Linear mixed model vs two-stage methods: Developing prognostic models of diabetic kidney disease progression

TL;DR

This study compares linear mixed models and two-stage methods for developing prognostic models of diabetic kidney disease progression, focusing on bias, accuracy, and applicability in real-world data with irregular measurements.

Contribution

It provides a comprehensive comparison of LMM and two-stage methods, highlighting when two-stage approaches are appropriate as alternatives in longitudinal prognostic modeling.

Findings

Two-stage methods can be suitable alternatives to LMM under certain conditions.

The comparison includes bias and mean squared error analyses.

Findings generalize to other disease progression studies.

Abstract

Identifying prognostic factors for disease progression is a cornerstone of medical research. Repeated assessments of a marker outcome are often used to evaluate disease progression, and the primary research question is to identify factors associated with the longitudinal trajectory of this marker. Our work is motivated by diabetic kidney disease (DKD), where serial measures of estimated glomerular filtration rate (eGFR) are the longitudinal measure of kidney function, and there is notable interest in identifying factors, such as metabolites, that are prognostic for DKD progression. Linear mixed models (LMM) with serial marker outcomes (e.g., eGFR) are a standard approach for prognostic model development, namely by evaluating the time and prognostic factor (e.g., metabolite) interaction. However, two-stage methods that first estimate individual-specific eGFR slopes, and then use these as outcomes in a regression framework with metabolites as predictors are easy to interpret and implement for applied researchers. Herein, we compared the LMM and two-stage methods, in terms of bias and mean squared error via analytic methods and simulations, allowing for irregularly spaced measures and missingness. Our findings provide novel insights into when two-stage methods are suitable longitudinal prognostic modeling alternatives to the LMM. Notably, our findings generalize to other disease studies.