Joint Models for a Primary Endpoint and Multiple Longitudinal Covariate Processes

Joint models are formulated to investigate the association between a primary endpoint and features of multiple longitudinal processes. In particular, the subject-specific random effects in a multivariate linear random-effects model for multiple longitudinal processes are predictors in a generalized linear model for primary endpoints. Li, Zhang, and Davidian (2004, Biometrics60, 1-7) proposed an estimation procedure that makes no distributional assumption on the random effects but assumes independent within-subject measurement errors in the longitudinal covariate process. Based on an asymptotic bias analysis, we found that their estimators can be biased when random effects do not fully explain the within-subject correlations among longitudinal covariate measurements. Specifically, the existing procedure is fairly sensitive to the independent measurement error assumption. To overcome this limitation, we propose new estimation procedures that require neither a distributional or covariance structural assumption on covariate random effects nor an independence assumption on within-subject measurement errors. These new procedures are more flexible, readily cover scenarios that have multivariate longitudinal covariate processes, and can be implemented using available software. Through simulations and an analysis of data from a hypertension study, we evaluate and illustrate the numerical performances of the new estimators.

[1]  Marie Davidian,et al.  Conditional Estimation for Generalized Linear Models When Covariates Are Subject‐Specific Parameters in a Mixed Model for Longitudinal Measurements , 2004, Biometrics.

[2]  Raymond J. Carroll,et al.  Conditional scores and optimal scores for generalized linear measurement-error models , 1987 .

[3]  D. Ruppert,et al.  Measurement Error in Nonlinear Models , 1995 .

[4]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[5]  P. Heagerty,et al.  Misspecified maximum likelihood estimates and generalised linear mixed models , 2001 .

[6]  R Henderson,et al.  Joint modelling of longitudinal measurements and event time data. , 2000, Biostatistics.

[7]  Daowen Zhang,et al.  Likelihood and pseudo-likelihood methods for semiparametric joint models for a primary endpoint and longitudinal data , 2007, Comput. Stat. Data Anal..

[8]  S. Zeger,et al.  Joint analysis of longitudinal data comprising repeated measures and times to events , 2001 .

[9]  M. Wulfsohn,et al.  A joint model for survival and longitudinal data measured with error. , 1997, Biometrics.

[10]  B. Lindsay.  Using Empirical Partially Bayes Inference for Increased Efficiency , 1985 .

[11]  C Y Wang,et al.  Regression Analysis When Covariates Are Regression Parameters of a Random Effects Model for Observed Longitudinal Measurements , 2000, Biometrics.

[12]  Yijian Huang,et al.  Functional methods for logistic regression on random-effect-coefficients for longitudinal measurements , 2001 .

[13]  F. Brancati,et al.  Blood pressure in young adulthood and the risk of type 2 diabetes in middle age. , 2003, Diabetes care.

[14]  P. Diggle.  An approach to the analysis of repeated measurements. , 1988, Biometrics.

[15]  Lewis B. Sheiner,et al.  Heteroscedastic nonlinear regression , 1988 .

[16]  E. Vonesh,et al.  Linear and Nonlinear Models for the Analysis of Repeated Measurements , 1996 .

[17]  Marie Davidian,et al.  A Semiparametric Likelihood Approach to Joint Modeling of Longitudinal and Time‐to‐Event Data , 2002, Biometrics.

[18]  Anastasios A. Tsiatis,et al.  A semiparametric estimator for the proportional hazards model with longitudinal covariates measured with error , 2001 .

[19]  P. Diggle,et al.  Analysis of Longitudinal Data , 2003 .

[20]  G. Verbeke,et al.  The effect of misspecifying the random-effects distribution in linear mixed models for longitudinal data , 1997 .

[21]  T A Louis,et al.  Random effects models with non-parametric priors. , 1992, Statistics in medicine.