxifen, Alone or in Combination, CAROC Canadian Association of Radiologists and Osteoporosis Canada, APC average percent change, FRAX fracture risk assessment tool, CI confidence interval References included from left to right: 1994 WHO Report [95]; 1996 Canadian Guidelines [193]; development and validation of the osteoporosis risk assessment instrument to facilitate selection of women for bone densitometry [194]; 2002 Canadian guideline [84]; 2004 WHO report [88]; 2004 Canadian Task Force [175]; CAROC 10-year absolute fracture risk assessment [110]; 2005 Canadian guidelines – follow-up after treatment for breast cancer [176]; 2008 FRAX [195]           49 2.3.2 Association analysis From the initial survivor groups, we identified 7,632 eligible survivors with complete data during the period 2006-2008, to assess for the analysis of associations between potential clinical and socio-demographic factors, and BMD testing rates. Slightly more than half of the survivors were aged 75 years and over at the start of the observation period, and 13% of the group were living in rural areas (Table 2-2). More than 80% of survivors had at least one of five selected chronic diseases. The prevalence of osteoporosis diagnosis at the end of 2005 was 21.6%. Recent osteoporotic fracture history was found among 5% of survivors. Slightly less than 40% of survivors received initial hormonal treatments. The BMD testing rate within the three-year period 2006-2008 was 26.5%.       50 Table 2-2   Characteristics of female breast cancer survivors for associations between factors and bone mineral density testing rates during the period 2006-2008  N = 7632 % Attained age    65-74 3535 46.3 75+ 4097 53.7 Health Authority region   Vancouver Coastal  1791 23.5 Interior 1509 19.8 Fraser 2305 30.2 Vancouver Island 1711 22.4 Northern 316 4.1 SES   5 (highest)  1503 19.7 4 1456 19.1 3 1446 18.9 2 1623 21.3 1 (lowest)  1597 20.9 Unknown  7  < 0.1 Urban/rural residential status   Metropolitan (≥ 100,000) 5003 65.6 Large community (99,999 - 50,000) 468 6.1 Small community  (49,999 - 10,000) 1164 15.3 Rural (< 10,000) 997 13.1 Chronic disease count   0 1489 19.5 1 3406 44.6 2 2074 27.2 3-5 663 8.7 Osteoporosis    No 5983 78.4 Yes 1649 21.6 Previous BMD test (year 2003-2005)   No 5420 71.0 Yes 2212 29.0 Recent osteoporotic fracture    No 7249 95.0 Yes 383 5.0 Nursing home residence   No  7561 99.1 Yes 71 0.9 Age at initial breast cancer diagnosis (years)   < 50 502 6.6 50-59  1636 21.4 60-69 3243 42.5 ≥ 70  2251 29.5 SES socio-economic status, BMD bone mineral density  51 Significantly different BMD testing rates were associated with all identified factors except chronic disease count and stage at initial breast cancer diagnosis (Table 2-3). Interaction terms between osteoporosis diagnosis and other factors were examined but none was found. Significantly higher BMD testing rates were associated with either osteoporosis diagnosis (adjusted prevalence ratio (aPR)=2.39, 95% CI=2.12 to 2.69); or previous BMD tests in 2003-2005 (aPR=3.87, 95% CI=3.46 to 4.32). Significantly lower BMD testing rates were seen among survivors who were aged 75 and over (aPR=0.47); lived in the Fraser Health service region (0.72); lived in the Northern Health service region (0.66); had lower SES (range 0.66 to 0.78); lived in rural areas (0.70); had at least one selected chronic disease (range 0.62 to 0.79); had a recent osteoporotic fracture history (0.21); or were nursing home residents (0.05), compared with corresponding reference groups. BMD testing rates were 20-30% lower in survivors with low SES vs. high SES, in a dose-dependent manner (p <0.01). Among factors associated with breast cancer diagnosis, BMD testing rates were positively associated with the treatment combination of surgery/systemic/radiation (1.23); or tamoxifen treatment (1.29), compared with corresponding reference groups. Compared with survivors diagnosed with initial breast cancer 0-10 years ago, BMD testing rates were higher in survivors diagnosed more than 30 years ago 1.46), but lower in survivors diagnosed 11-20 (0.85) and 20-30 years ago (0.78).     52 Table 2-3   Associations between factors and bone mineral density testing rates in older female breast cancer survivors during the period 2006-2008   N =7625 a BMD testing rates b p value  aPR c 95% CI Attained age       65-74 3533 34.4  1.00 -- ≥ 75 4092 19.8 <0.01 d 0.47 0.42, 0.52 Health service region      Vancouver Coastal 1791 29.8  1.00 -- Interior 1506 26.5  1.01 0.84, 1.21 Fraser 2303 22.8  0.72 0.62, 0.83 Vancouver Island 1710 29.5  1.07 0.92, 1.26 Northern 315 20.6 <0.01 0.66 0.48, 0.91 SES      5 (highest)  1503 32.3  1.00 -- 4 1456 26.7  0.78 0.66, 0.92 3 1446 26.2  0.78 0.67, 0.92 2 1623 25.1  0.74 0.63, 0.87 1 (lowest)  1597 22.7 <0.01 d 0.66 0.56, 0.78 Urban/rural residential status      Metropolitan (≥ 100,000) 5000 27.0  1.00 -- Large community (99,999 - 50,000) 468 32.5  1.18 0.94, 1.49 Small community  (49,999 - 10,000) 1163 25.4  0.85 0.72, 1.00 Rural (< 10,000) 994 22.8 <0.01 0.70 0.58, 0.84 Chronic disease count      0 1487 34.9  1.00 -- 1 3402 26.9  0.79 0.69, 0.90 2 2074 21.7  0.62 0.53, 0.72 3-5 662 21.5 0.28 d 0.64 0.51, 0.80 Osteoporosis       No 5976 22.8  1.00 -- Yes 1649 40.2 <0.01 2.39 2.12, 2.69 Previous BMD test (2003-2005)      No 5414 17.9  1.00 -- Yes 2211 47.8 <0.01 3.87 3.46, 4.32 Recent osteoporotic fracture e       No 7242 27.6  1.00 -- Yes 383 6.3 <0.01 0.21 0.14, 0.32 Nursing home residence      No  7554 26.8  1.00 -- Yes 71 1.4 <0.01 0.05 0.01, 0.39 Age at initial breast cancer diagnosis      < 50 501 34.9  1.00 -- 50-59 1634 30.4  0.82 0.66, 1.02 60-69 3241 29.6  0.91 0.75, 1.12 ≥ 70 2249 17.5 <0.01 d 0.64 0.50, 0.81       53 Continued       N =7625 a BMD testing rates b p value  aPR c 95% CI Time since initial breast cancer diagnosis (years) 0-10 3875 28.7  1.00 -- 11-20 2936 24.1  0.85 0.76-0.95 21-30 678 23.8  0.78 0.65-0.95 30+ 136 33.1 <0.01 d 1.46 1.00-2.11 Stage at initial breast cancer diagnosis f      I 3443 27.5  1.00 -- II 2109 28.7  1.05 0.93-1.19 III 275 25.1 0.82 0.88 0.66-1.17 Initial breast cancer treatment(s) g      Surgery only  691 24.9  1.00 -- Surgery + Systemic 1026 27.1  1.03 0.82-1.29 Surgery + Radiation 1447 23.6  0.87 0.70-1.08 Surgery + Systemic + Radiation 2201 31.9 <0.01  1.23 1.01-1.50 Initial hormonal treatment(s) h      None  2488 24.9  1.00 -- Tamoxifen only   2881 30.2  1.29 1.14-1.46 AIs or ovary suppression  19 47.4 <0.01  2.46 0.97-6.22 Values in bold and italic indicate statistical significance BMD bone mineral density, aPR adjusted prevalence ratio, CI confidence interval, Ref reference, SES socio-economic status, AIs aromatase inhibitors a Survivors with unknown SES status were excluded for the entire analysis  b Calculated as portions of survivors with at least one BMD test c  PR was adjusted for age, SES, urban/rural status of residence and health service region using a logistic regression model d p for trend e  Included hip, spine and wrist fracture f  Survivors with unknown stage information were not included for this analysis  g Systemic treatment, radiation and systemic treatment with radiation were not included for multivariate analysis due to small sample sizes and a lack of clinical meanings h  Survivors with unknown hormonal treatment information were not included for this analysis    54 2.4 Discussion This is the first population-based study to evaluate utilization of BMD testing among older female breast cancer survivors; namely women aged 65 and over, and diagnosed with breast cancer for three or more years in BC, Canada. Improved survival rates over time lead to a fast-growing group of women diagnosed with breast cancer. In BC alone, the annual estimated number of women surviving breast cancer will increase by 80% from 2,600 in 2000 to 4,675 in 2028 [62]. These survivors are at higher risk of osteoporosis and osteoporotic fractures. BMD tests play a key role in screening for osteoporosis among survivors without osteoporosis diagnosis, and monitoring treatment effectiveness among survivors with osteoporosis diagnosis. Our results showed relatively stable proportions of survivors with ≥1 BMD test for each year from 2005 to 2008. Only 26.5% of survivors aged 65+ received ≥1 BMD test over the three-year period 2006-2008; however, one BMD test at a one- to three-year interval for women aged 65+, regardless of breast cancer diagnosis, was recommended by the Canadian guideline at that time [84, 176]. The utilization of BMD testing is sub-optimal, compared with other disease screening in the BC population [196], in older female breast cancer survivors in BC.   2.4.1 Trend analysis The osteoporosis prevalence rate in 2008 in our study was 25.6%, based on ICD-9 and ICD-10 codes. This is higher than the rate of 20-25% for women aged 60-69 years in the Canadian community based on actual BMD measurement using the World Health Organization Criteria [78], but lower than the self-reported rates of 33.4% in the US community and 27.7% in women with breast cancer history in the US [143]. For survivors without osteoporosis diagnosis, our study’s annual BMD testing rate in 2001 was 5.9%, compared with 13.3% in the US based on a 55 combination of ICD-9 codes and current procedural terminology [197]. For survivors with osteoporosis diagnosis, no other studies were found for rate comparison. These international differences could be explained by different methodologies in identifying osteoporosis diagnosis, survival time (a range from zero to five years), and measurement time used for rate calculation (a range from one to three years) and time frame used to calculate rates (ranges from one to three years and from 1997 to 2006).  For survivors with osteoporosis diagnosis in our study, we observed that the trends in proportion of survivors with ≥1 BMD test changed from being positive to stable in the year 2002.  The positive trend up to 2002 may be due to the 1996 Canadian guideline first recommending repeated BMD measurements to monitor bone loss for women with osteoporosis diagnosis [193]. The stable trend since 2003 could be because the 2002 Canadian guideline suggested a longer monitoring interval between repeated BMD tests to avoid mistaking random fluctuations for real changes [84]. For survivors without osteoporosis diagnosis in our study, we observed that the trends changed from being positive to weak positive in years 2001 and from being weak positive to stable 2005. The positive trend up to 2001 was possibly due to the 1994 WHO report and again, the 1996 Canadian guideline recommending screening of high-risk individuals for osteoporosis using BMD tests [95, 193]. From 2001 to 2005, the trend continued to grow at a slower rate possibly due to two factors. First, high-risk individuals were extended from “peri/postmenopausal women without hormonal therapy” to women with validated risk factors [84, 88, 175, 194]. Second, most fractures happened among individuals with normal BMD measurements [90, 198]. This led to a shift from identifying “high-risk individuals for osteoporosis based on BMD measurements” to “high-risk individuals for fractures using a 56 combination of BMD measurement and validated risk factors” [88]. Since 2005, the trend became stable due to two possible factors. The first fracture risk calculation tool – Canadian Association of Radiologist and Osteoporosis Canada (CAROC) - was adopted in 2005 [97, 110]. Each individual’s fracture risk would be calculated using the combination of BMD measurements and validated risk factors, such as age and gender. Individuals with low fracture risk were suggested to repeat BMD screening test at longer intervals, which could decrease proportions of survivors receiving BMD testing. On the other hand, the Canadian breast cancer follow-up guideline suggested screening breast cancer women treated with AIs for osteoporosis [176]. This should increase proportions of survivors receiving BMD testing.  Several non-guideline factors might also be associated with the trend changes in the proportion of survivors with ≥1 BMD test. The positive trends from 1995 to early 2000s possibly reflected increases in availability of DXA machines, awareness of higher osteoporosis risk associated with systematic adjuvant breast cancer treatments (such as chemotherapy and hormonal treatments) among survivors and physicians, and usage of systemic adjuvant breast cancer treatments in women with early-stage breast cancer. The positive trends might be associated with prioritized access to diagnostic tests for specialists including oncologists over family doctors in Canada [199]. However, our further internal analysis showed that only 12-17% of BMD tests were ordered by specialists. This suggests that specialists’ potentially privileged access to diagnostic tests had a minor impact on the usage of BMD tests in survivors in BC.   57 2.4.2 Association analysis In our study, multiple factors were associated with different BMD testing rates. Being ≥75 years of age was associated with lower BMD testing rates in our and the Snyder et al. study [113]. These survivors aged 75 and over are at significantly higher osteoporotic fracture risk while BMD test screening remains cost-effective up to age 80 [200]. Survivors under age 80 should be encouraged to have BMD tests. A direct relationship between SES and BMD testing rates was observed in survivors in our study. This is consistent with findings in the general population in the province of Manitoba, Canada [114]. Further studies are needed to better understand the nature of these associations. Our results showed urban-rural disparity in BMD testing rates, too. This could be due to lower DXA machine availability in rural BC areas. Chronic disease history was associated with lower BMD testing rates in our study. We observed an insignificant trend between chronic disease count and utilization of BMD testing while a negative trend was found in the Snyder et al. study [113]. This could result from different definitions and categories of chronic disease count.   In our study, survivors without osteoporosis diagnosis were less likely to have BMD tests, compared to survivors with osteoporosis diagnosis. Further analysis showed that around 50% of the survivors without osteoporosis did not have any BMD tests for six consecutive years from 2003 to 2008. These survivors may skip BMD tests as they underestimated their personal osteoporosis risk [119].   In our study, survivors residing in nursing homes or with recent osteoporotic fractures were significantly less likely to have BMD tests, while nursing home residence and recent 58 osteoporotic fractures are associated with significantly higher fracture risk [201, 202]. BMD testing might be skipped for these two groups due to patient or physician factors. For nursing home residents, the common patient factor is limited mobility or cognitive impairment, resulting in difficult transportation from a nursing home to a testing machine, or difficult maintaining a steady position during the test. The common physician factor is that physician may consider BMD tests futile due to unproven cost-benefit effectiveness, polypharmacy and short life expectancy [100]. Patients with recent osteoporotic fractures may be resistant to osteoporosis diagnosis with BMD testing while their physicians may initiate fracture-prevention treatment without BMD measurement, or skip BMD testing for frail patients [123].  In our study, high BMD testing rates were associated with several clinical factors. A higher BMD testing rate was observed among survivors who were diagnosed with breast cancer more than 30 years ago. This is likely because more than half of these survivors (53%) have been diagnosed with osteoporosis, and having an osteoporosis diagnosis is associated with high utilization of BMD testing. Higher BMD testing rates were observed among survivors who were diagnosed with breast cancer at younger ages or received a treatment combination of surgery, adjuvant systemic treatment, and radiation. These survivors were likely to receive aggressive hormonal treatment, such as ovarian suppression. This could also lead to higher utilization of BMD testing.   Although tamoxifen is not approved for treating osteoporosis in the general population or breast cancer survivors by the Food & Drug Administration, this therapy may preserve bone mass in breast cancer survivors[174], which could lessen physicians’ likelihood of ordering BMD tests. 59 However, tamoxifen treatment was associated with higher BMD testing rates in our study. This finding might be explained by our inability to distinguish survivors switching to AIs after tamoxifen from survivors receiving tamoxifen only.   2.4.3 Limitations and future directions The main limitation of this study is data availability. This prevents us from examining more recent BMD testing rates however is balanced by the completeness of this linked data set. In this study, we evaluated utilization of BMD testing in BC only from 1995 to 2008 due to data availability. Since 2002, we observed stable trends in BMD testing for survivor with or without osteoporosis diagnosis. Since 2008, utilization of BMD testing may have remained stable or changed. Utilization may rise due to greater awareness of the osteoporosis care gap and extending fracture risk assessment to any individuals 50+ [97]. But utilization may drop due to a higher availability of validated fracture risk assessment tools. In 2008, the Canadian version of WHO Fracture Risk Assessment Tool (FRAX) became available in Canada [195]. FRAX assessment without BMD measurements has been used to screen eligibility for DXA BMD tests in BC. This might reduce utilization of BMD testing. Future studies are needed to better understand more recent utilization patterns of BMD testing.  Chronic diseases, osteoporosis diagnosis and diagnostic fractures were identified using diagnostic ICD-9 and ICD-10 codes from outpatient services (MSP Payment Information File) and inpatient services (Discharge Abstracts Database) in this study. ICD-9 and ICD-10 codes are commonly used to identify diseases in data linkage studies, since the ICD classification is commonly required to document diagnoses in administrative records. Diagnostic codes selected 60 in this study were not validated but based on other published studies for consistency. Potential bias associated with mis-recording should be considered when interpreting the study results [203].   In this study, we only had a small number of survivors receiving initial hormonal therapy with AIs or ovary suppression to examine associations of factors and BMD testing rates during the period 2006-2008. The study group used here was women diagnosed with breast cancer prior to 1st, January 2003 who had survived three years or more as of 1st, January 2006. AIs were first introduced in BC around 2003 and were first recommended as first-line hormonal treatment for postmenopausal women in 2005 [167]. Postmenopausal survivors diagnosed before 2003 were likely to receive tamoxifen initially and switched to AIs later. These survivors would be identified as the tamoxifen group in this study.   Approximately 7% and 21% of selected survivors in our study were diagnosed with breast cancer before age 50 and at age 50-59 respectively. These survivors were at increased osteoporosis risk if they were pre-menopausal at time of breast cancer diagnosis and became amenorrheic after completing chemotherapy (chemo-induced amenorrhea). We were unable to identify these survivors as the chemo-induced amenorrhea status was not recorded in the data.   Approximately 30% of survivors had unknown initial breast cancer treatment. Those were survivors diagnosed more than 20 years ago; diagnosed at early stages that did not require treatments other than surgery; or receive cancer treatments in a few community clinics outside of BCCA administration.   61 2.5 Conclusion  BMD testing rates over the three-year period 2006-2008 for breast cancer survivors in BC, Canada are far lower than other disease screening.  Lower SES and rural residence were associated with low BMD testing rates. Low BMD testing rates were also associated with other factors, including advanced age, nursing home residence, having recent osteoporotic fractures, or not having previous BMD tests. These survivors with lower SES or in rural areas should be encouraged to have BMD tests as recommended by the Canadian guidelines.  62 Chapter 3: Promoting Bone Health Management in Women Diagnosed with Breast Cancer: A Pilot Randomized Controlled Trial (study 2)  3.1 Introduction Osteoporosis affects an estimated 200 million women globally, as osteoporotic fractures occur in one in every three women over age 50 in their lifetime [76, 79, 80]. Osteoporotic fractures lead to excessive mortality, impaired physical function, and more long-term nursing home stays [204-206]. The incidence and economic burden of osteoporotic fractures are increasing over time [207]. It is important to find an intervention to prompt bone health management which prevents osteoporotic fractures. Compared with women without breast cancer, women diagnosed with breast cancer are at higher risk of osteoporosis and osteoporotic fractures [147, 179] due to the negative effects of certain breast cancer treatments [66, 174]. Bone mineral density (BMD) testing, a key to good bone health management, can identify high-risk women before fractures occur. The majority of women, aged 65 and above, and not on osteoporosis medications, have a moderate fracture risk (10-20%) and should have one bone mineral density (BMD) test at a one- to three-year interval per Canadian osteoporosis guidelines [97].   Patient educational material improves BMD testing rates in high-risk patients with recent fractures, or aged ≥65 [128, 129]. It remains unclear whether patient educational material would improve BMD testing rates in another high-risk population – women diagnosed with breast cancer. In the British Columbia (BC) primary care setting, information is primarily delivered to patients by postal mail. With advances in communication technologies, there is a growing interest in conveying information by text messaging or email [208, 209]. Little is known whether a patient’s choice of delivery method for educational material – e.g., postal mail, email or 63 smartphone text messaging - affects patient behavior, such as BMD testing rates, differently, compared with postal mail. A randomized controlled trial (RCT) protocol has been designed to evaluate the effects of educational material and its delivery methods on BMD testing rates. In this study, we assessed the feasibility of the study protocol and pilot-tested the intervention effects.   3.2 Methods  3.2.1 Study protocol A randomized, unblinded, three-armed (parallel), controlled trial evaluating the effects of interventions, “educational material delivered by postal mail” and “educational material delivered by the patient choice of postal mail, email or smartphone text messaging”, on BMD testing rates in women aged 65 and over, and diagnosed with breast cancer for three or more years in BC, Canada.    