Sensitivity models for missing covariates in the analysis of survival data from multiple surveys
Hemming, Karla and Hutton, Jane L. (2007) Sensitivity models for missing covariates in the analysis of survival data from multiple surveys. Working Paper. University of Warwick. Centre for Research in Statistical Methodology, Coventry.
WRAP_Hemming_07-12w.pdf - Published Version - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Official URL: http://www2.warwick.ac.uk/fac/sci/statistics/crism...
Using individual patient data from five independent surveys, we evaluate regional variations in survival in cerebral palsy. The influence of four important variables measuring disability, which are only partially observed for many cases, are analysed. Results are compared between a naive complete case analysis; a full likelihood model in which the covariates are assumed to be missing at random and in which each of the binary predictor variables are modelled as independent Bernoulli random variables; a model in which the covariates are modelled by a conditionalwise sequence, accommodating dependencies between the likelihoods of having various mixtures of disabilities; and a model in which the likelihood of a predictor variable being observed is allowed to depend on the value of the covariate itself (NMAR). Fully parametric survival regression models are used and analysis carried out in BUGS. Results suggest that proportions recorded as having severe visual or cognitive impairments are substantially lower than the actual proportions severely impaired. Associations between the likelihood of a particular covariate being recorded and the likelihood of a more severe disability imply that life expectancies for those who are very severely impaired may be up to 20% less than inferences based on complete case analyses.
|Item Type:||Working or Discussion Paper (Working Paper)|
|Subjects:||Q Science > QA Mathematics|
|Divisions:||Faculty of Science > Statistics|
|Library of Congress Subject Headings (LCSH):||Survival analysis (Biometry), Cerebral palsy -- Mathematical models, Missing observations (Statistics)|
|Series Name:||Working papers|
|Publisher:||University of Warwick. Centre for Research in Statistical Methodology|
|Place of Publication:||Coventry|
|Number of Pages:||26|
|Status:||Not Peer Reviewed|
|Access rights to Published version:||Open Access|
|References:||Baker, S. G. (1994). Regression analysis of grouped survival data with incomplete covariates: non-ignorable missing data and censoring mechanisms. Biometrics 50, 821–826. Baker, S. G. and N. M. Laird (1988). Regression analysis for categorical variables with outcome subject to non-ignorable non response. Journal of the American Statistical Association 83, 62–69. Best, N., M. K. Cowles, and K. Vines (1997). CODA: Convergence diagnostics and output analysis software for Gibbs sampling output, version 0.4. Downloadable from http://www.mrc-bsu.cam.ac.uk/bugs/. Chen, M. H., J. G. Ibrahim, and S. R. Lipsitz (2002). Bayesian methods for missing covariates in cure rate models. Lifetime data analysis 8, 117–146. Cho, M. and N. Schenker (1999). Fitting the log-F accelerated failure time model with incomplete covariate data. Biometrics 55, 826–833. Evans, P. M., S. J. W. Evans, and E. Alberman (1990). Cerebral palsy: why we must plan for survival. Arch. Dis. Ch. 65, 1329–1333. Hemming, K., J. L. Hutton, S. Bonellie, and J. Kurinczuk (2007). Intra uterine growth and survival in cerebral palsy. Archives of Diseases in Childhood (in press). Hemming, K., J. L. Hutton, A. Colver, and M. J. Platt (2005). Regional variation in survival of people with cerebral palsy in the united kingdom. Pediatrics 116, 1383–1390. Hemming, K., J. L. Hutton, S. V. Glinianaia, S. Jarvis, andM. J. Platt (2006). A comparison of birthweight standards for europe. Developmental