Comparative analysis of two FFQ

Abstract

Objective: To examine the utility of a shorter FFQ compared with a longer FFQ, both of which are commonly used in Australia.

Design: Comparative study.

Setting: Community setting.

Subjects: One hundred and fifty-nine men (mean 55 (sem 7) years) screened for participation in an intervention study completed both the Commonwealth Scientific and Industrial Research Organisation FFQ and the Cancer Council of Victoria FFQ. Agreement between both questionnaires was assessed according to Bland-Altman plots and limits of agreement (LOA) and ordinary least products regression to test for the presence of fixed and proportional bias.

Results: There was good relative agreement between the methods for energy and macronutrients (Pearson's correlation coefficients: energy r = 0.7, protein r = 0.6, fat r = 0.8, carbohydrate r = 0.7, alcohol r = 0.8; P < 0.01). Mean group-level agreement for the majority of nutrients (70 %) fell between 80 % and 110 %. According to the criteria used (maximum LOA was 50-200 % and no significant proportional bias), there was acceptable agreement between the FFQ for energy and total saturated and monounsaturated fat, but not for protein, carbohydrate and fibre. Micronutrients that did not meet the agreement criteria including calcium, iron, thiamin, niacin, riboflavin and folate. When the data were analysed according to quintiles, the majority of subjects were either in exactly the same quintile or within one quintile for most nutrients, and 1-2 % were grossly misclassified by three or four quintiles.

Conclusions: We conclude that there is sufficient agreement between the instruments for group-level comparisons in men, but they are not interchangeable for estimation of individual intakes.