Development of an interview administered food-frequency questionnaire for use amongst women of South Asian ethnic origin in Britain

T. Kassam-Khamis, Aga Khan University
K. Nanchahal, London School of Hygiene and Tropical Medicine
P. Mangtani, London School of Hygiene and Tropical Medicine
I. dos Santos Silva, London School of Hygiene and Tropical Medicine
A. McMichael, London School of Hygiene and Tropical Medicine
A. Anderson, University of Glasgow Royal Infirmary

Abstract

Objectives: To develop an interview-administered food-frequency questionnaire (FFQ) to assess usual long-term intake of foods and nutrients amongst women from the largest South Asian communities residing in Britain.

Methods: A list of foods and dishes consumed by these South Asian communities was compiled from data collected in previous studies. Foods which contributed to interindividual variation in the intake of the various macronutrients or were rich in fat or fibre were included in the FFQ. Portion size for traditional foods were estimated using sets of eight colour photographs. A nutrient database was modified to make it appropriate for South Asian diets. The reliability of the FFQ was assessed in a group of volunteer South Asian women (n=14) of differing regional and religious backgrounds. A preliminary validation exercise was conducted in a group of Punjabi Muslim women (n=11) who had kept 7-day weighed records 2 years previously. ]

Results: The reliability of this FFQ was high, with most women being classified in the same tertile for energy and macronutrient intake in the two interviews. The main sources of energy and macronutrients identified by the FFQ were similar to those identified by the weighed records but the average nutrient intakes estimated by the FFQ were slightly higher. The level of agreement between the two dietary methods, measured by the ability of the FFQ to classify women in the same tertile of intake as the weighed records, was high for percentage energy intake from fat, protein and carbohydrates, but less so for absolute nutrient intakes. These results were, however, limited because of the small sample size.

Conclusions: This is the first FFQ specifically designed for South Asian communities in Britain. Despite the diversity of diets, these preliminary findings suggest it will be a useful and easy to administer tool in chronic disease epidemiology to obtain standardized information on long-term usual dietary intake from these communities. A more comprehensive validation of this FFQ is now underway.