Population-Based Trends in Osteoporosis Management after New Initiations of Long-Term Systemic Glucocorticoids ( )

Transcription

1 ORIGINAL Endocrine ARTICLE Care Population-Based Trends in Osteoporosis Management after New Initiations of Long-Term Systemic Glucocorticoids ( ) S. R. Majumdar, L. M. Lix, M. Yogendran, S. N. Morin, C. J. Metge, and W. D. Leslie From Department of Medicine (S.R.M.), University of Alberta, Edmonton Alberta, Canada T6G 2B7; School of Public Health (L.M.L.), University of Saskatchewan, Saskatoon Saskatchewan, Canada S7N 5A2; University of Manitoba (M.Y., C.J.M., W.D.L.), Winnipeg Manitoba, Canada R3T 2N2; and Department of Medicine (S.N.M.), McGill University, Montreal, Québec, Canada H3A 3R1 Objectives: Our objective was to describe changes in glucocorticoid-induced osteoporosis (GIOP) preventive care from including rates and correlates of bone mineral density (BMD) testing and osteoporosis treatment in new long-term glucocorticoid initiations. Methods: A population-based study of adults aged 20 yr or older in Manitoba, Canada, was conducted using linked healthcare databases. Subjects with new long-term ( 90 d) systemic glucocorticoid initiations were identified within each fiscal year. High-quality GIOP preventive care was defined by the composite of BMD testing or osteoporosis treatment within 6 months of starting glucocorticoids. For each initiation, we identified sociodemographic and clinical characteristics, prednisone dose equivalents, and prescriber specialty. Multivariable Poisson regression models were used to calculate adjusted incidence rate ratios (airr). Results: We studied 17,736 new long-term glucocorticoid initiations; one third were at least 10 mg prednisone daily, and most (64%) were prescribed by general practitioners. Overall, 6-month rates of BMD testing were 6%, osteoporosis treatment 22%, and the composite of testing or treatment 25%. From , there were modest increases in BMD testing (from 4 to 6%), osteoporosis treatment (from 15 to 24%), and testing or treatment [from 17 to 27%; airr 1.51; 95% confidence interval (CI) ]. High-quality GIOP preventive care varied significantly by age (16% for those 50 yr vs. 27% for those 70 yr; airr 0.57; 95% CI ), sex (13% for men vs. 34% for women; airr 0.40; 95% CI ), and prescriber (23% general practice vs. 44% rheumatology; airr 0.56; 95% CI ). Conclusions: Quality of GIOP preventive care has improved but remains suboptimal with only one quarter of those starting long-term glucocorticoids receiving BMD testing or osteoporosis treatment. Interventions to improve GIOP prevention, especially targeting younger patients, men, and nonspecialists, are needed. (J Clin Endocrinol Metab 97: , 2012) Long-term systemic glucocorticoid therapy is the most common cause of secondary osteoporosis, and up to 1% of adults are treated with glucocorticoids each year (1 5). Glucocorticoid-induced osteoporosis (GIOP) occurs rapidly (6 12% loss of bone mass within the first year), and the risk of fracture increases within 3 months of starting therapy (1 3). The first presentation of GIOP may be a fracture, and as many as 30 50% of patients exposed to glucocorticoids will eventually sustain a fracture (1 3). Furthermore, although sustained daily doses of mg prednisone (or equivalent) commonly lead to bone loss and fractures, doses as low as 2.5 mg prednisolone daily ISSN Print X ISSN Online Printed in U.S.A. Copyright 2012 by The Endocrine Society doi: /jc Received September 23, Accepted January 3, First Published Online February 1, 2012 Abbreviations: ADG, Aggregated diagnosis groups; airr, adjusted incidence rate ratio; BMD, bone mineral density; CI, confidence interval; DXA, dual-energy x-ray absorptiometry; GIOP, glucocorticoid-induced osteoporosis jcem.endojournals.org J Clin Endocrinol Metab, April 2012, 97(4):

2 J Clin Endocrinol Metab, April 2012, 97(4): jcem.endojournals.org 1237 have been reported to increase risk of vertebral and hip fractures (1, 3). Although GIOP is not entirely preventable, it is certainly ameliorable, and there are several relatively consistent guidelines with evidence-based recommendations that suggest bone mineral density (BMD) testing and prophylactic osteoporosis treatment in most if not all patients starting a course of daily 5 10 mg prednisone for 3 months or longer (1, 2). Previous studies that have examined the quality of guideline-concordant GIOP preventive care among patients starting a new course of long-term systemic glucocorticoids have reported a large care-gap