Demispan (finger web to midline) is the preferred index of stature in
the study of the elderly where height is unreliable because of age-rel
ated lass or distortion. As a consequence, two alternative mass indice
s, Demiquet (male mass.demispan(-2)) and Minder (female mass.demispan(
-1)), have been proposed for use in the elderly, in place of body mass
index calculated from height, which might be suspect. These gender-sp
ecific indices reflect the squared and linear relationship of height t
o mass in men and women respectively. To assess the independence of De
miquet and Minder from stature and the reliability of demispan, height
, demispan (second fingerweb to midline), and mass were recorded in a
random sample of independent Canadian men (182) and women (212) aged 5
5-86 years. Both Demiquet and Minder were highly correlated with mass,
r = 0.81 (p = 0.001) and r = 0.96 (p = 0.001), respectively. In both
gender groups these mass indices were independent of height and demisp
an. The demispan within- and between-observer errors were small -0.04(
-0.16 to 0.08) cm and - 0.09(-0.22 to 0.03)cm and the between-observer
intraclass correlation estimate of reliability was 0.96 (lower-bound
one-sided 95% CI 0.90). Demispan is a reliable and reproducible measur
e of stature in the elderly. Demiquet and Minder are useful mass indic
es, independent of stature and suitable for use in the study of the el
derly where the use of mass indices based on height may be questionabl
e.