Estimation of energy intake in clinical practice: a comparison between a food record protocol and a precoded food record book.

The aim of this study was to compare energy intake estimated from a clinical food record protocol (CFRP) with that from a precoded food record book (PFRB) as reference method. Food and fluid consumption were recorded in 10 older patients using a CFRP in parallel with a PFRB during a 6-day period. The results showed that there were no significant differences in mean energy intake estimated from the CFRP as compared with that estimated from the PFRB. The correlation coefficient between the calculated daily energy intake from the CFRP and PFRB was 0.96. The differences in energy intake (kcal/day) between the CFRP and PFRB, plotted against their mean value for 10 patients, showed that results were within the limits of agreement (mean +/- 2SD) for nine patients. The differences in each day's energy intake between the two methods plotted against their mean value showed that 97% of the estimated daily energy intake was within the limits of agreement. The weighted kappa between the two methods was 0.76. The CFRP would seem to be acceptable for the estimation of mean energy intake in the hospital setting.