Use of two-part regression calibration model to correct for measurement error in episodically consumed foods in a single-replicate study design: EPIC case study.

George O Agogo,Pietro Ferrari,Max Leenders,Sven Knüppel,Ingegerd Johansson,Fred A Van Eeuwijk,Emilio Sánchez-Cantalejo,Tonje Braaten,Christina Bamia,Hendriek Boshuizen,David C Muller,Pieter Van’T Veer,Hilko Van Der Voet,Zhong-Ke Gao

doi:10.1371/journal.pone.0113160

Abstract

In epidemiologic studies, measurement error in dietary variables often attenuates association between dietary intake and disease occurrence. To adjust for the attenuation caused by error in dietary intake, regression calibration is commonly used. To apply regression calibration, unbiased reference measurements are required. Short-term reference measurements for foods that are not consumed daily contain excess zeroes that pose challenges in the calibration model. We adapted two-part regression calibration model, initially developed for multiple replicates of reference measurements per individual to a single-replicate setting. We showed how to handle excess zero reference measurements by two-step modeling approach, how to explore heteroscedasticity in the consumed amount with variance-mean graph, how to explore nonlinearity with the generalized additive modeling (GAM) and the empirical logit approaches, and how to select covariates in the calibration model. The performance of two-part calibration model was compared with the one-part counterpart. We used vegetable intake and mortality data from European Prospective Investigation on Cancer and Nutrition (EPIC) study. In the EPIC, reference measurements were taken with 24-hour recalls. For each of the three vegetable subgroups assessed separately, correcting for error with an appropriately specified two-part calibration model resulted in about three fold increase in the strength of association with all-cause mortality, as measured by the log hazard ratio. Further found is that the standard way of including covariates in the calibration model can lead to over fitting the two-part calibration model. Moreover, the extent of adjusting for error is influenced by the number and forms of covariates in the calibration model. For episodically consumed foods, we advise researchers to pay special attention to response distribution, nonlinearity, and covariate inclusion in specifying the calibration model.

Highlights

Dietary variables are often measured with error in nutritional epidemiology
To fill the aforementioned gaps, we developed a two-part regression calibration model to adjust for the bias in the dietdisease association, due to measurement error in self-reported episodically consumed foods, when each subject in the calibration sub-study has only a single reference measurement
These slopes of the fitted lines were all close to the theoretical value of 1 for a generalized linear modeling (GLM) gamma model

Summary

Introduction

Dietary variables are often measured with error in nutritional epidemiology. In such studies, usual dietary intake is assessed with instruments such as, food frequency questionnaire and dietary questionnaire [1,2,3]. Usual dietary intake is assessed with instruments such as, food frequency questionnaire and dietary questionnaire [1,2,3] In these instruments, the queried period of intake ranges from several months to a year. The queried period of intake ranges from several months to a year As a result, these instruments are prone to error caused by difficulties to recall past intake of foods or food groups, the frequency of consumption, and the portion size. The random error attenuates the association between dietary intake and disease occurrence, whereas systematic error can either attenuate or inflate the association

Methods

Results

Conclusion