Multiple Imputation Approaches for Epoch-level Accelerometer data in Trials

TL;DR

This paper introduces epoch-level missing data classification and multiple imputation methods for accelerometer data in clinical trials, improving bias reduction and data utilization.

Contribution

It presents novel epoch-level missingness classification and two MI approaches, with the non-parametric method showing reduced bias in treatment effect estimates.

Findings

Non-parametric MI reduces bias in treatment effect estimates.

Epoch-level missingness classification improves data handling.

Application to PACE-UP Trial demonstrates practical utility.

Abstract

Clinical trials that investigate interventions on physical activity often use accelerometers to measure step count at a very granular level, often in 5-second epochs. Participants typically wear the accelerometer for a week-long period at baseline, and for one or more week-long follow-up periods after the intervention. The data is usually aggregated to provide daily or weekly step counts for the primary analysis. Missing data are common as participants may not wear the device as per protocol. Approaches to handling missing data in the literature have largely defined missingness on the day level using a threshold on daily wear time, which leads to loss of information on the time of day when data are missing. We propose an approach to identifying and classifying missingness at the finer epoch-level, and then present two approaches to handling missingness. Firstly, we present a parametric approach which takes into account the number of missing epochs per day. Secondly, we describe a non-parametric approach to Multiple Imputation (MI) where missing periods during the day are replaced by donor data from the same person where possible, or data from a different person who is matched on demographic and physical activity-related variables. Our simulation studies comparing these approaches in a number of settings show that the non-parametric approach leads to estimates of the effect of treatment that are least biased while maintaining small standard errors. We illustrate the application of these different MI strategies to the analysis of the 2017 PACE-UP Trial. The proposed framework of classifying missingness and applying MI at the epoch-level is likely to be applicable to a number of different outcomes and data from other wearable devices.