Estimation of food and macronutrient intake by household measures

Nine healthy women aged 21–57 years, ranging in weight from lean to obese, took part in a 12-d metabolic study on energy metabolism. They were provided with two different dietary regimes: a 'Standard' diet with 40% energy from fat and 45% energy from carbohydrate and a 'NACNE' diet with 30% energy from fat and 55% energy from carbohydrate. During both dietary periods they received a 3-d rotating menu. For 3 of the 6 days on each diet, all food and drink consumed was recorded by the women in units based on household measures. This record was used to estimate energy intake based on food composition tables and the subjects' assessment of the portion sizes in served meals. Estimated energy intake by the group averaged 97% of actual intake (range 83% to 112%) on the Standard diet, and 97% (range 67% to 115%) on the NACNE diet. There was a tendency to overestimate carbohydrate intake and underestimate fat intake on both diets. Protein was overestimated on the NACNE diet. There was little evidence of a relationship between body mass index (BMI) of subjects and the accuracy of estimation of total energy intake.