Bone Density Screening: A New Level of Evidence?

In this issue, Kern and colleagues (1) report that a cohort of older women and men who underwent bone mineral density (BMD) testing of the hip had a lower rate of hip fractures than did those in a cohort who did not get hip BMD testing. If Kern and colleagues' methods are sound, then they provide the first evidence from a trial (albeit not randomized) that screening lowers the risk for fracture.

Bone densitometry, invented by dentists in 1897 (2), became compact enough for clinical use during the 1980s (3). As soon as bone densitometry entered clinical practice, debates began about its appropriate clinical use. By the early 1990s, evidence from prospective studies showed that densitometry of the hip and spine satisfied rigorous criteria for a screening test for postmenopausal women (4). In particular, prospective studies showed that BMD strongly predicts fractures (5), and then clinical trials proved that treatments reduce the risk for fractures in postmenopausal women who have low hip or spine BMD (6-10). These studies suggest, but do not prove, that screening BMD reduces fracture rates. In 1998, a detailed analysis supported by the National Osteoporosis Foundation demonstrated that measuring hip BMD was cost-effective at least for white women 65 years of age or older with a hip T-score of −2.5 or less (11). This assumed that women with that level of hip bone density received pharmacologic treatment that reduced the risk for hip fractures by 50%. Subsequently, the National Osteoporosis Foundation (12), the U.S. Preventive Services Task Force (13), and specialty groups (14, 15) issued guidelines that …