Medicine and Childhood Neurology 48, 906–912. Hemming, K., J. L. Hutton, and P. O. D. Pharoah (2006). Long term survival for a cohort of children with cerebral palsy. Developmental Medicine and Childhood Neurology 48, 90–95. Herring, A. J. and J. G. Ibrahim (2001). Likelihood-based methods for missing covariates in the Cox proportional hazards model. Journal of the American Statistical Association 96, 292–302. Herring, A. J. and J. G. Ibrahim (2002). Maximum likelihood estimation in random effects cure rate models with non-ignorable missing covariates. Biostatistics 3, 387–405. Herring, A. J., J. G. Ibrahim, and S. R. Lipsitz (2002). Frailty models with missing covariates. Biometrics 58, 98–109. Huang, L., M.-H. Chen, and J. G. Ibrahim (2005). Bayesian analysis for generalised linear models with non-ignorably missing covariates. Biometrics 61, 767–780. Hutton, J. L., A. F. Colver, and P. C. Mackie (2000). Effect of severity of disability on survival in north east England cerebral palsy cohort. Arch. Dis. Ch. 83, 468–473. Hutton, J. L. and P. F. Monoghan (2002). Choice of parametric accelerated life and proportional hazards models for survival data: asymptotic results. Lifetime data analysis 8, 375–393. Hutton, J. L. and P. O. D. Pharoah (2002). Effect of cognitive, motor and sensory disabilities on survival in cerebral palsy. Arch. Dis. Ch. 86, 84–89. Ibrahim, J. G., S. R. Lipsitz, and M. H. Chen (1999). Missing covariates in generalised linear models when the missing data mechanism is non-ignorable. Journal of the Royal Statistical Society, B 61, 173–190. Jarvis, S., S. V. Glinianaia, M. G. Torrioli, M. J. Platt, M. Miceli, P. S. Jouk, A. Johnson, J. Hutton, K. Hemming, G. Hagberg, H. Dolk, and J. Chalmers (2003). Cerebral palsy and intrauterine growth in single births: European collaborative study. The Lancet 362, 1106–1111. Kwong, G. P. S. and J. L. Hutton (2003). Choice of parametric models in survival analysis: applications to monotherapy for epilepsy and cerebral palsy. Journal of the Royal Statistical Society: Series C (Applied Statistics) 52, 153–168. Lipsitz, S. R. and J. G. Ibrahim (1996a). A conditional model for incomplete covariates in parametric regression models. Biometrika 83, 916–922. Lipsitz, S. R. and J. G. Ibrahim (1996b). Using the EM algorithm for survival data with incomplete categorical covariates. Lifetime Data Analysis 2, 5–14. Meng, X. and N. Schenker (1999). Maximum likelihood estimation for linear regression models with right censored outcomes and missing predictors. Computational statistics and data analysis 29, 471–483. Rathouz, P. L. (2007). Identifiability assumptions for missing covariate data in failure time regression models. Biostatistics 8, 345–356. Rubin, D. B. (2002). Statistical analysis of missing data (second edition). Wiley interscience. Schluchter, M. D. and K. L. Jackson (1989). Log-linear analysis of censored survival data with partially observed covariates. Journal of the American Statistical Association 84, 42–52. Spiegelhalter, D. J., N. G. Best, B. P. Carlin, and A. van der Linde (2002). Bayesian measures of model complexity and fit (with discussion). Journal of the Royal Statistical Society: Series B (Statistical Methodology) 64, 583–639. Spiegelhalter, D. J., A. Thomas, and N. G. Best (1999). WinBUGS Version 1.2 User Manual. Downloadable from http://www.mrc-bsu.cam.ac.uk/bugs/. Stubbendick, A. L. and J. G. Ibrahim (2003). Maximum likelihood methods for nonignorable missing responses and covariates in random effects models. Biometrics 59, 1140–1150. Surman, G., S. Bonellie, J. Chalmers, A. Colver, H. Dolk, K. Hemming, A. King, J. Kurinczuk, J. Parkes, and M. J. Platt (2006). Ukcp: a collaborative network of cerebral palsy registers in the united kingdom. Journal of Public Health 28, 148–156.|
Actions (login required)