Energy, protein and fat intake estimated using a food recording electronic device compared with a weighed diary.

Twenty nine subjects kept a weighed diet diary for 7 days. This was coded using both detailed food composition tables and food groups. During the 7 days a Food Recording Electronic Device (FRED) with food group keyboard labelling was also used. This group of motivated subjects had no difficulty in using the device, and the use of food groups was not a source of error. Some technical problems were identified which led to a significant systematic underestimation of energy, protein and fat determinations using the FRED, although the correlation coefficients between the two methods were high. The technical problems were of a type which can be eliminated in future models of the FRED.