A latent class based imputation method under Bayesian quantile regression framework using asymmetric Laplace distribution for longitudinal medication usage data with intermittent missing values

ABSTRACT Evaluating the association between diseases and the longitudinal pattern of pharmacological therapy has become increasingly important. However, in many longitudinal studies, self-reported medication usage data collected at patients’ follow-up visits could be missing for various reasons. These pieces of missing or inaccurate/untenable information complicate determining the trajectory of medication use and its complete effects for patients. Although longitudinal models can deal with specific types of missing data, inappropriate handling of this issue can lead to a biased estimation of regression parameters especially when missing data mechanisms are complex and depend upon multiple sources of variation. We propose a latent class-based multiple imputation (MI) approach using a Bayesian quantile regression (BQR) that incorporates cluster of unobserved heterogeneity for medication usage data with intermittent missing values. Findings from our simulation study indicate that the proposed method performs better than traditional MI methods under certain scenarios of data distribution. We also demonstrate applications of the proposed method to data from the Prospective Study of Outcomes in Ankylosing Spondylitis (AS) cohort when assessing an association between longitudinal nonsteroidal anti-inflammatory drugs (NSAIDs) usage and radiographic damage in AS, while the longitudinal NSAID index data are intermittently missing.

[1]  R. Koenker,et al.  An interior point algorithm for nonlinear quantile regression , 1996 .

[2]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data: Little/Statistical Analysis with Missing Data , 2002 .

[3]  C. Mitchell Dayton,et al.  Applied Latent Class Analysis: Use of Categorical and Continuous Covariates in Latent Class Analysis , 2002 .

[4]  Stef van Buuren,et al.  MICE: Multivariate Imputation by Chained Equations in R , 2011 .

[5]  Michael J Daniels,et al.  Quantile regression in the presence of monotone missingness with sensitivity analysis. , 2016, Biostatistics.

[6]  R. Ramamoorthi,et al.  Posterior Consistency of Bayesian Quantile Regression Based on the Misspecified Asymmetric Laplace Density , 2013 .

[7]  T. Lancaster,et al.  Bayesian Quantile Regression , 2005 .

[8]  MinJae Lee,et al.  Multiple imputation for left‐censored biomarker data based on Gibbs sampling method , 2012, Statistics in medicine.

[9]  George B. Macready,et al.  Concomitant-Variable Latent-Class Models , 1988 .

[10]  Daniel S. Nagin,et al.  Advances in Group-Based Trajectory Modeling and an SAS Procedure for Estimating Them , 2007 .

[11]  G. Yin,et al.  Bayesian Quantile Regression for Longitudinal Studies with Nonignorable Missing Data , 2010, Biometrics.

[12]  M. West.  Outlier Models and Prior Distributions in Bayesian Linear Regression , 1984 .

[13]  MinJae Lee,et al.  A multiple imputation method based on weighted quantile regression models for longitudinal censored biomarker data with missing values at early visits , 2018, BMC Medical Research Methodology.

[14]  M. Kaptein,et al.  Multiple imputation of missing categorical data using latent class models: State of art , 2015 .

[15]  Nan Lin,et al.  Model selection in binary and tobit quantile regression using the Gibbs sampler , 2012, Comput. Stat. Data Anal..

[16]  Lena Osterhagen,et al.  Multiple Imputation For Nonresponse In Surveys , 2016 .

[17]  M. A. van 't Hof,et al.  Assessment of outcome in ankylosing spondylitis: an extended radiographic scoring system , 2004, Annals of the rheumatic diseases.

[18]  M. West.  On scale mixtures of normal distributions , 1987 .

[19]  Ofer Harel,et al.  Latent class regression: Inference and estimation with two‐stage multiple imputation , 2013, Biometrical journal. Biometrische Zeitschrift.

[20]  Raymond J Carroll,et al.  Multiple imputation in quantile regression. , 2012, Biometrika.

[21]  Daniel S Nagin,et al.  Group-based trajectory modeling in clinical research. , 2010, Annual review of clinical psychology.

[22]  Nicole A. Lazar,et al.  Statistical Analysis With Missing Data , 2003, Technometrics.

[23]  J. Reveille,et al.  Clinical, radiographic and functional differences between juvenile-onset and adult–onset ankylosing spondylitis: results from the PSOAS cohort , 2007, Annals of the rheumatic diseases.

