Online Program

Measurement error correction in longitudinal dietary intervention studies in the presence of nonignorable missing data (306770)

Keywords: 24-hour dietary recall, multiple imputation, dropout, sodium intake, biomarker

In lifestyle intervention trials, where the goal is to modify a participant’s diet, self-reported diet is a longitudinal outcome variable that is subject to measurement error. An additional analytic challenge in these studies is participant dropout that is likely related to the intensive nature of lifestyle interventions that often include self-monitoring, following a prescribed diet, coaching, and/or changing other health behaviors. We propose a statistical framework for correcting for measurement error in longitudinal self-reported dietary data by combining intervention data with auxiliary data from an external biomarker validation study where both self-reported and recovery biomarkers of dietary intake are available. In this setting, dietary intake measured without error in the intervention trial is missing data and multiple imputation is used to fill in the missing measurements. We use sensitivity analyses to address the influence of unverifiable assumptions involving dropout and its association with the measurement error process. We apply our methods to self-reported sodium intake from the PREMIER study, a multi-component lifestyle intervention trial.