Joint modeling of longitudinal drug using pattern and time to first relapse in cocaine dependence treatment data

An important endpoint variable in a cocaine rehabilitation study is the time to first relapse of a patient after the treatment. We propose a joint modeling approach based on functional data analysis to study the relationship between the baseline longitudinal cocaine-use pattern and the interval censored time to first relapse. For the baseline cocaine-use pattern, we consider both self-reported cocaine-use amount trajectories and dichotomized use trajectories. Variations within the generalized longitudinal trajectories are modeled through a latent Gaussian process, which is characterized by a few leading functional principal components. The association between the baseline longitudinal trajectories and the time to first relapse is built upon the latent principal component scores. The mean and the eigenfunctions of the latent Gaussian process as well as the hazard function of time to first relapse are modeled nonparametrically using penalized splines, and the parameters in the joint model are estimated by a Monte Carlo EM algorithm based on Metropolis-Hastings steps. An Akaike information criterion (AIC) based on effective degrees of freedom is proposed to choose the tuning parameters, and a modified empirical information is proposed to estimate the variance-covariance matrix of the estimators.

[1]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[2]  H. Müller,et al.  Functional Data Analysis for Sparse Longitudinal Data , 2005 .

[3]  Fang Yao,et al.  FUNCTIONAL PRINCIPAL COMPONENT ANALYSIS FOR LONGITUDINAL AND SURVIVAL DATA , 2007 .

[4]  Linda C. Sobell,et al.  Timeline Follow-Back A Technique for Assessing Self-Reported Alcohol Consumption , 1992 .

[5]  Jun Yan,et al.  Functional Association Models for Multivariate Survival Processes , 2005 .

[6]  Jane-ling Wang,et al.  Functional linear regression analysis for longitudinal data , 2005, math/0603132.

[7]  Jiawei Wei,et al.  Model selection using modified AIC and BIC in joint modeling of paired functional data , 2010 .

[8]  P. Rutigliano,et al.  The timeline followback reports of psychoactive substance use by drug-abusing patients: psychometric properties. , 2000, Journal of consulting and clinical psychology.

[9]  T. Cai,et al.  Hazard Regression for Interval‐Censored Data with Penalized Spline , 2003, Biometrics.

[10]  R. Carroll,et al.  Semiparametric Regression for Clustered Data Using Generalized Estimating Equations , 2001 .

[11]  A. Cnaan,et al.  Effectiveness of propranolol for cocaine dependence treatment may depend on cocaine withdrawal symptom severity. , 2001, Drug and alcohol dependence.

[12]  Paul H. C. Eilers,et al.  Flexible smoothing with B-splines and penalties , 1996 .

[13]  Raymond J Carroll,et al.  Generalized Functional Linear Models With Semiparametric Single-Index Interactions , 2010, Journal of the American Statistical Association.

[14]  Jianhua Z. Huang,et al.  Joint modelling of paired sparse functional data using principal components. , 2008, Biometrika.

[15]  R. Sinha.  How does stress increase risk of drug abuse and relapse? , 2001, Psychopharmacology.

[16]  Yu-Ru Su,et al.  MODELING LEFT-TRUNCATED AND RIGHT-CENSORED SURVIVAL DATA WITH LONGITUDINAL COVARIATES. , 2012, Annals of statistics.

[17]  R. Sinha,et al.  Stress-induced cocaine craving and hypothalamic-pituitary-adrenal responses are predictive of cocaine relapse outcomes. , 2006, Archives of general psychiatry.

[18]  R. Sinha,et al.  Cocaine Dependence Treatment Data: Methods for Measurement Error Problems With Predictors Derived From Stationary Stochastic Processes , 2011, Journal of the American Statistical Association.

[19]  Raymond J. Carroll,et al.  Measurement error in nonlinear models: a modern perspective , 2006 .

[20]  Ana-Maria Staicu,et al.  Generalized Multilevel Functional Regression , 2009, Journal of the American Statistical Association.

[21]  Galin L. Jones,et al.  Fixed-Width Output Analysis for Markov Chain Monte Carlo , 2006, math/0601446.

[22]  H. Müller,et al.  Modelling sparse generalized longitudinal observations with latent Gaussian processes , 2008 .

[23]  N. Breslow,et al.  Approximate inference in generalized linear mixed models , 1993 .

[24]  Jian Huang,et al.  A Spline‐Based Semiparametric Maximum Likelihood Estimation Method for the Cox Model with Interval‐Censored Data , 2010 .

[25]  P. Rosenberg,et al.  Hazard function estimation using B-splines. , 1995, Biometrics.

[26]  P. Hall,et al.  Properties of principal component methods for functional and longitudinal data analysis , 2006, math/0608022.

[27]  K. Carroll,et al.  One‐Year Follow‐Up Status of Treatment‐Seeking Cocaine Abusers: Psychopathology and Dependence Severity as Predictors of Outcome , 1993, The Journal of nervous and mental disease.

[28]  T. Louis.  Finding the Observed Information Matrix When Using the EM Algorithm , 1982 .

[29]  C. McCulloch.  Maximum Likelihood Algorithms for Generalized Linear Mixed Models , 1997 .

[30]  I. Meilijson.  A fast improvement to the EM algorithm on its own terms , 1989 .

[31]  C. Kooperberg,et al.  Hazard regression with interval-censored data. , 1997, Biometrics.

[32]  T. Hsing,et al.  Uniform convergence rates for nonparametric regression and principal component analysis in functional/longitudinal data , 2010, 1211.2137.

[33]  Bo Thiesson,et al.  Accelerating EM for Large Databases , 2001, Machine Learning.

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

[35]  F. Yao,et al.  Functional approach of flexibly modelling generalized longitudinal data and survival time , 2008 .

[36]  D. Hall.  Measurement Error in Nonlinear Models: A Modern Perspective , 2008 .

[37]  S. Ratcliffe,et al.  Joint Modeling of Longitudinal and Survival Data via a Common Frailty , 2004, Biometrics.

[38]  Arnab Maity,et al.  Reduced Rank Mixed Effects Models for Spatially Correlated Hierarchical Functional Data , 2010, Journal of the American Statistical Association.

[39]  Geoffrey J. McLachlan,et al.  Finite Mixture Models , 2019, Annual Review of Statistics and Its Application.

[40]  J. Rosenthal,et al.  Optimal scaling for various Metropolis-Hastings algorithms , 2001 .

[41]  Catherine A. Sugar,et al.  Principal component models for sparse functional data , 1999 .

[42]  R. Sinha,et al.  Gender differences in cardiovascular and corticoadrenal response to stress and drug cues in cocaine dependent individuals , 2006, Psychopharmacology.

[43]  Xiao-Li Meng,et al.  Using EM to Obtain Asymptotic Variance-Covariance Matrices: The SEM Algorithm , 1991 .

[44]  Rajita Sinha,et al.  The role of stress in addiction relapse , 2007, Current psychiatry reports.

[45]  J. Ibrahim,et al.  Model Selection Criteria for Missing-Data Problems Using the EM Algorithm , 2008, Journal of the American Statistical Association.