Estimating Time to Event From Longitudinal Categorical Data

The expanded disability status scale (EDSS) is an ordinal score that measures progression in multiple sclerosis (MS). Progression is defined as reaching EDSS of a certain level (absolute progression) or increasing EDSS by one point (relative progression). Survival methods for time to progression are not adequate for such data because they do not exploit the EDSS level at the end of follow-up. Instead, we suggest a Markov transitional model applicable for repeated categorical or ordinal data. This approach enables derivation of covariate-specific survival curves, obtained after estimation of the regression coefficients and manipulations of the resulting transition matrix. Large-sample theory and resampling methods are employed to derive pointwise confidence intervals, which perform well in simulation. Methods for generating survival curves for time to EDSS of a certain level, time to increase EDSS by at least one point, and time to two consecutive visits with EDSS greater than 3 are described explicitly. The regression models described are easily implemented using standard software packages. Survival curves are obtained from the regression results using packages that support simple matrix calculation. We present and demonstrate our method on data collected at the Partners Multiple Sclerosis Center in Boston. We apply our approach to progression defined by time to two consecutive visits with EDSS greater than 3 and calculate crude (without covariates) and covariate-specific curves.

[1]  Laurence L. George,et al.  The Statistical Analysis of Failure Time Data , 2003, Technometrics.

[2]  A. Agresti,et al.  Categorical Data Analysis , 1991, International Encyclopedia of Statistical Science.

[3]  Patrick J Heagerty,et al.  Marginalized Transition Models and Likelihood Inference for Longitudinal Categorical Data , 2002, Biometrics.

[4]  J. Kalbfleisch,et al.  The Analysis of Panel Data under a Markov Assumption , 1985 .

[5]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data , 1980 .

[6]  A. Zeghnoun,et al.  Assessment of short‐term association between health outcomes and ozone concentrations using a Markov regression model , 2003 .

[7]  P S Albert,et al.  A Markov model for sequences of ordinal data from a relapsing-remitting disease. , 1994, Biometrics.

[8]  C Confavreux,et al.  Relapses and progression of disability in multiple sclerosis. , 2000, The New England journal of medicine.

[9]  Susan A Gauthier,et al.  A model for the comprehensive investigation of a chronic autoimmune disease: the multiple sclerosis CLIMB study. , 2006, Autoimmunity reviews.

[10]  P. Albert,et al.  A two-state Markov chain for heterogeneous transitional data: a quasi-likelihood approach. , 1998, Statistics in medicine.

[11]  Benjamin Kedem,et al.  Prediction and Classification of Non-stationary Categorical Time Series , 1998 .

[12]  F. Harrell,et al.  Partial Proportional Odds Models for Ordinal Response Variables , 1990 .

[13]  J. Klein,et al.  Statistical Models Based On Counting Process , 1994 .

[14]  Rebecca A. Betensky,et al.  Simultaneous confidence intervals based on the percentile bootstrap approach , 2008, Comput. Stat. Data Anal..

[15]  James R. Kenyon,et al.  Analysis of Multivariate Survival Data , 2002, Technometrics.

[16]  Anders Odén,et al.  Prediction of outcome in multiple sclerosis based on multivariate models , 1994, Journal of Neurology.

[17]  Gary G. Koch,et al.  Categorical data analysis using the sas® system, 2nd edition , 2000 .

[18]  Gary G. Koch,et al.  Categorical Data Analysis Using The SAS1 System , 1995 .

[19]  H. J. Hansen,et al.  Multicentre, randomised, double blind, placebo controlled, phase III study of weekly, low dose, subcutaneous interferon beta-1a in secondary progressive multiple sclerosis , 2004, Journal of Neurology, Neurosurgery & Psychiatry.

[20]  H. Kaufmann,et al.  Regression Models for Nonstationary Categorical Time Series: Asymptotic Estimation Theory , 1987 .

[21]  Paul S. Albert,et al.  A Random Effects Transition Model For Longitudinal Binary Data With Informative Missingness , 2003 .

[22]  Marco Alfò,et al.  Longitudinal analysis of repeated binary data using autoregressive and random effect modelling , 2003 .

[23]  PARTIAL LIKELIHOOD ESTIMATION IN CATEGORICAL TIME SERIES WITH STOCHASTIC COVARIATES , 1998 .

[24]  Hsin-Chou Yang,et al.  Modeling Animals' Behavioral Response by Markov Chain Models for Capture–Recapture Experiments , 2005, Biometrics.

[25]  P. Diggle.  Analysis of Longitudinal Data , 1995 .

[26]  A. Agresti.  Modelling ordered categorical data: recent advances and future challenges. , 1999, Statistics in medicine.

[27]  B. Weinshenker,et al.  Not every patient with multiple sclerosis should be treated at time of diagnosis. , 2006, Archives of neurology.

[28]  J. Baskerville,et al.  The natural history of multiple sclerosis: a geographically based study. 7. Progressive-relapsing and relapsing-progressive multiple sclerosis: a re-evaluation. , 1999, Brain : a journal of neurology.

[29]  Gary G. Koch,et al.  Categorical Data Analysis , 2005 .

[30]  T. W. Anderson,et al.  Statistical Inference about Markov Chains , 1957 .

[31]  P. Adeleine,et al.  Early clinical predictors and progression of irreversible disability in multiple sclerosis: an amnesic process. , 2003, Brain : a journal of neurology.

[32]  J. Baskerville,et al.  The natural history of multiple sclerosis: a geographically based study. 5. The clinical features and natural history of primary progressive multiple sclerosis. , 1999, Brain : a journal of neurology.

[33]  B Bass,et al.  The natural history of multiple sclerosis: a geographically based study. 2. Predictive value of the early clinical course. , 1989, Brain : a journal of neurology.

[34]  B. Kedem,et al.  Regression Theory for Categorical Time Series , 2003 .

[35]  R. D'Agostino.  Adjustment Methods: Propensity Score Methods for Bias Reduction in the Comparison of a Treatment to a Non‐Randomized Control Group , 2005 .