Dietary assessment tools should be designed for the target population. We developed an FFQ designed to assess diet in South Asian women in Norway. The study objective was to evaluate this FFQ using 24-h dietary recalls as reference method.

Approximately 3 weeks after the participants (n 40) had filled in the FFQ, the first of three non-consecutive 24-h dietary recalls was completed. The recalls were telephone-based, unannounced and performed by a trained dietitian, with 2–3 weeks between each interview.

The DIASA 1 study, in Oslo, Norway.

Women of South Asian ethnic origin participating in the DIASA 1 study were invited to participate in the evaluation study.

The WebFFQasia significantly overestimated the absolute intake of energy, protein, fat and carbohydrates compared with the 24-h dietary recalls. Absolute intakes of sugar, starch and fibre did not differ significantly between the methods. For energy percentages (E%), there were no significant differences, except for monounsaturated fat. Correlations were strong for E% from sugar and saturated fat and moderate for E% from fibre, carbohydrate, total fat and protein. Fourteen food groups out of twenty three were not significantly different compared with the reference method, and sixteen groups showed strong to moderate correlations.

The WebFFQasia may be used to assess E% from habitual diet and can adequately estimate intakes and rank participants according to nutrient intake and main food categories at group level.