Relative validity of a short food frequency questionnaire for disadvantaged families in Hong Kong

Abstract

Individuals with lower socioeconomic status are more vulnerable in securing good nutritional quality. It was also found that people who had received a lower education level had greater difficulty in completing the conventional dietary assessment such as a food frequency questionnaire (FFQ). Previous studies have demonstrated the validity of a short FFQ in Hong Kong’s pregnant women, but its validity among a wider community was still unknown. For the present study, we aimed to validate a short FFQ among disadvantaged communities in Hong Kong. Amongst 103 individuals participating in a dietary intervention programme, their dietary data were collected by FFQs and three-day dietary records. Relative validity was assessed by correlation analysis, cross-tabulation, one-sample t-test, and linear regression. In general, water and total energy intake had significant correlations (0.77 for crude water intake and 0.87 for crude total energy intake) between values reported by FFQ and dietary records, good agreement (both with over 50% of observations falling into the same quartile), and insignificant differences between assessment methods reported by one-sample t-test and linear regression. Meanwhile, several nutrients had good agreement in terms of the values reported by FFQ and dietary records, such as energy from total fat, carbohydrates, total fat, cholesterol, phosphorus, and potassium. The results of this study demonstrated that the short version FFQ could be a convenient assessment tool of multiple dietary behaviors, especially in total energy and water intakes.