|
Activity Number:
|
491
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 7, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
SSC
|
| Abstract - #300264 |
|
Title:
|
Analysis of Survey-Based Usual Intake Nutrition Data: The Issue of Within-Person Variation
|
|
Author(s):
|
Kevin Dodd*+ and Joseph Goldman
|
|
Companies:
|
National Cancer Institute and U.S. Department of Agriculture
|
|
Address:
|
6130 Executive Blvd, Rockville, MD, 20892,
|
|
Keywords:
|
24-hour recall ; usual intake ; nonlinear mixed model ; balanced repeated replication
|
|
Abstract:
|
Statistical methods to estimate usual (i.e., long-run average) daily intake of a food or nutrient are required to properly conduct dietary surveillance. The dietary assessment portion of the NHANES includes up to two 24-hour recalls (24HR) per person. Day-to-day (within-person) variability in diet ensures that 24HRs measure usual intake with substantial error. For dietary components not consumed daily, 24HR reports may be zero, even for individuals with positive usual intake. Distributions of measured and usual intake are often very skewed. We present a nonlinear mixed model for usual intake and demonstrate its use in estimating the population distributions of usual intake for major food groups and nutrients. The balanced repeated replication method was used to approximate the standard errors of estimated quantiles of usual intake.
|
