|
|
Vol. 54, No. 9, September 2008, pp.1270 - 1276 Copyright © 2008 by The College of Family Physicians of Canada
Postfracture care for older womenGaps between optimal care and actual careColleen J. Metge, PhDAssociate Professor in the Faculty of Pharmacy and a Research Associate at the Manitoba Centre for Health Policy in the Department of Community Health Sciences in the Faculty of Medicine at the University of Manitoba in Winnipeg.
William D. Leslie, MD MSc FRCPC
Lori-Jean Manness
Marina Yogendran, MSc
C.K. Yuen, MD FRCSC FACOG FSOGC
Brent Kvern, MD CCFP FCFP for the Maximizing Osteoporosis Management in Manitoba Steering Committee
Correspondence to: Dr C.J. Metge, Faculty of Pharmacy, 50 Sifton Rd, University of Manitoba, Winnipeg, MB R3T 2N2; telephone 204 474-8407; fax 204 474-7617; e-mail c_metge{at}umanitoba.ca There is a serious gap in the care of patients with fractures characteristic of osteoporosis. Osteoporosis is increasingly being recognized as an important public health problem because of its age-related increase in prevalence and the morbidity, mortality, and economic consequences associated with it. Yet many family physicians appear to be unaware of the magnitude of this problem and the importance of identifying people at high risk for appropriate intervention, and of the process of diagnosis and management of the disease.1,2 Researchers currently estimate that 30% to 50% of women will experience fractures characteristic of osteoporosis during their lives.3 Womens lifetime risk of hip fractures is greater than the sum of their lifetime risk of having breast, endometrial, or ovarian cancer.4 The rate of premature death (at younger than 75 years) and substantially increased morbidity among patients with osteoporosis make it a particularly compelling public health problem. Women who have sustained major osteoporotic fractures have a 2-fold increase in ageadjusted risk of mortality.3 Hip fractures are the cause of up to 40% of fall-related hospitalizations among those 65 years old and older,5 and 40% of all nursing home admissions occur as a result of fractures among people older than 65 years.6 These morbidity and mortality rates are especially distressing given that effective prevention and treatment strategies are available for those at highest risk of osteoporotic fractures—that is, people needing secondary prevention because they have already experienced spine, hip, or other fractures characteristic of osteoporosis. Appropriate intervention can be very effective. Pharmacologic therapy has been shown to reduce risk of fracture by 30% to 60% in women at high risk. Additional nonpharmacologic intervention with calcium and vitamin D supplementation,7 exercise,8 smoking cessation, and fall prevention9 can further contribute to preventing fractures. Serious gaps between what could be done for postfracture patients and what is done in actual practice have been observed. A recent meta-analysis of 37 studies of diagnosis and treatment of osteoporosis and intervention for those who have sustained fragility fractures revealed that, in some studies, none of the fracture patients was investigated or treated for underlying osteoporosis.1 Studies have demonstrated that all patients, including those older than 75 years, can benefit from treatment, and yet older women have been least likely to receive bone mineral density (BMD) testing or appropriate treatment for osteoporosis.10,11 This gap in the care of fracture patients should be less evident in a publicly funded health care system such as Canadas. Bone mineral density testing as a medically necessary diagnostic procedure does not require payment from patients, and drug benefit programs are available to most Canadian residents 65 years old and older. As a prelude to developing interventions to improve diagnosis and treatment of osteoporosis, we examined the rates of investigation and treatment of osteoporosis among older women in an entire Canadian province during the first year after they had had fractures. METHODS We conducted a repeated historical cohort study from April 1, 1997, to March 31, 2002. Women in our cohort were 50 years old or older as of April 1st for each panel year, were residents of Manitoba, and had experienced fractures during that year. Patients who died, left the province, or moved into the province during the study period were excluded from the analysis to eliminate those with partial data. Manitoba has developed a system for building longitudinal files of individual patients use of health care services. Links between hospital, physician, and pharmacy databases and clinic-based data are possible through unique but anonymous