On the performance of random‐coefficient pattern‐mixture models for non‐ignorable drop‐out

Random-coefficient pattern-mixture models (RCPMMs) have been proposed for longitudinal data when drop-out is thought to be non-ignorable. An RCPMM is a random-effects model with summaries of drop-out time included among the regressors. The basis of every RCPMM is extrapolation. We review RCPMMs, describe various extrapolation strategies, and show how analyses may be simplified through multiple imputation. Using simulated and real data, we show that alternative RCPMMs that fit equally well may lead to very different estimates for parameters of interest. We also show that minor model misspecification can introduce biases that are quite large relative to standard errors, even in fairly small samples. For many scientific applications, where the form of the population model and nature of the drop-out are unknown, interval estimates from any single RCPMM may suffer from undercoverage because uncertainty about model specification is not taken into account.

[1]  Yang C. Yuan,et al.  Multiple Imputation for Missing Data: Concepts and New Development , 2000 .

[2]  D. Rubin.  Multiple imputation for nonresponse in surveys , 1989 .

[3]  N M Laird,et al.  Mixture models for the joint distribution of repeated measures and event times. , 1997, Statistics in medicine.

[4]  Geert Molenberghs,et al.  Discussion of Diggle, P. and Kenward, M. G.: 'Informative drop-out in longitudinal data analysis' , 1994 .

[5]  D. Bates,et al.  Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data , 1988 .

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

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

[8]  George E. P. Box,et al.  Empirical Model‐Building and Response Surfaces , 1988 .

[9]  D. Bates,et al.  Mixed-Effects Models in S and S-PLUS , 2001 .

[10]  R Little,et al.  Intent-to-treat analysis for longitudinal studies with drop-outs. , 1996, Biometrics.

[11]  J. Schafer,et al.  Computational Strategies for Multivariate Linear Mixed-Effects Models With Missing Values , 2002 .

[12]  M. Kenward,et al.  Informative Drop‐Out in Longitudinal Data Analysis , 1994 .

[13]  J M Taylor,et al.  Multiple Imputation and Posterior Simulation for Multivariate Missing Data in Longitudinal Studies , 2000, Biometrics.

[14]  Donald Hedeker,et al.  Application of random-efiects pattern-mixture models for miss-ing data in longitudinal studies , 1997 .

[15]  D. Rubin,et al.  Statistical Analysis with Missing Data. , 1989 .

[16]  M P Becker,et al.  A multiple imputation strategy for incomplete longitudinal data , 2001, Statistics in medicine.

[17]  P. Royston,et al.  Regression using fractional polynomials of continuous covariates: parsimonious parametric modelling. , 1994 .

[18]  J. Schafer,et al.  A comparison of inclusive and restrictive strategies in modern missing data procedures. , 2001, Psychological methods.

[19]  D. Rubin.  INFERENCE AND MISSING DATA , 1975 .

[20]  P. Diggle,et al.  Analysis of Longitudinal Data , 2003 .

[21]  N M Laird,et al.  Missing data in longitudinal studies. , 1988, Statistics in medicine.

[22]  Roderick J. A. Little,et al.  A Class of Pattern-Mixture Models for Normal Incomplete Data , 1994 .

[23]  Geert Molenberghs,et al.  Monotone missing data and pattern‐mixture models , 1998 .

[24]  Joseph L Schafer,et al.  Analysis of Incomplete Multivariate Data , 1997 .

[25]  J. Ware,et al.  Random-effects models for longitudinal data. , 1982, Biometrics.

[26]  N M Laird,et al.  An alternative parameterization of the general linear mixture model for longitudinal data with non‐ignorable drop‐outs , 2001, Statistics in medicine.

[27]  K. Bailey,et al.  Estimation and comparison of changes in the presence of informative right censoring: conditional linear model. , 1989, Biometrics.

[28]  Peter McCullagh,et al.  Some Statistical Properties of a Family of Continuous Univariate Distributions , 1989 .

[29]  R. Little,et al.  Pattern-mixture models for multivariate incomplete data with covariates. , 1996, Biometrics.

[30]  Roderick J. A. Little,et al.  Modeling the Drop-Out Mechanism in Repeated-Measures Studies , 1995 .

[31]  Geert Molenberghs,et al.  Strategies to fit pattern-mixture models. , 2002, Biostatistics.

[32]  M. Kenward.  Selection models for repeated measurements with non-random dropout: an illustration of sensitivity. , 1998, Statistics in medicine.