Bad medicine: osteoporosisBMJ 2010; 340 doi: http://dx.doi.org/10.1136/bmj.c643 (Published 03 February 2010) Cite this as: BMJ 2010;340:c643
“One in two women and one in five men over the age of 50 in the UK will fracture a bone, mainly as a result of osteoporosis,” says the National Osteoporosis Society. But osteoporosis is an abstract numerical concept. As judged against a young white person at maximum bone density, it is defined as the 1% of people with the lowest bone mineral density shown on dual energy x ray absorptiometry (DXA) scanning, while the lowest 16% are defined as having osteopenia (erroneously called preosteoporosis). By this definition one in eight young women currently have osteoporosis or osteopenia, and the proportion rises to more than half after the age of 50. This has prompted calls for more DXA scanning and doomsday predictions of a “silent epidemic.” These numbers have spawned a thousand emotive magazine articles and enormous public anxiety.
The truth is that osteoporosis is not a disease but merely a risk factor for fracture, particularly of the hip. Age over 80 is by far the single greatest risk. Also, there is an assumption that effective “treatment” exists. However, there is limited evidence of the effectiveness of the widely prescribed bisphosphonates in the primary prevention of hip fracture in people with no history of fracture, even in highly selective study populations of elderly people. For secondary prevention the small reduction in hip facture is again in highly selected elderly populations.
On closer inspection this research carries the cosmetic surgery scars of big pharma, with relative risk reductions, non-clinical outcomes, and composite end points. Furthermore I can find no mortality data and not even convincing evidence of reduced back pain. The treatment paradox of managing medical risk—that the individual patient is unlikely to benefit personally from treatment—is not even acknowledged. But the key issue is that these data should not and cannot be extrapolated to a younger population. This, however, is exactly what is happening: an overdiagnosis disease creep. In our practice a fifth of prescriptions of bisphosphonates are for patients younger than 60, with the youngest in their 20s.
A recent Canadian study noted an age adjusted decline in the number of hip fractures since 1985 of 30%, a decline that treatment doesn’t account for. Does this reflect a fundamental change in epidemiology, with the passing of the deprived generations of the early 20th century? The term osteoporosis is an age dependent concept; primary prevention is questionable in all but the most frail; and “osteopenia” should be struck from the medical lexicon. The wanton promotion of osteoporosis and treatment of the young is bad medicine, and that is even before we consider the drugs’ side effects.
Cite this as: BMJ 2010;340:c643