with rates of BMD testing or osteoporosis treatment far lower than would be expected for such a common iatrogenic condition (5 13). Even randomized trials of interventions to improve GIOP care among repeat glucocorticoid prescribers have achieved testing or treatment rates of only 40 50% (14 17). Previous studies of the GIOP care gap have had limitations that include small sample sizes (6 9), inclusion of both incident and prevalent glucocorticoid users simultaneously (6 13), biased or nonrepresentative populations (6 9), older data (to our knowledge, the most recent published audit was in 2003) (5 13), and/or an inability to examine longer-term temporal trends (5 12). Therefore, we conducted a decade-long populationbased cohort study in the province of Manitoba, Canada, in all adults newly initiating long-term systemic glucocorticoid therapy. Our objectives were to 1) describe temporal trends in the quality of GIOP preventive care from and 2) determine the rates and correlates of high-quality GIOP preventive care defined as the composite of either a BMD test or new osteoporosis treatment within 6 months of a new long-term systemic glucocorticoid initiation. Materials and Methods Study design We undertook a population-based cohort study of all adults 20 yr of age and older in Manitoba, Canada, who newly initiated long-term systemic glucocorticoids between fiscal years 1998 and Excluded from these analyses were those living in nursing homes and related institutions and those who did not have continuous healthcare coverage for the fiscal year of interest. Manitoba has a population of 1.3 million residents with universal healthcare coverage that includes comprehensive healthcare services regardless of age or income. The Health Research Ethics Board of the University of Manitoba approved the study, and Manitoba Health Information Privacy Committee approved data access. Data sources The main data source for this study was the Manitoba Population Health Research Data Repository (18). This is a comprehensive collection of continuously updated and longitudinal population-based anonymized health services datasets housed at Manitoba Centre for Health Policy (18). Health data in the Repository include sociodemographic characteristics, vital statistics, physician claims, hospital separations, and prescription drugs (including drug name and date/strength/quantity dispensed). These data are well validated and have been used extensively in previous research (18 21). A de-identified unique personal health information number permits linkage to clinical registries such as the Manitoba Bone Density Program Database (19 21). This is a population-based registry of all BMD tests performed with central dual-energy x-ray absorptiometry (DXA) in the province since All clinical DXA tests conducted in Manitoba are performed within this single comprehensive provincial program, and the registry has been previously described and validated and extensively used in research (19 21). The goal of the provincial program is to have BMD tests completed within 6 wk of physician request, and currently most BMD tests are completed within 4 5 wk of request, suggesting that overall access to testing is quite good. However, all of the central DXA scanners are located in urban centers, and patients in rural regions do have to travel greater distances to obtain a BMD test, and this may limit their access. New initiations of long-term systemic glucocorticoids For these analyses, we were interested only in new initiations, and so any subject who had been dispensed systemic glucocorticoids in the previous year was excluded. This is analogous to a new-user design (22). By convention, 90 d or more of glucocorticoid therapy in a given fiscal year was considered long-term treatment, and subjects with shorter courses of treatment were excluded. We converted all systemic glucocorticoid dispensations to daily prednisone equivalents (e.g. 20 mg hydrocortisone 5 mg prednisone) and then averaged this over the course of therapy to obtain average daily dose in prednisone equivalents (12). We then categorized the daily dose of glucocorticoids dispensed as low (less than 5 mg prednisone daily), medium (5 10 mg daily), or high (greater than 10 mg daily) dose. High-quality GIOP preventive care The primary dependent variable of interest was receipt of high-quality GIOP preventive care after the initiation of systemic glucocorticoids. We, as have others (14, 17), conceptualized this as the composite endpoint of a BMD test or a