Authors Challenge Memory-Based Dietary Assessments

Dietary recall, food frequency questionnaires, and other memory-based dietary assessment methods (M-BMs) are “pseudoscientific” and shouldn’t be used to set dietary guidelines and policies, concludes a special article in Mayo Clinic Proceedings.

Edward Archer, PhD, of the University of Alabama at Birmingham and colleagues evaluated the validity of using M-BMs for nutrition surveillance and epidemiologic nutrition research. Their critique focuses on the “What We Eat in America” and National Health and Nutrition Examination Surveys (WWEIA/NHANES), which relied on 24-hour recall and food-frequency questionnaires to assess diet.

The authors cite “many decades of evidence demonstrating that M-BMs have severe, intractable systematic biases that render the data implausible and, therefore, invalid.” Not only is it “indisputably false” that human memory can accurately or precisely reproduce past consumption, but M-BM protocols mimic procedures designed to produce false recall, they write.

The memories on which M-BM data are based cannot be independently confirmed or refuted; “as such, these data are pseudoscientific and inadmissible in scientific research,” according to the authors. They add that failure to measure and control for physical activity, cardiorespiratory fitness, and other confounders leads to equivocal inferences about the relationship between diet and health.

On the basis of this “overwhelming evidence,” Archer and colleagues conclude, “M-BM data cannot be used to inform national dietary guidelines.” They believe that continued funding of projects using these methods, such as WWEIA/NHANES, “constitutes an unscientific and major misuse of research resources” [Archer E, et al. The inadmissibility of What We Eat in America and NHANES dietary data in nutrition and obesity research and the scientific formulation of national dietary guidelines. Mayo Clin Proc 2015; doi: 10.1016/j.mayocp.2015.04.009].