Final Recommendation Statement
Osteoporosis: Screening, 2002
September 17, 2002
Recommendations made by the USPSTF are independent of the U.S. government. They should not be construed as an official position of the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services.
This statement summarizes the current U.S. Preventive Services Task Force (USPSTF) recommendations on screening for osteoporosis and the supporting scientific evidence, and it updates the 1996 recommendations contained in the Guide to Clinical Preventive Services, second edition.1 The complete information on which this statement is based, including evidence tables and references, is available in the article Screening for Osteoporosis: A Summary of the Evidence for the U.S. Preventive Services Task Force2 (which follows this recommendation) and in the Systematic Evidence Review3 on this topic. These documents can be obtained through the USPSTF Web site (www.uspreventiveservicestaskforce.gov), and through the National Guideline Clearinghouse (www.guideline.gov).
This was first released on the AHRQ Web site on September 17, 2002, and an abridged version of this recommendation also appeared in Ann Intern Med. 2002;137(6):526-528.
- Modeling analysis suggests that the absolute benefits of screening for osteoporosis among women aged 60-64 who are at increased risk for osteoporosis and fracture are comparable to those of routine screening in older women. The exact risk factors that should trigger screening in this age group are difficult to specify based on evidence. Lower body weight (weight < 70 kg ) is the single best predictor of low bone mineral density.4,5 Low weight and no current use of estrogen therapy are incorporated with age into the 3-item Osteoporosis Risk Assessment Instrument (ORAI).4,5 There is less evidence to support the use of other individual risk factors (for example, smoking, weight loss, family history, decreased physical activity, alcohol or caffeine use, or low calcium and vitamin D intake) as a basis for identifying high-risk women younger than 65. At any given age, African-American women on average have higher bone mineral density (BMD) than white women and are thus less likely to benefit from screening. Additional characteristics of screening tools are discussed in the "Accuracy and Reliability of Screening Tests" section.
- Among different bone measurement tests performed at various anatomical sites, bone density measured at the femoral neck by dual-energy x-ray absorptiometry (DXA) is the best predictor of hip fracture and is comparable to forearm measurements for predicting fractures at other sites. Other technologies for measuring peripheral sites include quantitative ultrasonography (QUS), radiographic absorptiometry, single energy x-ray absorptiometry, peripheral dual-energy x-ray absorptiometry, and peripheral quantitative computed tomography. Recent data suggest that peripheral bone density testing in the primary care setting can also identify postmenopausal women who have a higher risk for fracture over the short term (1 year). Further research is needed to determine the accuracy of peripheral bone density testing in comparison with dual-energy x-ray absorptiometry (DXA). The likelihood of being diagnosed with osteoporosis varies greatly depending on the site and type of bone measurement test, the number of sites tested, the brand of densitometer used, and the relevance of the reference range.
- Estimates of the benefits of detecting and treating osteoporosis are based largely on studies of bisphosphonates. Some women, however, may prefer other treatment options (for example, hormone replacement therapy, selective estrogen receptor modulators, or calcitonin) based on personal preferences or risk factors. Clinicians should review with patients the relative benefits and harms of available treatment options, and uncertainties about their efficacy and safety, to facilitate an informed choice.
- No studies have evaluated the optimal intervals for repeated screening. Because of limitations in the precision of testing, a minimum of 2 years may be needed to reliably measure a change in bone mineral density; however, longer intervals may be adequate for repeated screening to identify new cases of osteoporosis. Yield of repeated screening will be higher in older women, those with lower BMD at baseline, and those with other risk factors for fracture.
- There are no data to determine the appropriate age to stop screening and few data on osteoporosis treatment in women older than 85. Patients who receive a diagnosis of osteoporosis fall outside the context of screening but may require additional testing for diagnostic purposes or to monitor response to treatment.
