Dr Eisman has served as a consultant and received research support from Aventis, Eli Lilly and Company, Merck, Sharp & Dohme, Novartis, NPS Pharmaceuticals, Organon, Roche, and Servier. Dr Clapham is an employee of Merck, Sharp, & Dohme Australia and owns stock in Merck & Co., Inc. Dr Kehoe served as a consultant for Merck, Sharp, & Dohme.
Osteoporosis Prevalence and Levels of Treatment in Primary Care: The Australian BoneCare Study†
Article first published online: 13 SEP 2004
Copyright © 2004 ASBMR
Journal of Bone and Mineral Research
Volume 19, Issue 12, pages 1969–1975, December 2004
How to Cite
Eisman, J., Clapham, S. and Kehoe, L. (2004), Osteoporosis Prevalence and Levels of Treatment in Primary Care: The Australian BoneCare Study. J Bone Miner Res, 19: 1969–1975. doi: 10.1359/jbmr.040905
- Issue published online: 2 DEC 2009
- Article first published online: 13 SEP 2004
- Manuscript Accepted: 23 JUL 2004
- Manuscript Revised: 1 JUL 2004
- Manuscript Received: 16 FEB 2004
- primary care
The level of recognition and treatment of osteoporosis is not well characterized in primary care. In data from a large sample of postmenopausal women attending 927 primary care physicians, 29% of women reported one or more fractures after menopause. The great majority (72%) were not on any osteoporosis-specific therapy.
Introduction: Osteoporosis is often first recognized at the time of a low-trauma fracture. However, by this stage, the risk of subsequent fractures has already risen substantially. Moreover, in many countries, only a small proportion of women, who have already sustained fractures, receive a treatment shown to reduce this increased risk of further fractures.
Materials and Methods: This project was initiated to examine the prevalence of osteoporotic fractures, risk factors for osteoporosis, and use of antifracture therapy among postmenopausal Australian women. More than 88,000 women from 927 primary care physicians returned over 69,358 surveys. Of these, 57,088 reported the presence of a postmenopausal fracture or risk factors.
Results: Among these randomly selected postmenopausal women, 29% reported having had one or more low-trauma fractures after menopause (44% substantiated in current records). One-third of these women reported multiple fractures. The prevalence of all types of fractures, except rib and ankle, increased with age and low body weight. Those who reported fractures were also more likely to report early menopause, corticosteroid use, and a family history of osteoporosis. Moreover, those with vertebral fractures were more likely to record height loss, kyphosis, and back pain. Physical inactivity, low calcium intake, and smoking had no consistent relationship with any fracture outcome. Of the women who reported a fracture after menopause, only 28% were on any specific therapy for osteoporosis, and 7% were on calcium alone. Of those who had been told they had osteoporosis by a doctor, 40% were receiving specific osteoporosis therapy.
Conclusions: In this large study of postmenopausal Australian women attending primary care physicians, 29% reported at least one low-trauma fracture after menopause. Less than one-third of these women were on specific treatment for osteoporosis, and only 40% were ever told they had osteoporosis. Therefore, osteoporotic fractures are common in postmenopausal Australian women, and few, despite their substantially increased risk of further fractures, are on any specific anti-osteoporotic therapy. These data support the need for more effective education for the community and medical practitioners of the clinical significance of osteoporotic fractures and alternatives for treatment.
Osteoporosis is a major health care problem worldwide.(1–8) However, individuals who have already sustained fractures and thus are at considerably higher risk of further fractures may not be receiving treatment to reduce their risk of further fractures. Osteoporosis affects postmenopausal women and older men, with major societal costs because of morbidity and mortality related to these low-trauma fractures.(9) Recent controlled-trial studies of various medications able to prevent bone loss in postmenopausal women have shown clinically and statistically significant reductions in fracture events.(3,10–20) However, there is limited information on the extent to which postmenopausal women at risk of osteoporosis are receiving effective treatment to prevent the progression of the disease, even after a low-trauma event has occurred. In fact, in some preliminary U.S. studies, women at greatest risk of fracture seem only slightly more likely to receive treatment.(21) This is despite the existence of clinical guidelines regarding the treatment and management of women with osteoporosis, with or without previous fractures.(22–25)
This study was designed to examine the following: (1) the prevalence of low-trauma fractures and early signs of osteoporosis (i.e., kyphosis and/or height loss and back pain) among Australian women ≥60 years of age attending their primary care physician, and (2) the level of use of specific anti-osteoporosis therapy among these women.
This study collected data from a large random sample of women ≥60 years of age from primary care practices throughout Australia. The primary care setting is a key feature of this large study, because these physicians are at the front line for diagnosis and treatment of osteoporosis.
MATERIALS AND METHODS
Full-time general practitioners from across Australia were invited to participate in a program, BoneCare, about the diagnosis and treatment of osteoporosis. The 5% (927) of doctors that accepted, 72% trained within Australia, were generally representative of the 19,500 full-time practitioners in Australia. The program and patient data collection, developed by Merck Sharp & Dohme Australia, was accredited and approved by the Royal Australian College of General Practitioners. Recruitment was through the calendar year 1999. Each doctor was requested to select 100 female patients ≥60 years of age, based on their medical records, with surnames beginning with “A” and progressing through the alphabet until 100 were obtained. For many doctors, this reflected the entirety of elderly women in their practice. There were no inclusion or exclusion criteria other than age and gender. Patient ethnicity was not a criterion and, although the survey was available in five major languages, English, Chinese, Arabic, Vietnamese, Italian, and Greek, ethnicity was not recorded. Patients 59 years of age were allowed because of uncertainty around age 60; however, 790 (1.4%) with ages recorded as <59 years were not further analyzed.
Patient survey procedure
All selected patients were sent a cover letter describing the program, a consent form requesting participation and permission to access their medical records, and a one-page questionnaire by their general practitioner. The questionnaire covered the patient's birth date, height, weight, postmenopausal fractures, BMD tests, height loss (estimate from early adulthood), kyphosis, back pain, and a series of history and lifestyle factors relevant to osteoporosis. Participants returned completed surveys to an independent research group (Black Group Pty Ltd., Sydney, New South Wales, Australia) that managed data coding and entry. Research nurses coded all survey information, but entered only data in the database where patients reported symptoms of and/or risk factors for osteoporosis, specifically, postmenopausal low-trauma fractures, the presence of kyphosis (i.e., having noticed their back changing to become more rounded), height loss in centimeters since early adulthood, back pain, a family history of osteoporosis, onset of menopause before the age of 45, use of corticosteroids, low intake of calcium products (0–2 servings of dairy products/day), low level of exercise per week (0–2 times/week), and/or current/previous smoker. Fractures were coded as low trauma based on information provided by the medical records and by patients on how the injury happened. X-ray reports confirming vertebral height loss were considered evidence of vertebral fractures, A higher number of women than expected coded as “other fracture” had reported fractures at nonclassical osteoporotic fracture sites (i.e., not vertebra, hip, wrist, ankle, or rib). For all “fracture” patients, medical records were examined, where possible, at the doctor's office to substantiate patient reports of fracture. In many cases, verification of fracture was hindered by absence of record of X-ray in the current doctor's files. This may reflect the high mobility of the Australian population, their tendency to have more than one local doctor, and changing local doctors in relation to change of place of residence or work.
Anthropometric data for ∼18% of women who did not report any of the above symptoms or risk factors were not entered into the database (Fig. 1). For some women (166, 0.3%), extreme values (i.e., height <100 cm or >199 cm and weight <21 kg) were recorded; these values were set to “missing.” A further 7279 (13%) did not provide height, and 5267 (9%) did not provide weight.
Comparisons were assessed using ANOVA for continuous variables (e.g., age, height, weight, total number of fractures) and Pearson χ2 statistic for categorical variables (e.g., presence or absence of kyphosis, back pain, smoking, family history of osteoporosis) with p set at 0.01. Logistic regression and the Wald statistic were used to examine factors associated with fracture, height loss in centimeters (<4 or ≥4 cm), and BMD testing. All variables were initially entered and then removed using a stepwise deletion with p for removal set at 0.05.
Of the 88,041 osteoporosis surveys sent, 69,358 (78.8%) valid responses were returned (Fig. 1). A small percentage of additional surveys (0.8%, 690) were returned, but excluded from analysis because respondents were below the age criterion. Data for 57,088 patients who reported either symptoms of and/or risk factors for osteoporosis were available in the database for analysis. These patients were 59–105 years of age (mean, 71.8 ± 7.6 years), were 159.2 ± 7.4 cm in height, and weighed 67.3 ± 14.1 kg. Symptoms and risk factors are shown in Table 1.
Among these postmenopausal women, 20,248 reported fractures representing 29% of the total group. Of these women, 66% reported 1 fracture, 22% reported 2 fractures, and 12% reported 3–14 fractures. Of all reported fracture types, 38–67% were substantiated in the doctors' notes or in a X-ray report. Vertebral fractures were more likely to be substantiated (67%) than other fractures (hip, 48%; wrist, 38%; ankle, 41%; rib, 45%; any other low-trauma fracture, 44%). Overall, 34% of patients reported multiple fractures at the same and/or at additional sites (Table 2). Patients with rib and hip fractures were more likely to have reported additional fractures.
The prevalence of hip, vertebral, wrist, and “other” fractures, but not ankle or rib fractures, increased with lower weight and with advancing age (Fig. 2). Most fractures, including hip fractures, were reported by women <80 years of age, and the peak was in the 70- to 80-year range.
Symptoms and risk factors
Even women who did not report spine or other fractures commonly noted back pain, height loss, or kyphosis; only 36% denied any such symptoms. Women who reported fractures were more likely to report one or more of these major symptoms, with only 18% denying any such symptoms.
Back pain (OR, 1.32; 95% CI, 1.22–1.43), kyphosis (1.32; 1.22–1.42), and height loss (1.19; 1.11–1.27) were significantly more common among women with multiple fractures compared with those with only a single fracture, even after accounting for age, height, and weight. In addition, multiple fractures were more likely among patients with early menopause (1.23; 1.14–1.32), corticosteroid use (1.19; 1.08–1.31), and current/previous smoking (1.12; 1.02–1.22). Weight, family history of osteoporosis, low calcium intake, and physical inactivity were not associated with multiple fractures. Women with multiple fractures were more likely to report use of specific osteoporosis medication (1.61; 1.50–1.74)
Symptoms and risk factors associated with each fracture type for all valid returned surveys are shown in Table 3. In summary, kyphosis, height loss, and back pain were most strongly associated with the presence of vertebral fractures, although these symptoms were generally predictive of the presence of any type of fracture, with some exceptions for ankle and hip fractures. Early menopause was associated with fracture at all sites, while family history of osteoporosis was associated with fracture at all sites except hip. Corticosteroid use was associated with the presence of vertebra, rib, ankle, and other fractures, as was previous/current smoking. Physical inactivity was associated with all fracture sites except wrist, and low calcium intake was associated with the presence of wrist, rib, ankle, and other fractures.
Only 5049 (23%) of the 22,267 patients who reported height loss indicated the amount lost. The majority (84%) reported height loss of ≤6 cm (median, 3 cm). Twenty-seven percent reported a height loss of 4–5 cm, and 29% indicated height loss of ≤2 cm. Vertebral fracture (OR, 1.37; 95% CI, 1.13–1.66), kyphosis (1.48; 1.30–1.68), back pain (1.46; 1.26–1.68), early menopause (1.23; 1.07–1.42), and age (1.40; 1.35–1.47; in 5-year increments) were each associated with greater height loss.
Only 4559 women (8%) reported BMD tests, most (3589, 78%) of both spine and hip and most (80%) in the previous year. Approximately one-half had osteopenic T scores (−1 to −2.5), and approximately one-third had osteoporotic T scores (< −2.5).
Women with a previous fracture or other relevant history were more likely to have had a BMD test. However, the increased likelihood was generally modest except for previous fracture (OR, 2.31; 95% CI, 2.15–2.47), corticosteroid use (1.50; 1.45–1.74), osteoporotic family history (1.52; 1.40–1.65), kyphosis (1.50; 1.39–1.62), or height loss (1.39; 1.29–1.49).
Specific treatment for osteoporosis
Less than one-quarter of all women were receiving any specific anti-osteoporosis treatment (Table 4), with an additional 7% receiving calcium alone. The likelihood of receiving specific osteoporosis treatment increased modestly with various factors, but the greatest increase was in those who had had a BMD measurement (OR, 3.68; 95% CI, 3.43–3.95). Other factors that influenced treatment were age (OR for each 5 years decreasing; 1.39; 1.37–1.42); weight (OR for each 5 kg lower; 1.09; 1.08–1.11); height (OR for each 5 cm greater; 1.04; 1.02–1.05); presence of fracture(s) (1.51; 1.44–1.58); reports of height loss (1.30; 1.24–1.36); kyphosis (1.23; 1.17–1.30); back pain (1.50; 1.42–1.58); use of corticosteroids (1.79; 1.67–1.91); and family history of osteoporosis (1.46; 1.38–1.55).
Only 25% of women reported that a doctor had told them that they had osteoporosis; 79% were told in the previous 5 years. The women who reported fracture, kyphosis, height loss, or back pain were 1.5–3 times more likely to have been informed about osteoporosis. These women who were informed about osteoporosis were also more than twice as likely to be receiving “specific” medication (OR, 2.86; 95% CI, 2.73–2.99). However, only 40% of these “identified” women were receiving such medication.
