Return to publication list

Reproducibility of a short semi-quantitative food group questionnaire and its performance in estimating nutrient intake compared with a 7-day diet diary in the Million Women Study.

Roddam Andrew W-1, Spencer Elizabeth-1, Banks Emily-2, Beral Valerie-1, Reeves Gillian-1, Appleby Paul-1, Barnes Isobel-3, Whiteman David C-4, Key Timothy J-1,

Public Health Nutrition 2005; 8 :201-213.

1-Cancer Research UK Epidemiology Unit, Gibson Building, Radcliffe Infirmary, Oxford OX2 6HE UK
2-National Centre for Epidemiology and Population Health, Australian National University, Canberra, Australia
3-School of Applied Statistics, University of Reading, Reading, UK
4-Division of Population Studies and Human Genetics, Queensland Institute of Medical Research, Brisbane, Australia


Objectives To assess the short- and long-term reproducibility of a short food group questionnaire, and to compare its performance for estimating nutrient intakes in comparison with a 7-day diet diary.

Design Participants for the reproducibility study completed the food group questionnaire at two time points, up to 2 years apart. Participants for the performance study completed both the food group questionnaire and a 7-day diet diary a few months apart. Reproducibility was assessed by kappa statistics and percentage change between the two questionnaires; performance was assessed by kappa statistics, rank correlations and percentages of participants classified into the same and opposite thirds of intake.

Setting A random sample of participants in the Million Women Study, a population-based prospective study in the UK.

Subjects In total, 12 221 women aged 50-64 years.

Results In the reproducibility study, 75% of the food group items showed at least moderate agreement for all four time-point comparisons. Items showing fair agreement or worse tended to be those where few respondents reported eating them more than once a week, those consumed in small amounts and those relating to types of fat consumed. Compared with the diet diary, the food group questionnaire showed consistently reasonable performance for the nutrients carbohydrate, saturated fat, cholesterol, total sugars, alcohol, fibre, calcium, riboflavin, folate and vitamin C.

Conclusions The short food group questionnaire used in this study has been shown to be reproducible over time and to perform reasonably well for the assessment of a number of dietary nutrients..

Keywords Cross-Sectional Studies; Diet Records; Diet Surveys; Female; Food Habits; Great Britain; Humans; Middle Aged; Prospective Studies.