Censored mixed-effects models for irregularly observed repeated measures with applications to HIV viral loads

In some acquired immunodeficiency syndrome (AIDS) clinical trials, the human immunodeficiency virus-1 ribonucleic acid measurements are collected irregularly over time and are often subject to some upper and lower detection limits, depending on the quantification assays. Linear and nonlinear mixed-effects models, with modifications to accommodate censored observations, are routinely used to analyze this type of data (Vaida and Liu, J Comput Graph Stat 18:797-817, 2009; Matos et al., Comput Stat Data Anal 57(1):450-464, 2013a). This paper presents a framework for fitting LMEC/NLMEC with response variables recorded at irregular intervals. To address the serial correlation among the within-subject errors, a damped exponential correlation structure is considered in the random error and an EM-type algorithm is developed for computing the maximum likelihood estimates, obtaining as a byproduct the standard errors of the fixed effects and the likelihood value. The proposed methods are illustrated with simulations and the analysis of two real AIDS case studies.