Semiparametric Bayesian inference on skew–normal joint modeling of multivariate longitudinal and survival data

Tang, An-Min; Tang, Niansheng

doi:10.1002/sim.6373

Cited by 34 publications

(33 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Spline-based approaches, which allow also the random effects to be nonlinear functions in time, are mentioned by Song and Wang (2008) and were employed by Rizopoulos and Ghosh (2011) and Rizopoulos, Hatfield, Carlin, and Takkenberg (2014) as well as Brown, Ibrahim, and DeGruttola (2005) and Brown (2009). Tang and Tang (2015) also use P-splines in modeling longitudinal trajectories, but do so only in estimating the mean function, whereas we model also the individual trajectories as smooth functions of time. As the number of random effects increases with the number of knots, this number is limited in practice.…”

Section: Introductionmentioning

confidence: 99%

“…We aim to avoid the explicit choice of knots and number of basis functions using a penalized spline approach, where a larger number of knots is specified and smoothness penalties are employed (Lang & Brezger, 2004). Tang and Tang (2015) also use P-splines in modeling longitudinal trajectories, but do so only in estimating the mean function, whereas we model also the individual trajectories as smooth functions of time. This is similar in spirit to the specification of individual trajectories in Jiang, Wang, Sammel, and Elliott (2015), however we do not assume an underlying class membership for the random effects.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Flexible Bayesian additive joint models with an application to type 1 diabetes research

et al. 2017

View full text Add to dashboard Cite

The joint modeling of longitudinal and time-to-event data is an important tool of growing popularity to gain insights into the association between a biomarker and an event process. We develop a general framework of flexible additive joint models that allows the specification of a variety of effects, such as smooth nonlinear, time-varying and random effects, in the longitudinal and survival parts of the models. Our extensions are motivated by the investigation of the relationship between fluctuating disease-specific markers, in this case autoantibodies, and the progression to the autoimmune disease type 1 diabetes. Using Bayesian P-splines, we are in particular able to capture highly nonlinear subject-specific marker trajectories as well as a time-varying association between the marker and event process allowing new insights into disease progression. The model is estimated within a Bayesian framework and implemented in the R-package bamlss.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Flexible Bayesian additive joint models with an application to type 1 diabetes research

et al. 2017

View full text Add to dashboard Cite

show abstract

“…However, over the past 2 decades, statistical methods that can provide a more flexible modelling framework for both the time‐to‐event and longitudinal aspects have emerged. The resulting methodology is called joint modelling (Chi & Ibrahim, ; Rizopoulos, ; Tang & Tang, ; Tang, Tang, & Pan, ). Corresponding software for implementing joint modelling in analysis has also recently become available in mainstream statistical software packages (Gould et al, ).…”

Section: Introductionmentioning

confidence: 99%

A new method for analysing transition to psychosis: Joint modelling of time‐to‐event outcome with time‐dependent predictors

Yuen

Mackinnon

Nelson

2017

Int J Methods Psych Res

View full text Add to dashboard Cite

An active area in psychosis research is the identification of predictors of transition to a psychotic state among those who are assessed as being at high risk of psychosis. Many of the potential predictors are time dependent in the sense that they may change over time and are measured at a number of assessment time points. Examples are various psychopathological measures such as negative symptoms, positive symptoms, depression, and anxiety. Most research in transition to psychosis has not made use of the dynamic nature of these measures, probably because suitable statistical methods and software have not been easily available. However, a relatively new statistical methodology is well suited to include such time-dependent predictors in transition to psychosis analysis. This methodology is called joint modelling and has recently been incorporated in mainstream statistical software. This paper describes this methodology and demonstrates its usefulness using data from one of the pioneering studies on transition to psychosis.

show abstract

“…The integral computation becomes cumbersome with more than a couple of random effects 6,13 and not necessarily accurately solved. 14 This entails that most applications focused on only two longitudinal markers 11,15 and/or random intercepts only, 15,16 which is not sensible in complex diseases such as dementia. Contributions also mostly relied on a Bayesian estimation to circumvent the numerical issues [7][8][9][10]16 or a two-stage estimation.…”

mentioning

confidence: 99%

“…14 This entails that most applications focused on only two longitudinal markers 11,15 and/or random intercepts only, 15,16 which is not sensible in complex diseases such as dementia. Contributions also mostly relied on a Bayesian estimation to circumvent the numerical issues [7][8][9][10]16 or a two-stage estimation. 6 An additional problem in dementia as in many other diseases (especially in psychiatry and neurology) is that each domain is not directly measured; it is approached by multivariate markers.…”

mentioning

confidence: 99%

A joint model for multiple dynamic processes and clinical endpoints: Application to Alzheimer's disease

Proust‐Lima

Philipps

Dartigues

2019

Statistics in Medicine

View full text Add to dashboard Cite

As other neurodegenerative diseases, Alzheimer's disease, the most frequent dementia in the elderly, is characterized by multiple progressive impairments in the brain structure and in clinical functions such as cognitive functioning and functional disability. Until recently, these components were mostly studied independently because no joint model for multivariate longitudinal data and time to event was available in the statistical community. Yet, these components are fundamentally interrelated in the degradation process toward dementia and should be analyzed together. We thus propose a joint model to simultaneously describe the dynamics of multiple correlated components. Each component, defined as a latent process, is measured by one or several continuous markers (not necessarily Gaussian). Rather than considering the associated time to diagnosis as in standard joint models, we assume diagnosis corresponds to the passing above a covariate‐specific threshold (to be estimated) of a pathological process that is modeled as a combination of the component‐specific latent processes. This definition captures the clinical complexity of diagnoses such as dementia diagnosis but also benefits from simplifications for the computation of maximum likelihood estimates. We show that the model and estimation procedure can also handle competing clinical endpoints. The estimation procedure, implemented in an R package, is validated by simulations and the method is illustrated on a large French population‐based cohort of cerebral aging in which we focused on the dynamics of three clinical manifestations and the associated risk of dementia and death before dementia.

show abstract

Semiparametric Bayesian inference on skew–normal joint modeling of multivariate longitudinal and survival data

Cited by 34 publications

References 37 publications

Flexible Bayesian additive joint models with an application to type 1 diabetes research

Flexible Bayesian additive joint models with an application to type 1 diabetes research

A new method for analysing transition to psychosis: Joint modelling of time‐to‐event outcome with time‐dependent predictors

A joint model for multiple dynamic processes and clinical endpoints: Application to Alzheimer's disease

Contact Info

Product

Resources

About