Subgroup detection in linear growth curve models with generalized linear mixed model (GLMM) trees

TL;DR

This paper extends GLMM trees to longitudinal data, enabling subgroup detection in growth curve models with improved accuracy, speed, and flexibility over existing methods, demonstrated through simulations and real data.

Contribution

The paper introduces an extension of GLMM trees for longitudinal data, enhancing subgroup detection in growth curve models with better performance and computational efficiency.

Findings

Extended GLMM trees outperform original algorithms and LongCART.

Extended GLMM trees are as accurate as SEM trees.

They are faster and more flexible in modeling time series.

Abstract

Growth curve models are popular tools for studying the development of a response variable within subjects over time. Heterogeneity between subjects is common in such models, and researchers are typically interested in explaining or predicting this heterogeneity. We show how generalized linear mixed effects model (GLMM) trees can be used to identify subgroups with differently shaped trajectories in linear growth curve models. Originally developed for clustered cross-sectional data, GLMM trees are extended here to longitudinal data. The resulting extended GLMM trees are directly applicable to growth curve models as an important special case. In simulated and real-world data, we assess the performance of the extensions and compare against other partitioning methods for growth curve models. Extended GLMM trees perform more accurately than the original algorithm and LongCART, and similarly accurate as structural equation model (SEM) trees. In addition, GLMM trees allow for modeling both discrete and continuous time series, are less sensitive to (mis-)specification of the random-effects structure and are much faster to compute.