dispensation for prescription osteoporosis medications. The latter included primarily oral bisphosphonates, but we also considered calcitonin, raloxifene, and teriparatide as appropriate medications to prevent GIOP. However, we did not consider estrogen or other hormone therapies as appropriate because, over the time course of our study, none of the available guidelines recommend hormone replacement for GIOP prevention (1, 2) and because it is not possible to know the specific indication for these treatments (e.g. flushes, other menopausal symptoms, other health benefits vs. GIOP prevention). Potential correlates of interest First, we explored temporal trends for outcomes across fiscal years. Because visual inspection of these data suggested nonlinearity for the temporal trends, we grouped years as follows: (reference period), , and

3 1238 Majumdar et al. Osteoporosis Management Among Glucocorticoid Users J Clin Endocrinol Metab, April 2012, 97(4): Then we examined sociodemographic characteristics including age in 10-yr bands, sex, location of residence (urban vs. rural), and household income [top two quintiles (high income) vs. lower three quintiles (low income) based on census dissemination areas] (21). We defined burden of comorbidity using the Johns Hopkins ACG(r) Case-Mix System (version 9) (23). The number of aggregated diagnosis groups (ADG) was compared as follows: zero to two ADG (reference) vs. three to five ADG vs. more than five ADG. Although we could not determine the specific indication for glucocorticoid therapy, we did identify the specialty of the physicians who prescribed glucocorticoids and classified them into discrete categories of general practitioners, rheumatologists, or nonrheumatologist internal medicine specialists. Analyses First, we plotted the receipt of high-quality GIOP preventive care (the composite endpoint of BMD testing or osteoporosis treatment within 6 months) over fiscal years. Then we examined variables of interest according to the presence or absence of highquality GIOP preventive care received. Preliminary analyses suggested the outcomes of interest were relatively rare and best considered as count data. Therefore, we conducted Poisson multivariable regression analyses and included all available covariates in our adjusted models: fiscal year period, age, sex, income quintile category, residence location, burden of comorbidity, glucocorticoid dose, and prescriber specialty. Because individual initiation episodes were not necessarily statistically independent across fiscal years (e.g. a patient dispensed glucocorticoids in 1998 might again be dispensed them in 2002 and again in 2004 and thus contribute three new initiations), we used generalized estimating equations to account for statistical clustering. The coefficients for these models are adjusted incidence rate ratio (airr) with 95% confidence intervals (CI). This output can be interpreted as a relative percent change or difference. We then repeated all analyses examining the individual endpoints of BMD testing and osteoporosis treatment. Last, we undertook some exploratory analyses. First, we compared older age ( 65 yr) women with younger men. Second, we examined quality of care in a high-risk subgroup: those with previous history of major osteoporotic fracture, defined as any fracture of the hip, vertebrae, humerus, or wrist. Many individuals in this subgroup would have already been eligible for BMD testing or osteoporosis treatment regardless of glucocorticoid initiation. Third, we redefined BMD test as receipt up to 2 yr after glucocorticoid initiation rather than within 6 months to help determine whether the reason rates of BMD testing were low might be related to difficult or limited access related to distance or wait times. Fourth, we recalculated rates of BMD testing and osteoporosis treatment after excluding patients younger than 50 yr of age (i.e. premenopausal women and younger men) under the assumption that physicians might consider this group at such a very low risk of fracture that no GIOP preventive care would be required. Analyses were conducted using SAS version 9.2 (SAS Institute, Cary, NC). Results From , there were 17,736 new long-term systemic glucocorticoid initiations among 15,825 unique subjects in the province of Manitoba. Overall, 71% of those dispensed glucocorticoids were 50 yr of age or older [mean age 60.0 (SD 17.8) yr], 58% were women, and 65% were urban residents. General practitioners prescribed most of the glucocorticoids, whereas rheumatologists