Statistical approaches to relationships between sitting height and leg length in adults

Background: Relationships between sitting height (SH) and leg length (LL) in adults are almost always studied in terms of ratios such as the Cormic Index (CI), SH/stature, rather than as primary variables. They are affected by genetics and childhood nutrition. Aim: To characterize these relationships and test whether the CI is ideal as an index of relative LL. Subjects and methods: Regression and reduced major axis (RMA) equations were calculated for 1653 men and women of European descent. For other population groups the RMA parameters were calculated from published means and standard deviations of SH and LL. Results: Linear and 'allometric' (power) equations fit the data equally well. For people of European origin the RMA equations for men and women do not differ significantly. Corresponding equations for other populations differ in line with published CIs. Conclusions: The linear equations suggest that LL tends to vary in proportion to SH minus a quantity similar to head height. A new index of relative LL may therefore be preferable to the CI for some research purposes to reflect this, but there is otherwise no strong reason to abandon the use of the CI.