3.2.2 Recruitment and randomization  Inclusion criteria were (1) female aged 65-75 on July 1st, 2015 and diagnosed with Stage 0-III breast cancer in 2010-2012, (2) a valid current address in BC, (3) no chemotherapy or radiotherapy for any cancers, (4) fluent English in reading and speaking, (5) no BMD testing with DXA over the three-year period from July 1st, 2012 to June 30th, 2015, (6) no osteoporosis medication in the past 12 months before being recruited, (7) signed consent, and (8) returned their pre-study questionnaire with their choice in delivery methods for educational materials.   64 We randomly selected 398 survivors fulfilled the inclusion criteria (1)-(3) from the provincial BC Cancer Registry [210], a provincial dataset that includes information of BC residents diagnosed with cancer (Figure 3-1). Each selected survivor’s demographic, cancer diagnosis, and family doctor information was also retrieved from the same registry. During the initial screening, survivors were excluded and not invited if they did not have identifiable actively practicing family doctors on the BC Cancer Registry and the College of Physicians and Surgeons of BC website (www.cpsbc.ca/physician_search) (n=49), For the 349 survivors with a practicing family doctor, a questionnaire on the potential subjects eligibility was faxed to the physician. The family doctors indicated that 44 survivors were unfit (primarily due to poor health) for our study. An invitation package comprising an invitation letter, duplicated consent forms, and pre-study questionnaires was postal-mailed to each of the remaining 305 survivors. Up to three follow-up phone calls were made at different times of the day, on weekdays and weekends, to the survivors who did not respond within two weeks after their mail out dates. Of the 305 survivor invited, 251 were excluded if they were unable to be reached by both postal mail and phone calls (n=69), were not interested in participating this study (n=84), or were ineligible based on our inclusion criteria (4)-(8) (n=98). The criteria (4) and (5) were confirmed by participants’ reports. The criteria (6) was confirmed by medication dispensing records reviewed by a pharmacist through PharmaNet, a province-wide network linking all BC pharmacies that records outpatient prescription medications dispensed to any individual anywhere in BC (www.bcpharmacists.org/pharmanet).   65 From February to May 2015, we successfully recruited 54 participants who were then randomized in a 1:1:1 ratio by blocks of three or six into three groups: control group and two intervention groups. The randomization sequence was computed before the recruitment by statistician (JS) and saved as a hard copy by OT. The research assistant recruiting participants contacted OT by email after each participant was recruited. Both OT and JS were blinded to the participants.  66 Recruitment, Consent, Eligibility Assessment  Potentially eligible survivors selected from the  BC Cancer Registry  N=398         • No identifiable actively practicing family doctors (N=49)  Pre-screening by family doctor N=349      • No longer with family doctors (N=19)  • Unfit for this study (N=25) a  Invitation package (including pre-study questionnaire package) sent by postal mail N= 305       • Unable to contact (N=69) b  Responded  N= 236 ** Response rate: 77.4% (236/305)    Excluded (N=182) • Did not meet inclusion criteria  (N=98)      BMD tests prior to the study:56     On osteoporosis medication:24    Language barrier:10    Others:8  • Declined to participate  (N=84)       Allocation  Recruited eligible participants for block randomization, block =3 or 6 N=54 ** Recruitment rate: 13.6% (54/398) ** Participation rate: 39.1 %  (54/138)         Control, no intervention c N=18  Educational material delivered by postal mail,  N=19  Educational material delivered by patient choice d N=17  Follow-up        Lost to follow-up (N=0)  Lost to follow-up (N=0)  Lost to follow-up (N=0)              Analysis   Analyzed  (N=18)   Excluded from analysis (N=0)    Analyzed  (N=19)  Excluded from analysis (N=0)    Analyzed  (N=17)  Excluded from analysis (N=0) Figure 3-1 Consolidated Standards of Reporting Trails (CONSORT) flow diagram a  In poor health condition, nursing home residence, language barrier and other reasons, which were determined by their family doctors  b By postal mail or phone call c  No intervention during the study period. Educational materials sent after the study’s completion d Postal mail, email or smartphone text messaging  67 3.2.3 Interventions  The two interventions consisted of (1) educational material being delivered by postal mail, and (2) educational material being delivered by patient choice of postal mail, email or smartphone text messaging.   The educational material comprised of two parts: (1) three pages of information on osteoporosis, potential effects of breast cancer treatments on bones, BMD testing, lifestyle advice on exercise, calcium intake, and vitamin D intake, and advice to review osteoporosis risk with her family doctor (Appendix C); and (2) one double-sided page of risk factors based on the 2010 Canadian osteoporosis guidelines and the fracture risk assessment tool (FRAX) developed by the World Health Organization (Appendix D) [97, 195]. Education material was sent to to participants in the intervention groups immediately after being randomized and to control group participants after study completion.  The educational material were created by me, were edited by a material development expert from the Centre of Excellence in Cancer Prevention (https://cancerprevent.ca/), reviewed by three female volunteers aged 65-75 without any breast cancer diagnosis, and then finally reviewed by my committee members.   3.2.4 Self-reported participant questionnaires Pre-study and post-study questionnaire packages were postal-mailed to collect information from participants during recruitment (baseline) and at six months. The packages included a set of four individual self-report questionnaires: (1) a general or outcome questionnaire, (2) a Godin Leisure-Time Exercise Questionnaire (GLTEQ) [211], (3) a Vitamin D & Sun (VIDSUN) 68 questionnaire [212], and (4) a Calcium Assessment Tool (CAT) [213]. The general and the outcome questionnaires collected information on demographics, osteoporosis, choice of delivery method for the educational material, and BMD testing with DXA; and were pilot-tested on two volunteers. The GLTEQ, CAT and VIDSUN questionnaire are validated tools. The GLTEQ included four items to assess an individual’s exercise habits in a typical week. The GLTEQ is a relative easy-to-use tool and has been recommended for cancer survivor research by the Division of Cancer Epidemiology & Genetics of the National Cancer Institute [214]. The VIDSUN questionnaire incorporated five items to evaluate an individual’s risk of vitamin D deficiency. The CAT included 27 items to measure an individual’s daily calcium intake in a typical week. The VIDSUN questionnaire and CAT have been validated in this age group [212, 213].  3.2.5 Procedure / intervention outcome measures The feasibility of the protocol was evaluated by the effectiveness of the recruitment strategy, the representativeness of the recruited participants, and the completeness of outcome measures.   Each participant’s outcomes were measured at six months after she was randomized. Outcomes were measured using participant questionnaires, BMD testing reports retrieved from family doctor’s office and imaging facilities, and medication dispensing records (Table 3-1).   Our primary intervention outcome was BMD testing rates – the proportion of participants who had BMD tests during their six-month follow-up period based on questionnaire responses and BMD testing reports. Five secondary intervention outcomes were either evaluated once at six months (new osteoporosis diagnosis and initiation of osteoporosis medication from questionnaires, BMD and medication dispensing records) or measured as changes from baseline 69 (exercise, risk of vitamin D deficiency and calcium intake from questionnaires). Each participant’s exercise habits in a typical week were calculated as a leisure score index using the formula [(units of strenuous exercise*9) + (moderate*5) + (light*3)] [211]. VIDSUN questionnaire responses were scored first. Each participant’s scores were then tallied and categorized as at high or low risk of vitamin D deficiency [212]. Each participant’s daily calcium intake was calculated as mg per day based on the combination of food consumption, the calcium content in that food, and calcium supplements [213].  Table 3-1   Data sources for outcome measures  Participant questionnaire Family doctor BMD imaging facility Medication dispensing record Primary outcome     BMD testing rate      Secondary outcomes     Newly diagnosed osteoporosis   BMD reports  BMD reports  Initiation of osteoporosis medication      Changes in weekly exercise      Risk of vitamin D deficiency     Changes in daily calcium intake       3.2.6 Statistical analysis The characteristics among the three groups of the participants, non-participants who declined to participate in this study and other non-participants were compared using the chi-square test. The characteristics of the study participants were evaluated using descriptive analysis, either proportions, means with standard deviation (SD) or median with range (for skewed distributions). The changes on leisure score index and daily calcium intake were evaluated for each group using the means of individual differences between baseline and six months with 95% confidence intervals (CI). The effect of the educational material and its delivery method on BMD 70 testing rates were estimated using rate differences with 95% CI between the control group and the combined intervention groups (postal mail and patient choice groups), and between postal mail group and patient choice group. All analyses were performed Statistical Analysis System version 9.3 (SAS Institute Inc., Cary, NC).   3.2.7 Ethics and clinical trial registration  Ethics was approved by the Clinical Ethics Board of the University of BC (H15-00849). This study protocol has been registered at www.clinicaltrials.gov with the registration number NCT02484131.  3.3 Results  3.3.1 Feasibility of the study protocol  The response rate, defined as the proportion of women who responded to our invitation, was 77.4%. The participation rate, defined as the proportion of eligible participants who consented to participate, was 39.1% with an overall recruitment rate, defined as the proportion of participants from the original 398 survivors, of 13% (Figure 3-1). Similar distributions of age at diagnosis, stage of cancer at diagnosis, treatment and region of service were observed among the 54 participants, 84 non-participants who declined to participate in this study, and 260 other non-participants (Table 3-2). The primary and five secondary intervention outcomes were measured for 98% and 78-100% of the 54 recruited participants respectively. One or more missing values were noted in 4-19% of the returned GLTEQ, VIDSUN, and CAT questionnaires at baseline and six months (Table 3-3).    71   Table 3-2   Representativeness of participants  Participants Non-participants   p value b      (N=54) Others a (N=260) Declined to participate (N=84) Age 70.3 ± 3.7 69.6 ± 6.9 70.6 ± 3.1 0.32 Breast cancer stage In situ 8  (14.8) 35  (13.5) 9  (10.7)  I  27  (50.0) 126  (48.5) 50  (59.5)  II 15  (27.8) 74  (28.5) 20  (23.8)  III 3  (5.6) 18  (6.9) 5  (6.0)  Unknown 1 (1.9) 7  (2.7) 0  (0) 0.42 Initial chemotherapy  Y  12  (22.2) 74  (28.5) 20  (23.8)  N  39  (72.2) 163  (62.7) 60  (71.4)  Unknown 3  (5.6) 23  (8.9) 4  (4.8) 0.43 Initial hormonal therapy  Y 39  (72.2) 167  (64.2) 61  (72.6)  N 12  (22.2) 69  (26.5) 19  (22.6)  Unknown 3  (5.6) 24  (9.2) 4  (4.8) 0.47  Health service region VCHA 9  (16.7) 54  (20.8) 17  (20.2)  FHA 14  (25.9) 94  (36.2) 34  (40.5)  VIHA 12  (22.2) 63  (24.2) 14  (16.7)  IHA 16  (29.6) 43  (16.5) 16  (19.1)  NHA 3  (5.6) 6  (2.3) 3  (3.6) 0.28 VCHA Vancouver Coast Health Authority, FHA Fraser Health Authority, VIHA Vancouver Island Health Authority, IHA Interior Health Authority, NHA Northern Health Authority  a  Ineligible for this study, were unable to contact by postal mail or phone calls, or did not have identifiable actively practicing family doctors  b  Calculated using the chi-square test    72   73 Table 3-3   Response patterns in questionnaires Questionnaire Items  Completed both  At baseline (pre-study questionnaire) At six-month  (post-study questionnaire)     (N) % (N / total N) 0 missing 1 missing ≥ 2 missing 0 missing 1 missing ≥ 2 missing No response GLTEQ a 4 91 (49/54) 52 1 1 51 0 1 2 VIDSUN b 5 87 (47/54) 52 1 2 48 2 2 2 CAT c 27 78 (42/54) 48 3 3 44 2 6 2 GLTEQ Godin Leisure-Time Exercise Questionnaire, VIDSUN Vitamin D & Sun Questionnaire, CAT Calcium Assessment Tool  a  GLTEQ evaluated weekly exercise amount b VIDSUN evaluated risk of vitamin D deficiency c  CAT measured daily calcium intake  3.3.2 Study participants  Most of the 54 participants self-reported that they were Caucasian (96%), completed secondary school or higher (87%), received hormonal therapy (69%), and had ≤45 minutes of mild exercise (60%) (Table 3-4). The average daily calcium intake was 915 mg per day. The median FRAX risk without BMD measures was 13.5% (range 8.5 to 45%). No major differences were observed across the three treatment groups. Postal mail (69%) was the most popular choice for delivering educational material, followed by email (31%) regardless of the participants’ group assignment.    74 Table 3-4   Characteristics of study participants   Control group (Total N=18) Educational material delivered by postal mail (Total N=19) Educational material delivered by patient choice a (Total N=17  N (%) N (%) N (%) Demographic Factors  Age  (years, mean ± SD) 70.5 ± 3.9 69.8 ± 3.5 71.1 ± 4.0 Ethnicity     Caucasian  18  (100) 18  (95) 16  (94) Non-Caucasian 0  (0) 1  (5) 1  (6) Marital    Married / common law / living with a partner 11  (61) 12  (63) 15  (88) Single, widowed, divorced, separated 7  (39) 7  (37) 2  (12) Education    Did not complete secondary (high) school  3  (17) 1  (5) 3  (18) Completed secondary (high) school  15  (83) 18  (95) 14  (82) Choice for educational material delivery asked before the randomization  Mail  11  (61) 13  (68) 13  (76) Email  7  (39) 6  (32) 4  (24) Cell phone texting 0 (0)  0 (0)  0 (0) Osteoporosis: risk Factors & previous BMD tests FRAX risk b  Median (range)  13.5 (4.8, 45)  14 (5.2, 32)  11.5 (7, 22)  BMI (kg/m2, mean ± SD) 28.8 ± 6.3 26.7 ± 6.2 27.0 ± 10.7 Smoking     No  17  (94) 15  (79) 15  (88) Yes 1  (6) 4  (21) 2  (12) Drinking     No 8  (44) 11  (58) 9  (53) Yes  10  (56) 8  (42) 8  (47) Weekly exercise (leisure score index, mean ± SD) c 24.8 ± 17.0 27.9 ± 23.3 24.0 ± 27.5 Vitamin D deficiency    Low risk 10  (56) 13  (68) 11  (65) High risk   7  (39) 6  (32) 5 (29) Unknown  1  (6) 0 (0) 1  (6) Calcium intake (mg per day, mean ± SD) d 915 ± 370 911 ± 510 920 ± 511     75 Continued      Control group  (Total N=18)  Educational material delivered by postal mail  (Total N=19)  Educational material delivered by patient choice a (Total N=17)  N (%) N (%) N (%) Long term steroid usage     No 17  (94) 15  (79) 15  (88) Yes 0 (0) 4  (21) 1  (6) Unknown 1  (6) 0 (0) 1  (6) Previous fracture     No  10  (56) 12  (63) 13  (77) Yes 7  (39)  7  (37) 4  (24)  Unknown  1  (6) 0 (0) 0 (0) Parent’s fracture    No  11  (61) 13  (68) 13  (76) Yes 6  (33) 5  (26) 3  (18) Unknown 1  ( 6) 1 (5) 1  (6) Previous BMD test    No  7  (39) 3  (16) 4  (24) Yes 6  (28) 14  (74) 8  (47) Unknown 5  (33) 2  (11) 5  (29) Breast Cancer Treatment  Hormonal therapy     None 7   (39) 4  (21) 6  (35) Tamoxifen only  4  (22) 7  (37) 2  (12) AIs only  3  (17) 5  (26) 5  (29) Tamoxifen with AIs 4  (22) 3  (16) 4  (24) Ovarian suppression or ovary removal     No  17  (94) 19  (100) 17  (100) Yes 1  (6) 0  (0) 0  (0) BMI body mass index, SD Standard Deviation, BMD bone mineral density, AIs Aromatase inhibitors a  Postal mail, email or smartphone text messaging  b Developed by World Health Organization   c  Two participants who did not complete the exercise section of the pre-study questionnaires were excluded from the analysis  d Six participants who did not complete the calcium intake section of the pre-study questionnaires were excluded from the analysis 76 3.3.3 Primary outcome – bone mineral density testing rates  Although no formal statistical testing was conducted, there was a suggestion of higher BMD testing rates in the groups receiving educational material by mail (26%, 95% CI=10 to 49) and patient choice (18%, 95% CI=5 to 41), compared with the control group (6%, 95% CI=0.3 to 25) (Table 3-5). The BMD testing rate was 17% (95% CI=6 to 33) higher in the groups where educational material was delivered by either postal mail or patient choice compared with the control group. The BMD testing rate was 8.7% (95% CI= -33.9 to 18.9) lower for the patient choice compared with the postal mail group.  3.3.4 Secondary outcomes  Four of the nine participants (44%) who had BMD test were newly diagnosed with osteoporosis. Of the four, two (50%) initiated the osteoporosis medication risedronate which was consistently reported by both the participants’ self-reports and medication dispensing records (Table 3-5). Among all 54 participants during the six-month follow-up period, the leisure exercise index increased by 2.8 (95% CI= -13.8 to 19.5) for the educational material delivered by postal mail group, by 4.7 (-2.5 to 11.9) for the educational material delivered by patient choice group, but decreased by 0.9 (-8.3 to 6.5) for the control group. The daily calcium intake increased by 139 mg (-170 to 449) for the postal mail group and by 45 mg (-133 to 224) for the patient choice group, but decreased by 3 mg (-282 to 277) for the control group. The risk for vitamin D deficiency remained unchanged for 39 (72%) of the 54 participants.    77 Table 3-5 Outcome measures after six month follow-up period   Control group Total N=18 Educational material delivered by postal mail a Total N=19 Educational material delivered by patient choice Total N=17  N (%) N (%) N (%) Primary outcome Having BMD testing with DXA, N (%) 1  (6) 5  (26) 3 (18)  Reported by study participants  1  (6) 4  (21) 2 (12) BMD reports retrieved from family doctor or imaging facility 1  (6) 4  (21) 3b  (18) Unknown  0  (0)  1  (3) 0  (0) Secondary Outcomes  Newly diagnosed osteoporosis, N (%) 1  (6) 3  (16) 0  (0) Initiating osteoporosis medications, N (%) Reported by study participants  1  (6) 1  (5) 0  (0) Medication dispensing records  1  (6) 1  (5) 0  (0) Changes in weekly exercise c  - Leisure score index (95% CI) -0.9 (-8.3, 6.5) 2.8 (-13.8, 19.5) 4.7 (-2.5, 11.9) Risk of vitamin D deficiency    Remains low risk  8  (44) 12  (63) 7 (41) Remains high risk  4  (22) 5  (26) 3 (18) Low -> High risk 0  (0) 1  (5) 2 (12) High -> Low risk   3  (17) 1  (5) 1  (6) Incomplete /unknown  3  (17) 0  (0) 4  (24) Changes in calcium intake d  - mg per day (95% CI) -3 (-282, 277) 139 (-170, 449) 45 (-133, 224) BMD bone mineral density, DXA Dual-energy X-Ray absorptiometry, CI Confidence interval a Postal mail, email or smartphone text messaging b One patient reported no DXA test on questionnaire while her BMD testing report was retrieved from her family doctor’s office  c Five participants who did not complete the Godin Leisure-Time Exercise Questionnaires at either or both baseline and six months were excluded from the analysis  d Twelve participants who did not complete the Calcium Assessment Tool at either or both baseline and six months were excluded from the analysis     78 3.4 Discussion  The feasibility of this RCT protocol was evaluated on participants from throughout the province of BC. This study is the first study to pilot-test the effects of educational material on women diagnosed with breast cancer, a population at higher risk for osteoporotic fractures. The educational material intervention used in this study had a promising positive effect on BMD testing rates (17% increase), weekly exercise, and daily calcium intake in women aged 65 and over, and diagnosed with breast cancer for three or more years in BC. The study protocol is feasible for a future large-scale study. Per the online sample size calculator (www.stat.ubc.ca), a minimum of 56 participants per group is required to achieve a statistical power of 0.95 with α=0.05 (one sided test) to detect a 17% increase in BMD testing rates using educational material intervention with a two-parallel-group design in a future large-scale study.  The study participants had a slightly higher average osteoporosis risk (15%, without BMD measurement) compared to the Canadian population at age 70 (around 13%, with BMD measurement) while the negative effects of cancer treatments on bone health are not considered using FRAX [215]. Obesity (body mass index ≥30) and infrequent exercise were more prevalent in our study than in BC residents in the same age group [216]. Being obese is associated with higher breast cancer risk [217], and might also be linked with inadequate exercise and reduced metabolism associated with cancer treatment [218]. The potential barriers to exercise include fatigue associated with cancer treatment [219], lack of priority, self-discipline, and procrastination [220]. Inadequate average daily calcium intake was observed for our study participants per the 2010 Canadian Osteoporosis guidelines [97] while the intake was similar compared to BC residents in the same age group. One-third of our study participants were at high 79 risk of vitamin D deficiency while the assessment tool (VIDSUN questionnaire) has high sensitivity (%) and low specificity (%) [212]. Interventions, such as educational material in this study, should be considered to improve calcium intake and exercise in women diagnosed with breast cancer. Email was the preferred choice for delivering educational material by one-third of participants in this study, which suggests a potential willingness to adopt technology for health-related issues in this specific group.     3.4.1 Educational material The educational material created for this study included basic information only. There was a focus to make information easy to read and understandable for women from all educational backgrounds. Readers were encouraged to discuss details with relevant health-affiliated professionals, such as family doctors for lifestyle modification, dieticians for calcium or vitamin D rich food, physiotherapists for exercise, and pharmacists for calcium or vitamin D supplements. HealthLink BC URL (www.healthlinkbc.ca/) was provided to readers looking for more details on healthy lifestyles. A toll-free 8-1-1 number was provided for a free dietician consultation in BC.   