prescribed the least (64% of all initiations vs. 9%). Most initiations were for medium-dose (36%) and high-dose (32%) courses, and mean duration of therapy was (SD 93.7) days (Table 1). Overall, within 6 months of initiating glucocorticoids, 6% of subjects had a BMD test, 22% were treated for osteoporosis, and 25% had received high-quality GIOP preventive care (i.e. the composite of BMD testing or osteoporosis treatment). Temporal trends in GIOP preventive care From , there were modest improvements in the quality of GIOP preventive care; BMD testing rates improved from 4 to 6%, osteoporosis treatment starts from 15 to 24%, and testing or treatment rates from 17 to 27% (Fig. 1). Between 1998 and 2008, there was a 51% adjusted relative increase in BMD testing or osteoporosis treatment within 6 months of initiation (airr 1.51; 95% CI ), although these improvements seemed to reach a plateau (Fig. 1 and Table 2). Correlates of high-quality GIOP preventive care: univariable analyses Variables stratified according to the receipt of BMD testing, osteoporosis treatment, and the composite endpoint representing high-quality GIOP preventive care are presented in Table 1. Descriptive analyses for sociodemographic characteristics revealed that those individuals who received BMD testing or osteoporosis treatment were older and more likely to be female, had higher incomes, and were more likely to reside in urban locations compared with those who did not receive high-quality GIOP preventive care. In terms of clinical characteristics, those prescribed higher doses of glucocorticoids and those cared for by rheumatologists were more likely to receive highquality GIOP preventive care than were those who received lower doses or who were cared for by nonrheumatologists. Findings were nearly identical when BMD testing alone and osteoporosis treatment alone were examined separately (Table 1). Multivariable analyses In analyses adjusted for all covariates presented in Table 1, older age, female sex, and rheumatologist prescriber were independently associated with higher incidence rates of BMD testing or osteoporosis treatment (Table 2). Conversely, patients younger than 50 yr of age (compared with those 70 yr of age and older; airr 0.57; 95% CI

4 J Clin Endocrinol Metab, April 2012, 97(4): jcem.endojournals.org 1239 TABLE 1. Characteristics of 17,736 new systemic glucocorticoid initiations ), men (compared with women; airr 0.40; 95% CI ), and initiation of glucocorticoids by a general practitioner (vs. a rheumatologist; airr 0.56; 95% CI ) were each independently associated with lower rates of BMD testing or osteoporosis treatment. Findings were similar when the composite endpoint was disaggregated and analyses considered BMD testing and osteoporosis treatment as separate endpoints (Table 2). FIG. 1. Temporal trends in osteoporosis (OP) management among new initiations of long-term systemic glucocorticoids ( ). BMD testing (per 100 person-years) Osteoporosis treatment (per 100 person-years) Tested or treated (quality GIOP care) (per 100 person-years) (n 4981) (n 6345) (n 6410) Sociodemographics Age 50 yr (n 5065) yr (n 2875) yr(n 3361) yr (n 6435) Sex Male (n 7415) Female (n 10,321) Income Low income quintiles (n 7437) High income (n 10,035) Residence Rural (n 6273) Urban (n 11,463) Clinical characteristics Burden of comorbidity 2 ADG (n 1764) ADG (n 5937) ADG (n 10,035) Glucocorticoid dose Low ( 5 mg) (n 6296) Medium (5 10 mg) (n 6232) High ( 10 mg) (n 5208) Prescriber specialty General practice (n 11,347) Nonrheumatologist internists (n 4860) Rheumatologist (n 1529) Exploratory analyses First, we verified that older women received better GIOP preventive care than younger men: rates of testing or treatment of 37% over the entire study compared with rates of only 13% among younger men (P 0.001; Fig. 2). Second, among the individuals with previous fracture, rates of osteoporosis testing or treatment were 32% compared with rates of 25% among those who did not have previous fracture (P 0.001; Fig. 2). Third, redefining the time horizon for BMD testing from 6 months to up to 2 yr after initiation of glucocorticoids provided somewhat more encouraging results; BMD test rates increased from 6 to 16% with the alternate time window, whereas the rates of testing or treatment increased from 25 to 32% with the alternate time window. Fourth, excluding the 5065 premenopausal women and younger men from analyses yielded rates of GIOP preventive care that were very similar to that seen when studying the entire population at risk: BMD testing within 6 months of glucocorticoid initiation of 6% (vs. 6% in main analysis), osteoporosis treatment of 26% (vs. 22% in the main analysis), and

