scholarly journals A tractable Bayesian joint model for longitudinal and survival data

2021 ◽  
Author(s):  
Danilo Alvares ◽  
Francisco J. Rubio
Keyword(s):  
Survival Data ◽  
Joint Model ◽  
Longitudinal And Survival Data ◽  
Bayesian Joint Model

scholarly journals A new joint model for longitudinal and survival data with a cure fraction

2004 ◽  
Vol 91 (1) ◽  
pp. 18-34 ◽  
Author(s):  
Ming-Hui Chen ◽  
Joseph G Ibrahim ◽  
Debajyoti Sinha
Keyword(s):  
Survival Data ◽  
Joint Model ◽  
Cure Fraction ◽  
Longitudinal And Survival Data

scholarly journals A joint model for mixed and truncated longitudinal data and survival data, with application to HIV vaccine studies

Biostatistics ◽  
2017 ◽  
Vol 19 (3) ◽  
pp. 374-390 ◽  
Author(s):  
Tingting Yu ◽  
Lang Wu ◽  
Peter B Gilbert
Keyword(s):  
Longitudinal Data ◽  
Survival Data ◽  
Joint Model ◽  
Hiv Vaccine ◽  
Likelihood Inference ◽  
Likelihood Method ◽  
Parameter Estimates ◽  
Computationally Efficient ◽  
Approximate Likelihood ◽  
Longitudinal And Survival Data

SUMMARY In HIV vaccine studies, a major research objective is to identify immune response biomarkers measured longitudinally that may be associated with risk of HIV infection. This objective can be assessed via joint modeling of longitudinal and survival data. Joint models for HIV vaccine data are complicated by the following issues: (i) left truncations of some longitudinal data due to lower limits of quantification; (ii) mixed types of longitudinal variables; (iii) measurement errors and missing values in longitudinal measurements; (iv) computational challenges associated with likelihood inference. In this article, we propose a joint model of complex longitudinal and survival data and a computationally efficient method for approximate likelihood inference to address the foregoing issues simultaneously. In particular, our model does not make unverifiable distributional assumptions for truncated values, which is different from methods commonly used in the literature. The parameters are estimated based on the h-likelihood method, which is computationally efficient and offers approximate likelihood inference. Moreover, we propose a new approach to estimate the standard errors of the h-likelihood based parameter estimates by using an adaptive Gauss–Hermite method. Simulation studies show that our methods perform well and are computationally efficient. A comprehensive data analysis is also presented.

scholarly journals Complete Case versus Inverse Probability Weighting Methods of Fitting Incomplete Longitudinal and Survival Data Joint Models

2019 ◽  
pp. 1-10
Author(s):  
D. O. Nyaboga ◽  
A. Mwangi ◽  
D. Lusweti
Keyword(s):  
Missing Data ◽  
Survival Data ◽  
Joint Model ◽  
Information Criteria ◽  
Joint Models ◽  
Full Data ◽  
Complete Case ◽  
Inverse Probability ◽  
Longitudinal And Survival Data ◽  
The Impact

Missing data is a common problem in real word studies especially clinical studies. However, most people working with such data, often drop missing cases from individuals with incomplete observations that occur when patients do not complete the treatment or miss their scheduled visits. This may lead to misleading results and ultimately affect the decision of whether an intervention is good or bad for the patients under treatment. The comparison of Complete Case (CC) and Inverse Probability Weights (IPW) techniques of handling missing data in various models has been addressed, however little has been done to compare these methods when applied to joint models of longitudinal and time to event data. Therefore, this paper seeks to investigate the impact of assuming CC analysis on clinical data with missing cases, comparing it with IPW method when fitting joint models of longitudinal and survival data setting full data model as the baseline model. This paper made use of randomized aids clinical trial data. The model with Deviance Information Criteria (DIC) close to that of full data joint model is considered the best. From the results, joint models from full data, CC and IPW had DIC of 10603.94, 8410.33 and 10600.95 respectively. The joint model obtained from IPW data had a DIC too close to that of full data joint model as compared to model from CC data.