Exercise has shown positive effects on BMD [139]. Moderate-to-vigorous exercise reduces hip fractures by 38-45% [221]. Exercise recommendations vary significantly among different guidelines. Our educational material suggests 10-15 minutes of exercise once or twice a day at least 3-5 times per week, which is higher than the amount suggested by the Canadian Task Force (three times per week, for at least 20-30 minutes each time; for all adults) [175], but lower than the amount suggested in the 2010 Canadian Osteoporosis Guideline (4-7 days per week for 20 to 80 60 minutes; for all adults) [97] and the 2011 World Health Organization Guideline (≥150 minutes of moderate-intensity aerobic exercise per week or  ≥75 minutes of vigorous-intensity aerobic exercise per week; for adults aged ≥65 years) [222].   3.4.2 Bone mineral density testing rate The absolute increase in BMD testing rate during our six-month follow-up period starting from the delivery of educational material, was lower than the increase seen in high-risk patients with recent fractures (22-51%) [128], but higher than the increase seen in patients aged 65 and over (18%) in primary care settings [129]. This could be explained by differences in sample sizes,  interventions (patient reminder, physician reminder, or educational material), medical care settings (emergency department, hospital or primary care clinic), follow-ups ranging from 4 to 16 months, and perceived osteoporotic fracture risk (risk is underestimated in older women) [118, 120].   In this study, the proportion of participants who had BMD tests and who did not during the six-month follow-up period in age, FRAX score, calcium intake, and leisure score index were similar between three groups. Only two of the 14 participants who did not have BMD tests prior to this study, had BMD tests during their six-month follow-up period. This could be because women without fractures tend to underestimate their osteoporosis risk, or women may skip BMD testing due to a potential perceived vulnerability associated with an osteoporosis diagnosis.  81 3.4.3 Secondary outcome measures The common questions returning unanswered were associated with supplemental intake of vitamin D and calcium in the CAT and VIDSUN questionnaires. We identified common barriers including poor vision (preventing these participants from reading the small fonts on supplement labels), confusion due to different measurement units of vitamin D (e.g. international units and micrograms), and uncertainty in measurement units of calcium rich food (e.g. cups and cubes). These questions may require modification or clarification in a future large-scale study.   3.4.4 Limitation The percentage of non-Caucasian participants was low in our study compared to the population proportion of 27% in BC [223]. This could primarily result from language barrier as our educational material was only available in English. The other potential barriers include logistical challenges, cultural barriers, and mistrust of research [224].   The overall recruitment rate in this study was only 13% (54/398). This is primarily because we excluded 190 survivors who were ineligible or their family doctors were not identifiable by the study team. Despite this, the primary outcome measure was obtained for at least 90% of all 54 participants. We did not experience any major events associated with the study protocol. Our approach is a very low cost way to reach a large number of women participants.   3.5 Conclusion The protocol is feasible for a large-scale study with minor questionnaire modification. There was broad acceptance of the educational material intervention by the study participants. The data suggest that the educational material will increase BMD testing rates.  Given the importance of 82 diagnosing and treating osteoporosis and the low current rates of BMD testing in Canada, this area is a priority for future interventions. 83 Chapter 4: Aromatase Inhibitors are Associated with a Higher Fracture Risk than Tamoxifen: A Systematic Review and Meta-analysis (study 3) 4.1 Introduction Adjuvant systemic treatments, such as chemotherapy and hormonal treatment, have been used widely to treat breast cancer [225]. Hormonal treatment is recommended in women with hormone receptor-positive breast cancer, accounting for at least two-thirds of all breast cancer cases [149, 150]. The two most common hormonal treatments are tamoxifen and aromatase inhibitors (AIs).    Tamoxifen, a selective estrogen receptor modulator (SERM), was introduced in the 1970s. Tamoxifen is currently recommended to treat early and advanced stage breast cancer in pre-menopausal women, and post-menopausal women at lower risk of cancer relapse [151]. Tamoxifen is also an optional treatment in women with stage zero (in situ) breast cancer [152]. Tamoxifen reduces the available estrogen to cancer cells by competitively inhibiting the binding of estrogen to estrogen receptors on breast tissues. The effect of tamoxifen on bone tissues is inconsistent across studies and seems to differ by menopausal status. Tamoxifen caused a BMD decrease in healthy pre-menopausal women but a BMD increase in healthy post-menopausal women [155]. In women diagnosed with breast cancer, tamoxifen preserves bone mass in pre-menopausal women, and either slightly increases or decreases BMD in post-menopausal women [156-161]. Tamoxifen may have a beneficial effect on bone health in women diagnosed with breast cancer. However, tamoxifen has not 84 been approved for the treatment or prevention of osteoporosis in any populations by the US Food & Drug Administration.  AIs were introduced in the early 2000s. AIs are currently recommended to treat early and advanced stage breast cancer in post-menopausal women at higher risk of cancer relapse. AIs reduce the circulating estrogen levels by inhibiting the aromatase enzyme from converting androgen to estrogen in non-ovarian tissues. AIs are suggested for women at higher cancer relapse risk due to its potential negative effect on bone health. AIs significantly increase bone loss [159, 169] and are associated with higher fracture risks in several major trials [170, 171]. However, AI-associated fracture risk has not been reviewed systematically.  The initial goal of this study was to determine whether there are BMD changes and additional fracture risks associated with adjuvant systemic breast cancer treatments, compared with loco-regional treatments (i.e. surgery and radiation therapy) or no breast cancer treatment in women aged 65 and under. Fractures however have a higher clinical impact on healthcare systems than BMD changes. Tamoxifen and AIs are used to treat breast cancer more often than other adjuvant systemic treatments. Hence, we focused our research questions on the differential fracture risks associated with tamoxifen and AIs in younger women aged 65 years and under, and diagnosed with non-metastatic breast cancer.    85 4.2 Method This was a systematic review with meta-analysis study using aggregate data from RCTs and cohort studies on fracture risks associated with tamoxifen and AIs in younger women aged 65 years and under, and diagnosed with non-metastatic breast cancer. We registered the review protocol at PROSPERO (registration number CRD42015015604; www.crd.york.ac.uk/PROSPERO/). We reported study results using criteria from the Preferred Reporting Items for Systematic Review and Meta-analysis Protocols (PRISMA) [226]. Article search was conducted by the first author. Study selection (NR/OT for title/abstract screening; WH/OT for full-text article review), study quality evaluation (WH/OT), and data extraction (WH/OT) were performed independently by two reviewers using Excel spreadsheets. Disagreements between reviewers were resolved by discussion. Persistent disagreements between reviewers were arbitrated by another designated team member (MD).  4.2.1 Search strategy We searched PubMed, MEDLINE, CINHAL, EMBASE, and Cancerlit databases for article published from January 1st, 1970 to May 1st, 2015. We included search terms “breast” and “wom*n OR female” and “tumor OR cancer OR neoplasm OR malignanc?” and “fracture OR BMD OR densit? OR densitometr? OR absorptiometry?”. Studies were then limited to human studies and English language articles. Review articles were then excluded. The reference lists of the included articles were hand-searched. Approximately 20% of included and excluded articles at each step of the article search were randomly reviewed to ensure proper article search strategies.  86 4.2.2 Study selection Articles were initially screened by title and abstract, followed by full article reviews (Figure 4-1). Articles fulfilling the inclusion criteria: (1) RCTs or cohort studies [227], (2) women diagnosed with non-metastatic breast cancer, (3) at least one participant aged 65 years and under at baseline, (4) breast cancer treatments of tamoxifen, AIs or both, and (5) fracture outcomes, were selected. We defined the outcomes in this study as count of fracture events or participants with fractures. Articles reporting pathological fractures or any specific fracture type (e.g. spine fracture only) were excluded.   4.2.3 Study quality assessment We evaluated the methodological quality of the selected articles using two separate assessment tools suggested by the Cochrane Collaboration Review Group. RCTs were evaluated using the Cochrane risk of bias assessment tool. Each RCT was assessed and rated as “low risk of bias”, “high risk of bias” or “unknown risk of bias” in the seven domains of potential bias: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other biases (e.g. funding source, conflict of interest, etc.) [228, 229]. Cohort studies were evaluated using the Newcastle-Ottawa Scale with a range of zero to nine stars. Each cohort study was evaluated in three categories – the selection category with four items, the comparability category with only a single item, and the outcome category with three items. Each cohort study was awarded a maximum of one star per item within the selection and outcome categories, and a maximum of two stars for the single item within the comparability category [230, 231].    87 4.2.4 Data extraction  Articles reporting data at same follow-up times from the same independent study were collated (ID 5, 16, 18, 21, 30). We extracted data from each included study on method, participant, treatment, fracture outcome, and factors controlled for multivariable regression models as follows:  • Method: study design, study period, follow-up duration  • Participant: total number, age, breast cancer stage, proportion of post-menopausal women, non-interventional breast cancer treatments  • Treatment: dosage, treatment duration, prior tamoxifen treatment  • Fracture outcome: definition of fractures, count of fracture events (allowing more than one fracture event per participant), count of participants who developed fractures, and relative measures including odds ratios (ORs), risk ratios (RRs), incidence rate ratios (IRRs), and/or hazard ratios (HR) using Cox regression models • Factors controlled for multivariable regression models  There were two articles (ID 12, 34) each reporting combined data from two independent studies. Data from the Austrian Breast and Colorectal Cancer Study Group trial 8 (ABCSG-8), and Arimidex-Nolvadex-95 (ARNO-95) trial were combined in article 12. Data from the Tamoxifen and Exemestane Trial (TEXT), and Suppression Ovarian Functions (SOFT) trial were combined in article 34 [232, 233]. Extracted data from each independent study, ABCSG-8, ARNO-95, TEXT, and SOFT trial were inadequate for meta-analysis. The authors of both articles were contacted by email but we were unable to obtain additional information on any of these four studies.  88 4.2.5 Data synthesis Meta-analyses were undertaken to estimate the differential fracture risks of tamoxifen and AIs, and risks between tamoxifen and AI. Each fracture risk was stratified by three to five factors of menopausal status (pre-menopausal only, mixture of both pre- and post-menopausal, and post-menopausal only), prior tamoxifen treatment (yes vs. no), study design (RCTs vs. cohort study), AI treatment duration (≤48 months vs. 60 months) and AI drug (steroidal vs. non-steroidal vs. any) using subgroup analysis. Menopausal status was determined using age in the two cohort studies with missing menopausal status information (ID 4, 35).   The time effect on differential fracture risk between tamoxifen and AI was evaluated by ranges of follow-up durations (12-36, >36-60, >60-84, >84 months) and treatment period (on- and post-Tam/AI treatment). Meta-analyses were conducted independently for each range of follow-up duration and treatment period. The Tam/AI-treatment period was defined as the time period when women were receiving tamoxifen or AIs during the study period.  For each independent study with serial follow-up data, the article with the longest follow-up duration was included for each individual meta-analysis to avoid double counting of study participants. For studies with multiple treatment arms, the arms were either grouped as a single pair-wise comparison (ID 13, 14) or a three group comparison with each other (ID 35, 36. 37) of tamoxifen, AIs, and control group (no tamoxifen alone, no AIs alone, and no combination of tamoxifen and AIs). Articles with double-zero events (zero-cell counts in both intervention arms) were excluded from meta-analysis [234]. 89 4.2.6 Statistical analysis  Meta-analyses were restricted to studies reporting counts of participants with fractures and not fracture events. For RCTs included in meta-analysis, Relative Risks (RRs) with 95% confidence intervals (CIs) were calculated. For cohort studies included for meta-analysis, published adjusted Hazard Ratios (aHRs) with 95% confidence intervals (CIs) were used first. RRs were calculated for the cohort studies without available aHRs. Adjusted HRs were treated as adjusted RRs due to the low incidence of fracture outcomes. Overall differential fracture risk was pooled as weighted RRs using a generic inverse variance method with random effects models. The weight of each study was based on the inverse of that study’s variance. Statistical significance of the pooled RRs was evaluated using Chi-Square tests. Statistical heterogeneity was evaluated using Cochrane’s Q statistic and quantified as I2 measures. Sensitivity tests were conducted when combining RRs and aHRs. All statistical tests were performed using RevMan 5.2 analysis software (The Cochrane Collaboration Copenghagen, Denmark) [235].  4.3 Results  There were 4,004 articles identified, of which 2,078 were duplicate articles (Figure 1). This left 1,926 unique articles for title/abstract screening. Of them, 1,649 were excluded leaving 277 articles for full article review. A total of 43 articles covering 21 independent studies fulfilled our selection criteria and proceeded to methodological quality assessment.    90 Identification  Initial article search  (n=4,004) OVID (1117) PubMed (1180)  EMBASE (1422)  CINHAL (281)  Hand search (4)    Excluded duplicates (n=2078)         Study selection  Title / abstract screening  (n= 1926)    Excluded (n=1649)  - Publication type: review, comment, highlight and etc. (290)  - Study design: cross-sectional or non-randomized study (23) - Population, intervention or outcome (1334)  - Unable to find full abstract (2)        Full article review (n=277)   Excluded (n=234) -Population     - Intervention   - Outcome       - Unable to find full abstract   Non-breast cancer (12)  Breast cancer with osteoporosis or fractures (3) Recurrent, relapsing, or metastatic breast cancer (11)  Age >65 (5)   Lack of appropriate comparison groups (28)  BMD only, no fracture information (102)  Rib fractures associated with radiotherapy (37) Pathological fractures (2)  Incomplete data (4)  No fracture information (22)  Spin fracture only or hip fracture only (4)   (4)        Eligibility  Full article included (n= 43)  RCT=16 (7 studies had serial follow-up articles)   Cohort = 5 (3 studies with multiple treatment arms for more one comparisons)         Data  extraction  &  analysis  Tam vs. Control a 3 RCTs 3 Cohort     AIs vs. Control b 3 RCTs  4 Cohort  Aromatase inhibitors (AIs) vs. Tamoxifen (Tam) 10 RCTs 4 Cohort         Subgroup analysis  - Menopausal status - Prior tamoxifen treatment   - Study design  Subgroup analysis - Menopausal status - Prior tamoxifen treatment  - Study design - AI drug - AI treatment duration   Subgroup analysis - Menopausal status - Prior tamoxifen treatment  - Study design - AI drug - AI treatment duration  Meta-analyses - Follow-up duration 12-36 months  - Follow-up duration >36-60 - Follow-up duration >60-84 - Follow-up duration >84 - Tam/AI treatment - Post-Tam/AI treatment Figure 4-1 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analysis) flow diagram for systematic review of the fracture risks associated with breast cancer treatments AIs aromatase inhibitors, RCT randomized controlled trial, Tam tamoxifen  a No tamoxifen b No AIs   91 4.3.1 Characteristics of included studies Sixteen RCTs, four retrospective cohort studies, and one prospective cohort study were included (Table 4-1). All RCTs were designed to evaluate primary outcome of efficacy and secondary outcome of safety including fractures using intent-to-treat analysis with the exception of one study (ID 7). All cohort studies were designed to evaluate fracture outcomes. Seven of the 16 RCTs reported serial follow-up data. Eight of the 16 RCTs involved post-menopausal women only.   Mean or median age ranged from 43 to 67 years. Treatment dose was unknown in four cohort studies (ID 4, 11, 35, 36). Doses of tamoxifen were 20 mg per day in almost all studies, but one (ID1) of 30 mg per day and two of 20-30mg per day (ID 12, 15). Doses of AIs were consistent across all studies as follows: anastrozole (1 mg per day), letrozole (2.5 mg per day), and exemestane (25 mg per day). Treatment duration ranged from 12 to 72 months while follow-up duration ranged from 12 to 128 months. About 17-25% crossover was reported in a few studies (ID 25, 26). Fracture outcomes were measured as any self-reported fracture (15 studies), self-reported osteoporotic/minimal-trauma fracture (ID 1, 36), self-reported hospitalized fracture (ID 32), any fracture event in medical records (ID 11), or any fracture using data linkage (ID 4, 35). 92 Table 4-1   Summary of studies      Study Information Study  participants (safety population)  Treatment Published fracture outcomes - fracture Meta-analysis  Factors        ID Study name Author, year (ref) Design Country Data source Median follow-up duration                                Post- Age                 menopausal (Years) a                 (%)      Prior tamoxifen b (duration) No.   Arms       Duration  Type Participants      Fracture   with fractures      events Risk measure (95% CI) Risk Measure used in meta-analysis (95% CI)   Adjusted  No.             per 1000 PY Tamoxifen vs. Control / placebo (reference) 1 Kristense, 1994 [162] RCT Denmark Self-report 44 M 57 (NR, 65) 100 N 20  23 Tam  Control  24 M Osteo-porotic 0 / -- 0 / --  -- --  -- -- 2 Love, 1994 [163] RCT USA Self-report Mean  60.5 M 58 ± 4  100 N 70  70 Tam   Placebo 24 M Any 6 / 7 8 / 10  --  -- Calculated   RR  0.75 (0.27, 2.05)  -- 3 Sacco, 2003 [236] RCT Italy Self-report 52 M 61 ± 6  95 Y (24 M) 943  958 Tam Control 36 M Any 8 / -- 10 / -- -- -- Calculated   RR 0.81 (0.32, 2.05)  - Aromatase inhibitors (AIs) vs. Control / placebo (reference)  4 Mincey,  2006 [237] Cohort  US Data-linkage Range  1998- 2005 66 ± 11 64 ± 13 <100 c N 1354  11,014 AIs Control -- Any 183 / -- 1132 / -- 86 / -- 63.6 / -- aHR 1.21 (1.03, 1.43) IRR 1.35 (1.16, 1.58) Published   aHR 1.21 (1.03, 1.43)  Age, 1, 2, 3, 4 5 MA 17  Goss, 2003 [238]; DeGrendele, 2003 [239] RCT Multiple , 9 Self-report 2.4 Y 62 (NR)   100 Y (60 M) 2154 2145 AIs (Let)  Placebo  60 M Any 77 / -- 63 / -- -- -- -- -- 6 MA 17  Goss, 2005 [240]    30 M    2572 2577 AIs (Let) Placebo   137 / -- 119 / -- -- -- Calculated   RR 1.15  (0.94, 1.41)  -- 7 MA 17  Goss, 2008 [241]     1.1 Y d (after unblinding)    1579 e 804 AIs (Let)  Placebo    82 / -- 25 / -- -- -- -- -- 8 Norwegian Lonning, 2005 [242]  RCT Norway Self-report 24 M 59 (46-73)   100 N 73  74 AIs (Exe) Placebo 24 M Any 4 / -- 5 / -- -- --  -- -- 9 Norwegian Geisler, 2006 [243]     36 M    73  74 AIs (Exe) Placebo   4 / -- 5 / - --  -- Calculated   RR  0.81 (0.23, 2.90)  -- 10 NSABP (B-33) Mamounas, 2008 [244]  RCT USA / Canada Self-report Till  April 2004 60 (NR)    100 Y (57-66 M) 783  779 AIs (Exe)  Placebo  60 M Any 28 / -- 20 / -- --  -- Calculated  RR 1.39 (0.79, 2.45) -- Aromatase inhibitors (AIs)  vs. Tamoxifen (reference) 11 Koopal,  2015 [245] Cohort  Netherland Charts + X-ray Post-Tam/AI (3.1 Y) 52 ± 7 (pre-m) 71 ± 10 (post-m) 0  N  39 92  AIs  Tam  5.7 - 6 Y Any 4 / -- 24 / -- --   -- Calculated   RR 0.39 (0.15, 1.06) -- 12 ABCSG – 8 /  ARNO 95 Jakesz, 2005 [232]  RCT Germany / Austria,  Self-report 28 M 62 (41-80)    100 Y (24 M)   1602 1597 AIs (Ana) Tam   36 M Any 34 / -- 16 / -- -- OR 2.14 (1.14, 4.17)   -- 13 ABCSG-12 Grant, 2009 [246]  RCT Austria Self-report 47.8 M 45 (26-57)  0  N  903  900 AIs (Ana) Tam 36 M  Any 12 / -- 12 / -- -- -- Calculated   RR 1.00 (0.45, 2.21)  -- 14 ABCSG-12 Grant, 2011 [247]    62 M    903 900 AIs (Ana) Tam   13 / -- 12 / -- -- -- Calculated   RR 1.08 (0.50-2.35) --       93 Continued      Study Information Study  participants (safety population)  Treatment Published fracture outcomes - fracture Meta-analysis  Factors        ID Study name Author, year (ref) Design Country Data source Median follow-up duration                                Post- Age                 menopausal (Years) a                 (%)      Prior tamoxifen b (duration) No.   Arms       Duration  Type Participants      Fracture   with fractures      events Risk measure (95% CI) Risk Measure used in meta-analysis (95% CI)   Adjusted  No.             per 1000 PY Aromatase inhibitors (AIs)  vs. Tamoxifen (reference) 15 ARNO 95 Kaufmann, 2007 [248] RCT Germany  Self-report   30.1 M 61 (46-74) 100 Y (24 M) 445 452 AIs (Ana)  Tam 36 M Any 10 / -- 10 / -- -- -- Calculated   RR 1.02 (0.43, 2.42)  -- 16 ATAC Buzdar, 2002 [249];  Fisher, 2002 [250]; Baum, 2002[251] RCT  Multiple, 21 Self-report 33.3 M 64 ± 9 100 N  3092  3094 AIs (Ana)  Tam  60 M  Any 183 / -- 115 / -- -- -- Calculated   RR 1.59 (1.27,2.00)   -- 17 ATAC Baum,  2003 [252]    42 M    3092 3093 AIs (Ana)  Tam   219 / -- 137 / -- -- -- Calculated   RR 1.60 (1.30, 1.97) -- 18 ATAC Howell, 2005 [253]; Cuzick, 2007 [254]    68  M    3092  3094 AIs (Ana)  Tam   340 / -- 237 / -- 22.6 / -- 15.6 / --   OR  1.49 (1.25, 1.77) HR 1.44 (1.21, 1.68) Calculated   RR 1.44 (1.23, 1.68) -- 19 ATAC Arimidex, 2008 [255]    100 M    3092  3094 AIs (Ana)  Tam   -- -- --  -- --      On Tam/AI    3092 3094 AIs (Ana)  Tam   -- / 375 -- / 234 -- / 29.3 --  / 19 IRR 1.55 (1.31-1.83)  -- --      Post Tam/AI    2496 2419 AIs (Ana)  Tam   -- / 146 -- / 143 -- / 15.6 -- / 15.1 IRR 1.03 (0.81, 1.31) -- -- 20 ATAC Cuzick, 2010 [170]     120 M    3092 3094 AIs (Ana)  Tam   -- -- --  -- --      On Tam/AI     3092 3094    451 / -- 351 / -- -- -- Calculated   RR 1.29 (1.13, 1.46)  --      Post-Tam/AI    2223 2246    110 / -- 112 / -- -- -- Calculated   RR 0.99 (0.77, 1.28)  -- 21 BIG 1-98 Thurlimann, 2005 [256]; Monnier, 2005 [257] RCT Multiple, 27  Self-report 25.8 M 61 (38-90) 100 N 3975 3988 AIs (Let) Tam 60 M Any 225 / -- 159 / -- 22 / -- 15 / -- OR  1.44 Calculated   RR 1.42 (1.16, 1.73)   -- 22 BIG 1-98 Crivellari, 2008 [258]    40.4 M    2448 2447  AIs (Let) Tam   196 / -- 132 / -- -- -- -- -- 23 BIG 1-98 Coates, 2007 [259]    51 M    2448 2447 AIs (Let) Tam   211 / -- 141 / -- -- -- Calculated    RR 1.50 (1.22, 1.84)  -- 24 BIG 1-98 Rabaglio, 2009, [171]     On Tam/AI 60.3 M  f    2448  2447   AIs (Let) Tam   228 / -- 160 / -- 25.2 / 27.1 18.1 / 18.7 HR 1.38 (1.13, 1.69)  aHR  1.40 (1.14, 1.71)   Calculated   RR 1.42 (1.17, 1.73) Age, 5, 6, 7, 8, 9, 10                    94 Continued      Study Information Study  participants (safety population)  Treatment Published fracture outcomes - fracture Meta-analysis  Factors        ID Study name Author, year (ref) Design Country Data source Median follow-up duration                                Post- Age                 menopausal (Years) a                 (%)      Prior tamoxifen b (duration) No.   