[24]  Alexander Hehmeyer,et al.  Nonparametric Bayesian Multiple Imputation for Incomplete Categorical Variables in Large-Scale Assessment Surveys , 2013 .

[25]  Geert Molenberghs,et al.  A Multiple-Imputation-Based Approach to Sensitivity Analyses and Effectiveness Assessments in Longitudinal Clinical Trials , 2014, Journal of biopharmaceutical statistics.

[26]  A. Calin,et al.  The Bath Ankylosing Spondylitis Radiology Index (BASRI): a new, validated approach to disease assessment. , 1998, Arthritis and rheumatism.

[27]  Xingdong Feng,et al.  Multiple Imputation for M-Regression With Censored Covariates , 2012 .

[28]  A. Mackinnon,et al.  The use and reporting of multiple imputation in medical research – a review , 2010, Journal of internal medicine.

[29]  Jerome P. Reiter,et al.  Nonparametric Bayesian Multiple Imputation for Missing Data Due to Mid-Study Switching of Measurement Methods , 2012 .

[30]  Jørn Wetterslev,et al.  Comparison of Results from Different Imputation Techniques for Missing Data from an Anti-Obesity Drug Trial , 2014, PloS one.

[31]  H. Lian,et al.  Bayesian quantile regression for longitudinal data models , 2012 .

[32]  Ivana Komunjer,et al.  Quasi-maximum likelihood estimation for conditional quantiles , 2005 .

[33]  D. Rubin.  Multiple Imputation After 18+ Years , 1996 .

[34]  Mulugeta Gebregziabher,et al.  Latent class based multiple imputation approach for missing categorical data. , 2010, Journal of statistical planning and inference.

[35]  Kim-Hung Li,et al.  Imputation using Markov chains , 1988 .

[36]  Xuming He,et al.  Posterior Inference in Bayesian Quantile Regression with Asymmetric Laplace Likelihood , 2016 .

[37]  M. Dougados,et al.  NSAIDs in ankylosing spondylitis. , 2002, Clinical and experimental rheumatology.

[38]  J Zhang,et al.  Review of guidelines and literature for handling missing data in longitudinal clinical trials with a case study , 2006, Pharmaceutical statistics.

[39]  H. Kozumi,et al.  Gibbs sampling methods for Bayesian quantile regression , 2011 .

[40]  Johannes Lenhard,et al.  Computation and Simulation , 2017 .

[41]  Ruibin Xi,et al.  Bayesian regularized quantile regression , 2010 .

[42]  Jerome P. Reiter,et al.  Bayesian multiple imputation for large-scale categorical data with structural zeros , 2013 .

[43]  Constantine Frangakis,et al.  Multiple imputation by chained equations: what is it and how does it work? , 2011, International journal of methods in psychiatric research.

[44]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[45]  Pulak Ghosh,et al.  Simultaneous Bayesian Estimation of Multiple Quantiles with an Extension to Hierarchical Models , 2012 .

[46]  Dirk Van den Poel,et al.  bayesQR: A Bayesian Approach to Quantile Regression , 2017 .

[47]  Katherine J. Lee,et al.  The rise of multiple imputation: a review of the reporting and implementation of the method in medical research , 2015, BMC Medical Research Methodology.

[48]  Andrew GelmanyJanuary,et al.  Prior distribution , 2000 .

[49]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .

[50]  C. Anderson‐Cook,et al.  Group-Based Modeling of Development , 2006 .

[51]  Xiao-Li Meng,et al.  Multiple-Imputation Inferences with Uncongenial Sources of Input , 1994 .

[52]  Manouchehr Hessabi,et al.  Harmonization, data management, and statistical issues related to prospective multicenter studies in Ankylosing spondylitis (AS): Experience from the Prospective Study Of Ankylosing Spondylitis (PSOAS) cohort , 2018, Contemporary clinical trials communications.

[53]  A. Kottas,et al.  A Bayesian Nonparametric Approach to Inference for Quantile Regression , 2010 .

[54]  A. Cats,et al.  Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. , 1984, Arthritis and rheumatism.

[55]  S. Normand,et al.  A Bayesian Two‐Part Latent Class Model for Longitudinal Medical Expenditure Data: Assessing the Impact of Mental Health and Substance Abuse Parity , 2011, Biometrics.