Statistical issues related to semiquantitative food-frequency questionnaires

Area of Science:

• Epidemiology
• Nutritional Science
• Biostatistics

Background:

• Semiquantitative food-frequency (SQFF) questionnaires are widely used in epidemiological studies for dietary data collection.
• Their self-administered nature and minimal time commitment make them attractive for large-scale research.
• Concerns exist regarding the statistical validity and application of data obtained through these methods.

Purpose of the Study:

• To critically evaluate the validation status of SQFF methods in dietary assessment.
• To examine the suitability of SQFF data for comparing population mean intakes and estimating usual nutrient intake distributions.
• To discuss statistical issues and potential misuse of dietary data from SQFF questionnaires.

Main Methods:

• Review of statistical issues pertinent to dietary data from frequency questionnaires.
• Analysis of data from multiple validation studies to assess measurement error.
• Illustrative examples to clarify potential problems in statistical analyses.

Main Results:

• SQFF methods may not be as well-validated as commonly suggested by investigators.
• Measurement error inherent in SQFF data can significantly impact statistical analyses.
• Potential limitations exist for using SQFF data to accurately compare mean intakes or estimate usual intake distributions across populations.

Conclusions:

• The statistical validity and appropriate application of SQFF dietary data require careful consideration.
• Researchers must be aware of measurement error and its consequences for epidemiological findings.
• Guidelines are needed to prevent the misuse of SQFF-derived dietary information in scientific research.