identifiers. Computerized provincial government health databases capture claims for physician services, hospitalizations, and pharmaceutical dispensings for each person in the system. Databases include information on patients identities, dates of services, services provided, drugs dispensed, and diagnoses classified under the World Health Organizations International Classification of Diseases, 9th revision, Clinical Modification (ICD-9-CM) codes.12 All clinical bone densitometry in Manitoba is done under a single program that maintains uniform testing indications, requisitions, and reporting. Criteria for testing are broadly consistent with most published guidelines and emphasize the importance of female sex, older age, previous fragility fractures, and other clinical risk factors. The Manitoba Bone Density Program maintains a population-based database that includes all dual-energy x-ray absorptiometry results. It is more than 99% complete and accurate as judged by chart audit.13 Each womans annual medical record between April 1, 1997, and March 31, 2002, was assessed for the presence of any ICD-9-CM fracture codes (ICD-9-CM 805, 807–829). Vertebral fractures without cord injury (ICD-9-CM code 805) and hip fractures (ICD-9-CM code 820) have been consistently associated with osteoporosis and were analyzed as a specific fracture category designated type 1 fractures. All other fractures (ie, not hip or spine) were designated as type 2 fractures. We examined postfracture care during the first 12 months after type 1 or type 2 fractures. The availability of a province-wide BMD testing database allowed us to identify when BMD scans were done. Use of pharmacologic therapy postfracture was determined by examining dispensing of recognized osteoporosis drugs available from the provinces Drug Programs Information Network, which captures prescriptions dispensed for outpatient use.14 Dispensing of at least 1 prescription for a bisphosphonate or selective estrogen receptor modulator for all age groups and hormone replacement therapy (HRT) for women aged 65 or older was included in our analysis. During the years of our study, HRT use by women older than 65 years with fractures was taken to be evidence of its use for prevention of osteoporosis. This is consistent with the general view that HRT was a first-line agent for prevention and treatment of osteoporosis among postmenopausal women that was held until after publication of the Womens Health Initiative study in July 2002.15
The denominators for calculating yearly fracture rates were based on the population of women resident in Manitoba for the complete year of the panel and are reported per 1000 women. Rates of BMD assessment and pharmacologic treatment for osteoporosis were calculated based on the proportion of women who had 1 of these interventions within 12 months of fracture and are reported as percentages. Analyses are stratified by type of fracture (type 1 and type 2) and age group (50 to 64 years, 65 to 74 years, and 75 years and older). Rates are reported for each panel year and are compared using The study was approved by the Health Research Ethics Board at the University of Manitoba and by Manitobas Health Information Privacy Committee. RESULTS During the first year (April 1, 1997, to March 31, 1998), there were 162 009 women 50 years old or older in the province of Manitoba, and by the final year (April 1, 2001, to March 31, 2002), there were 175 072 women 50 years old or older (Table 1). Annualized fracture rates were stable during this time (9.6 to 10.5 per 1000 women for type 1 fractures, and 26.5 to 28.6 per 1000 women for type 2 fractures). As expected, fracture rates were strongly related to age and were much higher among women 75 years old and older (Figure 1). Within each age group, fracture rates were stable over the 5 years of the study.
Figure 2 shows the proportion of women with type 1 fractures (hip or spine) or type 2 fractures (not hip or spine) who received either BMD assessment or pharmacologic treatment after the fractures. For all ages combined, BMD assessment after type 1 fractures rose from 2.6% in 1997–1998 to 4.6% in 2001–2002 (P for trend .0004). During the same period, pharmacologic treatment rates increased from 4.9% in 1997–1998 to 17.6% in 2001–2002 (P for trend < .0001). When BMD assessments and pharmacologic treatments were considered together, 6.9% of women with type 1 fractures received these interventions within 1 year of having fractures in 1997–1998, and 20.5% received them in 2001–2002. Women 50 to 64 years old had significantly higher rates of intervention after type 1 fractures (26.4%) than women 75 years old and older did (17.9%, P <.0001).
