Linear Mixed Models for Comparing Dynamic Treatment Regimens on a Longitudinal Outcome in Sequentially Randomized Trials

TL;DR

This paper introduces a mixed effects modeling approach to compare dynamic treatment regimens in longitudinal outcomes within SMARTs, aiding personalized treatment strategies.

Contribution

It develops a novel mixed effects model specifically designed for comparing DTRs in longitudinal data from SMARTs, which was not previously available.

Findings

Method successfully applied to autism SMART data.

Provides a framework for marginal mean comparison of DTRs.

Enhances analysis of sequential treatment strategies.

Abstract

A dynamic treatment regimen (DTR) is a pre-specified sequence of decision rules which maps baseline or time-varying measurements on an individual to a recommended intervention or set of interventions. Sequential multiple assignment randomized trials (SMARTs) represent an important data collection tool for informing the construction of effective DTRs. A common primary aim in a SMART is the marginal mean comparison between two or more of the DTRs embedded in the trial. This manuscript develops a mixed effects modeling and estimation approach for these primary aim comparisons based on a continuous, longitudinal outcome. The method is illustrated using data from a SMART in autism research.