5 1240 Majumdar et al. Osteoporosis Management Among Glucocorticoid Users J Clin Endocrinol Metab, April 2012, 97(4): TABLE 2. Independent correlates of high-quality GIOP preventive care: Multivariable Poisson regression analyses testing or treatment rates of 29% (vs. 25% in the main analysis). Discussion BMD testing [airr (95% CI)] In a large population-based cohort we found that the quality of GIOP preventive care improved 51% over the decade from , although our study also suggests FIG. 2. Associations between quality of glucocorticoid preventive care and age, sex, and previous fracture history: exploratory analyses. Osteoporosis treatment [airr (95% CI)] Tested or treated (quality GIOP care) [airr (95% CI)] (reference) ( ) 1.56 ( ) 1.57 ( ) ( ) 1.50 ( ) 1.51 ( ) Age 50 yr (reference) yr 1.45 ( ) 2.15 ( ) 1.88 ( ) yr 1.33 ( ) 2.15 ( ) 1.83 ( ) 70 yr 0.91 ( ) 2.13 ( ) 1.75 ( ) Sex Male (reference) Female 2.04 ( ) 2.78 ( ) 2.51 ( ) Income Low income (reference) High income quintiles 1.27 ( ) 1.06 ( ) 1.09 ( ) Residence Rural (reference) Urban residence 1.39 ( ) 0.96 ( ) 1.00 ( ) Burden of comorbidity 2 ADG (reference) ADG 1.27 ( ) 1.25 ( ) 1.23 ( ) 5 ADG 1.35 ( ) 1.50 ( ) 1.46 ( ) Glucocorticoid dose Low ( 5 mg) (reference) Medium (5 10 mg) 1.35 ( ) 1.15 ( ) 1.17 ( ) High ( 10 mg) 1.31 ( ) 1.36 ( ) 1.32 ( ) Prescriber specialty General practice (reference) Nonrheumatologist internists 1.78 ( ) 0.99 ( ) 1.12 ( ) Rheumatologist 2.31 ( ) 1.81 ( ) 1.79 ( ) that these improvements have reached a plateau. Indeed, although there is not a statistically significant difference in osteoporosis treatment rates from when compared with rates in , visually, our results might even suggest the possibility that postinitiation treatment rates may be decreasing. It is noteworthy that this same phenomenon was also recently observed in the setting of postfracture osteoporosis care, suggesting that this may be part of an as yet unexplained secular trend (24). More concerning, in absolute terms, GIOP prevention remains suboptimal because only 25% of those starting long-term glucocorticoids received a BMD test or were treated for osteoporosis within 6 months. Several sociodemographic and clinical characteristics were independently associated with lower rates of BMD testing or osteoporosis treatment, most notably, younger age, male sex, and initiation by general practitioners. Several studies examining quality of GIOP preventive care have been published in the past, and all previous studies have reported a major care gap between guideline recommendations and actual clinical practice (5 13). These previous studies have varied in size (from ,000), populations studied (specialist clinics, hospitalized patients, Medicare recipients, postmenopausal women only,

6 J Clin Endocrinol Metab, April 2012, 97(4): jcem.endojournals.org 1241 managed care populations), outcomes reported (BMD testing, calcium and vitamin D, prescription osteoporosis medication, composite of BMD testing or osteoporosis treatment, or any intervention), location (Canada, United States, or United Kingdom) and vintage ( ) (5 13). Regardless of these differences, the results have been remarkably similar, fairly stable over time, and mirror our findings; BMD testing and osteoporosis treatment after initiation of long-term glucocorticoids are underused particularly among younger patients and men. This is noteworthy given that most studies in the quality of care literature for other conditions demonstrate that it is usually older patients and women that are at risk of receiving suboptimal care (25). In previous studies, the rates of BMD testing after glucocorticoid initiation have ranged from a low of 0% (in young men) to a high of 64% (in older women cared for by rheumatologists) (5 13); in our study, overall rates of BMD testing were only 6% with little difference between young men and older women. Underuse of BMD testing was present across time and in all ages and both sexes, and because most BMD tests that are requested in the province are completed within 6 wk, it suggests this is not about limited access, except perhaps in the case of rural residents. Rates of BMD testing after glucocorticoid initiation were 7% among urban individuals compared with 4% among rural individuals, and this small difference was statistically significant in adjusted analyses (airr 1.39; 