A Markov regression random‐effects model for remission of functional disability in patients following a first stroke: A Bayesian approach

Few attempts have been made to model the dynamics of stroke-related disability. It is possible though, using panel data and multi-state Markov regression models that incorporate measured covariates and latent variables (random effects). This study aimed to model a series of functional transitions (following a first stroke) using a three-state Markov model with or without considering random effects. Several proportional hazards parameterizations were considered. A Bayesian approach that utilizes the Markov Chain Monte Carlo (MCMC) and Gibbs sampling functionality of WinBUGS (a Windows-based Bayesian software package) was developed to generate the marginal posterior distributions of the various transition parameters (e.g. the transition rates and transition probabilities). Model building and comparisons was guided by reference to the deviance information criteria (DIC). Of the four proportional hazards models considered, exponential regression was preferred because it led to the smallest deviances. Adding random effects further improved the model fit. Of the covariates considered, only age, infarct size, and baseline functional status were significant. By using our final model we were able to make individual predictions about functional recovery in stroke patients.

[1]  Carlo Caltagirone,et al.  Early and long-term outcome of rehabilitation in stroke patients: the role of patient characteristics, time of initiation, and duration of interventions. , 2003, Archives of physical medicine and rehabilitation.

[2]  José Luis González,et al.  Quality of Life Among Stroke Survivors Evaluated 1 Year After Stroke: Experience of a Stroke Unit , 2000, Stroke.

[3]  A. Ashburn,et al.  People with stroke living in the community: Attention deficits, balance, ADL ability and falls , 2003, Disability and rehabilitation.

[4]  S Lin,et al.  An algorithm for Monte Carlo estimation of genotype probabilities on complex pedigrees , 1994, Annals of human genetics.

[5]  D. Balding,et al.  Measuring gametic disequilibrium from multilocus data. , 2001, Genetics.

[6]  Valen E. Johnson.  A Bayesian x2 test for goodness-of-fit , 2004 .

[7]  D. Wagner,et al.  Predicting Outcome in Ischemic Stroke: External Validation of Predictive Risk Models , 2003, Stroke.

[8]  A. Gelman.  Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper) , 2004 .

[9]  Y. Ohashi,et al.  A Bayesian hierarchical survival model for the institutional effects in a multi-centre cancer clinical trial. , 1998, Statistics in medicine.

[10]  L. Tabár,et al.  Estimation of Sojourn Time in Chronic Disease Screening Without Data on Interval Cases , 2000, Biometrics.

[11]  Amy Ming-Fang Yen,et al.  SAS macro program for non-homogeneous Markov process in modeling multi-state disease progression , 2004, Comput. Methods Programs Biomed..

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

[13]  A. Mackenzie,et al.  Predictors of quality of life following stroke , 2002, Disability and rehabilitation.

[14]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[15]  R. Cook.  A Mixed Model for Two‐State Markov Processes Under Panel Observation , 1999, Biometrics.

[16]  Y. Beloosesky,et al.  The importance of brain infarct size and location in predicting outcome after stroke. , 1995, Age and ageing.

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

[18]  James K. Lindsey,et al.  Parametric Statistical Inference , 1996 .

[19]  C. Viboud,et al.  A Bayesian MCMC approach to study transmission of influenza: application to household longitudinal data , 2004, Statistics in medicine.

[20]  Walter R. Gilks,et al.  A Language and Program for Complex Bayesian Modelling , 1994 .

[21]  L. Sharples,et al.  Use of the Gibbs sampler to estimate transition rates between grades of coronary disease following cardiac transplantation. , 1993, Statistics in medicine.

[22]  G. Raab,et al.  Modelling human immunodeficiency virus infection and acquired immune deficiency syndrome cases in Scotland: data sources, prior information and Bayesian estimation , 1998, Journal of the Royal Statistical Society. Series A,.

[23]  E W Lee,et al.  The analysis of correlated panel data using a continuous-time Markov model. , 1998, Biometrics.

[24]  R. Lin,et al.  Effectiveness of low-dose ASA in prevention of secondary ischemic stroke, the ASA Study Group in Taiwan. , 1997, Thrombosis research.

[25]  Mahoney Fi,et al.  FUNCTIONAL EVALUATION: THE BARTHEL INDEX. , 1965 .

[26]  D. Cox.  Regression Models and Life-Tables , 1972 .

[27]  L. Tabár,et al.  Estimation of mean sojourn time in breast cancer screening using a Markov chain model of both entry to and exit from the preclinical detectable phase. , 1995, Statistics in medicine.

[28]  Tony Hsiu-Hsi Chen,et al.  Location and size of infarct on functional outcome of noncardioembolic ischemic stroke , 2006, Disability and rehabilitation.

[29]  R. Jacobs,et al.  Bayesian inference for hierarchical mixtures-of-experts with applications to regression and classification , 1996, Statistical methods in medical research.

[30]  I. König,et al.  Predicting functional outcome and survival after acute ischemic stroke , 2002, Journal of Neurology.

[31]  N. Welton,et al.  Estimation of Markov Chain Transition Probabilities and Rates from Fully and Partially Observed Data: Uncertainty Propagation, Evidence Synthesis, and Model Calibration , 2005, Medical decision making : an international journal of the Society for Medical Decision Making.

[32]  P. Burney,et al.  Predicting the Disability of First‐Time Stroke Sufferers at 1 Year: 12‐Month Follow‐up of a Population‐Based Cohort in Southeast England , 1994, Stroke.

[33]  C. Warlow.  Epidemiology of stroke , 1998, The Lancet.

[34]  R H Jones,et al.  Multi-state models and diabetic retinopathy. , 1995, Statistics in medicine.

[35]  M. Yen,et al.  A case–cohort study for the disease natural history of adenoma–carcinoma and de novo carcinoma and surveillance of colon and rectum after polypectomy: implication for efficacy of colonoscopy , 2003, British Journal of Cancer.

[36]  W. Alves,et al.  Ischemic stroke: relation of age, lesion location, and initial neurologic deficit to functional outcome. , 1998, Archives of physical medicine and rehabilitation.

[37]  J Wakefield,et al.  The hierarchical Bayesian approach to population pharmacokinetic modelling. , 1994, International journal of bio-medical computing.

[38]  E. W. Olsen,et al.  Recurrence in affective disorder: analyses with frailty models. , 1999, American journal of epidemiology.