The rate of BMD assessment among women with type 2 fractures increased over the study period (P for trend <.0001) and reached a high of 7.0% in 2000–2001. The proportion of women in this group who received pharmacologic treatment showed a similar pattern (P for trend <.0001) with the highest rate (15.5%) in 2001–2002. As with women with type 1 fractures, the rate of intervention among women with type 2 fractures was substantially higher among women 50 to 64 years old (25.5%) than among those 75 years old or older (16.8%, P <.0001). DISCUSSION This population-based analysis found large gaps in the assessment and treatment of women older than 50 years with fractures characteristic of established osteoporosis (type 1 fractures) or of increased risk of osteoporosis (type 2 fractures). In contrast to other studies that have relied on chart audits, patient reports, or selected patient recruitment, the population-based data set upon which this analysis was based was free from selection bias at both physician and patient levels. Although some improvement was observed in clinical management of osteoporosis over the 5 years of observation, even in the projects final year, 4 out of 5 women received no pharmacologic treatment within the year following hip or vertebral fractures, and fewer than 1 in 10 underwent BMD assessment. Similar gaps were noted in those at increased risk of osteoporosis based on having had type 2 fractures. The very large gap between optimal care and actual care identified in Manitoba is comparable to that reported by others.16–18 Canadas recent efforts to define, collect, and report data on chronic diseases uniformly19 could reveal gaps similar to those found here. It might be, however, that family physicians need to be convinced that these women need treatment or, at least, that they as physicians need to reorient their practices to improve delivery of chronic care. The population-based nature of the study effectively allowed us to document the collective practice patterns of approximately 1000 physicians in Manitoba and gave us critical baseline data for design and implementation of strategies to improve diagnosis and treatment of osteoporosis. Interventions based on data demonstrating need could address the following points. The occurrence of fractures in postmenopausal women, one of the entry criteria for this study, has been shown to be one of the strongest independent risk factors for future fractures.20 Women who have sustained vertebral fractures have a 4-fold increased risk of future vertebral fractures, and the risk escalates with the number of previous fractures. There is compelling evidence that timely treatment of these patients using approved antiresorptive therapies can have a substantial clinical effect and reduce future risk of fractures by 40% to 60%.21 Our findings show also that the overall rate of assessment and treatment of women in the 50- to 64-year-old age group is significantly higher than in the 2 older groups, demonstrating a tendency to focus attention on younger women who are actually at lowest risk. In reality, this understates the discrepancy, as use of HRT was counted for the older women but not for the women 50 to 64 years old. Older women sustain the largest number of fractures and thus are at greatest risk, yet they receive the least clinical intervention. This observation reflects a potential inequity in the treatment of women older than 65 years that has been observed for other chronic conditions also.22 Women aged 70 have an average life expectancy of 12 more years and can benefit from appropriate treatment that has been shown to reduce fracture rates within 1 year of initiation.23
Limitations
Future research
Conclusion
Acknowledgments This study was undertaken in partnership with the University of Manitoba, the Government of Manitoba, and Merck Frosst Canada Ltd. Although the study was wholly funded by Merck Frosst Canada Ltd, the partnership maintained control over the concept, design, implementation, analysis, and authorship of the study. The authors are indebted to Marilyn Krelenbaum for her assistance in preparing the manuscript and to Health Information Services at Manitoba Health for providing data to the University of Manitoba. The results and conclusions of this study are those of the authors; no official endorsement by Manitoba Health is intended or should be inferred. Footnotes All of the authors made the 3 types of contributions required by the International Committee of Medical Journal Editors. They contributed substantially to concept and design of the study, or acquisition of data, or analysis and interpretation of data; they either drafted the article or revised it critically for important intellectual content; and they gave final approval to the version to be published. The Maximizing Osteoporosis Management in Manitoba project is a public-private partnership between the University of Manitoba, Manitoba Health, and Merck Frosst Canada Ltd. The project was funded through an unrestricted grant from the Patient Health Management Department of Merck Frosst Canada Ltd. Dr Metge has received speaking and consulting fees from and Ms Manness is an employee of Merck Frosst Canada Ltd. This article has been peer reviewed. Cet article a fait lobjet dune révision par des pairs. References
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||