95% CI ). However, overall quality of GIOP preventive care was no different in urban residents when compared with rural residents (airr 1.00; 95% CI ). In previous studies, the rates of prescription osteoporosis treatment after glucocorticoid initiation have ranged from a low of 2% (in young men) to a high of 84% (in hospitalized postmenopausal women) (5 13). A major limitation of most of these studies is overestimating rates of GIOP prevention in postmenopausal women because osteoporosis treatment is already used to a greater degree in this subgroup, and so these GIOP prevention-related activities may in fact be unrelated to initiating glucocorticoids. Our new-user design minimizes (22), but does not eliminate, this bias by considering only new initiations and excluding anyone who had received glucocorticoids in the previous year. In addition to the new-user design, the major strengths of our study are the recency and duration of the data and the large population-based sample of community-dwelling men and women. However, there are also several limitations. First, although we know the specialty of the prescribers, we have no information on the actual indication for glucocorticoid therapy. Second, we do not have any information on other measures that might be used to prevent GIOP, such as weight-bearing exercises or over-thecounter calcium and vitamin D supplements. Third, we have no data with respect to primary nonadherence, i.e. prescriptions written but not filled. Fourth, we do not know the results of BMD tests and assumed that a BMD test without osteoporosis treatment reflected high quality of GIOP preventive care. Fifth, we cannot calculate absolute fracture risk using FRAX or other related methods because these data are available only for those individuals who had a BMD test, representing only 6% of those we studied. But even among those individuals initiating glucocorticoids who had a previous history of major fracture, the rates of GIOP preventive care were only 32 vs. 25% for the entire population. Sixth, because of the study design, we did not follow individuals over time and capture total glucocorticoid exposure or remote use of BMD tests or osteoporosis treatments or identify adverse clinical events related to potentially preventable fractures. Last, our cohort was drawn from one Canadian province where all subjects had universal healthcare and drug coverage; our findings may not necessarily generalize well to other settings or healthcare systems. In conclusion, although quality of GIOP care has improved slowly over time, rates of improvement have leveled off. About three quarters of those who initiate glucocorticoids still do not receive a BMD test or osteoporosis treatment within 6 months, and this has both important clinical consequences as well as medico-legal ramifications. Younger patients and men are less likely to receive high-quality GIOP preventive care, although nonspecialists are less likely to deliver it. Interventions to improve GIOP prevention need to be developed and tested in controlled trials, and these interventions might consider targeting young patients, men, and general practitioners. Acknowledgments We acknowledge the Manitoba Centre for Health Policy for use of data contained in the Population Health Research Data Repository (HIPC project 2008/ ). The results and conclusions are those of the authors, and no official endorsement by the Manitoba Centre for Health Policy, Manitoba Health, or other data providers is intended or should be inferred. This article has been reviewed and approved by the members of the Manitoba Bone Density Program Committee. Address all correspondence and requests for reprints to: Dr. Sumit R. Majumdar, Professor, Department of Medicine, University of Alberta, 2F1.24 Walter Mackenzie Health Sciences Centre, University of Alberta Hospital, th Street, Edmonton, Alberta, Canada T6G 2B7. majumdar@ualberta.ca.

