Comparative Analysis of Approaches for Assessing Energy Intake Underreporting by Female Bariatric Surgery Candidates

Abstract

Objective: To test six variations in the Goldberg equation for evaluating the underreporting of energy intake (EI) among obese women on the waiting list for bariatric surgery, considering variations in resting metabolic rate (RMR), physical activity, and food intake levels in group and individual approaches.Methods: One hundred obese women aged 20 to 45years (33.3 6.08) recruited from a bariatric surgery waiting list participated in the study. Underreporting assessment was based on the difference between reported energy intake, indirect calorimetry measurements and RMR (rEI:RMR), which is compatible with the predicted physical activity level (PAL). Six approaches were used for defining the cutoff points. The approaches took into account variances in the components of the rEI:RMR = PAL equation as a function of the assumed PAL, sample size (n), and measured or estimated RMR.Results: The underreporting percentage varied from 55% to 97%, depending on the approach used for generating the cutoff points. The ratio rEI:RMR and estimated PAL of the sample were significantly different (p = 0.001). Sixty-one percent of the women reported an EI lower than their RMR. The PAL variable significantly affected the cutoff point, leading to different proportions of underreporting. The RMR measured or estimated in the equation did not result in differences in the proportion of underreporting. The individual approach was less sensitive than the group approach.Conclusion: RMR did not interfere in underreporting estimates. However, PAL variations were responsible for significant differences in cutoff point. Thus, PAL should be considered when estimating underreporting, and even though the individual approach is less sensitive than the group approach, it may be a useful tool for clinical practice.