Implications of Self-Reported Dietary Intake Measured with Error in Longitudinal Lifestyle Intervention Trials

• Activity Number: 246
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 2:00 PM to 5:50 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Juned Siddique*
• Companies: Northwestern University
• Keywords: measurement error ; simulation ; longitudinal ; intervention ; clinical trial

Abstract

Lifestyle intervention studies often use self-reported measures of diet, and obtaining accurate measurement of diet and its change overtime is a major challenge due to measurement error. Measurement error in intervention studies can result in biased estimates of the treatment effect in addition to reduced power to detect treatment effects. Reasons for this measurement error include memory limitations and poor quantification of portion sizes. Dietary intervention studies involve repeated dietary assessments over time and produce additional measurement issues. Repeated monitoring may induce participants to change their reporting behavior in response to the burden of the assessment procedure. Participants may also modify their reporting behavior during the monitoring period to appear compliant with the dietary recommendations of the intervention (i.e. compliance bias). Using simulations motivated by data from existing studies, we investigate the implications of longitudinal outcomes measured with error in treatment studies. We focus on settings where measurement error is differential, with respect to both treatment and time.