DIETQ is a semi-quantitative food frequency questionnaire. Questions are asked about consumption of a large number of food items: breads, breakfast cereals, meats/meat products, fish, eggs, cheese, milk, cream, biscuits, cakes, puddings, confectionery/snack items, fruits, vegetables, fat spreads, cooking fats/oils, tea/coffee, sugar, soft drinks and alcoholic drinks. For many items, questions are included about the amount consumed: breads, eggs, cheese, milk, cream, fruits, fat spreads, tea/coffee, sugar and alcoholic drinks. For other items, the nutritional analysis program uses average portion sizes.

Nutrient intakes calculated from the questionnaire have been compared with those calculated from 7-day weighed food intake records in 722 adults. Mean daily nutrient intakes of the group were almost identical. Correlation coefficents were reasonably high for most nutrients, as shown in the following table. This indicates a good level of agreement between the two methods.

Nutrient	Correlation Coefficient
Energy (MJ)	0.47
Protein (g)	0.36
Total Fat (g)	0.44
Saturates (g)	0.58
P/S ratio	0.77
Starch (g)	0.52
Total Sugars (g)	0.59
Fibre (g)	0.62
Vitamin C (mg)	0.53
Alcohol (g)	0.74

The table below shows the percentage of subjects classified in the same third for each nutrient distribution and the percentage classified in opposite thirds. Classification of subjects is reasonably good for the food intake questionnaire, being better for some nutrients than for others, i.e. a reasonably high proportion are classified in the same third of the distribution by the questionnaire and only a small proportion are classified in opposite thirds.

Nutrient	% in Same Third	% in Opposite Third
Energy (MJ)	47	9
Protein (g)	48	13
Total Fat (g)	47	11
Saturates (g)	56	7
P/S ratio	67	1
Starch (g)	52	8
Total Sugars (g)	52	7
Fibre (g)	56	5
Vitamin C (mg)	46	11
Alcohol (g)	65	4

As with any food intake questionnaire, its validation depends not only on the questionnaire itself, but also upon how the responses are converted to nutrient intakes.

Ideal uses for the questionnaire are when conducting case/control comparisons or examining the effect of dietary advice. In such analyses, nutrient intakes calculated from the questionnaire are likely to be exactly the same as those obtained from weighed intake records (Fehily et al, 1987, Fehily et al, 1988; Burr et al, 1989). In other cases, the usefulness of the questionnaire will depend upon the nutrient of interest, being better for some nutrients than for others (Broadfield et al, 2003; Bolton-Smith et al, 1991; Fehily et al, 1988).

REFERENCES

Bolton-Smith C et al. Brit J Nutr. 1991;65:337-346.
Broadfield E et al. Brit J Nutr 2003;90:215-220.
Burr M et al. Eur Heart J 1989;10:558-567.
Fehily A et al. Food Sci Nutr 1988;42F:77-78.
Fehily A et al. Hum Nutr: Appl Nutr 1987;41A:319-326.