A Measurement Error Approach to Assess the Association Between Diet Diversity, Nutrient Intake, and Mean Probability of Adequacy

• Activity Number: 193
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Social Statistics Section
• Author(s): Maria L. Joseph*+ and Alicia Carriquiry
• Companies: Iowa State University and Iowa State University

Abstract

Costs associated with the extensive methods involved in collecting nutrient intake information, make it inaccessible to resource-poor countries. Quantifying the association between usual nutrient intake/adequacy and more simple and obtainable measures of diet diversity, called food group diversity indexes (FGIs), would allow inferences to be made about the nutritional status of population-level micronutrient adequacy at a fraction of the cost. By implementing Fuller's error-in-the-equation measurement error model (EEM) and simple linear models, we assess this association while accounting for the error in the observed quantities. Through simulation we support the use of the EEM on such data. Significant associations are detected between usual FGI and usual calcium intake for some of the indexes when the more complex EEM is used. The simple linear model also detects significant associations between the mean probability of adequate nutrient intake and FGI, as well the best linear unbiased predictor of these measures.