Non-ignorable missing covariate data in parametric survival analysis

Doctoral thesis English OPEN
Boyd, Katherine
  • Subject: QA

Within\ud any epidemiological study missing\ud data is\ud almost inevitable.\ud This\ud missing\ud data is often ignored; however,\ud unless we can assume quite\ud restrictive mechanisms, this will\ud lead to biased estimates. Our\ud motivation\ud are data collected to study the long-term\ud effect of severity of disability\ud upon\ud survival in children with cerebral palsy (henceforth CP). The analysis of\ud such an old\ud data set brings to light\ud statistical difficulties. The main issue in\ud this data is the amount of missing covariate data. We raise concerns about\ud the mechanism causing data to be missing.\ud We present a flexible\ud class of joint models for the survival times and the\ud missing\ud data mechanism which allows us to vary the mechanism causing\ud the missing\ud data. Simulation studies prove this model to be both\ud precise\ud and reliable in estimating survival with missing data. We show that long\ud term survival in the moderately\ud disabled is high and, therefore, a large\ud proportion will\ud be\ud surviving to times when they require care specifically\ud for\ud elderly CP sufferers.\ud In\ud particular, our models suggest that survival\ud from diagnosis is considerably higher than has been previously estimated\ud from this data.\ud This thesis contributes to the discussion of possible methods for dealing\ud with\ud NMAR data.\ud
  • References (19)
    19 references, page 1 of 2

    Anderson, R, Borgan, 0., Gill, R. & Kielding, N. (1989),StatisticalModels Basedon CountingProcesseSsp,ringer-Verlag,New York.

    Chen, H. & Little, R. (1999), 'Proportional hazards regression with missing covariates', Journalof theAmerican Statistical Association94,896-908.

    Cho, A& Schenker,N. (1999),'Fitting the log-f acceleratedfailure time model with incompletecovariatedata', Biometrics55,826-833.

    Collett, D. (1999), Modelling Survival Data in Medical Res-earclCi, hapman & Hall / CRC,London.

    Copas, J. & Li, H. 0 997), 'Inference for non-random samples', Journal of the RoyalStatistical Society,SeriesB 59,55-95.

    Copas,J.& Shi,J. (2000),'Meta analysis,funnel plots and sensitivity analysis', Biostatistics1,247-262.

    Copas, J. & Shi, J. (2001), 'A sensitivity analysis for publication bias in systernatic reviews', Statistical Methodsin Medical Research10,251-265.

    Cox, D. (1972), 'Regression models and life tables', fourizal of the Royal Statistical Society,B 34,187-220.

    Cox, D. & Oakes, D. (1984), Analysis of Survival Data, Chapman & Hall / CRC, London.

    Crowder,M., Kimber,A., Smith, R. & Sweeting,T.(1991),StatisticaAlnalstl-si of Reliability Data, Chapman & Hall, London.

  • Metrics
    0
    views in OpenAIRE
    0
    views in local repository
    19
    downloads in local repository

    The information is available from the following content providers:

    From Number Of Views Number Of Downloads
    Warwick Research Archives Portal Repository - IRUS-UK 0 19
Share - Bookmark