Food composition table-based estimation of energy and major nutrient intake in comparison with chemical analysis: a validation study in Korea.

This study was initiated to examine the accuracy of conventional food composition table-based estimation of intakes of energy, protein, lipid and carbohydrate, in comparison with chemical analysis. For this purpose, 66 women (at the ages of 29 to 54 years) in three locations in Jeju Island, Republic of Korea, volunteered to offer 24-hour food duplicate samples. A half of them were house-wives, and the remaining half were farmers or fishers. The duplicate samples were subjected 1) to the chemical analysis for daily intake of energy, protein, lipid and carbohydrate after official methods in Korea (measured values), and 2) to the estimation of intakes of the same items taking advantage of Korean Food Composition Tables (estimated values). The two sets of the results, i.e., the measured and estimated values, were compared by paired and unpaired t-test, and linear regression analysis. The estimated values correlated closely with the measured values, irrespective of energy or the three major nutrients. A close agreement was observed for energy intake (the estimated/measured ratio of > 98%), and it was also the case for protein intake (101%). Under- and over-estimation was observed, however, in regard to carbohydrate (by - 8%) and lipid intake (by + 24%), respectively. It was concluded that the Korean Food Composition Tables are sufficiently accurate when applied for estimation of total energy intake as well as protein intake. Cares should be taken, however, in applying the tables for estimation of lipid and carbohydrate intake, because there may be the risk of over- and under-estimation for the former and the latter, respectively.