Estimation of Body Height in Korean Elderly

ABSTRACT

BACKGROUND: Height is essential variable for assessing nutritional status, estimating glomerular filtration rate, and drug dosage besides in epidemiologic study. In old age, spinal deformity and thinning of the intervertebral discs may produce a spurious decrease in height. In the nonambulants, standard methods of height measurement may be impractical. The present study has been designed to predict the height using regression analysis by long bone measurement. METHODS: One hundred twelve persons in one elderly welfare facility were studied. Standing height was measured. Total arm length, upper arm length, forearm length, tibial length in erect and supine position were measured. A nomogram has been prepared from these data to predict height. RESULTS: Multiple regression analysis was used to measure relationship between measured height, age, sex, and each of the measured parameters. Among the parameters in the erect position, height was best predicted by erect tibial length(ETL)(p<0.01). Among the parameters in the supine position, height was best predicted by supine total arm length(STAL)(p<0.05). Estimation of height was calculated by linear regression analysis. Estimation of height(cm)=93.036+2.383 x ETL(cm)-0.233 x age-5.747 x sex(R2=.815). Estimation of height(cm)=87.985 + 1.775 x STAL(cm)-0.151 age-6.787 x sex(R2=.803). CONCLUSION: In the present study it has been shown that erect tibial length and supine total arm length may be a good predictor of total height. The nomogram may provide the prediction of height in the nonambulant elderly.