Is macronutrient composition of dietary intake data affected by underreporting? Results from the EPIC-Potsdam study

Objective: To investigate whether subjects with low reported relative energy intake differ from those with higher relative energy intake according to characteristics such as obesity, physical activity, and macronutrient composition of the diet. Design: Cross-sectional data from a cohort study employing a semi-quantitative food frequency questionnaire (SFFQ). To determine energy intake relative to BMR the ratio of reported energy intake (EI) to BMR was used and categorized by quintiles. Setting: East German (Potsdam) cohort of the EPIC study (European Prospective Investigation into Cancer and Nutrition) Subjects: 2862 women and 2356 men taking part in the EPIC-Potsdam study from January ist to December 31st 1995. Results: A significant declining trend could be observed for BMI, percentage of body fat, and body weight from the lowest to the highest quintile of EI/BMR. BMR was slightly decreasing, whereas physical activity was slightly increasing with quintiles of EI/BMR. Absolute macronutrient intake was directly related to EI/BMR. Percent macronutrient intake indicated lower fat intake, and higher carbohydrate and protein intake in low energy reporters. Energy adjusted macronutrient intake by the residual method showed no dependencies on EI/BMR. Conclusion: Underestimation of energy intake is related to obesity and affects the relation of macronutrients in the reported diet. This implies, that the assumption of adequate ranking of subjects by a SFFQ cannot be maintained. Energy adjusted intake values according to the residual method should be employed in diet-disease risk analysis since they were found to be independent of the methodological influence of underreporting.