7 1242 Majumdar et al. Osteoporosis Management Among Glucocorticoid Users J Clin Endocrinol Metab, April 2012, 97(4): This work was supported by an unrestricted research grant from Amgen Canada Ltd. The funding source had no access to the data before publication, no input into the writing of the manuscript, and no input into the decision to publish the results. S.R.M. holds the Endowed Chair in Patient Health Management from the Faculties of Medicine and Dentistry and Pharmacy and Pharmaceutical Sciences (University of Alberta) and receives salary support from the Alberta Heritage Foundation for Medical Research AIHS (Health Scholar). L.M.L. receives salary support from the Canadian Institutes of Health Research (New Investigator) and the University of Saskatchewan Centennial Chair Program. S.N.M. is chercheur-clinicien boursier des Fonds de la Recherche en Sante du Quebec. Disclosure Summary: The authors have nothing to disclose. References 1. Weinstein RS 2011 Glucocorticoid induced bone disease. N Engl J Med 365: Hansen KE, Wilson HA, Zapalowski C, Fink HA, Minisola S, Adler RA 2011 Uncertainties in the prevention and treatment of glucocorticoid induced osteoporosis. J Bone Miner Res 26: Kanis JA, Johansson H, Oden A, Johnell O, de Laet C, Melton III LJ, Tenenhouse A, Reeve J, Silman AJ, Pols HA, Eisman JA, McCloskey EV, Mellstrom D 2004 A meta-analysis of prior corticosteroid use and fracture risk. J Bone Miner Res 19: Yood RA 2006 Prevention of glucocorticoid induced osteoporosis: why are we doing so poorly? J Rheum 33: Saag KG, Gehlbach SH, Curtis JR, Youket TE, Worley K, Lange JL 2006 Trends in prevention of glucocorticoid induced osteoporosis. J Rheum 33: Walsh LJ, Wong CA, Pringle M, Tattersfield AE 1996 Use of oral corticosteroids in the community and the prevention of secondary osteoporosis: a cross sectional study. BMJ 313: Peat ID, Healy S, Reid DM, Ralston SH 1995 Steroid induced osteoporosis: an opportunity for prevention? Ann Rheum Dis 54: Aagaard EM, Lin P, Modin GW, Lane NE 1999 Prevention of glucocorticoid induced osteoporosis: practice at an urban county hospital. Am J Med 107: Yood RA, Harrold LR, Fish L, Cernieux J, Emani S, Conboy E, Gurwitz JH 2001 Prevention of glucocorticoid induced osteoporosis: experience in a managed care setting. Arch Intern Med 161: Ettinger B, Chidambaran P, Pressman A 2001 Prevalence and determinants of osteoporosis drug prescription among patients with high exposure to glucocorticoid drugs. Am J Manag Care 7: Feldstein AC, Elmer PJ, Nichols GA, Herson M 2005 Practice patterns in patients at risk for glucocorticoid induced osteoporosis. Osteoporos Int 16: Che M, Ettinger B, Nguyen MT, Pressman AR, Johnston J 2006 High dose corticosteroid exposure and osteoporosis intervention in adults. Ann Allergy Asthma Immunol 97: Curtis JR, Westfall AO, Allison JJ, Becker A, Casebeer L, Freeman A, Spettell CM, Weissman NW, Wilke S, Saag KG 2005 Longitudinal patterns in the prevention of osteoporosis in glucocorticoid treated patients. Arthritis Rheum 52: Solomon DH, Katz JN, La Tourette AM, Coblyn JS 2004 Multifaceted intervention to improve rheumatologists management of glucocorticoid induced osteoporosis: a randomized controlled trial. Arthritis Rheum 51: Naunton M, Peterson GM, Jones G, Griffin GM, Bleasel MD 2004 Multifaceted educational program increases prescribing of preventive medication for corticosteroid induced osteoporosis. J Rheumatol 31: Chitre MM, Hayes W year results of a member and physician intervention to reduce risk associated with glucocorticoid induced osteoporosis in a health plan. J Manag Care Pharm 14: Curtis JR, Westfall AO, Allison J, Becker A, Melton ME, Freeman A, Kiefe CI, MacArthur M, Ockershausen T, Stewart E, Weissman N, Saag KG 2007 Challenges in improving the quality of osteoporosis care for long term glucocorticoid users: a prospective randomized trial. Arch Intern Med 167: Roos NP, Shapiro E 1999 Revisiting the Manitoba Centre for health policy and evaluation and its population-based health information system. Med Care 37:JS10 JS Leslie WD, Tsang JF, Caetano PA, Lix LM 2007 Effectiveness of bone density measurement for predicting osteoporotic fractures in clinical practice. J Clin Endocrinol Metab 92: Leslie WD, Caetano PA, Macwilliam LR, Finlayson GS 2005 Construction and validation of a population-based bone densitometry database. J Clin Densitom 8: Leslie WD, Lix LM, Prior HJ, Derksen S, Metge C, O Neil J 2007 Biphasic fracture risk in diabetes: a population-based study. Bone 40: Ray WA 2003 Evaluating medication effects outside of clinical trials: new user designs. Am J Epidemiol 158: Smith NS, Weiner JP 1994 Applying population-based case mix adjustment in managed care: the Johns Hopkins Ambulatory Care Groups. Manag Care Q 2: Leslie WD, Giangregorio LM, Yogendran M, Azimaee M, Morin S, Metge C, Caetano P, Lix LM 8 April 2011 A population-based analysis of the post-fracture care gap : the situation is not improving. Osteoporos Int /s McAlister FA, Oreopoulos A, Norris CM, Graham MM, Tsuyuki RT, Knudtson M, Ghali WA 2007 Exploring the treatment-risk paradox in coronary disease. Arch Intern Med 167: