Osteoporosis in Postmenopausal Women: Diagnosis and Monitoring

Objectives:

This report examines the evidence on the effectiveness of various strategies for diagnosing and monitoring postmenopausal women with osteoporosis. Specifically, it addresses: (1) the role of risk factors in identifying high-risk women and guiding their initial treatment, (2) the advantages and disadvantages of various techniques for bone measurement in predicting risk of hip or spine fracture, (3) the effectiveness of bone measurement tests for monitoring response to treatment and for guiding treatment change, (4) the role of markers of bone turnover in diagnosis and treatment management, (5) the evaluation of patients with osteoporosis for secondary causes, and (6) the costs and cost-effectiveness of various diagnostic strategies for osteoporosis.

Search Strategy:

The authors conducted a MEDLINE search (covering the years 1966 to 2000), supplemented by searches of HealthSTAR (covering 1975 to 2000) of papers published in English, reviewed reference lists of review articles, and sought guidance from local and national experts.

Selection of Studies:

The authors included abstracts relevant to one or more topic areas that had original data about postmenopausal women and osteoporosis. Two reviewers read each abstract to determine its eligibility. Articles were excluded if they did not provide sufficient information to determine the methods for selecting subjects and for analyzing data. For some topics, additional eligibility criteria were applied. For all topics combined, the authors retrieved 10,174 citations. After reviewing these citations for possible relevance, 530 articles about risk factors, 123 about bone measurement testing, 23 about monitoring, 277 about biochemical markers, and 53 about costs were selected for further review. An additional 242 studies were retrieved after reviewing the reference lists of studies and/or by suggestion of others. The search yielded no papers with data for the secondary causes topic.

Data Collection and Analysis:

From full-text published studies of fracture or bone density prediction or bone measurement methods, the authors extracted selected information about the patient population, interventions, clinical endpoints, study design, and study quality, and used this information to construct evidence tables. Additional reviews assessed the internal validity of studies of risk factors and the diagnostic performance of bone measurement tests and biochemical markers, summarized recommendations for testing for secondary causes of osteoporosis, reviewed studies about cost and cost-effectiveness, and compared diagnostic strategies.

Main Results:

Epidemiologic studies report clinical risk factors for osteoporosis and fractures, but few studies evaluate how to use them to identify individual women at risk for fracture, and no studies provide evidence that treatment decisions based on clinical risk factors lead to better or worse fracture outcomes than those based on bone measurement tests. Because of differences between bone measurement techniques, and because individuals have different rates of bone loss at different sites, no one test can exclude osteoporosis at the most important fracture sites -- hip, spine, and wrist. Dual-energy X-ray absorptiometry (DXA) of the femoral neck is the best validated test to predict hip fracture. Other techniques predict hip fracture less accurately or have not been evaluated in prospective studies.

Recent results from clinical trials raise questions about the value of repeated, annual densitometry tests for patients on therapy to prevent osteoporosis or bone loss; moreover, there is no evidence from clinical trials that adjusting therapy based on serial densitometry at any interval improve outcomes. Markers of bone turnover correlate poorly with bone measurement tests and are not good predictors of fractures.

Cost and cost-effectiveness studies, which are based solely on economic models, suggest targeting treatment to women with the lowest bone density and including a risk factor score or less expensive (and more widely available) technology to determine which women should receive hip DXA. The authors' supplementary analysis on cost-effectiveness favors a sequential strategy of quantitative ultrasound at the heel followed by densitometry of those identified by ultrasound as high risk over densitometry alone. In high-risk populations, ultrasound alone may also be cost-effective.

Conclusions:

Application of the results from epidemiologic studies to diagnosis and monitoring strategies for individual patients in the clinical setting is currently based on extrapolation from models or, for most questions addressed in this review, is simply lacking. To be more useful for clinicians and patients, future research should focus on the application of these data to the clinical setting.