Accuracy of reporting food energy intake: influence of ethnicity and body weight status in South African women

The current study sought to identify characteristics that may be associated with the misreporting of food energy intake (EI) in urban South African women. A total of 198 women (61 black, 76 of mixed ancestry, 61 white) completed a quantified food frequency questionnaire, from which daily energy and macronutrient intake were calculated. Body composition (body mass index [BMI], percentage of body fat), body image (Feel-Ideal Difference index and Body Shape questions) and socio-economic status (SES) (household density and asset index) were also measured. Food EI in relation to estimated basal metabolic rate ratio that was less than 1.05 represented under-reporting, whereas a ratio greater than 2.28 represented over-reporting. Results suggested that 26% of the participants under-reported, 64% adequately reported and 10% over-reported. Participants who under-reported had a higher BMI (p < 0.01) and higher percentage of body fat (p < 0.05) than those who adequately and over-reported. The majority of under-reporters were black (38%) versus 21% under-reporters of mixed ancestry and 20% white under-reporters (p < 0.01). Eighty-three per cent of black under-reporters were obese. On the other hand, a majority (63%) of overweight women of mixed ancestry and a majority (50%) of white normal-weight women under-reported their food EI. Under-reporters reported a lower intake of dietary fat (p < 0.01) and a higher intake of dietary protein (p < 0.01) than adequate or over-reporters. Food EI reporting was not influenced by SES or body image. In conclusion, results suggest that food EI reporting is influenced by body size, and may be ethnic-specific in South African women.