[Use of knee height to correct the body height of elderly Hispanics].

Loss of stature in certain elderly subjects can be attributed to diseases such as osteoporosis, as well as to age and generational effects. In addition, many elders cannot stand straight for accurate measurement. For these cases, total height can be estimated with regression equations based on knee height. The aims of this study were, firstly, to evaluate the applicability of regression equations based on knee height for estimation of stature and, secondly, to document the differences between measured and estimated height in a group of elderly Hispanics with postural problems (n = 166) in comparison with a group of elderly Hispanic without postural problems (n = 270). Using both, estimated and measured height, we also calculated the body mass index (BMI) of both groups of elders. Statistical analyses were done with paired t-tests, within sex and study group. Within the group with postural problems, estimated height was higher than the measured height for both men (p < or = 0.001) and women (p < or = 0.001). There were no significant differences between measured and estimated height in the group without postural problems. Furthermore, in the group with postural problems, BMI values calculated with estimated height were lower than those estimated with the measured height, and these differences were also significant for both men (p < or = 0.001) and women (p < or = 0.001). With the aging of the Latin American population, there is a need for more nutrition and health research among elders. In order to do this we need to develop and use methods and criteria appropriate for each population.