An efficient multiple imputation algorithm for control-based and delta-adjusted pattern mixture models using SAS

TL;DR

This paper introduces a simple, efficient imputation algorithm for control-based and delta-adjusted pattern mixture models in clinical trial analysis, implemented easily in SAS, improving handling of missing data in longitudinal studies.

Contribution

The paper presents a novel, straightforward imputation method for PMMs that leverages MMRM, compatible with standard software, and modifies copy reference procedures to prevent bias after dropout.

Findings

The algorithm is easily implemented in SAS PROC MI.

It effectively imputes missing data in clinical trial models.

Modified copy reference reduces bias post-dropout.

Abstract

In clinical trials, mixed effects models for repeated measures (MMRM) and pattern mixture models (PMM) are often used to analyze longitudinal continuous outcomes. We describe a simple missing data imputation algorithm for the MMRM that can be easily implemented in standard statistical software packages such as SAS PROC MI. We explore the relationship of the missing data distribution in the control-based and delta-adjusted PMMs with that in the MMRM, and suggest an efficient imputation algorithm for these PMMs. The unobserved values in PMMs can be imputed by subtracting the mean difference in the posterior predictive distributions of missing data from the imputed values in MMRM. We also suggest a modification of the copy reference imputation procedure to avoid the possibility that after dropout, subjects from the active treatment arm will have better mean response trajectory than subjects who stay on the active treatment. The proposed methods are illustrated by the analysis of an antidepressant trial.