Broken Stick Model for Irregular Longitudinal Data

Many longitudinal studies collect data that have irregular observation times, often requiring the application of linear mixed models with time-varying outcomes. This paper presents an alternative splits quantitative analysis into two steps. The first step converts irregularly observed a set repeated measures through broken stick model. second estimates parameters scientific interest from measurements at subject level. model approximates each subject's trajectory by series connected straight lines. breakpoints, specified user, divide time axis consecutive intervals common to all subjects. Specification requires just three variables: time, measurement and subject. is special case model, as B-spline grouping factor. main assumptions are: Subjects are exchangeable, trajectories between breakpoints straight, random effects follow multivariate normal distribution, unobserved missing random. R package brokenstick v2.5.0 offers tools calculate, predict, impute visualize estimates. supports optimization methods, including options constrain variance-covariance matrix effects. We demonstrate six applications model: Detection critical periods, estimation time-to-time correlations, profile analysis, curve interpolation, multiple imputation personalized prediction future outcomes matching.