Epidemiology and Clinical Consequences
One-half of all postmenopausal women will have an osteoporosis-related fracture during their lives, including 25 percent who will develop a vertebral deformity6 and 15 percent who will suffer a hip fracture.7 Risk for fracture increases steadily as bone density declines, with no threshold. The commonly used definition of osteoporosis, derived from the World Health Organization (WHO) recommendations for epidemiologic studies, defines a BMD more than 2.5 standard deviations (SD) below the mean for a young healthy adult woman as osteoporosis, and a BMD between 1 and 2.5 SD below the mean as osteopenia. Based on the WHO criteria and DXA measurements at the femoral neck, population-based studies estimate that 41 percent of white women older than 50 have osteopenia.8 When bone density is measured at the hip, spine, and wrist, 15 percent of white women aged 50-59 and 70 percent of white women older than 80 have osteoporosis by WHO criteria at at least one site.9
The prevalence of osteoporosis in Mexican-American women is similar to the prevalence in white women. While rates of osteoporosis in African-American women are approximately one-half those of the other groups, they are still substantial (8 percent among women older than 50). Including all races, an estimated 14 million women older than 50 have osteopenia, and over 5 million have osteoporosis.10 The actuarial risk of a 65-year-old white woman sustaining a fracture by age 90 is 16 percent for the hip, 9 percent for distal forearm, and 5 percent for proximal humerus.9 Sixteen percent of postmenopausal women have osteoporosis of the lumbar spine.11
The USPSTF examined two components of screening:
- The accuracy of risk factors or risk assessment instruments for identifying women at risk for osteoporosis or fracture.
- The accuracy of different bone density measurement techniques for identifying women at risk for fracture who can benefit from osteoporosis treatment.
Predicting Risk for Osteoporosis or Fracture
The USPSTF evaluated both individual risk factors and prescreening assessment tools that incorporate two or more of the risk factors. Risk for osteoporosis increases steadily and substantially with age. Relative to women aged 50-54, the odds of having osteoporosis were 5.9-fold higher in women aged 65-69 and 14.3-fold higher in women aged 75-79, in a study of over 200,000 postmenopausal women.12 Low body weight or body-mass index (BMI) and not using estrogen replacement were also consistently associated with osteoporosis but to a lesser degree than age. Other risk factors for fracture or low bone density found in some, but not all, studies include white or Asian ethnicity, history of fracture, family history of osteoporotic fracture, history of falls, low levels of physical activity, smoking, excessive alcohol or caffeine use, low calcium or vitamin D intake, and the use of various medications.
Specific instruments to assess risk for low bone density or fractures generally have moderate-to-high sensitivity and low specificity. The best validated instruments include the 3-item ORAI and the 6-item Simple Calculated Osteoporosis Risk Estimation tool (SCORE). The ORAI uses age, weight, and current use of hormone replacement therapy to identify women at risk for osteoporosis and has a sensitivity of 94 percent and specificity of 41 percent.4 The SCORE has a sensitivity of 91 percent and specificity of 40 percent in one validation population (n = 259), but it has much lower specificity in an older population.11
Among eight studies of prediction instruments for fracture risk, most had only modest sensitivity and specificity. The best performing model for hip fracture outcomes included age, gender, height, use of a walking aid, current smoking, and weight and had a sensitivity of 70 percent with specificity of 84 percent.13
Measurements of Bone Density
To date, bone density measured at the femoral neck by DXA is the best predictor of hip fracture and is comparable to forearm measurements for predicting fractures at other sites. Recent prospective studies have evaluated QUS measurements at the heel.14,15 While QUS measurements are not highly correlated with DXA measurements, a result in the osteoporotic range on either test is associated with an increased short-term probability of hip fracture. Several other radiologic methods that measure bone density at peripheral sites2 (including sites in the hand, heel, wrist, and forearm) include single photon absorptiometry, quantitative computed tomography, single-energy x-ray absorptiometry, and peripheral quantitative computed tomography.
In a study of over 200,000 women in a primary care setting, women diagnosed with osteoporosis by peripheral bone density measurements were 4 times more likely to have fractures than women with normal bone density over the subsequent year. The likelihood of being diagnosed with osteoporosis varies greatly depending on the site and type of bone measurement test, the number of sites tested, the brand of densitometer, and the relevance of the reference range.
Effectiveness of Early Treatment
No controlled studies have evaluated the effect of screening on fractures or fracture-related morbidity. The Task Force reviewed the evidence to determine whether treatment for osteoporosis or low bone density in asymptomatic patients reduced fractures.
