Food-frequency questionnaires: a review of their design, validation and utilisation

Area of Science:

• Nutrition science
• Epidemiology
• Dietary assessment methodology

Background:

• Food-frequency questionnaires (FFQ) are widely used for dietary assessment.
• Existing literature on FFQ design, utilization, and validation requires synthesis.
• Understanding optimal FFQ characteristics is crucial for accurate dietary research.

Purpose of the Study:

• To conduct a semi-systematic literature review on FFQ design, utilization, and validation.
• To identify key factors influencing FFQ accuracy and application.
• To develop expert consensus and recommendations for FFQ development and use.

Main Methods:

• Searched databases for FFQ studies from 1980 to 1999.
• Reviewed 227 validation studies and 164 utilization studies.
• Evaluated design parameters like portion size description, administration method, and number of items.
• Assessed validation techniques, comparing FFQ to other methods and biomarkers.

Main Results:

• Larger FFQs and interviewer administration generally yielded higher nutrient correlations.
• Self-defined portion sizes showed higher mean correlations.
• Calcium and fat showed highest correlation coefficients; Vitamin A and vegetables showed lowest.
• FFQs were predominantly used in cross-sectional surveys, often disease-specific.

Conclusions:

• FFQ design significantly impacts data quality; larger, interviewer-administered questionnaires with clear portion size descriptions are preferable.
• Current validation practices often rely on correlation coefficients, with limited use of biomarkers.
• Expert consensus provides recommendations for optimizing FFQ development, validation, and application in nutritional research.