Many studies have demonstrated significant differences in bone mineral
density between various racial groups, Although it has been suggested
that differences in body weight contribute to such interracial variat
ion, the artifactual effect of the skeletal size inherent in projectio
nal absorptiometry methods has been largely ignored, We have measured
bone mineral density by dual-energy X-ray absorptiometry in the lumbar
spine and at three femoral sites in 200 premenopausal women of Chines
e, Indian, European, or Polynesian origin (50 of similar mean age in e
ach group), In the Chinese and Indian women the measured bone mineral
density measurements (g/cm(2)) were similar, but significantly less, a
t all sites, than those of European women (p less than or equal to 0.0
05), The European women were, however, significantly taller than both
the Chinese and Indian women (p < 0.0001), and when the scale artifact
of absorptiometry was removed by dividing the measured bone mineral d
ensity either by the height of the subject, or by the square root of t
he area over which the X-ray beam was projected, then the differences
in mean bone mineral density between the Chinese, Indian, and European
women were almost completely eliminated, The Polynesian women were si
gnificantly more obese (as judged from mean body mass index) than all
the other groups (p < 0.0001) and had significantly greater bone miner
al density at all sites than all the other groups both before (p < 0.0
001) and after (p < 0.0001) correcting for the scale artifact, Analysi
s of covariance suggested that 10-40% of the remaining increase in bon
e density could be accounted for by their greater obesity, the balance
representing a true interracial difference, This effect was most obvi
ous in the lumbar spine and least evident at the femoral sites, We con
clude that there is a true interracial difference in bone mineral dens
ity between Polynesians and the other races studied, and that this is
increased further by the greater body weight of Polynesians, The diffe
rences between Europeans and Asians are substantially attributable to
a measurement artifact arising from the smaller skeletal dimensions of
Asians, This needs to be borne in mind by clinicians assessing bone d
ensity in Asian patients.