Arms       Duration  Type Participants      Fracture   with fractures      events Risk measure (95% CI) Risk Measure used in meta-analysis (95% CI)   Adjusted  No.             per 1000 PY Aromatase inhibitors (AIs)  vs. Tamoxifen (reference) 25 BIG 1-98 Mouridsen, 2009 [260]     71 M    1540  1534  AIs (Let) Tam   -- --  -- -- --      on Tam/AI (Y1-2)    1540  1534  AIs (Let) Tam   65 / -- 50 / -- -- -- -- --      on Tam/AI (Y 3-5) g    1540 1534  AIs (Let) Tam   90 / -- 67 / -- -- -- -- --      on Tam/AI  (Y 1-5)    1540 1534  AIs (Let) Tam   150 / -- 112 / -- -- -- -- -- 26 BIG 1-98 Colleoni, 2011 [261]    74 M     2448 2447   AIs (Let) Tam   244 / -- 165 / -- --  -- Calculated   RR 1.48 (1.22, 1.79) -- 27 HOBOE Nuzzo, 2012 [262] RCT Italy Self-report 12 M 50 (29-80)   46  N  148 152 AIs (Let)  Tam 60 M Any 0 / 0 0 / 0 -- --  -- -- 28 IES Coombes, 2004 [263]   RCT Multiple, 37 Self-report 30.6 M 64 ± 8 100 Y ( 2.4 Y) 2305  2329 AIs (Exe)  Tam 2-3 Y Any  72 / -- 53 / -- --  -- Calculated   RR 1.37 (0.97, 1.95) -- 29 IES Coleman, 2007 [264]     58 M    2320 2338   AIs (Exe)  Tam   162 / 188 115 / 143 17.6 / 20.1 13.2 / 16.0 OR 1.45 (1.13-1.87)  Calculated   RR 1.42 (1.13, 1.79)   -- 30 IES Bliss, 2012 [265]; Clomean, 2010 [158]    91 M    2319 2338  AIs (Exe)  Tam    249 / 280 190 / 214 -- OR h 1.36 (1.04, 1.76)  Calculated   RR 1.32 (1.10, 1.58)    --      On Tam/AI     2319 2338 AIs (Exe)  Tam    113 / 117  86 / 83 -- / 21 -- / 12.3 OR h 1.39  (0.94, 2.06)  HR h  1.39 (0.96, 2.01)  Calculated   RR 1.37 (1.04, 1.81)   --      Post Tam/AI     2105 2036 AIs (Exe)  Tam    144 / 163 117 / 128 -- / 20.3 -- / 20.6 OR h 1.20 (0.86, 1.69)  HR h 1.20 (0.89, 1.63) Calculated   RR 1.19 (0.94, 1.51) -- 31 ITA Boccardo. 2005 [266] RCT Italy Self-report 36 M 63 (38-77) 100 Y (28 M)  223 225 AIs (Ana) Tam 2-3 Y  Any 2 / -- 2 / --  -- -- Calculated   RR 1.01 (0.14, 7.10)  -- 32 ITA Boccardo, 2013 [267]    128 M    223 225  AIs (Ana)  Tam  Hospital events,  4 / -- 4 / --   -- -- Calculated   RR 1.01 (0.26, 3.98)  -- 33 N-SAS BC03 Aihara, 2010 [268] RCT Japan Self-report 42 M 60 ± 7 100 Y (1-4 Y) 347 349 AIs (Ana)  Tam 1-4 Y Any 5 / -- 9 / -- --   -- Calculated   RR 0.56 (0.19, 1.65)  --       95 Continued      Study Information Study  participants (safety population)  Treatment Published fracture outcomes - fracture Meta-analysis  Factors        ID Study name Author, year (ref) Design Country Data source Median follow-up duration                                Post- Age                 menopausal (Years) a                 (%)      Prior tamoxifen b (duration) No.   Arms       Duration  Type Participants      Fracture   with fractures      events Risk measure (95% CI) Risk Measure used in meta-analysis (95% CI)   Adjusted  No.             per 1000 PY Aromatase inhibitors (AIs)  vs. Tamoxifen (reference) 34 TEXT / SOFT (IBCSG) Pagani, 2014 [233] RCT Multiple  Self-report 68 M 43 ± NR  0  N  2318 2325 AIs (Exe)  Tam 60 M  Any 158 / -- 120 / -- --   -- -- -- Multiple treatment arms  35 Ligibel, 2012 [269] Cohort  US Data linkage 30 M 67 ± NR <100 c N Total 44,026  Tam Control --  Any -- 26.8 / -- 38.1 / --  aHR 0.93 (0.82-1.06) Published   aHR 0.93 (0.82-1.06) Age, 1, 2, 3, 11, 12, 13, 14, 15          Total 44,026 AIs  Control    33.3 / -- 38.1 / -- aHR 1.13 (1.02-1.25) Published   aHR 1.13 (1.02-1.25)          Total 44,026 AIs  Tam    33.3 / -- 26.8 / --  -- 36 Robinson, 2014 [270] Cohort  Australia Self-report Mean  5.7 Y 57 (27-87) 35  N  393 252 Tam Control --  Minimal trauma  56 / -- 30 / -- --  -- Calculated   RR 1.20 (0.79-1.81) --          306 252 AIs  Control   46 / -- 30 / --  OR 1.31 (0.80, 2.14) Calculated   RR 1.26 (0.82-1.94) --          306  393 AIs  Tam   46 / -- 56 / --   Calculated   RR 1.05 (0.74, 1.51) -- 37 Xu, 2014 [271] Cohort  China Self-report 32.5 M 56 ± 8 61 ± 9 76-88 N  52 89 Tam Control -- Any 1 /  -- 1 / -- -- aHR 2.64 (0.14, 48.73)  Published   aHR 2.64 (0.14, 48.73) 10, 16, 17         61 ± 7 61 ± 9   70  89 AIs  Control   9 / -- 1 / --  aHR 20.08 (1.7, 234.1) Published   aHR 20.08 (1.7, 234.1)       61 ± 7 56 ± 8   70 52 AIs  Tam   9 / -- 1 / --  -- Calculated   RR 6.69 (0.87, 51.14) a  Mean ± SD or median (range)   b  Tamoxifen treatment prior to the study c  Menopausal status was determined based on age range d  Information of fracture outcome was collected for 1.1 years after unblinding on October 2003 e  1579 participants crossed over from placebo group after unblinding f  Fracture data obtained from participants on medications only g  25.2% crossover  h 99% confidence intervals  Abbreviations: aHR adjusted hazard ratio, AIs Aromatase inhibitors, Ana anatrozole, CI confidence interval, Exe Exemestane, HR hazard ratio, IRR incidence ratio, Let letrozole, M month, No number, NR not recorded, OR odds ratio, pre-m pre-menopausal, post-m post-menopausal, ref reference, RCT randomized controlled trail, RR risk ratio, SD Standardized deviation, Tam Tamoxifen, Y year,   Study abbreviations: ABCSG Austrian Breast & Colorectal Cancer Study Group, ARNO Arimidex-Nolvadex, ATAC Arimidex, Tamoxifen, Alone or in Combination, BIG Breast International Group, HOBOE Hormonal Bone Effects, IES Intergroup Exemestane Study, ITA Italian Tamoxifen Anastrozole, NSABP National Surgical Adjuvant Breast and Bowel Project, SOFT Suppression of Ovarian Function Trial, TEXT Tamoxifen and Exemestane  Adjusted factor: 1 Charlson comorbidity index, 2 residential regions, 3 health plan, 4 income, 5 body mass index, 6 smoking, 7 osteoporosis, 8 fracture history, 9 hormonal replacement therapy, 10 bisphosphonates, 11 index year, 12 urban/rural status, 13 drug class, 14 education, 15 % of black, 16 age of diagnosis, 17 age of menopause 96 4.3.2 Study quality assessment  Study quality assessment was summarized in Table 4-2 and Table 4-3. High risk of bias was observed primarily in domains of blinding of participants, blinding of outcome assessors, incomplete data, and other biases (e.g. funding) among RCTs. Unblinding of participants and their outcome assessment was observed in at least half of the RCTs that were either open RCTs or unblinded during their study periods. Financial support from pharmaceutical companies was noted in at least 80% of the RCTs. High percentages of incomplete data were observed in studies with longer follow-up durations. The quality of all cohort studies was consistently high with either seven or nine out of a maximum of nine stars.    97 Table 4-2   Summary of risk of bias assessment for the included randomized controlled trials   A    B     C     D     E     F    G   A    B     C     D    E     F     G   A  Random sequence generation (selection bias)  B  Allocation concealment (selecitn bias)  C  Blinding of participants and personnel (performance bias)  D  Blinding of outcome asessment (detection bias)  E  Incomplete outcome data (attrition bias)  F  Selective reporting (reporting bias)  G  Other bias   Low risk of bias  Unknown risk of bias  High risk of bias    98 Table 4-3   Summary of Newcastle-Ottawa Scale assessment for the included cohort studies Study Selection (maximum 4 stars) Comparability (maximum 2 stars) Outcome (maximum 3 stars) Total score  (out of 9)  1 2 3 4 5 6 7 8  Xu, 2014  --     --   7 Robinson, 2014   --   --   7 Koopal, 2015      --    7 Mincey, 2006         9 Ligibel, 2012         9 MAX Maximum 1. Representativeness of exposed cohort 2. Selection of non-exposed cohort 3. Ascertainment of exposure 4. Outcome not present at start of study 5. Comparability of cohorts on the basis of the design or analysis 6. Assessment of outcome 7. Was follow-up long enough for outcomes to occur 8. Adequacy of follow up of cohort  4.3.3 Tamoxifen Three RCTs and three cohort studies compared fracture outcomes between women treated and not treated with tamoxifen (Table 4-4 and Figure 4-2). One RCT with double-zero events was excluded from this meta-analysis. This analysis included 37,783 participants. Fracture risk did not differ between tamoxifen and no-tamoxifen groups (pooled RR=0.95, 95% CI=0.84 to 1.07). The statistical heterogeneity was low with an I2 measure of 0% (p=0.72). No statistical significance was reported in subgroup analyses by menopausal status (p=0.65), prior tamoxifen treatment (p=0.74) or study design (p=0.58).  4.3.4 Aromatase inhibitors   Three RCTs and four cohort studies compared fracture outcomes between women treated and not treated with AIs. All seven studies were included in this meta-analysis (Table 4-4, Figure 4-3). Data from the longest follow-up durations were selected for the two included studies (ID 6, 9). 99 This analysis included 59,258 participants. A 17% (95% CI =1.07 to 1.28) higher fracture risk was observed in the AI group than the no-AI group. Statistical heterogeneity was low with an I2 measure of 8% (p=0.37). No statistical significance was noted in subgroup analyses by menopausal status (p=0.88), prior tamoxifen treatment (p=0.99), study design (p=0.88), AI treatment duration (p=0.57), or AI drug (p=0.93). Sensitivity analyses excluding the Xu et.al. study (ID 37) resulted in a similar estimate of 16% RR increase with a zero I2 measure across all analyses.  4.3.5 Comparison of aromatase inhibitors and tamoxifen  Ten RCTs and four cohort studies compared fracture outcomes between women treated with AIs and treated with tamoxifen (Table 4-4, Figure 4-4). Four studies (ID 12, 27, 34, 35) were excluded due to either missing data, double-zero events, or reporting combined data from more than one independent study. Data from the longest follow-up duration was selected for the five included studies (ID 14, 18, 26, 30, 32). This analysis included 20,403 participants. A 35% (95% CI=1.21 to 1.51) higher fracture risk was observed in the AI group compared with the tamoxifen group. The statistical heterogeneity was low with an I2 measure of 12% (p=0.43). No statistical significance was observed in subgroup analysis by menopausal status (p=0.75), prior tamoxifen treatment (p=0.5), study design (p=0.68), AI drug (p=0.83), or AI treatment duration (p=0.19) (Table 4-4). Sensitivity analyses excluding the Xu et.al. study (ID 37) resulted in a similar estimate of 36% RR increase with a low I2 measure (range 0-7) across all analyses.   100 Table 4-4   Meta-analysis including subgroup analysis of aromatase inhibitors, tamoxifen, and control groups on fractures Treatment arms Study  (N) Participant (N) Pooled RR (95%CI) p for effect I2 (%) a   p for subgroup differences ID of article included Tam vs. control (no-Tam) b Total effect 5 37,783 0.95 (0.84, 1.07) 0.39 0 0 2, 3, 35, 36, 37 Subgroup analysis         Menopausal status      0.65  Pre-menopausal  0  -- -- --  -- Pre- / post-menopausal  4  0.95 (0.84, 1.08) 0.42 0  3, 35, 36, 367 Post-menopausal  1  0.75 (0.27, 2.05) 0.57 --  2 Prior tamoxifen treatment      0.74  No  4  0.95 (0.84, 1.07)  0.41 0  2, 35, 36, 37 Yes  1  0.81 (0.32, 2.05)  0.66 --  3 Study design      0.58  RCT 2  0.78 (0.40, 1.55) 0.48 0  2, 3 Cohort  3  0.95 (0.84, 1.08) 0.45 0  35, 36, 37 AIs vs. control (no-AIs) b Total effect 7 59,258 1.17 (1.07, 1.28) <0.01 8  4, 6, 9, 10, 35, 36, 37 Subgroup analysis         Menopausal status      0.88  Pre-menopausal 0  -- -- --  -- Pre- / post-menopausal  4  1.19 (1.01, 1.41) 0.04 49  4, 35, 36, 37 Post-menopausal  3  1.17 (0.97, 1.41) 0.10  0  6, 9, 10  Prior tamoxifen treatment      0.99  No  5  1.18 (1.02, 1.37) 0.03 35  4, 9, 35, 36, 37 Yes 2  1.18 (0.97. 1.42)  0.09 0  6, 10 Study design      0.88  RCT 3  1.17 (0.97. 1.41) 0.10  0   6, 9, 10 Cohort  4  1.19 (1.01, 1.41) 0.04 49   4, 35, 36, 37  AI treatment duration      0.57  ≤48 months 2  1.18 (0.97, 1.42)  0.09 0  6, 10 60 months 1  0.81 (0.23, 2.90)  0.75 --  9          101 Continued Treatment arms Study  (N) Participant (N) Pooled RR  (95% CI) p for effect I2 (%) a   p for subgroup differences ID of article included AIs vs. control (no-AIs) b AI drug      0.93  Non-steroidal (letrozole and anastrozole)  1  1.15 (0.94, 1.41) 0.16 --  6 Steroidal (exemestane)  2  1.27 (0.76, 2.14) 0.36 0  9, 10  Any AI  4  1.19 (1.01, 1.41)  0.04 49  35, 36, 37 AIs vs. Tam b Total effect 9 20,403 1.35 (1.21, 1.51)  <0.01 12  14, 15, 18, 26, 30, 32, 33, 36, 37 Subgroup analysis         Menopausal status      0.75  Pre-menopausal 1  1.08 (0.5, 2.35)  0.85 --  14 Pre- / post-menopausal  2  2.00 (0.36, 11.21) 0.43 67  36, 37 Post-menopausal  6  1.39 (1.26, 1.54)  <0.01 0   15, 18, 26, 30, 32, 33 Prior tamoxifen treatment      0.5  No  5  1.38 (1.18, 1.62)  <0.01 27  13, 18, 26, 36, 37 Yes  4  1.27 (1.07, 1.51)  <0.01 0  15, 30, 32, 33 Study design      0.68  RCT 7  1.39 (1.26, 1.53) <0.01 0  14, 15, 18, 26, 30, 32, 33 Cohort  2  2.00 (0.36, 11.21) 0.43 67  36, 37 AI treatment duration      0.19   ≤48 months 5  1.26 (1.07, 1.50)  <0.01 0  14, 15, 30, 32, 33 60 months 2  1.45 (1.29, 1.64)  <0.01 0  18, 26 AI drug      0.76  Non-steroidal (letrozole and anastrozole) 6  1.41 (1.26, 1.59)  <0.01 0   14, 15, 18, 26, 32, 33 Steroidal (exemestane)  1  1.32 (1.10, 1.58) <0.01 --  30  Any AI  2  2.00 (0.36, 11.21) 0.43 67  36, 37 Values in bold and italic indicate statistical significance Tam tamoxifen, AI aromatase inhibitor, RR risk ratio a  For heterogeneity  b Reference group 102  Figure 4-2 Forest plot of comparison for fracture risk between women treated with tamoxifen and not treated with tamoxifen (control) by study design subgroups Tam tamoxifen, IV inverse variance, CI confidence interval, SE standard error, RCT randomized controlled trial The large diamond at the bottle of the table represents the pooled risk ratio of all studies. The width of the diamond represents with 95% CI. Results of study quality assessment were included.  Risk of bias: A Random sequence generation (selection bias), B Allocation concealment (selecitn bias), C Blinding of participants and personnel (performance bias), D Blinding of outcome asessment (detection bias), E Incomplete outcome data (attrition bias), F Selective reporting (reporting bias), G Other bias   Low risk of bias    Unknown risk of bias   High risk of bias  103  Figure 4-3 Forest plot of comparison for fracture risk between women treated with AIs and not treated with AIs (control) by study design subgroups AI aromatase inhibitor, IV inverse variance, CI confidence intervals, SE standard error, RCT randomized controlled trial The large diamond at the bottle of the table represents the pooled risk ratio of all studies. The width of the diamond represents with 95% CI. Results of study quality assessment were included.  Risk of bias: A Random sequence generation (selection bias), B Allocation concealment (selecitn bias), C Blinding of participants and personnel (performance bias), D Blinding of outcome asessment (detection bias), E Incomplete outcome data (attrition bias), F Selective reporting (reporting bias), G Other bias   Low risk of bias    Unknown risk of bias   High risk of bias 104  Figure 4-4 Forest plot of comparison for fracture risk between women treated with AIs and treated with tamoxifen by study design subgroups AI aromatase inhibitor, Tam tamoxifen, IV inverse variance, CI confidence intervals, SE standard error, RCT randomized controlled trial The large diamond at the bottle of the table represents the pooled risk ratio of all studies. The width of the diamond represents with 95% CI. Results of study quality assessment were included.  Risk of bias: A Random sequence generation (selection bias), B Allocation concealment (selecitn bias), C Blinding of participants and personnel (performance bias), D Blinding of outcome asessment (detection bias), E Incomplete outcome data (attrition bias), F Selective reporting (reporting bias), G Other bias   Low risk of bias    Unknown risk of bias   High risk of bias 105 4.3.6 Comparison of aromatase inhibitors and tamoxifen – time effect   Twenty articles from ten independent studies were included for these meta-analyses (Table 4-5, Figure 4-5, Figure 4-6). Compared with the tamoxifen group, increased AI-associated fracture risk showed a downward trend from 47% (pooled RR=1.47, 95% CI=1.28 to 1.68) to 32% (pooled RR=1.32, 95% CI=1.1 to 1.57) when the range of follow-up duration increased from 12-36 months to > 84 months. Compared with the tamoxifen group, AI-associated fracture risk increased by 33% (pooled RR=1.33, 95% CI=1.21 to 1.47) during the Tam/AI treatment period, but did not increase (pooled RR=0.99; 95% CI=0.72 to 1.37) during the post-Tam/AI treatment period. Sensitivity analysis excluding the Koopal et. al. study (ID 11) resulted in a similar RR estimate (pooled RR=1.09, 95% CI=0.92 to 1.31) with a reduction of I2 measure by 56% for the post-Tam/AI treatment period.   Table 4-5   Meta-analysis of aromatase inhibitors and tamoxifen on fractures at different ranges of follow-up duration and treatment phases  Study (N) Participant (N) Pooled RR (95% CI) p for effect I2 (%) a  ID of included articles Aromatase inhibitors (AIs) vs. Tamoxifen b Range of follow-up duration (months) 12-36  6 20,250 1.47 (1.28, 1.68) <0.01 0 15, 16, 21, 28, 31, 37 > 36-60  5 18,237 1.46 (1.27, 1.68) <0.01 15 13, 17, 23, 29, 33 > 60-84 4 13,583 1.39 (1.23, 1.57) <0.01 7 14, 18, 26, 36 > 84 2 5,105 1.32 (1.10, 1.57) <0.01 0 30, 32 Treatment period       Tam/AI treatment 3 13,917 1.33 (1.21, 1.47) <0.01 0 20, 25, 30 Post-Tam/AI treatment  3 8,741 0.99 (0.72, 1.37)  0.96 60 11, 20, 30  a For heterogeneity b Reference group   106  Figure 4-5 Forest plot of comparison for fracture risk between women treated with aromatase inhibitors and tamoxifen (during treatment period) Tam tamoxifen, IV inverse variance, CI confidence interval, SE standard error, RCT randomized controlled trial The large diamond at the bottle of the table represents the pooled risk ratio of all studies. The width of the diamond represents with 95% CI. Results of study quality assessment were included.    Figure 4-6 Forest plot of comparison for fracture risk between women treated with aromatase inhibitors and tamoxifen (during treatment period) Tam tamoxifen, IV inverse variance, CI confidence interval, SE standard error, RCT randomized controlled trial The large diamond at the bottle of the table represents the pooled risk ratio of all studies. The width of the diamond represents with 95% CI. Results of study quality assessment were included.   107 4.4 Discussion  Osteoporosis is a significant global health issue. Osteoporotic fractures are associated with excessive mortality, impaired physical function, and more long term nursing home stays [204-206]. The impact of osteoporosis is higher on women diagnosed with breast cancer than women without breast cancer. Most adjuvant systemic breast cancer treatments result in faster bone loss and a consequential higher fracture rate in women diagnosed with breast cancer [147, 174]. The study data showed that fracture risk did not differ between women treated and not treated with tamoxifen. AI-associated fracture risk was 17% and 35% higher than the risks in the no-AI group and tamoxifen group respectively. Compared with the tamoxifen group, increased AI-associated fracture risk trended down when the range of follow-up duration increased. AI-associated fracture risk increased by 30% during the Tam/AI treatment period but did not increase during the post-Tam/AI treatment period when compared with the tamoxifen group.  4.4.1 Tamoxifen Our results showed that fracture risk did not differ between the tamoxifen and no-tamoxifen groups. This finding is consistent with the fact that tamoxifen has no effect on reducing vertebral or hip fractures in general populations [141, 164]. By contrast, tamoxifen treatment for one-year increased the risk of trochanteric fractures (HR=2.12, 95% CI=1.12 to 4.01) among 1,716 post-menopausal women with non-metastatic breast cancer during the 12-year follow-up in the Danish Breast Cancer Cooperative Group (DBCG) trial [272]. While evidence shows that tamoxifen may preserve BMD, tamoxifen has not been approved for the treatment or prevention of osteoporosis in any population by the US Food and Drug Administration. Women, who 108 receive tamoxifen breast cancer treatment, should still receive BMD testing recommended for women diagnosed with breast cancer.   Our results showed that menopausal status did not affect fracture risk between women treated and not treated with tamoxifen. However, 80% of articles included in our analysis involved a mixture of pre- and post-menopausal women. There was only one article available for the post-menopausal subgroup and none for the pre-menopausal subgroups.   4.4.2 Aromatase inhibitors  AIs are more effective than tamoxifen (Tam) at reducing mortality, reducing cancer recurrences, and prolonging disease-free survival [273, 274]. AIs are given alone for 5 years to patients at higher cancer relapse risk, or in sequence for 2-3 years before or after tamoxifen (sequential AI-Tam or sequential Tam-AI) for patients at lower cancer relapse risk [275]. Sequential treatments, compared with either tamoxifen or AIs alone, reduce the exposure times of both tamoxifen and AIs, which may reduce the long-term side effects associated with either tamoxifen or AIs, such as fracture risk.  Our analysis showed that AI-associated fracture risk increased by 17 and 35% when compared with the no-AI and tamoxifen groups respectively. This finding is consistent with higher fracture risks observed in major trials with an AI intervention.   When comparing AI with tamoxifen groups, differential fracture risks were higher without a statistical difference in the prior tamoxifen treatment subgroup (pooled RR=1.38, 95% CI=1.18 109 to 1.62) than the no prior tamoxifen treatment subgroup (pooled RR=1.27, 95% CI=1.07 to 1.51). This might be because prior tamoxifen treatment may reduce AI-associated fracture risk. Or it may be because follow-up time was longer in the prior tamoxifen subgroup (30-128 months) than the no prior tamoxifen subgroup (32-74 months), and fracture risk decreased when follow-up duration increased.   