Testing the utility of a 24-hour semiquantitative food frequency questionnaire for estimating fat intake in a free living population

Abstract

Validation of a new dietary assessment instrument requires direct measures of food consumption which can best be achieved in a controlled environment. Although direct observation is the “gold standard” for validation, it does not provide an indication of how well the dietary assessment method performs in a free-living population. A previous study established the validity of a 24-hour semiquantitative food frequency questionnaire (SFFQ) by comparing actual fat intake values obtained by plate waste measures to fat intake values estimated by the SFFQ. This study tests the utility of the SFFQ in a free-living population by comparing estimated fat intake values obtained by randomly administered 24-hour recalls and the SFFQ. The overall mean fat intake estimated by the SFFQ was not significantly different from the overall mean fat intake estimated by 24-hour recalls. However, for female subjects, fat intake estimated by 24-hour recalls was significantly lower than estimates using the SFFQ. Repeated measures of dietary intake using the SFFQ and 24-hour recalls provide estimates of fat intake in the free-living population which are not significantly different. However, the SFFQ is faster and less expensive to administer to large groups of people since it does not require a trained interviwer.