Available trials that reported fracture outcomes have examined the efficacy of bisphosphonates (alendronate and risendronate), estrogen, and selective estrogen receptor modulators (raloxifene) and calcitonin. A meta-analysis16 of 11 randomized trials17-27 involving a total of 12,855 women, found that alendronate significantly reduced vertebral fractures (RR, 0.52; 95 percent CI, 0.43-0.65), forearm fractures (RR, 0.48; 0.29-0.78), hip fractures (RR, 0.63; 0.43-0.92), and other nonvertebral fractures (RR, 0.51; 0.38-0.69). There were nonsignificant trends toward reduction in hip fractures. No randomized trial of treatment for osteoporosis has demonstrated an impact on mortality. One trial in women aged 70-79 with very low bone density (T-score less than -3) reported that risendronate reduced the risk for hip fracture (RR, 0.60; 95 percent CI, 0.40-0.90).28
There are no direct comparisons of alendronate and estrogen or raloxifene that report fracture outcomes. Estrogen, either alone or with progestin, consistently improves bone density in randomized trials. The effects of estrogen and the selective estrogen receptor modulators on fractures are reviewed in more detail in a separate report.13 Only a few small randomized clinical trials of estrogen indicate mixed results for fracture outcomes, but these studies are methodologically limited. Observational studies report a 25-30 percent reduction in the risk for hip fracture with estrogen use. A good-quality study of raloxifene reported a reduced risk for vertebral fractures (RR, 0.59; 95 percent CI, 0.50-0.70).29
The benefits of treating osteoporosis are larger in women at higher risk for fracture than in women at lower risk. The Fracture Intervention Trial (FIT) was conducted with 2 different groups of participants: 2,027 high-risk women who had T-scores of -1.6 or lower and pre-existing vertebral fractures, and 4,432 women with comparable T-scores but no pre-existing vertebral fracture. Over 3 years of treatment in high-risk women, alendronate reduced the risk for hip fracture (1.1 percent vs. 2.2 percent in the placebo group; relative hazard [RH], 0.49 [0.23-.099]) and the risk for any clinical fracture (18.2 percent vs. 13.6 percent; RH, 0.72 [0.58-0.90]). Among women with no pre-existing fracture, only the subgroup of patients who had a T-score less than -2.5 had a significant reduction in all clinical fractures from treatment, from 19.6 percent to 13.1 percent (RR, 0.64; 0.50-0.82). Alendronate had no effect on fractures among lower risk women who had T-scores between -1.6 and -2.5.
These results suggest that treatment will produce larger benefits in women with more risk factors for fracture, such as those who are older, have very low bone density, or have pre-existing vertebral fractures. FIT, as well as other therapy trials, enrolled highly selected patients thus limiting the generalizability of their results to asymptomatic women detected in a typical primary care setting.
There is little evidence regarding which patients are likely to benefit from screening and treatment. It is not known whether women who have a similar overall risk for fracture, but different bone densities, will benefit similarly from treatment. This uncertainty is clinically important because the lack of accepted criteria for initiating treatment remains a problem.
To estimate the benefits of routine screening for women in different age groups, the USPSTF used estimates from recent studies to project the number of fractures that would be prevented over 5 years from screening and treatment of a hypothetical cohort of 10,000 postmenopausal women.2 For women aged 55-59, more than 4,000 would need to be screened to prevent 1 hip fracture and more than 1,300 to prevent 1 vertebral fracture. For women older than 60, the number needed to screen to prevent 1 hip fracture is 1,856 for women aged 60-64, 731 for women aged 65-69, and 143 for women aged 75-79. The benefits of screening improve substantially in older women because osteoporosis is both more prevalent and more likely to lead to a fracture in older women.
In all age groups, the number needed to screen to prevent fractures is lower in women with important risk factors than it is in women who do not have risk factors. For women aged 60-64 who have a risk factor that increases the risk of osteoporosis by 100 percent and fracture by 70 percent, the number needed to screen is 1,092 and the number need to treat is 72 to prevent 1 hip fracture. These numbers are comparable to those of women aged 65-69 without risk factors.2 These estimates rely on many assumptions that may not apply for specific populations.