scholarly journals Joint Modeling of  Longitudinal Ordinal Data on  Quality of Life and Survival

2021 ◽  
Author(s):  
◽  
Kemmawadee Preedalikit
Keyword(s):  
Quality Of Life ◽  
Survival Data ◽  
Latent Variable ◽  
Ordinal Data ◽  
Likelihood Function ◽  
Joint Model ◽  
Hiv Progression ◽  
Bayesian Joint Model ◽  
Lower Hazard

<p>Joint models for longitudinal and survival data have been widely discussed in the literature. This thesis proposes a joint model using a stereotype model for the longitudinal ordinal responses and a Cox proportional hazards model for survival time. Our current joint model has a new feature since no literature has examined the joint model under the stereotype model. The stereotype model can improve the fit by adding extra score parameters, but it still has the advantage of requiring only a single parameter to describe the effect of a predictor on the item response levels. We give an example to model longitudinal ordinal data and survival data for patients being followed up after treatments. The main focus is on modeling both the quality of life data and the survival data simultaneously with a goal of understanding the association between the two processes over time. These two models are linked through a latent variable that characterizes the quality of life of an individual and is assumed to underlie the hazard rate. In other words, the latent variable serves as a shared variable in the joint model. We present the joint model in two different aspects: one based on a Bayesian approach and the other one a semiparametric approach using the EM algorithm. For the Bayesian approach, the latent variable is treated as a continuous variable and is assumed to have a multivariate normal distribution. The partial survival likelihood function is used in the survival component of the Bayesian joint model, while the full likelihood function is considered in the semiparametric joint model. In the latter approach the baseline hazard is assumed to be a step function and has no parametric form. The latent variable in the semiparametric joint model is then treated as a discrete variable. We illustrate our methodologies by analyzing data from the Staccato study, a randomized trial to compare two treatment methods, for Human Immunodeficiency Virus (HIV) infection of Thai patients on Highly Active Antiretroviral Therapy (HAART), in which the quality of life was assessed with a HIV Medical Outcome Study (MOS-HIV) questionnaire. Furthermore, we extend the study further to the case of multiple failure types in the survival component. Thus, the extension of the joint model consists of the stereotype model and the competing risks model. The Bayesian method is employed to estimate all unknown parameters in this extended joint model. The results we obtained are consistent for both the Bayesian joint model and the semiparametric joint model. Both models show that patients who had a better quality of life were associated with a lower hazard of HIV progression. Patients on continuous treatment also had a lower hazard of HIV progression compared with patients on CD4-guided interruption treatment.</p>

A Semiparametric Joint Model for Longitudinal and Survival Data with Application to Hemodialysis Study

Biometrics ◽  
2009 ◽  
Vol 65 (3) ◽  
pp. 737-745 ◽  
Author(s):  
Liang Li ◽  
Bo Hu ◽  
Tom Greene
Keyword(s):  
Survival Data ◽  
Joint Model ◽  
Longitudinal And Survival Data

scholarly journals Profile Likelihood Estimation Applied To The Semi-parametric Models With Survival Data

2021 ◽  
Author(s):  
Khandoker Mohammad
Keyword(s):  
Survival Data ◽  
Profile Likelihood ◽  
Bootstrap Method ◽  
Score Function ◽  
Likelihood Estimation ◽  
Joint Model ◽  
Likelihood Score ◽  
Likelihood Approach ◽  
R Packages ◽  
Longitudinal And Survival Data

