The validity of the 24 hour recall for estimating the energy and selected nutrient intakes of a group of rural Malawian preschool children

Publisher Website
Google Scholar
Abstract
The validity of the 24‐hour recall (24 HR) for estimating group aggregate and individual daily energy, protein, calcium, iron, zinc and vitamin C intakes of 29 rural Malawian children (4–6 years) was assessed by comparing weighed record (WR) and 24 HR dietary data collected for the same day of food intake. There were no significant differences comparing group median energy and nutrient intakes estimated using the two techniques. The percentage differences between group mean intakes ranged from 1% (vitamin C) to 16% (Ca). In general, these differences were reduced when morning meal intakes were omitted to eliminate the systematic bias affecting estimates of porridge intake. Spearmans and intra‐class correlation coefficients ranged from 0.28 (vitamic C) to 0.55 (Zn), and from 0.14 (vitamin C) to 0.51 (Zn), respectively. Less than 35% of the nutrient intakes estimated using the 24 HR were within ± 10% of those estimated via the WR (range = 12% for vitamin C and Ca to 33% for protein), and over 50% of the subjects consumed more than ± 20% of the 24 HR estimates for most nutrients. Reasons for the low percentage agreement between the WR and 24 HR results included errors in reporting snack food consumption, the use of average recipes, and imprecision in the recall of quantities of main meal dishes consumed. These data suggest that the 24 HR could be substituted for the WR, when estimates to within ± 10% of actual group mean intakes of energy, protein, iron, and zinc are required.