Potential Adverse Effects of Screening and Treatment
There are several potential harms of screening, although the empirical data for them are few. Women who undergo screening with bone density tests are more likely to begin hormone replacement therapy than women who do not. However, women who were diagnosed with osteoporosis after screening reported increased fears and anxiety in one study. Other potential harms may arise from inaccuracies and misinterpretations of bone density tests. Clinicians may have difficulty in using test results to provide accurate information to the patients because techniques used to measure bone density vary, test results are reported as T-scores, and information on how to integrate bone density results with other clinical predictors has not been clearly defined.2
In the alendronate treatment trials, gastrointestinal side effects occurred in about 25 percent of patients taking alendronate, but this was usually not higher (or only slightly higher) than the rate for placebo. Higher rates were observed among Medicare enrollees taking alendronate. In the FIT-II trial, the rates of ulcer disease were higher in the alendronate treatment group, with 2.2 percent developing ulcer disease, as opposed to 1.2 percent in the placebo group (P<0.05).30 The long-term adverse effects of alendronate are unknown. Harms of hormone replacement therapy include venous thromboembolic events, endometrial cancer, and cholecystitis, all with relative risks of approximately 2.0.12 Both raloxifene and tamoxifen are associated with thromboembolic events, leg cramps, and hot flashes.2
In 1998, the National Osteoporosis Foundation, in collaboration with other professional organizations, issued screening guidelines recommending bone density testing for all women aged 65 or older and younger postmenopausal women who have had a fracture or who have one or more risk factors for osteoporosis.31 Collaborating groups included the American Academy of Orthopaedic Surgeons, the American College of Obstetricians and Gynecologists, the American Geriatrics Society, the American College of Radiology, the American College of Rheumatology, the American Academy of Physical Medicine and Rehabilitation, the American Association of Clinical Endocrinologists, the Endocrine Society, and the American Society of Bone and Mineral Research. The American Association of Clinical Endocrinologists released revised guidelines in 2001.32 A 2000 Consensus Development Conference sponsored by the U.S. National Institutes of Health concluded that the value of universal osteoporosis screening was not yet established.33 The conference panel recommended an individualized approach to screening, noting that bone density measurement is appropriate when it will aid the patient's decision to institute treatment. The Canadian Task Force on Preventive Health Care is currently revising its recommendations on screening for osteoporosis.
2. Nelson HD, Helfand M, Woolf SH, et al. Screening for postmenopausal osteoporosis: A review of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2002;137:529-41. (Available on the AHRQ Web site at https://www.uspreventiveservicestaskforce.org).
3. Nelson HD, Helfand M. Screening for Postmenopausal Osteoporosis. Systematic Evidence Review No. 17. (Prepared by the Oregon Health & Science University Evidence-based Practice Center under Contract No. 290-97-0018.) Rockville, MD: Agency for Healthcare Research and Quality, Research and Quality, 2002. (Available at: http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=hscps2ed1996&part=A14728)
4. Cadarette SM, Jaglal SB, Kreiger N, et al. Development and validation of the Osteoporosis Risk Assessment Instrument to facilitate selection of women for bone densitometry. Can Med Assoc J 2000;162:1289-94.
12. Siris ES, Miller PD, Barrett-Connor E, et al. Identification and fracture outcomes of undiagnosed low bone mineral density in postmenopausal women: Results from the National Osteoporosis Risk Assessment. JAMA 2001;286:2815-22.
13. Nelson HD, Helfand M. Hormone Replacement Therapy and Osteoporosis. Systematic Evidence Review No 12. (Prepared by the Oregon Health & Science University Evidence-based Practice Center under Contract No. 290-97-0018.) Rockville, MD: Agency for Healthcare Research and Quality 2002. (Available at https://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=es12)
17. Black DM, Cummings SR, Karpf DB, et al. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Fracture Intervention Trial Research Group. Lancet 1996:1535-41.
18. Adami S, Passeri M, Ortolani S, et al. Effects of oral alendronate and intranasal salmon calcitonin on bone mass and biochemical markers of bone turnover in postmenopausal women with osteoporosis. Bone 1995:383-90.
19. Bone HG, Downs RW Jr, Tucci JR, et al. Dose-response relationships for alendronate treatment in osteoporotic elderly women. Alendronate Elderly Osteoporosis Study Centers. J Clin Endocrinol Metab 1997:265-74.