<p><b>In this thesis, we have investigated the efficiency of profile likelihood in the estimation of parameters from the Cox Proportional Hazards (PH) cure model and joint model of longitudinal and survival data. For the profile likelihood approach in the joint model of longitudinal and survival data, Hsieh et al. (2006) stated “No distributional or asymptotic theory is available to date, and even the standard errors (SEs), defined as the standard deviations of the parametric estimators, are difficult to obtain”. The reason behind this difficulty is the estimator of baseline hazard which involves implicit function in the profile likelihood estimation (Hirose and Liu, 2020). Hence finding the estimated SE of the parametric estimators from the Cox PH cure model and joint model using profile likelihood approach is a great challenge. Therefore, bootstrap method has been suggested to get the estimated standard errors while using the profile likelihood approach (Hsieh et al., 2006).</b></p> <p>To solve the difficulty, we have expanded the profile likelihood function directly without assuming the derivative of the profile likelihood score function and obtain the explicit form of the SE estimator using the profile likelihood score function. Our proposed alternative approach gives us not only analytical understanding of the profile likelihood estimation, but also provides closed form formula to compute the standard error of the profile likelihood maximum likelihood estimator in terms of profile likelihood score function. To show the advantage of our proposed approach in medical and clinical studies, we have analysed the simulated and real-life data, and compared our results with the output obtained from the smcure, JM(method: ’Cox-PH-GH’) and joineRML R-packages. The outputs suggest that the bootstrap method and our proposed approach have provided similar and comparable results. In addition, the average computation times of our approach are much less compared to the above mentioned R-packages.</p>

scholarly journals Joint Modeling of  Longitudinal Ordinal Data on  Quality of Life and Survival

2021 ◽  
Author(s):  
◽  
Kemmawadee Preedalikit
Keyword(s):  
Quality Of Life ◽  
Survival Data ◽  
Latent Variable ◽  
Ordinal Data ◽  
Likelihood Function ◽  
Joint Model ◽  
Hiv Progression ◽  
Bayesian Joint Model ◽  
Lower Hazard

<p>Joint models for longitudinal and survival data have been widely discussed in the literature. This thesis proposes a joint model using a stereotype model for the longitudinal ordinal responses and a Cox proportional hazards model for survival time. Our current joint model has a new feature since no literature has examined the joint model under the stereotype model. The stereotype model can improve the fit by adding extra score parameters, but it still has the advantage of requiring only a single parameter to describe the effect of a predictor on the item response levels. We give an example to model longitudinal ordinal data and survival data for patients being followed up after treatments. The main focus is on modeling both the quality of life data and the survival data simultaneously with a goal of understanding the association between the two processes over time. These two models are linked through a latent variable that characterizes the quality of life of an individual and is assumed to underlie the hazard rate. In other words, the latent variable serves as a shared variable in the joint model. We present the joint model in two different aspects: one based on a Bayesian approach and the other one a semiparametric approach using the EM algorithm. For the Bayesian approach, the latent variable is treated as a continuous variable and is assumed to have a multivariate normal distribution. The partial survival likelihood function is used in the survival component of the Bayesian joint model, while the full likelihood function is considered in the semiparametric joint model. In the latter approach the baseline hazard is assumed to be a step function and has no parametric form. The latent variable in the semiparametric joint model is then treated as a discrete variable. We illustrate our methodologies by analyzing data from the Staccato study, a randomized trial to compare two treatment methods, for Human Immunodeficiency Virus (HIV) infection of Thai patients on Highly Active Antiretroviral Therapy (HAART), in which the quality of life was assessed with a HIV Medical Outcome Study (MOS-HIV) questionnaire. Furthermore, we extend the study further to the case of multiple failure types in the survival component. Thus, the extension of the joint model consists of the stereotype model and the competing risks model. The Bayesian method is employed to estimate all unknown parameters in this extended joint model. The results we obtained are consistent for both the Bayesian joint model and the semiparametric joint model. Both models show that patients who had a better quality of life were associated with a lower hazard of HIV progression. Patients on continuous treatment also had a lower hazard of HIV progression compared with patients on CD4-guided interruption treatment.</p>

Sign in / Sign up

Export Citation Format

Share Document