Background: Although osteoporosis is less common in men than in women, the aging of the population is expected to significantly increase the incidence of osteoporotic fractures in men over the next 20 years. Recent studies show that treatment with bisphosphonates in men with osteoporosis reduces the risk of vertebral and hip fractures at rates comparable with those in postmenopausal women, suggesting that early detection and treatment of osteoporosis in men could potentially prevent fractures. However, it is uncertain how physicians should identify men who would most likely benefit from dual-energy x-ray absorptiometry (DEXA) testing. Shepherd and colleagues developed and validated a clinical prediction rule to identify older men at increased risk of osteoporosis.

The Study: The prediction rule was developed from an analysis of the National Health and Nutrition Examination Survey (NHANES) III, a nationally representative U.S. population sample that collected information on demographic characteristics (e.g., age, race/ethnicity, education), health-related behaviors (e.g., tobacco use, physical activity), and medical histories of 2,995 men 50 years or older who underwent DEXA testing. World Health Organization criteria were used to define osteoporosis based on racial and ethnicity-specific normal bone density values for men 20 to 29 years of age; men with T scores of less than -2.5 were considered to have osteoporosis.

The study sample was evenly divided using random assignment into two groups: a development group and a validation group. Multivariable logistic regression was used to identify independent associations between clinical variables and the presence of osteoporosis in the development group. The Male Osteoporosis Risk Estimation Score (MORES) was constructed from the "best fitting model" (greatest predictive ability for osteoporosis) that also contained the fewest clinical variables: age, weight, and a history of chronic obstructive pulmonary disease (see accompanying table). The instrument was then tested in the validation group.

Results: In the validation group, a MORES of 6 or more points had a sensitivity of 0.95 (95% confidence interval [CI], 0.81 to 0.99) and a specificity of 0.61 (95% CI, 0.57 to 0.64) for osteoporosis. These values were similar across racial and ethnic groups, except specificity was slightly higher in non-Hispanic white persons. Using a technique previously developed for the U.S. Preventive Service Task Force's recommendation on screening for postmenopausal osteoporosis, the authors estimated the number of men needed to screen with DEXA in various age groups to prevent one hip fracture. If all men 50 years and older received screening, 595 men would need to be screened to prevent one hip fracture over 10 years. However, if only men with a MORES of 6 or more were screened, the number needed to screen was 279.

Conclusion: The authors conclude that the MORES appears to be a simple and useful tool for determining which men are at increased risk of osteoporosis and which may benefit from screening. They recommend that this instrument be evaluated further in other clinical samples.