Joint modelling of time-to-event and multivariate longitudinal outcomes: recent developments and issues

Abstract

BackgroundAvailable methods for the joint modelling of longitudinal and time-to-event outcomes have typically only allowed for a single longitudinal outcome and a solitary event time. In practice, clinical studies are likely to record multiple longitudinal outcomes. Incorporating all sources of data will improve the predictive capability of any model and lead to more informative inferences for the purpose of medical decision-making.MethodsWe reviewed current methodologies of joint modelling for time-to-event data and multivariate longitudinal data including the distributional and modelling assumptions, the association structures, estimation approaches, software tools for implementation and clinical applications of the methodologies.ResultsWe found that a large number of different models have recently been proposed. Most considered jointly modelling linear mixed models with proportional hazard models, with correlation between multiple longitudinal outcomes accounted for through multivariate normally distributed random effects. So-called current value and random effects parameterisations are commonly used to link the models. Despite developments, software is still lacking, which has translated into limited uptake by medical researchers.ConclusionAlthough, in an era of personalized medicine, the value of multivariate joint modelling has been established, researchers are currently limited in their ability to fit these models routinely. We make a series of recommendations for future research needs.Electronic supplementary materialThe online version of this article (doi:10.1186/s12874-016-0212-5) contains supplementary material, which is available to authorized users.

Increased Moraxella and Streptococcus species abundance after severe bronchiolitis is associated with recurrent wheezing

“…Next, we performed longitudinal analyses by constructing joint models for the longitudinal microbial data to quantify their relationship with the onset of recurrent wheezing. 20 Specifically, we created a longitudinal mixed-effects generalized linear model to predict relative genus abundances over time using the 8 most abundant taxa and clinical covariates as the fixed effects, time as the random slope, and subject as the random intercept. These relative abundances and their interaction by time point were both scaled such that a 10% change in abundance was equivalent to a unit change and incorporated as time-dependent covariates into a Cox proportional hazards model with the outcome recurrent wheezing by age 3 years.…”

A Comparison of Joint Models for Longitudinal and Competing Risks Data, with Application to an Epilepsy Drug Randomized Controlled Trial

“…The literature on this topic is extensive, with comprehensive reviews given by Hogan and Laird (1997), Tsiatis and Davidian (2004) and Gould et al (2015). Reviews specific to joint latent class models and multivariate longitudinal data are given by Proust-Lima et al (2012) and Hickey et al (2016) respectively.…”

Review and Comparison of Computational Approaches for Joint Longitudinal and Time‐to‐Event Models

“…This necessitates joint models with multiple longitudinal variables. Hickey et al () review developments in multivariate joint models including software implementations. Hickey et al .…”

“…This necessitates joint models with multiple longitudinal variables. Hickey et al (2016) review developments in multivariate joint models including software implementations. Hickey et al mentioned a new package for multivariate joint models, sjmsoft for R, available from the author's website (Brown, 2005).…”