Height prediction from anthropometric length parameters in Thai people.

Abstract

Height is an important clinical parameter. However, there were no specific measurements available for particular clinical situations. Although many anthropometric measurements were suggested, no formula was recommended in Thailand. The objective of this study was to develop a formula for height prediction with acceptable validity. Two thousand volunteers were included and were divided consecutively according to both age and gender. Model and validation groups were further separated independently. Linear regression was analyzed to create a predictive formula. Ten parameters were included and analyzed. Of these, demispan, sitting height and knee height were selected with a correlation coefficient of more than 0.5 and significant F test in all age groups and genders. All single parameters and the highest predictive value of double (sitting and knee height) and triple regression models (demispan, sitting and knee height) were proposed and these were modified into a simple formula. After validation of both formulas the correlation, quantitative error and relative error were comparable. The simple formula had more than 90% precision with an error of up to 10 cm in the validation group (89.7 to 99.0% in range). Of these, knee height had the least predictive error in all subgroups. The double and triple models had decreased error only in the younger group. In summary, anthropometric parameters with demispan, sitting height, knee height and combination could be applied to height prediction in the adult Thai with acceptable error. These formulas should be applied only in people who could not be directly measured.