We did not include or compare fracture risk between sequential AI-Tam and sequential Tam-AI treatments in this study due to limited available data. However, the BIG-98 trial showed sequential AI-Tam treatment reducing fracture risk by 22% (calculated RR=0.78, 95% CI=0.62 to 0.99) compared with the sequential Tam-AI treatment in approximately 3,000 participants during the 45-month follow-up [260].   Longer AI treatment duration did not affect fracture risk in our study, but increased fracture risk by 47% in the Amir et. al. study in 2011 [276]. This could be explained primary by different data synthesis methods. Our study evaluated the effect of AI treatment duration on differential fracture risk between AIs and tamoxifen. The Amir el. al. study evaluated differential fracture risk of AI treatment duration [276].  A steroid AI (exemestane) with irreversible binding properties may affect bone health differently than non-steroidal AIs (letrozole and anastrozole) with reversible binding properties [277]. Our results showed no difference between steroidal and non-steroidal AI subgroups when evaluating differential fracture risks of AIs, and between AIs and tamoxifen. This finding is consistent with findings from two other major trials; a bone sub-study of the Tamoxifen Exemestane Adjuvant 110 Multinational (TEAM) in Japan [278] and MA.27 [279] comparing non-steroidal anastrozole with steroidal exemestane.   Management of AI-associated bone loss is inconsistent to screen for bone loss. A baseline BMD test before initiating AI treatment and follow-up BMD tests at one- to three-year intervals have been suggested [97, 280]. However, an optimal interval for serial BMD testing remains uncertain. Is a shorter screening interval, such as annually, more helpful in identifying women at high fracture risk associated with AIs? Current risk assessment tools, such as the World Health Organization Fracture Risk Assessment tool, do not take AI-associated fracture risk into consideration, which can lead to underestimated fracture risk for women diagnosed with breast cancer and treated with AIs. Should AIs be considered for any fracture risk assessment tool?   4.4.3 Aromatase inhibitors vs. tamoxifen, time effect  While extracting and synthesizing data, we noted that fracture risk was not consistent over time. The RR decreased from 1.60 to 1.44 when the follow-up duration increased from 42 to 68 months in the Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial [252, 253]. The IRRs decreased significantly from 1.55 during the Tam/AI treatment period to 1.03 during the post-tam/AI treatment period in The ATAC trial [255]. In response to this we evaluated the time effect on fracture risk by conducting four individual meta-analyses for four ranges of follow-up duration (12-36, >36-60, >60-84, and >84 months) and two individual meta-analyses for Tam/AI treatment and post-Tam/AI treatment periods.   111 Our results showed that the increased AI-associated fracture risk decreased from 47% (95% CI=1.28 to 1.68) to 32% (95% CI=1.10 to 1.57), when compared with the tamoxifen group and the range of follow-up duration increased from 12-36 to > 85 months. AI-associated fracture risk increased by 33% (95% CI=1.21 to 1.47) during the Tam/AI treatment period but did not increase during the post-Tam/AI treatment period, when compared with tamoxifen. These two findings were consistent as fracture risk decreased over time when more participants entered their post-Tam/AI treatment periods. However, it remains unclear what caused the differences in fracture risks between the treatment and post-treatment periods. It may be due to the independent effect of AI on fracture risk, the independent effect of tamoxifen on fracture risk or both effects combined. The fracture incidence rates (per 1,000 person-years) in the AI group decreased significantly from 29.3 (95% CI=26.5 to 32.4) during the treatment period to 15.6 (95% CI=13.2 to 18.3) during the post-treatment period while rates in the tamoxifen group were stable (treatment period: 19.0, 95% CI=16.7 to 21.5; post-treatment period: 15.1, 95% CI=12.8 to 17.8) in the ATAC trial (ID 19). Contrasting this, the fracture incidence rates (per 1,000 person-years) in the AI group were stable during both treatment period (21.0, 95% CI=14.5 to 27.5) and post-treatment period (20.3, 95% CI=13.7 to 26.9), while rates in the tamoxifen group increased from 12.3 (95% CI=7.3 to 17.3) during the treatment period to 20.6 (95% CI=13.8 to 27.4) during the post- treatment period in the Intergroup Exemestane Study (IES) [158, 170].   4.4.4 Study methodology  Similar estimates between RCTs and cohort subgroups were observed for fracture risk in our study and for effects of other non-cancer drugs in other studies [281, 282]. This is likely because both RCTs and cohort studies included in this study had large participant populations, sufficient 112 follow-up time, and low risk of bias [283]. Most included cohort studies reported relative measures adjusted for confounders, which further reduced selection bias. While at least 50% of included RCTs were unblinded to outcome assessment, this had a minimal effect on assessing objective outcomes including fractures.   Risk differences, defined as differences in proportions of participants with fractures, between two treatments were not analyzed in this study due to significant variation in fracture rates (10 times), heterogeneous participant groups, and baseline risk between studies. Number needed to treat, the average number of participants who need to be treated to prevent one fracture, was not estimated for the same reason.   All selected RCTs and cohort studies in this study reported relative measures as ORs, HRs or IRRs. RRs were selected to estimate effect sizes for several reasons. RRs are more appropriate measures and easier to interpret than ORs. ORs could exaggerate effect sizes, especially for common events or being misinterpreted as RRs [284, 285]. RRs were favored over HRs and IRRs as RRs could be recalculated for almost all included articles except one. HRs were not re-calculable at aggregate data level. IRRs could be recalculated while incidence rates were only available from one-third of the included articles. Published aHRs instead of calculated RRs were selected from three included cohort studies (ID 4, 35, 36) for our meta-analysis. This was because adjusted ratios provide a better effect size estimate with less bias, when compared with unadjusted ratios. Both RRs and aHRs were pooled using a generic inverse variance method.   113 A generic inverse variance method with random effects model was selected in this study to account for different risk measures and heterogeneity across the included studies. The generic inverse variance method is able to pool different relative measures when only ratios with CIs were available. Although we chose random effects models in this study, statistical heterogeneity was low (<15%) in the majority of our analyses except the analysis for post-Tam/AI treatment period and some subgroup analyses. Effect sizes were almost identical using either random or fixed effects models based on our internal analysis.  Heterogeneity is inevitable in meta-analysis. Clinical heterogeneity involving study participant, treatment, and outcome was evaluated by reviewing study methods and assessing clinical relevance. Methodological heterogeneity involving study design and risk of bias was evaluated using Cochrane’s Q statistic and qualified as I2 measures. I2 measures (%) indicate proportions of variability across studies. Mild to moderate statistical heterogeneity (27-67%) was noted in our meta-analyses. This statistical heterogeneity decreased significantly to 0-7% after excluding the Xu et. al. study (ID 37) or the Koopal et. al. study (ID 11). This statistical heterogeneity associated with the Xu et. al. study and the Koopal et. al. study could be explained primarily by uncontrolled confounders due to a lack of reported adjusted relative measures.   There was clinical heterogeneity in age of diagnosis, age of menopause, and proportion bisphosphonate usage especially in the Xu et. al study (ID 37). Age at cancer diagnosis was four years younger in the tamoxifen group than the AI and control treatment groups. Proportions of participants on bisphosphonates was 18% in the AIs group but <1% in the control and tamoxifen groups. Other differences in the Xu et. al. study, compared with the most of the included studies 114 in this review, were setting (one hospital vs. national / multi-national), sample size (211 vs. 2000-44,000), and ethnicity (Chinese vs. Caucasians).   While distributions in characteristics in each treatment group were missing in the Koopal et. al. study, there were significant variations between the pre-menopausal and post-menopausal groups in average age of cessation of hormonal treatments (52 vs.72), follow-up time (2.5 vs. 3.4 years), proportion of chemotherapy (88 vs. 20%), and proportion of bisphosphonate usage (36% vs. 24%). These significant variations were likely to confound fracture risk estimates. Other differences in the Koopal et. al. study, compared with most selected studies in this review, were setting (one medical center vs. national/multi-national) and sample size (300 vs. 2000-44,000).   High risk of bias was noted in open or unblinded RCTs, which accounted for at least 50% of the selected RCTs, and could impact outcome assessment. Unblinding to outcome assessment is likely to affect subjective outcomes including pain and fatigue, but not objective outcomes like fractures. The high risk of bias associated in open or unblinded RCTs has minimal effects on our findings.   The numbers of study participants were identical or the same across serial follow-up articles of each independent study with the exception of the Breast International Group (BIG) 1-98 trial. This was primarily associated with its protocol involving two-arm and four-arm options (Figure 4-7 There were 1,835 participants recruited for the two-arm option (tamoxifen and AIs) in 1998-2000 and 6,193 participants recruited for the four-arm option (tamoxifen only, AIs only, sequential tamoxifen-AI, and sequential AI-tamoxifen) in 1999-2003. The first article (ID 21) 115 published in 2005, included participants from all six arms including sequential treatment groups still on their first treatments only. The participants receiving sequential treatments were excluded in the following four articles (ID 22, 23, 24, 26). The fifth article (ID 25) published in 2009, only involved participants from the tamoxifen and AI groups of the four-arm option.     Figure 4-7 Consort diagram of Breast International Group 1-98 trial  Reprinted with permission. L letrozole, T tamoxifen  4.4.5 Limitation This review was limited by the relative low numbers of available articles on certain subgroups, especially pre-menopausal groups. When comparing AIs with tamoxifen, fracture risks did not differ among subgroups of pre-menopausal, a mixture of pre- and post-menopausal, and post-menopausal women. Only two included studies (ID 13, 34) involved 100% pre-menopausal women. However, the TEXT/SOFT study (ID 34) was not included in our reported meta-analysis as it reported combined data from two independent studies TEXT and SOFT. An internal 116 analysis including data from the TEXT/SOFT study was conducted. It resulted in a similar RR estimate with a slightly narrower 95% CI of 1.24 to 1.48.   4.5 Conclusion  Fracture risk is significantly higher in women treated with AIs, especially during the treatment period. While tamoxifen may preserve BMD, tamoxifen is not associated with fracture risk reduction. Women who receive tamoxifen or AI breast cancer treatment should receive BMD tests as recommended for women diagnosed with breast cancer. Optimal osteoporosis management programs, especially during the treatment period, are needed for this group of women.   117 Chapter 5: Discussion and Conclusion The primary goal of this thesis is to provide a better understanding of osteoporosis, specifically bone mineral density (BMD) testing and the effects of breast cancer treatments on fracture risk, in women diagnosed with breast cancer. This thesis was developed on two main concepts. First, women diagnosed with breast cancer are at higher fracture risk compared with women without breast cancer. Second, BMD testing is recommended to high risk populations – by age (old women aged ≥65) or risk factors (younger women aged <65 with risk factors; breast cancer treatment is not consistently considered a risk factor for BMD testing eligibility). Studies were designed to understand (1) utilization BMD testing in older women who should receive BMD testing as recommended; and (2) the effects of breast cancer treatments on fracture risk in younger women as these women will only receive BMD testing when risk factors exist.  This thesis is the first study to evaluate BMD testing and pilot-test the effect of patient educational material to improve BMD testing rates at a population level - in women aged ≥65 and diagnosed with breast cancer for three or more years in British Columbia, Canada. This thesis also systematically reviewed the effects of the two most common hormonal treatments, tamoxifen and aromatase inhibitors (AIs), in women aged ≤65 and diagnosed with non-metastatic breast cancer.   118 5.1 Key Findings  5.1.1 Utilization of bone, mineral density testing among women diagnosed with breast cancer in British Columbia, Canada The goal of study 1 was to evaluate the trends in proportion of women with at least one BMD per calendar year from 1995 to 2008 and identify factors associated with different BMD testing rates, using a cross-sectional methodology and population based data-linkage in older female breast cancer survivors; namely women aged 65 years and over, and diagnosed with breast cancer for three or more years in British Columbia (BC), Canada.   During the period from 1995 to 2008, the number of survivors almost doubled from 4,974 to 9,662, prevalence of osteoporosis diagnosis increased from 6% to 25.6%, and proportions of women with at least one BMD test increased from 1.0% to 10.1%. The proportions were increasing annually from 1995 to 2005 and became relatively stable from 2005 to 2008.  Associations between socio-demographic, clinical factors, and BMD testing rates over the three-year period 2006-2008 were estimated as adjusted prevalence ratios (aPR) using log-binomial models in 7,625 survivors. Lower SES (aPR=0.66 to 0.78) or rural residency (aPR=0.70) were associated with a 20-30% lower utilization of BMD testing, compared with the highest SES or urban residency respectively. There was a significantly lower likelihood of having a BMD test observed in survivors who were aged 75 years and over (aPR=0.47, 95% CI=0.42 to 0.52), were nursing home residents (aPR=0.05, 95% CI=0.01 to 0.39), had recent osteoporotic fractures (aPR=0.21, 95% CI=0.14 to 0.32), or did not have previous BMD tests (aPR=0.26, 95% CI=0.23 to 0.29). 119 5.1.2 Promoting bone health management in women diagnosed with breast cancer: a pilot randomized controlled trial   The goal of study 2 was to test a randomized controlled trial (RCT) study protocol designed to (1) improve bone health management, especially BMD testing rates during a six-month follow-up period, with educational material; and (2) assess whether delivery methods (postal mail vs. patient’s choice of postal mail, email or smartphone text messaging) of educational material affect bone health management differently in the 54 recruited women aged ≥65 and diagnosed with breast cancer for three or more years in BC, Canada.  The feasibility of the study protocol was evaluated. The recruitment strategy worked well. The response rate, defined as the proportion of women who responded to our invitation, was 77.4%. The participation rate, defined as the proportion of eligible participants who consented to participate, was 39.1%. Representativeness of the recruited group was high based on similar distributions of five factors among the 54 participants, 84 non-participants who declined to participate in this study, and 260 other non-participants. Outcome measures were obtained for at least 90% of the 54 participants. No major issues associated with the study protocol were identified.   Although no formal statistical testing was conducted, there was a suggestion of higher BMD testing rates in the groups receiving educational material by mail (26%, 95% CI=10 to 49) and patient choice (18%, 95% CI=5 to 41), compared with the control group (6%, 95% CI=0.3 to 25). BMD testing rate was 17% (95% CI=6 to 33) higher in the groups where educational material was delivered by either postal mail or patient choice compared with the control group. 120 BMD testing rate was 8.7% (95% CI= -33.9 to 18.9) lower in the patient choice group compared with the postal mail group.  5.1.3 Aromatase inhibitors are associated with a higher fracture risk than tamoxifen: a systematic review and meta-analysis The objectives of study 3 was to systematically evaluate published evidence of bone fracture risks associated with the two most common adjuvant systemic breast cancer treatments, tamoxifen and aromatase inhibitors (AIs).  A total of 43 articles covering 21 independent studies were included in our evaluation. Fracture risk did not differ between women treated and not treated with tamoxifen (pooled RR=0.95, 95% CI=0.84 to 1.07). AI-associated fracture risk was 17% and 35% higher than the risk in the no-AI (pooled RR=1.17, 95% CI=1.07 to 1.28) and tamoxifen (pooled RR=1.35, 95% CI=1.21 to 1.51) groups respectively. No statistical significance was reported in subgroup analyses by menopausal status, prior tamoxifen treatment, study design, AI drug, or AI treatment duration for comparisons of tamoxifen vs. no-tamoxifen, AIs vs. no-AIs, and AIs vs. tamoxifen. Different time effects on fracture risk were observed when comparing AIs with tamoxifen. Compared with the tamoxifen group, the increased AI-associated fracture risk dropped from 47% to 32% when the range of follow-up duration increased from 12-36 months to >85 months. Compared with the tamoxifen group, AI-associated fracture risk was significantly higher (RR=1.33, 95% CI=1.21 to 1.47) during the treatment period but not (pooled RR=1.09, 95% CI=0.91 to 1.31) during the post-Tam/AI treatment period.  121 5.2 Strengths and Limitations  5.2.1 Strengths A major strength of this thesis is that the three study research questions were developed based on the eligibility criteria for BMD testing. Three different study methodologies were then selected based on what was best suited for each of the research questions.   Older women aged 65 and over are eligible for BMD testing regardless of other non-age risk factors. Study 1 provided in-depth knowledge on the utilization of BMD testing in older women diagnosed with breast cancer. Utilization of BMD was well evaluated from 1995 when dual energy X-ray absorptiometry (DXA) became a billable medical service in BC, to 2008 using secondary data-linkage. The advantage of using provincial data include (1) a larger sample size than other studies using non-data-linkage methods; (2) high representativeness of study group with low selection bias; (3) relative low cost as the data has been recorded for administrative purposes; (4) no recall bias that is commonly associated with self-reported data; (5) data was available for more than 10 years which permitted trend analysis. Proportions of women with at least one BMD test were evaluated by calendar year for 13 consecutive years instead of a one-time measurement. This provided a better understanding on how the proportions changed with time in BC. The proportions were stratified by osteoporosis diagnosis as BMD testing is used for osteoporosis screening in women without osteoporosis diagnosis and for treatment effectiveness monitoring in women with osteoporosis. This information was important for health service planning as different strategies are required to improve utilization for each group. Relevant guidelines and historical non-guideline local factors, which might influence the utilization of BMD testing, were integrated into the discussion section. These provided an exhaustive review 122 and a more complete picture of the utilization of BMD testing over a decade period in BC. The study results showed low BMD testing rates while BMD testing is an important tool to identify women at high fracture risk before fractures occur. This led to the development of study 2.   Study 2 evaluated the feasibility of a study protocol and pilot-tested a potential educational material intervention to improve BMD testing rates with a pilot randomized controlled trial (RCT) design. The pilot study design provided valuable information and identified potential issues before conducting a large-scale study. These findings can be applied to a future study protocol to enhance the likelihood of success in a large-scale study. Potential eligible women for this study were selected from the provincial BC Cancer Agency registry. This allowed me to conduct this population-based study and evaluate participants from throughout the province of BC. This approach is a low-cost technique to reach a large number of women. This easy, relatively inexpensive, centralized approach also supported the potential to develop a province-wide bone health management plan for women diagnosed with breast cancer, similar to other province-wide programs we have, such as the cervical cancer screening program and breast cancer screening program. Educational material was edited specifically for this unique population. A strength of educational material is its ability to promote knowledge and awareness of osteoporosis and osteoporotic fracture, which can lead to permanent lifestyle changes. With the positive effects of educational material on BMD testing, it should be considered for future bone health management. Email and text messaging were available options for delivering educational materials. Surprisingly, almost one-third of study participants preferred receiving educational material by email over traditional postal mail. This suggests a potential willingness to adopt technology for health-related issue in this older women group. Other technologies, 123 especially ones with easy access and the ability to deliver a large amount of information, to deliver educational material can be considered in the future. These would include online learning models.    