DIRECT MEASURED AND ALTERNATIVE ANTHROPOMETRIC INDICES IN GEORGIAN HEALTHY ELDERLY POPULATION: RELIABILITY/VALIDITY OF ASSESMENT TOOLS.

The aim of the research was standard and alternative measurement of anthropometric indices in Georgian healthy elderly population for determining the reliability/validity of the widely used and recognized assessment tools. Our research is the first study of the anthropometry as the part of of nutritional status assessment in Georgian elderly. For anthropometric assessment we first time used 1. Standard measurements (weight, height, WC, HC, MAC /MUAC, TSF, BMI, WHR, AMA using standard methods and tools for measurable values and standard formulas for calculating values), and 2. Alternative measurements (demispan, demispan equivalent height and BMI (DEH, DE BMI), Mindex for women, Demiquet for men. using Bassey's equations). Our study group formed for the nutritional status assessment in Georgian healthy elderly (Study group ≥ 60 y (n=75): men (n=14); women (n=61). Subgroup <75 y (n=64): men (n=9); women (n=55); Subgroup 75+ y (n=11); men (n=5); women (n=6)). Based on BMI range, the data from 78.7% of participants' were above the norm, 21.3% - normal and 0 - below the norm. The findings of our study have shown that BMI is not height-dependent variable, moreover, BMI as mindex and demiquet is a mass/weight-dependent anthropometric characteristic. BMI, demiquet and mindex had nearly same correlation with other anthropometric parameters (especially for women). High correlation between direct measured/calculated BMI and DE BMI were confirmed by Pearson coefficient 0.979 (p<0.01). The Cohen's Unweighted Kappa index was 0.8518 (0.81 - 1.00 almost perfect or perfect agreement). According to the Paired Samples Test results, there were no difference between BMI and DE BMI. Statistically significant difference between the BMI and DE BMI was determined only for men (0.003 - difference is significant at p<0.05). However, there were only 14 men in our research group and the use of this method recommended in a larger group. Graphical Bland and Altman agreement analysis demonstrated the absence of the trends in differences between BMI and DE BMI. The findings of our research confirm that Demi-span is a reliable measure of stature in Georgian healthy elderly. Demiquet and Mindex are useful indices for our population and suitable for use in clinical practice and in the study of the Georgian elderly when the use of standard weight to height indices may be questionable. Considering to results of statistical analysis, strength of correlation and agreement between direct measured BMI and alternative BMI confirms that the Bassey's equation is valid for Georgian healthy elderly, Demispan equivalent height and BMI is reliable and, accordingly, useful in case of lack of standard calculations or to avoid errors of direct measurements. However, further studies with higher subject numbers are needed to confirm our findings. Additional studies covering other communities in Georgia with different socioeconomic and ethnic composition, also, with different diseases would be necessary to obtain a better anthropometric characterization of the Georgian elderly.