Multiple imputation approach for interval‐censored time to HIV RNA viral rebound within a mixed effects Cox model

Abstract

We present a method to fit a mixed effects Cox model with interval‐censored data. Our proposal is based on a multiple imputation approach that uses the truncated Weibull distribution to replace the interval‐censored data by imputed survival times and then uses established mixed effects Cox methods for right‐censored data. Interval‐censored data were encountered in a database corresponding to a recompilation of retrospective data from eight analytical treatment interruption (ATI) studies in 158 human immunodeficiency virus (HIV) positive combination antiretroviral treatment (cART) suppressed individuals. The main variable of interest is the time to viral rebound, which is defined as the increase of serum viral load (VL) to detectable levels in a patient with previously undetectable VL, as a consequence of the interruption of cART. Another aspect of interest of the analysis is to consider the fact that the data come from different studies based on different grounds and that we have several assessments on the same patient. In order to handle this extra variability, we frame the problem into a mixed effects Cox model that considers a random intercept per subject as well as correlated random intercept and slope for pre‐cART VL per study. Our procedure has been implemented in R using two packages: truncdist and coxme, and can be applied to any data set that presents both interval‐censored survival times and a grouped data structure that could be treated as a random effect in a regression model. The properties of the parameter estimators obtained with our proposed method are addressed through a simulation study.

Multiparameter one‐sided tests for nonlinear mixed effects models with censored responses

“…In addition to censoring in responses, longitudinal data often exhibit other complexities such as dropouts and missing data. Formal methods to address dropouts and missing data include multiple imputations, 5 weighted estimating equations, 21 latent Markov models, 22 Bayesian approaches, 23 and likelihood‐based methods 24,25 . In particular, for likelihood‐based methods, it is conceptually straightforward to develop constrained tests, but the computation can become more challenging.…”

“…It is well‐known that ad‐hoc methods, such as imputing the censored values by half the detection limit, will lead to biased results 3,4 . Formal methods for censored data (and missing data in general) include multiple imputation methods 5 and likelihood‐based methods. Likelihood‐based methods are particularly attractive since possible missing data mechanism models may be specified and combined with the outcome model for parameter estimation and hypothesis testing 6 .…”

Occupational Risk Factors for SARS-CoV-2 Seropositivity in Healthcare Workers

“…Independent interval-censored data, which can result in biased estimates for time-varying covariates as intervals increase in length, 30 was accounted for by multiple imputation 31,32 as extended to mixed effects Cox models in R Statistical Software (https://www.r-project.org). 33 Community incidence rates and covariates from individual survey data were derived from data corresponding to the imputed survival time; surveys that occurred more than 28 days after the imputed survival time were considered at risk of bias and excluded from analysis for that assessment interval.…”

The association of apolipoprotein B with chronic kidney disease in the Chinese population

“…We determined the correlation between baseline ApoB level and the subsequent occurrence of CKD under the Cox proportional hazards regression model. The mixed-effects Cox regression models was applied in sensitivity analysis II ( 27 ).Two-sided P < 0.05 was considered statistically significant. We used SPSS Statistics (version 25.0, IBM, Armonk, NY, USA) and R-4.0.2 (R Foundation for Statistical Computing, Vienna, Austria) for data analysis.…”