The validity of a food frequency questionnaire in assessing the nutrient intake of New Zealand adults

Abstract

In 1993 a semi-quantitative food frequency questionnaire was designed to assess the dietary intakes of the New Zealand adult population. The aim of this study was to determine the validity of this food frequency questionnaire in assessing the dietary intakes of a sample of eighty Dunedin adults aged 25-49 years.

The performance of the questionnaire was assessed by comparison with seven day diet records as the reference method. The nutrients used in the comparison were the ones of special consideration for New Zealanders according to the Nutrition Taskforce (1991) ie, energy, protein, total fat, saturated and polyunsaturated fatty acids, carbohydrate, total sugars, starch, dietary fibre, calcium, iron, zinc, thiamin and ascorbic acid. Folate and pyridoxine were also measured for completeness. The questionnaire was assessed on its ability to determine nutrient group mean intakes and its ability to correctly classify individuals into quintiles of intake. Two possible analyses of the questionnaire were investigated, one of which involved 'adjusting' the questionnaire for the total number of servings of fruit, vegetables and meat, estimated by each subject, to be consumed on a weekly basis.

The mean nutrient intakes for most nutrients were less than 10% different between the two methods. Thiamin, folate, and ascorbic acid were the only nutrients to produce means differing by more than 30%. Correlations between the nutrient intake values from the diet record and those from the food frequency questionnaire ranged from -0.20 for thiamin to 0.70 for zinc. After adjusting for the total energy intake the correlation coefficients generally improved, though they were still low compared to accepted values. For most nutrients at least 60% of the subjects when classified by the diet record, fell into the same or adjacent quintile when classified by the food frequency questionnaire. For the statistical measures compared, the questionnaire analysis which did not 'adjust' for the total number of servings of fruit, vegetables and meat, produced better results than the 'adjusted' questionnaire.

These data indicate that in an adult sample, a simple semi-quantitative food frequency questionnaire is a valid dietary assessment method in assessing certain aspects of an adults dietary intake.