Study 3 systematically synthesized differential fracture risks associated with hormonal treatment of tamoxifen and AIs, and risks between tamoxifen and AIs using a meta-analysis methodology. Systematic review and meta-analysis were selected as it can (1) review all published studies using vigorous methodological assessment tools; (2) pool data from multiple high-quality independent studies which increase statistical power of outcome measures; and (3) provide the highest strength of evidence compared with other study types, such as cohort study or randomized controlled trials. The two breast cancer treatments, tamoxifen and AIs, reviewed in this study have a higher impact than other breast cancer treatments. Both are given to almost two-thirds of women diagnosed with breast cancer. Each of these two mediations is the alternative to the other when one is not well tolerated by patients or contraindicated in post-menopausal women. To well evaluate the effects of tamoxifen and AIs on differential fracture risk, three comparison groups were selected: (1) tamoxifen vs. no-tamoxifen (2) AIs vs. no-AI, and (3) tamoxifen vs. AIs. Different studies were included in each comparison which allowed me to cross-examine the effects between tamoxifen and AI. Moreover, the time effects on differential fracture risk between tamoxifen and AIs were evaluated by two time factors – follow-up duration and treatment period. Results were consistent with randomized controlled trial as the fracture risk decreased with time. To be more specific, the differential fracture risk between tamoxifen and AIs became insignificant when treatment discontinued. The study results 124 are generalizable to all younger women diagnosed with breast cancer and take hormonal treatment, and likely to influence clinical practice guidelines.   5.2.2 Limitations This thesis was limited by two factors. First, the link between study 1 and study 2 was weak due to a seven-year gap. Study 2 was developed based on the findings from study 1 in 2015 which analyzed data from 1995 to 2008. Second, the phases of cancer care differed between studies. Both study 1, evaluating utilization of BMD testing and study 2, pilot-testing interventions to improve BMD testing rates were conducted in women who have already completed their initial breast cancer treatments (diagnosed with breast cancer for three or more years), while study 3 showed a significantly higher AI-associated fracture risk during the treatment period. This inconsistency in cancer treatment phases came about because study 2 and study 3 were developed concurrently.  Study 1 was significantly limited by the fact that the data was old, and the lack of availability of data on treatment factors that might have affected BMD testing rates. The data was only available till year 2008, which was almost ten years old as of this publication. Results from study 1 may not reflect the current utilization of BMD testing, and therefore may not be as relevant to the development of strategies for current care improvement. While the proportions of women with at least one BMD test per calendar year had been relatively stable from 2005 to 2008, utilization of BMD testing after 2008 may be influenced by several factors. Utilization of BMD testing after 2008 may increase due to the increasing usage of AIs and the increasing awareness of fracture risk associated with systemic adjuvant breast cancer treatments. On the contrary, 125 utilization of BMD testing after 2008 may decrease due to the increasing use of validated fracture risk assessment tools, such as the Canadian version of the World Health Organization Fracture Risk Assessment Tool released in 2008. Developing study 2 based on the results from study 1 with a seven-year gap weakened the link between study 1 and study 2.   Some important factors that could affect BMD testing utilization or identification of risk groups, such as AI usage and chemo-induced amenorrhea were also not available in the data of study 1. The Breast Cancer Outcome Unit (BCOU) data only includes information on initial hormonal treatments. We were unable to identify women who switched from initial tamoxifen treatment to subsequent AI treatment. The BCOU data records menstrual status at initial diagnosis of breast cancer but not changes in menstrual status after the completion of chemotherapy. Women with chemo-induced amenorrhea were at higher fracture risk but were not identifiable in this study. Five selected non-cancer chronic diseases, osteoporosis and osteoporotic fractures were identified using International Classification of Disease, Ninth Revision (ICD-9) and Tenth Revision (ICD-10) codes. All codes were commonly used in data-linkage studies. Lack of accurate recording could lead to potential bias due to misclassification [203]. Also, osteoporotic fractures were measured at different time periods for women who had BMD tests (within six-month of a BMD test) or not had BMD tests (three year study period). This could lead to a potential estimation bias. However, the impact would be expected to be low due to the low fracture rate of 6.3%.    Study 2 was subject to potential bias due to low representativeness of the study group and a low overall recruitment rate. The proportion of non-Caucasian participants was less than 5% in our 126 study compared to the population proportion of 27% in BC [223]. This could be primarily explained by the exclusion of potential participants due to language barriers, as educational material was available in English only. The other potential causes of a non-representative study group are  logistical challenges, cultural barriers, and mistrust of research [224]. The overall recruitment rate in this study was only 13% (54/398). However, only three-quarters of the original 398 women were invited to participate in this study. Of all women invited, about one-third were ineligible for this study.   Study 3 was limited by low numbers of studies available for subgroup analyses, such as factors of menopausal status, AI treatment duration and AI drug. No studies reporting 100% pre-menopausal women were available for estimating differential fracture risks of tamoxifen and AIs. Only one study reporting 100% pre-menopausal women was available for estimating different fracture risk between tamoxifen and AIs. Another study involving 100% pre-menopausal women, The Tamoxifen and Exemestane Trial (TEXT)/Suppression Ovarian Functions (SOFT) study (ID 34), was not included in our reported meta-analysis as it reported combined data from two independent studies TEXT and SOFT. Potential bias should be considered when interpreting subgroup analyses with low numbers of studies.   5.3 Conclusion  Increased risk of fractures is reported in women diagnosed with breast cancer and treated with aromatase inhibitors, while screening for osteoporosis with bone mineral density testing is sub-optimal. There is a need for better bone health management programs which should include educational materials. 127 5.4 Implications, Applications and Future Research  Women diagnosed with breast cancer are at higher risk of osteoporosis and osteoporotic fractures because most adjuvant breast cancer treatments cause estrogen deficiency while estrogen plays a key role in bone heath. Osteoporosis is a major public health issue with significant care gaps while breast cancer is the most common female cancer worldwide. Osteoporosis has a significant impact with more deaths, physical disabilities, and economic burdens [132-134]. The disability burden of osteoporotic fractures is higher than most common cancers with the exception of lung cancer [88]. Costs of hospital care for osteoporotic fractures are higher than for breast cancer, myocardial infarction, and stroke [286]. Osteoporosis has become a more critical issue in women diagnosed with breast cancer than the female general population. A better understanding of bone health management is urgently needed for future care planning, which is the main objective of this thesis.   5.4.1 Older women aged 65 years and over who were diagnosed with breast cancer  In older women (aged 65 years and over) who were diagnosed with breast cancer, BMD tests at one- to three-year intervals are recommended. BMD testing plays a key role in identifying women at higher fracture risk before fractures occur. Study 1 (Chapter Two) showed that utilization of BMD testing remains sub-optimal, especially in women with lower SES or living in rural areas. The proportions of women with at least one BMD test per calendar year were under 15% over the period from 1995 to 2008. This is significantly lower than screening rates for other common chronic diseases, such as diabetes (73-80%) or cholesterol (82.4%) in Ontario, another Canadian province [287]. Study 2 suggested that educational material has a great potential to improve bone health management, especially BMD testing rates.  128 All findings imply that this group of older women should be encouraged to receive BMD testing as recommended by the most current Canadian guidelines. Patient educational material could potentially improve bone health management, especially BMD testing rates and physical activity in this group of women.   The patient educational material developed for this thesis, comprised of two parts: (1) three pages of information on osteoporosis, potential effects of breast cancer treatments on bones, BMD testing, lifestyle advice to promote bone health, and advice to review osteoporosis risk with one’s family doctor (Appendix C); and (2) one double-sided page of risk factors based on the 2010 Canadian osteoporosis guidelines and fracture risk assessment tool (FRAX) developed by the World Health Organization (Appendix D) [97, 195]. This educational material targets primarily patient barriers, especially underestimated personal perceived risk of osteoporosis [118-120]. The double-sided page of risk factors could prompt family doctors’ knowledge on osteoporosis risk factors. While it is important to prompt patients’ awareness in bone health management, physicians should be encouraged to recommend BMD testing, FRAX evaluation and healthy lifestyles based on the relevant guidelines. Healthy lifestyles include exercise, adequate calcium and vitamin D intake, avoid excessive alcohol drinking and fall prevention. Among lifestyles, exercise should be recommended per the World Health Organization Guidelines ((≥150 minutes of moderate-intensity aerobic exercise per week or ≥75 minutes of vigorous-intensity aerobic exercise per week; for adults aged ≥65 years) for all adults aged ≥65, regardless to their chronic disease history [222].   129 This thesis identified a significant care gap in utilization of BMD testing, factors associated with low BMD testing rates, and a potential educational material intervention to improve BMD testing rates and bone health-associated lifestyles. Several important questions still need to be answered in future studies. While the pilot RCT suggested that patient educational material developed for this thesis had positive effects on bone health management, the question of the most effective way to deliver this material, especially with newer communication technologies and targeting disparate subgroups with lower BMD testing rates, needs further investigation. Newer communication technologies, such as internet communication, with high availability and relative low-cost should be considered in disparate subgroups, including but not limited to women with lower SES, living in rural areas, or with previous osteoporotic fracture.   5.4.2 Younger women aged 65 years and under who were diagnosed with breast cancer  In younger women aged 65 and under, and diagnosed with breast cancer, BMD testing is only recommended to high-risk individuals. The definitions of high risk vary between guidelines. While most adjuvant systemic breast cancer treatments cause estrogen deficiency and accelerated bone loss, BMD testing is not universally indicated for women who receive these treatments. For example, BMD testing is not indicated for a 60-year-old woman who receives AIs without additional osteoporosis risk factors based on the most current guidelines in BC in 2011 [178]. A better understanding of fracture risks associated with adjuvant systemic breast cancer treatments can potentially alter BMD eligibility criteria for younger women. Study 3 demonstrated from current literature that tamoxifen has no effect on fracture risk while AIs increase fracture risk. AIs increased fracture risk over tamoxifen. This increased AI-associated fracture risk, compared with tamoxifen, decreases over time and is primarily seen only while women are receiving their AI treatment. 130 These findings have several implications. First, while tamoxifen may preserve BMD, tamoxifen does not reduce fracture risk; therefore, younger women who receive tamoxifen as part of breast cancer treatment should still receive BMD testing as recommended by guidelines. Second, increased AI-associated fracture risk should be taken into consideration for fracture risk assessment and BMD testing eligibility. Third, increased AI-associated fracture risk, compared with tamoxifen, is significantly higher during the treatment period but not the post-treatment period. Better bone health management programs, especially during the treatment period are needed for women who received AI breast cancer treatment.   Future research is needed to (1) evaluate fracture risk associated with sequential treatments of both tamoxifen and AIs; (2) to identify an optimized bone health management plan for women who are receiving AI treatment. First, AIs could be given alone for five years or in sequence for 2-3 years before or after tamoxifen (sequential AI-tamoxifen or sequential tamoxifen-AI). Sequential treatments, compared with either tamoxifen or AIs alone, reduce the exposure times of both tamoxifen and AIs, which may reduce the long-term side effects associated with either tamoxifen or AIs, such as fracture risk. This lead to two important questions: “do fracture risks differ between sequential tamoxifen-AI and AI-tamoxifen treatments?” and “is sequential treatment (either tamoxifen-AI or AI-tamoxifen) associated with lower fracture risk than an AI treatment alone?”  Second, women who receive AI treatment are at higher fracture risk, especially during the treatment period. A BMD testing at one- to three-year intervals is recommended for this group by the Canadian guideline [97]. The optimal interval for BMD testing for this particular group remains unclear. Is a shorter screening interval, such as annual BMD testing as suggested for cancer populations by some international guidelines [174], better 131 than a two- or three-year interval for screening this special population for osteoporosis before fractures occur? Also, should AI be considered in any fracture risk assessment tools such as the World Health Organization Fracture Risk Assessment tool?   5.4.3 Women diagnosed with breast cancer  Most adjuvant systemic breast cancer treatments accelerate bone loss and increase fracture risks in women diagnosed with breast cancer. Optimal bone health management could potentially prevent fractures from occurring. Bone health management should include lifestyle advice, screening with BMD testing, pharmacological treatment, and monitoring. Bone health management, especially for women who receive breast cancer treatments with negative effects on bones, should be initiated when breast cancer diagnosis is made and continue through to end of life. To achieve optimal bone health care through the breast cancer diagnosis, treatment and post-treatment phases, requires coordination and share of care between oncologists and family doctors [288-290]. While multiple shared care models between oncologists and family doctors have been proposed [288], it remains unclear which model is more suitable for women diagnosed with breast cancer in BC. Future research is needed to (1) understand current care-share patterns on bone health between oncologists and family doctors; and (2) identify potential barriers associated with care-share or care coordination between oncologists and family doctors.     132 Bibliography 1. Canadian Cancer Society's Advisory Committee on Cancer Statistics. and desLibris - Documents, Canadian Cancer Statistics 2015 - Special Topic Predictions of the Future Burden of Cancer in Canada, 2015, Canadian Cancer Society: Toronto, ON. p. 1 online resource. 2. Torre, L.A., et al., Global cancer statistics, 2012. CA Cancer J Clin, 2015. 65(2): p. 87-108. 3. Hulka, B.S. and P.G. Moorman, Breast cancer: hormones and other risk factors. Maturitas, 2001. 38(1): p. 103-113. 4. Ghafoor, A., et al., Trends in Breast Cancer by Race and Ethnicity. CA Cancer J Clin, 2003. 53(6): p. 342-355. 5. Chlebowski, R.T., et al., Ethnicity and Breast Cancer: Factors Influencing Differences in Incidence and Outcome. Journal of the National Cancer Institute, 2005. 97(6): p. 439-448. 6. Nichols, H.B., et al., Declining Incidence of Contralateral Breast Cancer in the United States From 1975 to 2006. Journal of Clinical Oncology, 2011. 29(12): p. 1564-1569. 7. McCormack, V.A. and I. dos Santos Silva, Breast Density and Parenchymal Patterns as Markers of Breast Cancer Risk: A Meta-analysis. Cancer Epidemiology Biomarkers &amp; Prevention, 2006. 15(6): p. 1159-1169. 8. El-Wakeel, H. and H.C. Umpleby, Systematic review of fibroadenoma as a risk factor for breast cancer. The Breast, 2003. 12(5): p. 302-307. 9. Colditz, G.A. and B. Rosner, Cumulative Risk of Breast Cancer to Age 70 Years According to Risk Factor Status: Data from the Nurses' Health Study. Am J Epidemiol, 2000. 152(10): p. 950-964. 10. Kenney, L.B., et al., Breast cancer after childhood cancer: A report from the childhood cancer survivor study. Annals of Internal Medicine, 2004. 141(8): p. 590-597. 11. Guibout, C., et al., Malignant Breast Tumors After Radiotherapy for a First Cancer During Childhood. Journal of Clinical Oncology, 2005. 23(1): p. 197-204. 12. Pukkala, E., et al., Breast cancer in Belarus and Ukraine after the Chernobyl accident. International Journal of Cancer, 2006. 119(3): p. 651-658. 13. Henderson, T.O., et al., Surveillance for Breast Cancer in Women Treated with Chest Radiation for a Childhood, Adolescent or Young Adult Cancer: A Report from the Children's Oncology Group. Annals of Internal Medicine, 2010. 152(7): p. 444-W154. 14. Hsieh, C.C., et al., Age at Menarche, Age at Menopause, Height and Obesity as Risk-Factors for Breast-Cancer - Associations and Interactions in an International Case-Control Study. International Journal of Cancer, 1990. 46(5): p. 796-800. 15. Ritte, R., et al., Height, age at menarche and risk of hormone receptor-positive and -negative breast cancer: A cohort study. International Journal of Cancer, 2013. 132(11): p. 2619-2629. 16. Collaborative Group on Hormonal Factors in Breast, C., Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52 705 women with breast cancer and 108 411 women without breast cancer. The Lancet, 1997. 350(9084): p. 1047-1059. 17. Kelsey, J.L., M.D. Gammon, and E.M. John, Reproductive Factors and Breast-Cancer. Epidemiologic Reviews, 1993. 15(1): p. 36-47. 133 18. Rosner, B., G.A. Colditz, and W.C. Willett, Reproductive Risk Factors in a Prospective Study of Breast Cancer: The Nurses' Health Study. Am J Epidemiol, 1994. 139(8): p. 819-835. 19. Collaborative Group on Hormonal Factors in Breast, C., Breast cancer and breastfeeding: collaborative reanalysis of individual data from 47 epidemiological studies in 30 countries, including 50 302 women with breast cancer and 96 973 women without the disease. The Lancet, 2002. 360(9328): p. 187-195. 20. Shah, N.R., J. Borenstein, and R.W. Dubois, Postmenopausal hormone therapy and breast cancer: a systematic review and meta-analysis. Menopause (New York, N.Y.), 2005. 12(6): p. 668-678. 21. Gierisch, J.M., et al., Oral Contraceptive Use and Risk of Breast, Cervical, Colorectal, and Endometrial Cancers: A Systematic Review. Cancer Epidemiology Biomarkers &amp; Prevention, 2013. 22(11): p. 1931-1943. 22. Harvie, M., L. Hooper, and A.H. Howell, Central obesity and breast cancer risk: a systematic review. Obesity Reviews, 2003. 4(3): p. 157-173. 23. van den Brandt, P.A., et al., Pooled Analysis of Prospective Cohort Studies on Height, Weight, and Breast Cancer Risk. Am J Epidemiol, 2000. 152(6): p. 514-527. 24. Green, J., et al., Height and cancer incidence in the Million Women Study: prospective cohort, and meta-analysis of prospective studies of height and total cancer risk. Lancet Oncol, 2011. 12(8): p. 785-794. 25. Lahmann, P.H., et al., Body size and breast cancer risk: Findings from the European prospective investigation into cancer and nutrition (EPIC). International Journal of Cancer, 2004. 111(5): p. 762-771. 26. Johnson, K.C., et al., Active smoking and secondhand smoke increase breast cancer risk: the report of the Canadian Expert Panel on Tobacco Smoke and Breast Cancer Risk (2009). Tobacco Control, 2011. 20(1): p. e2. 27. Gaudet, M.M., et al., Active Smoking and Breast Cancer Risk: Original Cohort Data and Meta-Analysis. Journal of the National Cancer Institute, 2013. 105(8): p. 515-525. 28. Bagnardi, V., et al., Light alcohol drinking and cancer: a meta-analysis. Annals of Oncology, 2013. 24(2): p. 301-308. 29. Cao, Y., et al., Light to moderate intake of alcohol, drinking patterns, and risk of cancer: results from two prospective US cohort studies. The BMJ, 2015. 351: p. h4238. 30. Allen, N.E., et al., Moderate Alcohol Intake and Cancer Incidence in Women. Journal of the National Cancer Institute, 2009. 101(5): p. 296-305. 31. Chen, W.Y., et al., Moderate alcohol consumption during adult life, drinking patterns, and breast cancer risk. JAMA, 2011. 306(17): p. 1884-1890. 32. Megdal, S.P., et al., Night work and breast cancer risk: A systematic review and meta-analysis. European Journal of Cancer, 2005. 41(13): p. 2023-2032. 33. Lang, J.E. and H.M. Kuerer, Breast ductal secretions: clinical features, potential uses, and possible applications. Cancer Control, 2007. 14(4): p. 350-9. 34. Newton, P., D.R. Hannay, and R. Laver, The presentation and management of female breast symptoms in general practice in Sheffield. Fam Pract, 1999. 16(4): p. 360-5. 35. Ayoade, B.A., A.O. Tade, and B.A. Salami, Clinical Features and Pattern of Presentation of Breast Diseases in Surgical Outpatient Clinic of a Suburban Tertiary Hospital in South-West Nigeria. Nigerian Journal of Surgery : Official Publication of the Nigerian Surgical Research Society, 2012. 18(1): p. 13-16. 134 36. Koo, M.M., et al., Typical and atypical symptoms in women with breast cancer: Evidence of variation in diagnostic intervals from a national audit of cancer diagnosis. Poster presented at: National Cancer Reserach Institute Cancer Conference; Nov. 2016; Liverpool, UK 2016. 37. Irvin, W., H.B. Muss, and D.K. Mayer, Symptom Management in Metastatic Breast Cancer. The Oncologist, 2011. 16(9): p. 1203-1214. 38. Tonelli, M., et al., Recommendations on screening for breast cancer in average-risk women aged 40-74 years. CMAJ, 2011. 183(17): p. 1991-2001. 39. Siu, A.L., Screening for Breast Cancer: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med, 2016. 164(4): p. 279-96. 40. Independent, U.K.P.o.B.C.S., The benefits and harms of breast cancer screening: an independent review. The Lancet. 380(9855): p. 1778-1786. 41. Gøtzsche, P.C. and K.J. Jørgensen, Screening for breast cancer with mammography. Cochrane Database of Systematic Reviews, 2013(6). 42. Miller, A.B., et al., Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. BMJ : British Medical Journal, 2014. 348. 43. Bevers, T.B., et al., NCCN clinical practice guidelines in oncology: breast cancer screening and diagnosis. J Natl Compr Canc Netw, 2009. 