In this study of a large random sample of postmenopausal women attending general practitioners in Australia, 57,088 women >59 years of age had either a fracture or one or more of the risk factors for osteoporosis, representing 82% of women in this age group. However, women with relevant symptoms or history could have been more likely to return the survey, representing 65% of all women sent surveys (88,041). Almost one-third had an osteoporotic fracture, yet >80% of women known to have had a presumed osteoporotic fracture after menopause had neither been investigated for possible osteoporosis nor were on any specific treatment to reduce their increased risk of further fractures. Thus, <20% of those who responded and reported having a postmenopausal fracture were on specific treatment, and only 8% had had a BMD measurement. Importantly, although fracture prevalence increased with age, the bulk of fractures reported (42%) was by women in the 70- to 79-year age group.
These data support the observations in other samples that only a small proportion of women with postmenopausal osteoporosis and prevalent fractures, who are thus at high risk of future fractures, are on any specific anti-osteoporotic therapy. The observed prevalence of treatment is remarkably similar to that observed in a recent prospective study,(26) with low overall treatment rates after a fracture of 24%. In our large random sample, an explanation of the presence of osteoporosis by their doctor, especially when supported by a BMD measurement, was associated with a 2- to 3-fold higher likelihood of a woman receiving specific anti-osteoporotic therapy. This implies that lack of recognition or concern explains, in large part, the low levels of treatment. This low level of treatment may reflect a low level of recognition of the extent of osteoporosis as a problem in “younger” women, in whom the majority of fractures occur.
This study has two limitations. First, ∼18% of the women who did not report any symptoms or risk factors did not have their anthropometric or treatment data entered in the study database. Consequently, specific osteoporosis treatment could have been used in this group, but this is unlikely because these therapies (e.g., bisphosphonates, raloxifene, or calcitriol) are subsidized by the Australian Pharmaceutical Benefits Scheme, but only after a “low-trauma” fracture. Second, the study's reliance on patient self-report has limitations, particularly regarding the accuracy of reporting about fracture, kyphosis, and height loss. Because patients in the Australian health care system can easily change or attend different doctors, as noted above, collateral confirmation was not always available. Elderly women may over-report some fractures; however, their self-report is reported to be relatively accurate for important osteoporotic fractures including the hip, wrist, and humerus.(27) It is of interest that 67% of vertebral fractures in this study were confirmed, because these are generally less reliably reported than other peripheral fractures, such as wrist and hip, which are usually highly reliable.(28) Given data and confirmation studies in other self-report studies, these fracture data are likely to be accurate to within 10%. Importantly, the fractures in this study were clinical fractures (i.e., those known to the individual), thus vertebral fractures would be underestimated; perhaps one-quarter achieve clinical recognition.(27,29)
A strength of this study is that these women did seek regular medical attention and attended a local primary care doctor. Therefore, they would be expected to have better access to investigation and management. If anything, these data may overestimate the levels of specific osteoporosis treatment.
Clinical “recognition” of osteoporosis, as communicated to the woman, especially if accompanied by a BMD test, was associated with significantly higher but still relatively low levels of treatment (40%). To some extent, the ease with which Australian women (and men) can move between primary health care providers might have led to fewer of their doctors being aware of their fracture history. In U.S. studies,(30,31) ethnic factors may have contributed to access to care. This does not seem to be the case in this study, because financial and access issues are unlikely to be influencing these overall poor outcomes because the Australian government-supported universal health care system covers the largest part of costs for both investigation and treatment in individuals who have already suffered a fracture. Although the barriers to the uptake of investigations and treatment are not clear, it seems likely that neither the women nor their doctors consider osteoporosis to be an important health issue.
In summary, this large random sample of postmenopausal women attending primary care physicians in Australia found a high prevalence of postmenopausal fractures with low levels of investigation and limited use of specific anti-osteoporosis treatments. The finding that a given diagnosis of osteoporosis, especially if accompanied by a BMD test, was associated with higher, albeit still low levels of treatment, show the need for more effective education of both the community and their medical practitioners. The clinical significance of osteoporotic fracture with respect to further fractures and associated morbidity and mortality, as well as the efficacy of standard anti-osteoporotic treatment, needs to be made more explicit to help reduce health care costs and adverse outcomes.
- 11998 Drug therapy: Treatment of postmenopausal osteoporosis. N Engl J Med 338: 736–746.
