Errors in reported dietary intakes. I. Short-term recall.

Area of Science:

• Nutrition Science
• Epidemiology
• Biostatistics

Background:

• Accurate quantification of nutrient intake is crucial for epidemiological studies on chronic diseases.
• Reproducibility and variability of dietary assessment methods significantly impact study power and sensitivity.
• Existing methods for assessing nutrient intake variability require further evaluation.

Purpose of the Study:

• To quantify the reproducibility and between-subject variability of nutrient intake data using three distinct assessment methods.
• To compare the performance of dietary history, frequency, and typical day methods in assessing nutrient intake.
• To explore the implications of these findings for the design and analysis of nutritional epidemiology studies.

Main Methods:

• Employed analysis of variance to assess within-subject, between-subject, and between-method variation for key nutrients (calories, protein, fat, vitamin A).
• Utilized the intraclass correlation coefficient to measure the ability of each method to distinguish interpersonal variation from within-person error.
• Collected dietary intake data from 44 male and 17 female healthy subjects aged 45-75 years via two interviews conducted 1-3 months apart.

Main Results:

• Dietary history method generally yielded higher nutrient intake estimates compared to frequency and typical day methods.
• The frequency method demonstrated less within-person error and less interpersonal variation for absolute intakes compared to the dietary history method.
• The typical day method proved unreliable for vitamin A due to occasional extremely high, unreproducible intakes.
• Females exhibited less within-person error and smaller interpersonal intake variation than males across all methods.

Conclusions:

• The frequency method offers superior reproducibility and reduced within-person error for nutrient intake assessment compared to dietary history and typical day methods.
• Methodological choices in dietary assessment significantly influence the power and sensitivity of studies investigating diet-disease relationships.
• Findings underscore the importance of selecting appropriate dietary assessment tools to minimize error and enhance the reliability of nutritional epidemiology research.