21. Hosking D, Chilvers CE, Christiansen C, et al. Prevention of bone loss with alendronate in postmenopausal women under 60 years of age. Early Postmenopausal Intervention Cohort Study Group. NEJM 1998:485-92.
22. Liberman UA, Weiss SR, Broll J, et al. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. The Alendronate Phase III Osteoporosis Treatment Study Group. NEJM 1995:1437-43.
23. McClung M, Clemmesen B, Daifotis A, et al. Alendronate prevents postmenopausal bone loss in women without osteoporosis. A double-blind, randomized, controlled trial. Alendronate Osteoporosis Prevention Study Group. Ann Intern Med 1998:253-61.
24. Greenspan SL, Parker RA, Ferguson L, et al. Early changes in biochemical markers of bone turnover predict the long-term response to alendronate therapy in representative elderly women: A randomized clinical trial. J Bone Miner Res 1998;13:1431-8.
25. Pols HA, Felsenberg D, Hanley DA, et al. 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. Fosamax International Trial Study Group. Osteo Intern 1999:461-8.
26. Cummings SR, Black DM, Thompson DE, et al. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. JAMA 1998:2077-82.
30. Cummings SR, Black DM, Thompson DE. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. JAMA 1998;280:2077-82.
32. American Association of Clinical Endocrinologists. 2001 Medical Guidelines for Clinical Practice for the Prevention and Management of Postmenopausal Osteoporosis. Available at: http://www.aace.com/clin/guidelines/osteoporosis2001.pdf. Accessed February 27, 2002.
Members of the U.S. Preventive Services Task Force are Alfred O. Berg, M.D., M.P.H., Chair, USPSTF (Professor and Chair, Department of Family Medicine, University of Washington, Seattle, WA); Janet D. Allan, Ph.D., R.N., C.S., F.A.A.N., Vice-chair, USPSTF (Dean, School of Nursing, University of Maryland Baltimore, Baltimore, M.D.); Paul Frame, M.D. (Tri-County Family Medicine, Cohocton, NY, and Clinical Professor of Family Medicine, University of Rochester, Rochester, NY); Charles J. Homer, M.D., M.P.H. (Executive Director, National Initiative for Children's Healthcare Quality, Boston, MA); *Mark S. Johnson, M.D., M.P.H. (Chair, Department of Family Medicine, University of Medicine and Dentistry of New Jersey-New Jersey Medical School, Newark, NJ); *Jonathan D. Klein, M.D., M.P.H. (Associate Professor, Department of Pediatrics, University of Rochester School of Medicine, Rochester, NY); Tracy A. Lieu, M.D., M.P.H. (Associate Professor, Department of Ambulatory Care and Prevention, Harvard Pilgrim Health Care and Harvard Medical School, Boston, MA); Cynthia D. Mulrow, M.D., M.Sc. (Clinical Professor and Director, Department of Medicine, University of Texas Health Science Center, and Director, National Program Office for Robert Wood Johnson Generalist Physician Faculty Scholars Program, San Antonio, TX); C. Tracy Orleans, Ph.D. (Senior Scientist and Senior Program Officer, The Robert Wood Johnson Foundation, Princeton, NJ); Jeffrey F. Peipert, M.D., M.P.H. (Director of Research, Women and Infants' Hospital, Providence, RI), Nola J. Pender, Ph.D., R.N., F.A.A.N. (Professor Emeritus, University of Michigan, Ann Arbor, MI); *Albert L. Siu, M.D., M.S.P.H. (Professor of Medicine, Chief of Division of General Internal Medicine, Mount Sinai School of Medicine, New York, NY); Steven M. Teutsch, M.D., M.P.H. (Senior Director, Outcomes Research and Management, Merck & Company, Inc., West Point, PA); Carolyn Westhoff, M.D., M.Sc. (Professor of Obstetrics and Gynecology and Professor of Public Health, Columbia University, New York, NY); and Steven H. Woolf, M.D., M.P.H. (Professor, Department of Family Practice and Department of Preventive and Community Medicine and Director of Research, Department of Family Practice, Virginia Commonwealth University, Fairfax, VA).
* These current members were not on the Task Force at the time this recommendation was voted.