7(10): p. 1060-96. 44. Compton, C.C., et al., Purposes and Principles of Cancer Staging, in AJCC Cancer Staging Atlas: A Companion to the Seventh Editions of the AJCC Cancer Staging Manual and Handbook, C.C. Compton, et al., Editors. 2012, Springer New York: New York, NY. p. 3-22. 45. Bartlett, J.M., et al., Estrogen receptor and progesterone receptor as predictive biomarkers of response to endocrine therapy: a prospectively powered pathology study in the Tamoxifen and Exemestane Adjuvant Multinational trial. J Clin Oncol, 2011. 29(12): p. 1531-8. 46. Pertschuk, L.P., et al., Immunocytochemical estrogen and progestin receptor assays in breast cancer with monoclonal antibodies. Histopathologic, demographic, and biochemical correlations and relationship to endocrine response and survival. Cancer, 1990. 66(8): p. 1663-70. 47. Harvey, J.M., et al., Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to adjuvant endocrine therapy in breast cancer. J Clin Oncol, 1999. 17(5): p. 1474-81. 48. Arteaga, C.L., et al., Treatment of HER2-positive breast cancer: current status and future perspectives. Nat Rev Clin Oncol, 2011. 9(1): p. 16-32. 49. Goldhirsch, A., et al., Meeting Highlights: International Expert Consensus on the Primary Therapy of Early Breast Cancer 2005. Annals of Oncology, 2005. 16(10): p. 1569-1583. 50. Elston, C.W. and I.O. Ellis, Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology, 1991. 19(5): p. 403-10. 51. Scarth, H., et al., Clinical practice guidelines for the care and treatment of breast cancer: mastectomy or lumpectomy? The choice of operation for clinical stages I and II breast cancer (summary of the 2002 update). CMAJ, 2002. 167(2): p. 154-5. 135 52. Cantin, J., et al., Clinical practice guidelines for the care and treatment of breast cancer: 13. Sentinel lymph node biopsy. CMAJ, 2001. 165(2): p. 166-73. 53. Truong, P.T., et al., Clinical practice guidelines for the care and treatment of breast cancer: 16. Locoregional post-mastectomy radiotherapy. CMAJ, 2004. 170(8): p. 1263-73. 54. Whelan, T., et al., Clinical practice guidelines for the care and treatment of breast cancer: breast radiotherapy after breast-conserving surgery (summary of the 2003 update). CMAJ, 2003. 168(4): p. 437-9. 55. Levine, M. and Steering Committee on Clinical Practice Guidelines for the Care and Treatment of Breast Cancer, Clinical practice guidelines for the care and treatment of breast cancer: adjuvant systemic therapy for node-negative breast cancer (summary of the 2001 update). CMAJ, 2001. 164(2): p. 213. 56. Levine, M. and Steering Committee on Clinical Practice Guidelines for the Care and Treatment of Breast Cancer, Clinical practice guidelines for the care and treatment of breast cancer: adjuvant systemic therapy for node-positive breast cancer (summary of the 2001 update). The Steering Committee on Clinical Practice Guidelines for the Care and Treatment of Breast Cancer. CMAJ, 2001. 164(5): p. 644-6. 57. Goldhirsch, A., M. Colleoni, and R.D. Gelber, Endocrine therapy of breast cancer. Ann Oncol, 2002. 13 Suppl 4: p. 61-8. 58. Denduluri, N., et al., Selection of Optimal Adjuvant Chemotherapy Regimens for Human Epidermal Growth Factor Receptor 2 (HER2) -Negative and Adjuvant Targeted Therapy for HER2-Positive Breast Cancers: An American Society of Clinical Oncology Guideline Adaptation of the Cancer Care Ontario Clinical Practice Guideline. J Clin Oncol, 2016. 34(20): p. 2416-27. 59. Statistics Canada., Population projections for Canada, Provinces and Territories, 2009 to 2036., 2010, Statistica Canada: Ottawa, Ontario. 60. Canadian Cancer Society's Advisory Committee on Cancer Statistics. and desLibris - Documents, Canadian Cancer Statistics 2016 - Special Topic HPV-associated cancer., 2016, Canadian Cancer Society,: Toronto, Ontario. p. 1 online resource. 61. Canadian Cancer Society's Advisory Committee on Cancer Statistics. and desLibris - Documents, National Cancer Institute of Canada: Canadian Cancer Statistics 2000., 2000, Canadian Cancer Society,: Toronto, Ontario. p. p. 1 online resource. 62. Cancer Surveillance and Outcomes of Population Oncology. BC Cancer Statistics 2014  [cited 2016 /12/01]; Available from: http://www.bccancer.bc.ca/health-info/disease-system-statistics/bc-cancer-statistics/facts-and-figures. 63. Smith, J.L., et al., Assessment of the status of a National Action Plan for Cancer Survivorship in the USA. J Cancer Surviv, 2013. 7(3): p. 425-38. 64. McCabe, M.S., et al., American Society of Clinical Oncology statement: achieving high-quality cancer survivorship care. J Clin Oncol, 2013. 31(5): p. 631-40. 65. Institute of Medicine, From Cancer Patient to Cancer Survivor: Lost in Transition (An American Society of Clinical Oncology and Institute of Medicine Symposium), ed. M. Hewitt and P.A. Ganz2006, Washington, D.C.: National Academy Press. 66. Abdel-Razeq, H. and A. Awidi, Bone health in breast cancer survivors. J Cancer Res Ther, 2011. 7(3): p. 256-63. 67. Bird, B.R. and S.M. Swain, Cardiac toxicity in breast cancer survivors: review of potential cardiac problems. Clin Cancer Res, 2008. 14(1): p. 14-24. 136 68. Edwards, B.J., et al., Cancer therapy associated bone loss: implications for hip fractures in mid-life women with breast cancer. Clin Cancer Res, 2011. 17(3): p. 560-8. 69. Khan, N.F., et al., Long-term health outcomes in a British cohort of breast, colorectal and prostate cancer survivors: a database study. Br J Cancer, 2011. 105 Suppl 1: p. S29-37. 70. Patnaik, J.L., et al., Cardiovascular disease competes with breast cancer as the leading cause of death for older females diagnosed with breast cancer: a retrospective cohort study. Breast Cancer Res, 2011. 13(3): p. R64. 71. Schover, L.R., Premature ovarian failure and its consequences: vasomotor symptoms, sexuality, and fertility. J Clin Oncol, 2008. 26(5): p. 753-8. 72. Christiansen, C. and P. Riis, [Consensus Development Conference. Prevention and treatment of osteoporosis]. Nord Med, 1991. 106(5): p. 145-7. 73. N. I. H. Consensus Development Panel, Osteoporosis prevention, diagnosis, and therapy. JAMA, 2001. 285(6): p. 785-795. 74. Johnell, O. and J. Kanis, Epidemiology of osteoporotic fractures. Osteoporosis International, 2005. 16(2): p. S3-S7. 75. Reginster, J.Y. and N. Burlet, Osteoporosis: a still increasing prevalence. Bone, 2006. 38(2 Suppl 1): p. S4-9. 76. Kanis J.A. on behave of the World Health Organizaiton Scientific Group Assessment of Osteoporosis at the Primary Health Care Level. Technical Report. 2007. 77. Hanley, D.A. and R.G. Josse, Prevention and management of osteoporosis: consensus statements from the Scientific Advisory Board of the Osteoporosis Society of Canada. 1. Introduction. CMAJ, 1996. 155(7): p. 921-3. 78. Tenenhouse, A., et al., Estimation of the prevalence of low bone density in Canadian women and men using a population-specific DXA reference standard: the Canadian Multicentre Osteoporosis Study (CaMos). Osteoporos Int, 2000. 11(10): p. 897 - 904. 79. Melton, L.J., et al., How many women have osteoporosis? (Reprinted from Journal Bone & Mineral Research, vol 7, pg 1005, 1992). Journal of Bone and Mineral Research, 2005. 20(5): p. 886-892. 80. Kanis, J.A., et al., Long-term risk of osteoporotic fracture in Malmo. Osteoporos Int, 2000. 11(8): p. 669-74. 81. Gullberg, B., O. Johnell, and J. Kanis, World-wide projections for hip fracture. Osteoporos Int, 1997. 7(5): p. 407 - 413. 82. Bessette, L., et al., The care gap in diagnosis and treatment of women with a fragility fracture. Osteoporos Int, 2008. 19(1): p. 79-86. 83. Papadimitropoulos, E.A., et al., Current and projected rates of hip fracture in Canada. Canadian Medical Association Journal, 1997. 157(10): p. 1357-1363. 84. Brown, J.P. and R.G. Josse, 2002 clinical practice guidelines for the diagnosis and management of osteoporosis in Canada. CMAJ, 2002. 167(10 Suppl): p. S1-34. 85. Wilson, J.M. and Y.G. Jungner, [Principles and practice of mass screening for disease]. Bol Oficina Sanit Panam, 1968. 65(4): p. 281-393. 86. Schousboe, J.T., et al., Universal bone densitometry screening combined with alendronate therapy for those diagnosed with osteoporosis is highly cost-effective for elderly women. J Am Geriatr Soc, 2005. 53(10): p. 1697-704. 87. Tosteson, A.N.A., et al., Cost-effective osteoporosis treatment thresholds: the United States perspective. Osteoporosis International, 2008. 19(4): p. 437-447. 137 88. World Health Organization. WHO Scientific Group on the Assessment of Osteoporosis at Primary Health Care Level. Summary Meeting Report, Brussels, Belgium, 5-7 May 2004. 2004. 89. Marshall, D., O. Johnell, and H. Wedel, Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ, 1996. 312(7041): p. 1254-9. 90. Cranney, A., et al., Low bone mineral density and fracture burden in postmenopausal women. Canadian Medical Association Journal, 2007. 177(6): p. 575-580. 91. Wainwright, S.A., et al., Hip fracture in women without osteoporosis. Journal of Clinical Endocrinology & Metabolism, 2005. 90(5): p. 2787-2793. 92. Heaney, R.P., Pathophysiology of Osteoporosis Endocrinology and Metabolism Clinics of North America, 1998. 27(2): p. 255-265. 93. Kenny, A.M. and K.M. Prestwood, Osteoporosis. Pathogenesis, diagnosis, and treatment in older adults. Rheum Dis Clin North Am, 2000. 26(3): p. 569-91. 94. El Maghraoui, A. and C. Roux, DXA scanning in clinical practice. QJM, 2008. 101(8): p. 605-17. 95. WHO Study Group on Assessment of Fracture Risk and its Application to Screening for Postmenopausal Osteoporosis (1992 : Rome Italy), Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. WHO technical report series,1994, Geneva: World Health Organization. v, 129 p. 96. Schousboe, J.T., et al., Executive Summary of the 2013 International Society for Clinical Densitometry Position Development Conference on Bone Densitometry. Journal of Clinical Densitometry, 2013. 16(4): p. 455-466. 97. Papaioannou, A., et al., 2010 clinical practice guidelines for the diagnosis and management of osteoporosis in Canada: summary. CMAJ, 2010. 182(17): p. 1864-73. 98. Bolland, M.J., et al., Effect of osteoporosis treatment on mortality: a meta-analysis. J Clin Endocrinol Metab, 2010. 95(3): p. 1174-81. 99. Cranney, A., et al., Meta-analyses of therapies for postmenopausal osteoporosis. IX: Summary of meta-analyses of therapies for postmenopausal osteoporosis. Endocr Rev, 2002. 23(4): p. 570-8. 100. Wall, M., et al., Fracture risk assessment in long-term care: a survey of long-term care physicians. Bmc Geriatrics, 2013. 13: p. 109. 101. International Atomic Energy Agency, Dual Energy X Ray Absorptiometry for Bone Mineral Density and Body Composition Assessment. IAEA Human Health Series2011, Vienna: International Atomic Energy Agency. 102. Nayak, S., et al., Meta-analysis: Accuracy of quantitative ultrasound for identifying patients with osteoporosis. Annals of Internal Medicine, 2006. 144(11): p. 832-841. 103. Kanis, J.A., et al., The use of clinical risk factors enhances the performance of BMD in the prediction of hip and osteoporotic fractures in men and women. Osteoporosis International, 2007. 18(8): p. 1033-1046. 104. Humadi, A., R.H. Alhadithi, and S.I. Alkudiari, Validity of the DEXA diagnosis of involutional osteoporosis in patients with femoral neck fractures. Indian J Orthop, 2010. 44(1): p. 73-8. 105. McLeod, K.M., et al., Predictors of change in calcium intake in postmenopausal women after osteoporosis screening. J Nutr, 2007. 137(8): p. 1968-73. 138 106. Brennan, R.M., et al., Factors associated with treatment initiation after osteoporosis screening. Am J Epidemiol, 2004. 160(5): p. 475-83. 107. Dugard, M.N., T.J. Jones, and M.W. Davie, Uptake of treatment for osteoporosis and compliance after bone density measurement in the community. J Epidemiol Community Health, 2010. 64(6): p. 518-22. 108. Marci, C.D., et al., Bone mineral densitometry substantially influences health-related behaviors of postmenopausal women. Calcif Tissue Int, 2000. 66(2): p. 113-8. 109. Barr, R.J., et al., Population screening for osteoporosis risk: a randomised control trial of medication use and fracture risk. Osteoporos Int, 2010. 21(4): p. 561-8. 110. Siminoski, K., et al., Recommendations for bone mineral density reporting in Canada. Can Assoc Radiol J, 2005. 56(3): p. 178-88. 111. Schuit, S.C.E., et al., Fracture incidence and association with bone mineral density in elderly men and women: the Rotterdam Study. Bone, 2004. 34(1): p. 195-202. 112. Rimes, K.A. and P.A. Salkovskis, Prediction of psychological reactions to bone density screening for osteoporosis using a cognitive-behavioral model of health anxiety. Behaviour Research and Therapy, 2002. 40(4): p. 359-381. 113. Snyder, C.F., et al., Comparing care for breast cancer survivors to non-cancer controls: a five-year longitudinal study. J Gen Intern Med, 2009. 24(4): p. 469-74. 114. Demeter, S., et al., The effect of socioeconomic status on bone density testing in a public health-care system. Osteoporos Int, 2007. 18(2): p. 153-8. 115. Giangregorio, L.M., et al., Osteoporosis management among residents living in long-term care. Osteoporosis International, 2009. 20(9): p. 1471-1478. 116. Gupta, G. and W.S. Aronow, Underuse of Procedures for Diagnosing Osteoporosis and of Therapies for Osteoporosis in Older Nursing Home Residents. Journal of the American Medical Directors Association, 2003. 4(4): p. 200-202. 117. Giangregorio, L., et al., Fragility fractures and the osteoporosis care gap: an international phenomenon. Semin Arthritis Rheum, 2006. 35(5): p. 293-305. 118. Grover, M.L., et al., Fracture risk perception study: patient self-perceptions of bone health often disagree with calculated fracture risk. Womens Health Issues, 2014. 24(1): p. e69-75. 119. Siris, E.S., et al., Failure to perceive increased risk of fracture in women 55 years and older: the Global Longitudinal Study of Osteoporosis in Women (GLOW). Osteoporos Int, 2011. 22(1): p. 27-35. 120. Hsieh, C., et al., Health beliefs and attitudes toward the prevention of osteoporosis in older women. Menopause, 2001. 8(5): p. 372-6. 121. Morse, L.R., et al., Barriers to Providing Dual Energy X-ray Absorptiometry Services to Individuals with Spinal Cord Injury. Am J Phys Med Rehabil, 2009. 88(1): p. 57-60. 122. Curtis, J., et al., The Geographic Availability and Associated Utilization of Dual Energy X-ray Absorptiometry (DXA) Testing among Older Persons in the United States. Osteoporos Int, 2009. 20(9): p. 1553-1561. 123. Simonelli, C., et al., Barriers to osteoporosis identification and treatment among primary care physicians and orthopedic surgeons. Mayo Clin Proc, 2002. 77(4): p. 334-338. 124. Yarnall, K.S.H., et al., Primary Care: Is There Enough Time for Prevention? American Journal of Public Health, 2003. 93(4): p. 635-641. 139 125. Health Quality, O., Utilization of DXA Bone Mineral Densitometry in Ontario: An Evidence-Based Analysis. Ontario Health Technology Assessment Series, 2006. 6(20): p. 1-180. 126. Morris, C.A., et al., Patterns of bone mineral density testing: current guidelines, testing rates, and interventions. J Gen Intern Med, 2004. 19(7): p. 783-90. 127. Saag, K.G. Overcoming the Barriers: Strategies for Improving Osteoporosis Management. 2012. 128. Little, E.A. and M.P. Eccles, A systematic review of the effectiveness of interventions to improve post-fracture investigation and management of patients at risk of osteoporosis. Implementation Science, 2010. 5. 129. Laliberté, M.C., et al., Effectiveness of interventions to improve the detection and treatment of osteoporosis in primary care settings: a systematic review and meta-analysis. Osteoporosis International, 2011. 22(11): p. 2743-2768. 130. Gardner, M.J., et al., Interventions to improve osteoporosis treatment following hip fracture. The Journal of Bone & Joint Surgery, 2005. 87(1): p. 3-7. 131. Davis, J.C., et al., HipWatch: osteoporosis investigation and treatment after a hip fracture: a 6-month randomized controlled trial. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 2007. 62(8): p. 888-891. 132. Abrahamsen, B., et al., Excess mortality following hip fracture: a systematic epidemiological review. Osteoporos Int, 2009. 20(10): p. 1633-50. 133. Johnell, O. and J.A. Kanis, An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int, 2006. 17(12): p. 1726-33. 134. Melton, L.J., 3rd, Adverse outcomes of osteoporotic fractures in the general population. J Bone Miner Res, 2003. 18(6): p. 1139-41. 135. Wiktorowicz, M.E., et al., Economic implications of hip fracture: Health service use, institutional care and cost in Canada. Osteoporosis International, 2001. 12(4): p. 271-278. 136. Berger, A., Bone mineral density scans. BMJ, 2002. 325(7362): p. 484. 137. Leib, E., Diagnosis of osteoporosis in men, premenopausal women, and children. Journal of Clinical Densitometry, 2004. 7(1): p. 17-26. 138. Tang, B.M.P., et al., Use of calcium or calcium in combination with vitamin D supplementation to prevent fractures and bone loss in people aged 50 years and older: a meta-analysis. Lancet, 2007. 370(9588): p. 657-666. 139. de Kam, D., et al., Exercise interventions to reduce fall-related fractures and their risk factors in individuals with low bone density: a systematic review of randomized controlled trials. Osteoporosis International, 2009. 20(12): p. 2111-2125. 140. Kemmler, W., L. Haberle, and S. von Stengel, Effects of exercise on fracture reduction in older adults A systematic review and meta-analysis. Osteoporosis International, 2013. 24(7): p. 1937-1950. 141. Qaseem, A., et al., Pharmacologic treatment of low bone density or osteoporosis to prevent fractures: A clinical practice guideline from the American College of Physicians. Annals of Internal Medicine, 2008. 149(6): p. 404-W77. 142. Baim, S., et al., Precision assessment and radiation safety for dual-energy X-ray absorptiometry. Journal of Clinical Densitometry, 2005. 8(4): p. 371-378. 143. Peppone, L.J., et al., Bone health issues in breast cancer survivors: a Medicare Current Beneficiary Survey (MCBS) study. Support Care Cancer, 2014. 22(1): p. 245-51. 140 144. Chen, Z., et al., Osteoporosis and rate of bone loss among postmenopausal survivors of breast cancer. Cancer, 2005. 104(7): p. 1520-30. 145. Chen, Z., et al., Fracture risk among breast cancer survivors: results from the Women's Health Initiative Observational Study. Archives of Internal Medicine, 2005. 165(5): p. 552-8. 146. Tsa, C.H., et al., Fracture in Asian Women with Breast Cancer Occurs at Younger Age. PLoS ONE, 2013. 8(9). 147. Bell, R. and J. Lewis, Assessing the risk of bone fracture among postmenopausal women who are receiving adjuvant hormonal therapy for breast cancer. Curr Med Res Opin, 2007. 23(5): p. 1045-51. 148. Khosla, S., M.J. Oursler, and D.G. Monroe, Estrogen and the skeleton. Trends in Endocrinology & Metabolism. 23(11): p. 576-581. 149. Howlader, N., et al., US Incidence of Breast Cancer Subtypes Defined by Joint Hormone Receptor and HER2 Status. JNCI: Journal of the National Cancer Institute, 2014. 106(5): p. dju055-dju055. 150. Davidson, A., et al., Stage, treatment and outcomes for patients with breast cancer in British Columbia in 2002: a population-based cohort study. CMAJ Open, 2013. 1(4): p. E134-E141. 151. Gnant, M., C. Thomssen, and N. Harbeck, St. Gallen/Vienna 2015: A Brief Summary of the Consensus Discussion. Breast Care, 2015. 10(2): p. 124-130. 152. Goldhirsch, A., et al., Meeting Highlights: Updated International Expert Consensus on the Primary Therapy of Early Breast Cancer. Journal of Clinical Oncology, 2003. 21(17): p. 3357-3365. 153. Shapiro , C.L. and A. Recht Side Effects of Adjuvant Treatment of Breast Cancer. New England Journal of Medicine, 2001. 344(26): p. 1997-2008. 154. Turner, R.T., et al., Tamoxifen Prevents the Skeletal Effects of Ovarian Hormone Deficiency in Rats. Journal of Bone and Mineral Research, 1987. 2(5): p. 449-456. 155. Powles, T., et al., Effect of tamoxifen on bone mineral density measured by dual-energy x-ray absorptiometry in healthy premenopausal and postmenopausal women. Journal of Clinical Oncology, 1996. 14(1): p. 78-84. 156. Rodriguez-Rodriguez, L.-M., et al., Changes on bone mineral density after adjuvant treatment in women with non-metastatic breast cancer. Breast Cancer Research & Treatment, 2005. 93(1): p. 75-83. 157. Zidan, J., et al., Effects of tamoxifen on bone mineral density and metabolism in postmenopausal women with early-stage breast cancer. Medical Oncology, 2004. 21(2): p. 117-21. 158. Coleman, R.E., et al., Reversal of skeletal effects of endocrine treatments in the Intergroup Exemestane Study. Breast Cancer Research & Treatment, 2010. 124(1): p. 153-61. 159. Zaman, K., et al., Bone mineral density in breast cancer patients treated with adjuvant letrozole, tamoxifen, or sequences of letrozole and tamoxifen in the BIG 1-98 study (SAKK 21/07). Annals of Oncology, 2012. 23(6): p. 1474-81. 160. Kristensen, B., et al., Tamoxifen and Bone Metabolism in Postmenopausal Low-Risk Breast-Cancer Patients - a Randomized Study. Journal of Clinical Oncology, 1994. 12(5): p. 992-997. 141 161. Love, R.R., et al., Effects of tamoxifen on bone mineral density in postmenopausal women with breast cancer. New England Journal of Medicine, 1992. 326(13): p. 852-6. 162. Kristensen, B., et al., Tamoxifen and bone metabolism in postmenopausal low-risk breast cancer patients: a randomized study. Journal of Clinical Oncology, 1994. 12(5): p. 992-7. 163. Love, R.R., et al., Effect of tamoxifen on lumbar spine bone mineral density in postmenopausal women after 5 years. Archives of Internal Medicine, 1994. 154(22): p. 2585-8. 164. MacLean, C., et al., Systematic review: Comparative effectiveness of treatments to prevent fractures in men and women with low bone density or osteoporosis. Annals of Internal Medicine, 2008. 148(3): p. 197-213. 165. Miller, W.R., Aromatase inhibitors: mechanism of action and role in the treatment of breast cancer. Semin Oncol, 2003. 30, Supplement 14: p. 3-11. 166. Harris, A.L., T.J. Powles, and I.E. Smith, Aminoglutethimide in the Treatment of Advanced Post-Menopausal Breast-Cancer. Cancer Research, 1982. 42(8): p. 3405-3408. 167. Kennecke, H.F., et al., New guidelines for treatment of early hormone-positive breast cancer with tamoxifen and aromatase inhitibors BCMJ, 2006. 48(3): p. 121-126. 168. Chumsri, S., et al., Aromatase, aromatase inhibitors, and breast cancer. J Steroid Biochem Mol Biol, 2011. 125(1–2): p. 13-22. 169. Eastell, R., et al., Effect of anastrozole on bone mineral density: 5-year results from the anastrozole, tamoxifen, alone or in combination trial 18233230. Journal of Clinical Oncology, 2008. 26(7): p. 1051-1058. 170. Cuzick, J., et al., Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 10-year analysis of the ATAC trial. Lancet Oncology, 2010. 11(12): p. 1135-41. 171. Rabaglio, M., et al., Bone fractures among postmenopausal patients with endocrine-responsive early breast cancer treated with 5 years of letrozole or tamoxifen in the BIG 1-98 trial. Annals of Oncology, 2009. 20(9): p. 1489-98. 172. Vekariya, K.K., J. Kaur, and K. Tikoo, Alleviating anastrozole induced bone toxicity by selenium nanoparticles in SD rats. Toxicol Appl Pharmacol, 2013. 