Studies rely on accurate measurements of nutrient intakes in order to correctly understand potential associations. Problems with diet recall are well known and in this respect progress has been made in designing food intake questionnaires and food diary techniques that allow more accurate recording. Studies also often use food data tables to estimate nutrient intakes, however these can also provide inaccurate data. The main problems with data tables is that they are unable to reflect changes to the nutrient composition of food and are therefore reliant on the original collection of data being representative of the nutrient content as a whole. Because of these problems researchers have compared the nutrient contents of foods from established data tables with more recent analytical measuiurements. Evidence suggests that for some nutrients the established data tables may contain inaccurate data.
For example, researchers1 have compared the vitamin C content of foods in the US Department of Agriculture’s Revised Handbook Number 8 and the Health Habits and History Questionnaire (HHHQ) with more recent high pressure liquid chromatographical (HPLC) analysis of foods obtained in the Washington DC area between 1988 and 1989. The HPLC values obtained from the foods were lower than the data base suggested for orange juice, grapefruit, tomatoes, tomato juice and potatoes; which were the five food items with the highest vitamin C content. Of the 24 food in the study, 13 had lower values than both the data tables suggested with some foods such as tomatoes and broccoli having less than half the vitamin C content. When the HPLC values were substituted into two studies based on the HHHQ data, the subsequent intake of vitamin C was reduced for the subjects.
These results suggest that certain foods may have had their vitamin C content overestimated in established data tables, based on data collected with more recent analytical techniques. It is tempting therefore to speculate that other vitamins and minerals within certain foods may also have been inaccurately reported within these tables. One problem with vitamin C is that some of the data reported in the HHHQ and US Department of Agriculture tables may be based on studies that did not differentiated between reduced ascorbic acid and dehydroascorbic acid. If certain populations base their vitamin C intake on the values of a small selection of the foods that have been shown to be overestimated in tables, their vitamin C intake my be much lower than predicted. The author’s analysis of two previous studies using revised data and finding lower intakes supports this contention.
RdB
