In reference to the interesting trial by Sato and collegues,1 we would like to raise the following issues. According to the US National Osteoporosis Foundation guidelines, hip bone mineral density (BMD) is the best predictor of hip fractures.2 Use of metacarpal BMD measurement in the study may have preselected a group of individuals at slightly different risk for hip fractures. The severity of dementia in patients with Alzheimer disease3 and osteoporosis4 may have influenced their predisposition to fractures. Therefore, it will be useful to perform a stratified analyses based on degree of severity of preexisting osteoporosis and dementia. Also, the level of physical activity is linked with BMD.5 It will be informative to examine the level of daily physical exercise of the study subjects, as it may be a potential confounder of the results.
Kanna B, Roffe E. Prevention of Hip Fracture in Elderly Women With Alzheimer Disease. Arch Intern Med. 2006;166(10):1144–1145. doi:10.1001/archinte.166.10.1144-b