268(2): p. 212-20. 173. Lim, L.S., L.J. Hoeksema, and K. Sherin, Screening for osteoporosis in the adult U.S. population: ACPM position statement on preventive practice. Am J Prev Med, 2009. 36(4): p. 366-75. 174. Lustberg, M.B., R.E. Reinbolt, and C.L. Shapiro, Bone health in adult cancer survivorship. J Clin Oncol, 2012. 30(30): p. 3665-74. 175. Cheung, A.M., et al., Prevention of osteoporosis and osteoporotic fractures in postmenopausal women: recommendation statement from the Canadian Task Force on Preventive Health Care. CMAJ, 2004. 170(11): p. 1665-7. 176. Grunfeld, E., S. Dhesy-Thind, and M. Levine, Clinical practice guidelines for the care and treatment of breast cancer: follow-up after treatment for breast cancer (summary of the 2005 update). CMAJ, 2005. 172(10): p. 1319-20. 177. Guidelines and Protocols Advisory Committee. Bone Density Measurement in Women 2005. 178. Coimmittee, G.a.P.A., Osteoporosis: Diagnosis, Treatment and Fracture Prevention 2011, Doctor of BC: Victoria, BC  179. Chen, Z., et al., Fracture risk among breast cancer survivors - Results from the Women's Health Initiative Observational Study. Arch Intern Med, 2005. 165(5): p. 552-558. 142 180. Canadian Cancer Society's Adivisory Committee on Cancer Statistics Canadian Cancer Statistics 2013. 2013. 1 online resource. 181. Harvey, N., E. Dennison, and C. Cooper, Osteoporosis: impact on health and economics. Nat Rev Rheumatol, 2010. 6(2): p. 99-105. 182. von Elm, E., et al., Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ, 2007. 335(7624): p. 806-8. 183. BC Cancer Agency Registry Data and Breast Cancer Outcome Unit Data, B.C. Agency, Editor 2012, British Columbia Cancer Agency [Publisher]: Vancouver, BC  184. Medical Service Plan (MSP) Payment Information File 2012, Population Data BC [Publisher]. 185. Discharge Abstracts Database (Hospital Separations), 2012, Population Data BC [Publisher]. 186. Medical Services Plan (MSP) Consolidation File 2012, Population Data BC [Publisher]. 187. Bilkins, R. and S. Khan, PCCF + Version 5J* User's Guide 2011, Statistics Canada Ottawa, ON. 188. Statistics Canada, Postal Code Conversion File (PCCF), 2002 2002, Statistics Canada. Geography Division [Producer and Distributor]: Ottawa, Ontario. 189. Quan, H., et al., Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care, 2005. 43(11): p. 1130-9. 190. Statistics Canada, Table 102-0561 - Leading causes of death, total population, by age group and sex, Canada, 2017. 191. Statistical Methodology and Applications Branch, Surveillance Research Program, and National Cancer Institute, Joinpoint Regression Program, Version 4.3.1.0, April 2016  192. Barros, A.J. and V.N. Hirakata, Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol, 2003. 3: p. 21. 193. Clinical practice guidelines for the diagnosis and management of osteoporosis. Scientific Advisory Board, Osteoporosis Society of Canada,. CMAJ, 1996. 155(8): p. 1113-33. 194. Cadarette, S.M., et al., Development and validation of the Osteoporosis Risk Assessment Instrument to facilitate selection of women for bone densitometry. CMAJ, 2000. 162(9): p. 1289-94. 195. Kanis, J.A., et al., Development and use of FRAX in osteoporosis. Osteoporos Int, 2010. 21 Suppl 2: p. S407-13. 196. Public Health Agency of Canada., Canadian Chronic Dsieaes Indicators, Quick Stats, 2017 Edition., 2017, Public Health Agency of Canada Ottawa (ON). 197. Keating, N.L., et al., Surveillance testing among survivors of early-stage breast cancer. Journal of Clinical Oncology, 2007. 25(9): p. 1074-1081. 198. Siris, E.S., et al., Bone mineral density thresholds for pharmacological intervention to prevent fractures. Archives of Internal Medicine, 2004. 164(10): p. 1108-1112. 199. Moores, D.G., et al., Improving the Quality and Capacity of Canada's Health Services: Primary Care Physician Perspectives. Healthc Policy, 2007. 3(2): p. e145-61. 200. Schott, A.M., et al., Which screening strategy using BMD measurements would be most cost effective for hip fracture prevention in elderly women? A decision analysis based on a Markov model. Osteoporosis International, 2007. 18(2): p. 143-151. 143 201. Khatib, R., et al., Fracture risk in long term care: a systematic review and meta-analysis of prospective observational studies. Bmc Geriatrics, 2014. 14. 202. Klotzbuecher, C.M., et al., Patients with prior fractures have an increased risk of future fractures: A summary of the literature and statistical synthesis. Journal of Bone and Mineral Research, 2000. 15(4): p. 721-739. 203. Quan, H., et al., Assessing Validity of ICD-9-CM and ICD-10 Administrative Data in Recording Clinical Conditions in a Unique Dually Coded Database. Health Services Research, 2008. 43(4): p. 1424-1441. 204. Bonar, S.K., et al., Factors associated with short- versus long-term skilled nursing facility placement among community-living hip fracture patients. J Am Geriatr Soc, 1990. 38(10): p. 1139-44. 205. Cummings, S.R. and L.J. Melton, Epidemiology and outcomes of osteoporotic fractures. Lancet, 2002. 359(9319): p. 1761-7. 206. Chrischilles, E.A., et al., A model of lifetime osteoporosis impact. Arch Intern Med, 1991. 151(10): p. 2026-32. 207. Burge, R., et al., Incidence and economic burden of osteoporosis-related fractures in the United States, 2005-2025. Journal of Bone and Mineral Research, 2007. 22(3): p. 465-75. 208. Clark, S.J., et al., Parents' Experiences With and Preferences for Immunization Reminder/Recall Technologies. Pediatrics, 2011. 128(5): p. e1100-e1105. 209. Baptist, A.P., et al., Social Media, Text Messaging, and Email—Preferences of Asthma Patients between 12 and 40 Years Old. Journal of Asthma, 2011. 48(8): p. 824-830. 210. BC Cancer Agency Registry Data and Breast Cancer Outcomes Unit Data, B.C. Agency, Editor 2015, British Columbia Cancer Agency [Publisher]: Vancouver, BC  211. Godin, G. and R. Shephard, Godin leisure-time exercise questionnaire. Med Sci Sports Exerc, 1997. 29(6): p. 36-38. 212. Nabak, A.C., et al., Can a questionnaire predict vitamin D status in postmenopausal women? Public health nutrition, 2014. 17(4): p. 739-746. 213. Hung, A., et al., Validation of a calcium assessment tool in postmenopausal Canadian women. Maturitas, 2011. 69(2): p. 168-172. 214. Amireault, S., et al., The use of the Godin-Shephard Leisure-Time Physical Activity Questionnaire in oncology research: a systematic review. BMC Med Res Methodol, 2015. 15(1): p. 60. 215. Leslie, W.D., et al., Construction of a FRAX® model for the assessment of fracture probability in Canada and implications for treatment. Osteoporosis International, 2011. 22(3): p. 817-827. 216. Levy Milne, R., et al., British Columbia Nutrition Survey report on seniors' nutritional health, 2004, Ministry of Health Services,: Victoria, B.C. p. 1 online resource (viii, 51 p.). 217. Basen-Engquist, K. and M. Chang, Obesity and Cancer Risk: Recent Review and Evidence. Curr Oncol Rep, 2011. 13(1): p. 71-76. 218. Demark-Wahnefried, W., K. Campbell, and S.C. Hayes, Weight Management and its Role in Breast Cancer Rehabilitation. Cancer, 2012. 118(8 0): p. 10.1002/cncr.27466. 219. Blaney, J., et al., The Cancer Rehabilitation Journey: Barriers to and Facilitators of Exercise Among Patients With Cancer-Related Fatigue. Physical Therapy, 2010. 90(8): p. 1135-1147. 144 220. Rogers, L.Q., et al., Exercise stage of change, barriers, expectations, values and preferences among breast cancer patients during treatment: a pilot study. European Journal of Cancer Care, 2007. 16(1): p. 55-66. 221. Moayyeri, A., The Association Between Physical Activity and Osteoporotic Fractures: A Review of the Evidence and Implications for Future Research. Ann Epidemiol, 2008. 18(11): p. 827-835. 222. in Global Recommendations on Physical Activity for Health2010: Geneva. 223. Wan, D., et al., Breast Cancer Subtype Variation by Race and Ethnicity in a Diverse Population in British Columbia. Clin Breast Cancer. 16(3): p. e49-e55. 224. Quay, T.A., et al., Barriers and facilitators to recruitment of South Asians to health research: a scoping review. BMJ Open, 2017. 7(5). 225. Berry , D.A., et al., Effect of Screening and Adjuvant Therapy on Mortality from Breast Cancer. New England Journal of Medicine, 2005. 353(17): p. 1784-1792. 226. Moher, D., et al., Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Reviews, 2015. 4(1): p. 1. 227. Golder, S., Y.K. Loke, and M. Bland, Meta-analyses of Adverse Effects Data Derived from Randomised Controlled Trials as Compared to Observational Studies: Methodological Overview. PLoS Med, 2011. 8(5): p. e1001026. 228. Higgins, J.P.T., et al., The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ, 2011. 343. 229. Savović, J., et al., Evaluation of the Cochrane Collaboration’s tool for assessing the risk of bias in randomized trials: focus groups, online survey, proposed recommendations and their implementation. Systematic Reviews, 2014. 3: p. 37-37. 230. Zeng, X., et al., The methodological quality assessment tools for preclinical and clinical studies, systematic review and meta-analysis, and clinical practice guideline: a systematic review. Journal of Evidence-Based Medicine, 2015. 8(1): p. 2-10. 231. Wells G, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analysis. 2011  [cited 2015 /01/01]; Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. 232. Jakesz, R., et al., Switching of postmenopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years' adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial. Lancet, 2005. 366(9484): p. 455-62. 233. Pagani, O., et al., Adjuvant exemestane with ovarian suppression in premenopausal breast cancer. New England Journal of Medicine, 2014. 371(2): p. 107-118. 234. Higgins JPT, Green S, and [editors], Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0, Higgins JPT and Green S, Editors. 2011, The Cochrane Collaboration. 235. Review Manager (RevMan) [Computer Program]. Version 5.2, 2014, The Cochrane Collaboration Copenhagen: The Nordic Cochrane Centre. 236. Sacco, M., et al., Randomized trial of 2 versus 5 years of adjuvant tamoxifen for women aged 50 years or older with early breast cancer: Italian Interdisciplinary Group Cancer Evaluation Study of Adjuvant Treatment in Breast Cancer 01. Journal of Clinical Oncology, 2003. 21(12): p. 2276-81. 237. Mincey, B.A., et al., Risk of cancer treatment-associated bone loss and fractures among women with breast cancer receiving aromatase inhibitors. Clinical Breast Cancer, 2006. 7(2): p. 127-132. 145 238. Goss, P.E., et al., A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. New England Journal of Medicine, 2003. 349(19): p. 1793-802. 239. DeGrendele, H. and J.A. O'Shaughnessy, Benefit of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. Clinical Breast Cancer, 2003. 4(5): p. 311-312. 240. Goss, P.E., et al., Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: updated findings from NCIC CTG MA.17. Journal of the National Cancer Institute, 2005. 97(17): p. 1262-71. 241. Goss, P.E., et al., Late extended adjuvant treatment with letrozole improves outcome in women with early-stage breast cancer who complete 5 years of tamoxifen. Journal of Clinical Oncology, 2008. 26(12): p. 1948-1955. 242. Lonning, P.E., et al., Effects of exemestane administered for 2 years versus placebo on bone mineral density, bone biomarkers, and plasma lipids in patients with surgically resected early breast cancer. Journal of Clinical Oncology, 2005. 23(22): p. 5126-37. 243. Geisler, J., et al., Changes in bone and lipid metabolism in postmenopausal women with early breast cancer after terminating 2-year treatment with exemestane: a randomised, placebo-controlled study. European Journal of Cancer, 2006. 42(17): p. 2968-75. 244. Mamounas, E.P., et al., Benefit from exemestane as extended adjuvant therapy after 5 years of adjuvant tamoxifen: Intention-to-treat analysis of the national surgical adjuvant breast and bowel project B-33 trial. Journal of Clinical Oncology, 2008. 26(12): p. 1965-1971. 245. Koopal, C., et al., Fracture incidence in pre- and postmenopausal women after completion of adjuvant hormonal therapy for breast cancer. Breast, 2015. 24(2): p. 153-158. 246. Gnant, M., et al., Endocrine therapy plus zoledronic acid in premenopausal breast cancer. New England Journal of Medicine, 2009. 360(7): p. 679-691. 247. Gnant, M., et al., Adjuvant endocrine therapy plus zoledronic acid in premenopausal women with early-stage breast cancer: 62-month follow-up from the ABCSG-12 randomised trial. The Lancet Oncology, 2011. 12(7): p. 631-641. 248. Kaufmann, M., et al., Improved overall survival in postmenopausal women with early breast cancer after anastrozole initiated after treatment with tamoxifen compared with continued tamoxifen: The ARNO 95 study. Journal of Clinical Oncology, 2007. 25(19): p. 2664-2670. 249. Buzdar, A.U., 'Arimidex' (anastrozole) versus tamoxifen as adjuvant therapy in postmenopausal women with early breast cancer-efficacy overview. Journal of Steroid Biochemistry and Molecular Biology, 2003. 86(3-5): p. 399-403. 250. Fisher, M.D., J. O'Shaughnessy, and J.A. Sparano, Anastrozole may be superior to Tamoxifen as adjuvant treatment for postmenopausal patients with breast cancer. Clinical Breast Cancer, 2002. 2(4): p. 269-271. 251. Baum, M., et al., Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial.[Erratum appears in Lancet 2002 Nov 9;360(9344):1520]. Lancet, 2002. 359(9324): p. 2131-9. 252. Baum, M., et al., Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early-stage breast cancer: 146 results of the ATAC (Arimidex, Tamoxifen Alone or in Combination) trial efficacy and safety update analyses. Cancer, 2003. 98(9): p. 1802-10. 253. Howell, A., et al., Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years' adjuvant treatment for breast cancer. Lancet, 2005. 365(9453): p. 60-2. 254. Cuzick, J., The ATAC trial: the vanguard trial for use of aromatase inhibitors in early breast cancer. Expert Review of Anticancer Therapy, 2007. 7(8): p. 1089-94. 255. Arimidex, T.A.o.i.C.T.G., et al., Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 100-month analysis of the ATAC trial. Lancet Oncology, 2008. 9(1): p. 45-53. 256. Thurlimann, B., et al., A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. New England Journal of Medicine, 2005. 353(26): p. 2747-2757. 257. Monnier, A., The evolving role of letrozole in the adjuvant setting: First results from the large, phase III, randomized trial BIG 1-98. Breast, 2006. 15(1 SUPPL.): p. S21-S29. 258. Crivellari, D., et al., Letrozole compared with tamoxifen for elderly patients with endocrine-responsive early breast cancer: the BIG 1-98 trial. Journal of Clinical Oncology, 2008. 26(12): p. 1972-9. 259. Coates, A.S., et al., Five years of letrozole compared with tamoxifen as initial adjuvant therapy for postmenopausal women with endocrine-responsive early breast cancer: update of study BIG 1-98. Journal of Clinical Oncology, 2007. 25(5): p. 486-92. 260. Mouridsen, H., et al., Letrozole therapy alone or in sequence with tamoxifen in women with breast cancer. New England Journal of Medicine, 2009. 361(8): p. 766-776. 261. Colleoni, M., et al., Analyses adjusting for selective crossover show improved overall survival with adjuvant letrozole compared with tamoxifen in the BIG 1-98 study. Journal of Clinical Oncology, 2011. 29(9): p. 1117-1124. 262. Nuzzo, F., et al., Bone effect of adjuvant tamoxifen, letrozole or letrozole plus zoledronic acid in early-stage breast cancer: the randomized phase 3 HOBOE study. Annals of Oncology, 2012. 23(8): p. 2027-33. 263. Coombes, R.C., et al., A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. Women's Oncology Review, 2004. 4(2): p. 135-136. 264. Coleman, R.E., et al., Skeletal effects of exemestane on bone-mineral density, bone biomarkers, and fracture incidence in postmenopausal women with early breast cancer participating in the Intergroup Exemestane Study (IES): a randomised controlled study. Lancet Oncology, 2007. 8(2): p. 119-27. 265. Bliss, J.M., et al., Disease-related outcomes with long-term follow-up: An updated analysis of the intergroup exemestane study. Journal of Clinical Oncology, 2012. 30(7): p. 709-717. 266. Boccardo, F., et al., Switching to anastrozole versus continued tamoxifen treatment of early breast cancer: preliminary results of the Italian Tamoxifen Anastrozole Trial. J Clin Oncol, 2005. 23(22): p. 5138-47. 267. Boccardo, F., et al., Switching to anastrozole versus continued tamoxifen treatment of early breast cancer: long term results of the Italian Tamoxifen Anastrozole trial. European Journal of Cancer, 2013. 49(7): p. 1546-54. 147 268. Aihara, T., et al., Phase III randomized adjuvant study of tamoxifen alone versus sequential tamoxifen and anastrozole in Japanese postmenopausal women with hormone-responsive breast cancer: N-SAS BC03 study. Breast Cancer Research & Treatment, 2010. 121(2): p. 379-87. 269. Ligibel, J.A., et al., Risk of myocardial infarction, stroke, and fracture in a cohort of community-based breast cancer patients. Breast Cancer Research & Treatment, 2012. 131(2): p. 589-97. 270. Robinson, P.J., et al., Minimal-trauma fracture in women with breast cancer surviving for at least 5 years from diagnosis. Osteoporosis International, 2014. 26(2): p. 795-800. 271. Xu, L., et al., Aromatase inhibitors associated musculoskeletal disorders and bone fractures in postmenopausal breast cancer patients: a result from Chinese population. Medical Oncology, 2014. 31(9): p. 128. 272. Kristensen, B., et al., Femoral fractures in postmenopausal breast cancer patients treated with adjuvant tamoxifen. Breast Cancer Research & Treatment, 1996. 39(3): p. 321-6. 273. Dowsett, M., et al., Meta-Analysis of Breast Cancer Outcomes in Adjuvant Trials of Aromatase Inhibitors Versus Tamoxifen. Journal of Clinical Oncology, 2010. 28(3): p. 509-518. 274. Eisen, A., et al., Aromatase inhibitors in adjuvant therapy for hormone receptor positive breast cancer: A systematic review. Cancer Treat Rev, 2008. 34(2): p. 157-174. 275. Harbeck, N., C. Thomssen, and M. Gnant, St. Gallen 2013: Brief Preliminary Summary of the Consensus Discussion. Breast Care, 2013. 8(2): p. 102-109. 276. Amir, E., et al., Toxicity of Adjuvant Endocrine Therapy in Postmenopausal Breast Cancer Patients: A Systematic Review and Meta-analysis. JNCI: Journal of the National Cancer Institute, 2011. 103(17): p. 1299-1309. 277. McCloskey, E.V., Aromatase inhibitors and bone health. IBMS BoneKEy, 2006. 3(6): p. 5-13. 278. Aihara, T., et al., Effects of exemestane, anastrozole and tamoxifen on bone mineral density and bone turnover markers in postmenopausal early breast cancer patients: results of N-SAS BC 04, the TEAM Japan substudy. Oncology, 2010. 79(5-6): p. 376-81. 279. Goss, P.E., et al., Exemestane versus anastrozole in postmenopausal women with early breast cancer: NCIC CTG MA.27--a randomized controlled phase III trial. J Clin Oncol, 2013. 31(11): p. 1398-404. 280. Body, J.-J., Aromatase inhibitors-induced bone loss in early breast cancer. BoneKEy Reports, 2012. 1: p. 201. 281. Ioannidis, J.A., et al., Comparison of evidence of treatment effects in randomized and nonrandomized studies. JAMA, 2001. 286(7): p. 821-830. 282. Benson , K. and A.J. Hartz A Comparison of Observational Studies and Randomized, Controlled Trials. New England Journal of Medicine, 2000. 342(25): p. 1878-1886. 283. Vandenbroucke, J.P., When are observational studies as credible as randomised trials? The Lancet, 2004. 363(9422): p. 1728-1731. 284. Balasubramanian, H., et al., Odds ratio vs risk ratio in randomized controlled trials. Postgrad Med, 2015. 127(4): p. 359-367. 285. Knol, M.J., et al., Potential Misinterpretation of Treatment Effects Due to Use of Odds Ratios and Logistic Regression in Randomized Controlled Trials. PLoS One, 2011. 6(6): p. e21248. 148 286. Sirois, M.J., M. Cote, and S. Pelet, The burden of hospitalized hip fractures: patterns of admissions in a level I trauma center over 20 years. J Trauma, 2009. 66(5): p. 1402-10. 287. Wilson, S.E., et al., The effectiveness and efficiency of diabetes screening in Ontario, Canada: a population-based cohort study. BMC Public Health, 2010. 10(1): p. 506. 288. McCabe, M.S., et al., Risk-Based Health Care, the Cancer Survivor, the Oncologist and the Primary Care Physician. Semin Oncol, 2013. 40(6): p. 804-812. 289. Klabunde, C.N., et al., Physician Roles in the Cancer-Related Follow-Up Care of Cancer Survivors. Family Medicine, 2013. 45(7): p. 463-474. 290. Klabunde, C.N., et al., The Role of Primary Care Physicians in Cancer Care. Journal of General Internal Medicine, 2009. 24(9): p. 1029-1036.     149 Appendices Appendix A  Summary of datasets in this study Database  Data range for this study Population coverage  Data Description Information obtained for this study BC Cancer Registry [21] 1986 – 2011 BC residents diagnosed with cancer since 1985 (95%) All cancers diagnosed for BC residents  • Gender • Birth dates • Death dates  • Cancer diagnosis  Breast Cancer Outcome Unit (BCOU)  1989 - 2011 BC residents diagnosed with breast cancer and referred to one of the provincial treatment centers operated by BCCA since year 1989  Breast cancer treatment  • Breast cancer treatment  Consolidation file [Medical Service Registration & Premium Billing] [23] 1986 – 2011 All Canadians and legal immigrants who have been living in BC for at least 6 months  Demographic information  • Gender • Birth dates • Residential regions • Postal codes for urban/rural status • Socioeconomic status  • Active registration status for alive follow-up status MSP Payment Information File [22] 1986 – 2011 All Canadians and legal immigrants who have been living in BC for at least 6 months.   Medically necessary services  • BMD tests • Osteoporosis diagnosis • Bone fracture diagnosis  • Chronic disease diagnosis Discharge Abstracts Database  [Hospital Separation] [26] 1986 – 2011 All BC residents Hospital admissions and day surgeries  • Osteoporosis diagnosis • Bone fracture diagnosis   • Chronic disease diagnosis  BCCA British Columbia Cancer Agency, MSP Medical Service Payment      150 Appendix B  Summary of codes for variables Variable Diagnostic Codes Fee code b  ICD 9 code a ICD 10 code  Nursing home    00115, 13334 Osteoporosis  733 M80, M81  Osteoporotic fracture    Hip 820  S72  Spine 805  S22, S32   Forearm 813 S62  Chronic disease [189]    Myocardial infarction and coronary heart disease  410, 412 I21, I22, I25.2  Cerebrovascular disease 430-438 G45, G46, H34.0, I60-I69  Dementia  290 F00-F03, G30, F05.1, G30, G31.1   Chronic pulmonary disease 490-505 I27.8, I27.9, I40 –I47, J60-J67, J68.4, J70.1, J70.3  Diabetes 250  E10.0–E10.9, E11.0-E11.9, E12.0-E12.9, E13.0-E13.9, E14.0-E14.9   ICD -9 International Classification of Diseases, Ninth Revision, ICD-10 International Classification of Diseases, Tenth Revision a fee codes are specific for the province of British Columbia only   151 Appendix C  Educational Material   152   153    154    155 Appendix D  List of risk factors   156  

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.24.1-0355207/manifest

Comment

Related Items