Joint modelling compared with two stage methods for analysing longitudinal data and prospective outcomes: A simulation study of childhood growth and BP

What is it about?

There is a challenge in choosing which methods of analysis are appropriate to examine the relationship between growth trajectories and prospective dependent outcomes. This study uses simulations to explore four two-stage and two joint modelling methods for this purpose, and compares their performance in terms of bias and coverage. The two-stage method of using multilevel models to estimate growth parameters and relating these to outcome gives unbiased estimates of the conditional associations between growth and outcome. The two joint modelling methods are unbiased.

Featured Image

Why is it important?

Helps inform robust modelling of longitudinal exposure data.