Retrospective versus original information on diet: implications for epidemiological studies

Area of Science:

• Epidemiology
• Nutritional Science
• Public Health

Background:

• Accurate dietary assessment is crucial for epidemiological studies investigating diet-related diseases.
• Recall bias in dietary questionnaires can affect the reliability of findings.
• Understanding the impact of time on dietary recall is essential for study design.

Purpose of the Study:

• To evaluate the accuracy of retrospective dietary information compared to original reports.
• To assess how time influences the reliability of dietary recall in a case-control study setting.
• To identify potential biases in dietary assessment for epidemiological research.

Main Methods:

• A case-control study involving 131 hospital controls in Stockholm, Sweden.
• Participants completed a second postal questionnaire on food consumption four years after the initial assessment.
• Dietary intake data were dichotomized for comparative analysis.

Main Results:

• Retrospective dietary information overestimated previous food consumption in subjects with increased intake.
• Retrospective dietary information underestimated previous food consumption in subjects with decreased intake.
• High agreement was observed for subjects with unchanged food consumption; age influenced recall accuracy.

Conclusions:

• Retrospective dietary assessment introduces potential for differential misclassification in epidemiological studies.
• Changes in food consumption over time can significantly bias study results.
• Researchers must consider the temporal impact on dietary recall when designing or analyzing studies.