Multivariate t linear mixed models for irregularly observed multiple repeated measures with missing outcomes

• Published in: Biometrical journal Vol. 55; no. 4; pp. 554 - 571
• Main Author: Wang, Wan-Lun
• Format: Journal Article
• Language: English
• Published: Germany Blackwell Publishing Ltd 01.07.2013; Wiley - VCH Verlag GmbH & Co. KGaA
• Subjects: Acquired Immunodeficiency Syndrome - blood; Adult; AECM algorithm; Algorithms; Bayes Theorem; Clinical Trials as Topic; Damped exponential model; Data Interpretation, Statistical; Female; Human immunodeficiency virus; Humans; Linear Models; Longitudinal Studies; Middle Aged; Missing values; Multivariate Analysis; Outliers; Prediction; Pregnancy; RNA, Viral - blood; Young Adult; Outliers; Damped exponential model; AECM algorithm; Missing values; Prediction
• ISSN: 0323-3847; 1521-4036; 1521-4036
• DOI: 10.1002/bimj.201200001

Missing outcomes or irregularly timed multivariate longitudinal data frequently occur in clinical trials or biomedical studies. The multivariate t linear mixed model (MtLMM) has been shown to be a robust approach to modeling multioutcome continuous repeated measures in the presence of outliers or heavy‐tailed noises. This paper presents a framework for fitting the MtLMM with an arbitrary missing data pattern embodied within multiple outcome variables recorded at irregular occasions. To address the serial correlation among the within‐subject errors, a damped exponential correlation structure is considered in the model. Under the missing at random mechanism, an efficient alternating expectation‐conditional maximization (AECM) algorithm is used to carry out estimation of parameters and imputation of missing values. The techniques for the estimation of random effects and the prediction of future responses are also investigated. Applications to an HIV‐AIDS study and a pregnancy study involving analysis of multivariate longitudinal data with missing outcomes as well as a simulation study have highlighted the superiority of MtLMMs on the provision of more adequate estimation, imputation and prediction performances.