|
Activity Number:
|
340
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #305582 |
|
Title:
|
Measurement Error Correction Methods Based on Cumulative Average Intake for Nutritional Data
|
|
Author(s):
|
Weiliang Qiu*+ and Bernard Rosner
|
|
Companies:
|
Harvard Medical School and Harvard Medical School
|
|
Address:
|
181 Longwood Ave., Boston, MA, 02115,
|
|
Keywords:
|
measurement error ; regression calibration ; nutritional data
|
|
Abstract:
|
In nutritional studies, surrogate measures such as the food frequency questionnaire (FFQ) often are measured with systematic and random error. Rosner et al. (1989) derived interval estimates to correct for measurement error at one time point based on regression calibration. In this article, we extend this approach to the case where FFQs are available at several time points. We build a logistic regression model to relate disease to cumulative average intake based on six FFQs obtained between 1980--1998 and propose a method to correct the RRs and CIs for measurement error using multiple imputation to handle missing FFQs at particular time points. The approach is illustrated with data from the Nurses' Health Study, relating incident breast cancer between 1980--2000 to calorie-adjusted saturated fat intake.
|