- 21991 Determinants of disease and disability in the elderly: The Rotterdam Elderly Study. Eur J Epidemiol 7: 403–422., , ,
- 31999 Danish Osteoporosis Prevention Study (DOPS): Project design and inclusion of 2000 normal perimenopausal women. Maturitas 31: 207–219., , , , ,
- 41999 Health burden of hip and other fractures in Australia beyond 2000. Projections based on the Geelong Osteoporosis Study. Med J Australia 170: 467–470., , , , ,
- 51999 Age- and gender-specific rate of fractures in Australia: A population-based study. Osteoporos Int 10: 240–247., , , , , , ,
- 61994 Symptomatic fracture incidence in elderly men and women: The Dubbo Osteoporosis Epidemiology Study (DOES). Osteoporos Int 4: 277–282., , , , ,
- 71992 Incidence of clinically diagnosed vertebral fractures: A population-based study in Rochester, Minnesota, 1985–1989. J Bone Miner Res 7: 221–227., , ,
- 81997 The crippling consequences of fractures and their impact on quality of life. Am J Med 103: 12S–19S.
- 91999 Mortality after all major types of osteoporotic fracture in men and women: An observational study. Lancet 353: 878–882., , , ,
- 101996 Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures: Fracture Intervention Trial Research Group. Lancet 348: 1535–1541., , , , , , , , , , , , , ,
- 111999 Larger increases in bone mineral density during alendronate therapy are associated with a lower risk of vertebral fractures in women with postmenopausal osteoporosis. Fracture Intervention Trial Research Group. Arthritis Rheum 42: 1246–1254., , , , , , , ,
- 121995 Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. N Engl J Med 333: 1437–1443., , , , , , , , ,
- 132000 A randomized trial of nasal spray salmon calcitonin in postmenopausal women with established osteoporosis: The prevent recurrence of osteoporotic fractures study. PROOF Study Group. Am J Med 109: 267–276., , , , , , , , , , , , , ,
- 141999 Multinational, placebo-controlled, randomized trial of the effects of alendronate on bone density and fracture risk in postmenopausal women with low bone mass: Results of the FOSIT study. Osteoporos Int 9: 461–468., , , , , , , , , ,
- 151992 Treatment of postmenopausal osteoporosis with calcitriol or calcium. N Engl J Med 326: 357–362., , ,
- 161994 Effect of calcium and cholecalciferol treatment for three years on hip fractures in elderly women. BMJ 308: 1081–1082., , ,
- 171997 Effect of calcium and vitamin D suppementation on bone density in men and women 65 years of age or older. N Engl J Med 337: 670–676., , ,
- 181992 Effect of calcitonin given intranasally on bone mass and fracture rates in established osteoporosis: A done-response study. BMJ 305: 556–561., , ,
- 191993 The effect of postmenopausal estrogen therapy on bone density in elderly women. N Engl J Med 329: 1141–1146., , , , ,
- 201995 Estrogen replacement therapy and fractures in older women: Study of Osteoporotic Fractures Research Group. Ann Intern Med 122: 9–16., , , , ,
- 211998 Prevention of hip fractures in older women: A population-based perspective. Osteoporos Int 8: S8–S12.
- 221999 Interim report and recommendations of the World Health Organization task-force for osteoporosis. Osteoporos Int 10: 259–264., , , , , , , , , , , , , , , , , , , , , , , ,
- 231999 Diagnosis and management of osteoporosis in postmenopausal women: Clinical guidelines. International Committee for Osteoporosis Clinical Guidelines. Clin Ther 21: 1025–1044., , , , , , ,
- 241999 Improved strategies for diagnosis and treatment of osteoporosis. Menopause 6: 242–250.,
- 251998 Intervention thresholds for the treatment of osteoporosis: Comparison of different approaches to decision making. Osteoporos Int 8: S22–S27., , ,
- 262003 Low frequency of treatment of osteoporosis among postmenopausal women following a fracture. Arch Intern Med 163: 2052–2057., , , , , , , ,
- 271992 The accuracy of self-report of fractures in elderly women: Evidence from a prospective study. Am J Epidemiol 135: 290–299., , , , , ,
- 28the EPOS Study group 2000 Validity of self-report of fractures: Results from a prospective study in men and women across Europe. Osteoporos Int 11: 248–254., , , , , , , , , , , ,
- 291999 Number and type of vertebral deformities: Epidemiological characteristics and relation to back pain and height loss. Osteroporos Int 9: 206–213., , , , , ,
- 302003 Ethnic disparity in the treatment of women with established low bone mass. J Am Med Womens Assoc 58: 173–177., ,
- 312003 Racial disparities in osteoporosis prevention in a managed care population. South Med J 96: 445–451., , , , , , , ,