• Missing data: A statistical framework for practice

    Type Journal Article
    Author James R. Carpenter
    Author Melanie Smuk
    URL https://onlinelibrary.wiley.com/doi/abs/10.1002/bimj.202000196
    Volume 63
    Issue 5
    Pages 915-947
    Publication Biometrical Journal
    ISSN 1521-4036
    Date 2021
    Extra _eprint: https://onlinelibrary.wiley.com/doi/pdf/10.1002/bimj.202000196
    DOI 10.1002/bimj.202000196
    Accessed 10/21/2021, 7:58:56 AM
    Library Catalog Wiley Online Library
    Language en
    Abstract Missing data are ubiquitous in medical research, yet there is still uncertainty over when restricting to the complete records is likely to be acceptable, when more complex methods (e.g. maximum likelihood, multiple imputation and Bayesian methods) should be used, how they relate to each other and the role of sensitivity analysis. This article seeks to address both applied practitioners and researchers interested in a more formal explanation of some of the results. For practitioners, the framework, illustrative examples and code should equip them with a practical approach to address the issues raised by missing data (particularly using multiple imputation), alongside an overview of how the various approaches in the literature relate. In particular, we describe how multiple imputation can be readily used for sensitivity analyses, which are still infrequently performed. For those interested in more formal derivations, we give outline arguments for key results, use simple examples to show how methods relate, and references for full details. The ideas are illustrated with a cohort study, a multi-centre case control study and a randomised clinical trial.
    Short Title Missing data
    Date Added 10/21/2021, 7:58:56 AM
    Modified 10/21/2021, 8:06:36 AM

    Tags:

    • teaching-mds
    • multiple-imputation
    • missing
    • teaching-paper
  • Missing data in prediction research: A five step approach for multiple imputation, illustrated in the CENTER-TBI study

    Type Journal Article
    Author Benjamin Gravesteijn
    Author Charlie Sewalt
    Author Esmee Venema
    Author Daan Nieboer
    Author Ewout W Steyerberg
    URL https://www.liebertpub.com/doi/abs/10.1089/neu.2020.7218
    Publication Journal of Neurotrauma
    ISSN 0897-7151
    Date January 20, 2021
    Extra Publisher: Mary Ann Liebert, Inc., publishers
    DOI 10.1089/neu.2020.7218
    Accessed 1/26/2021, 8:29:03 PM
    Library Catalog liebertpub.com (Atypon)
    Abstract In medical research, missing data is common. In acute diseases such as traumatic brain injury (TBI), even well conducted prospective studies may suffer from missing data in baseline characteristics and outcomes. Statistical models may simply drop patients with any missing values, potentially leaving a selected subset of the original cohort. Imputation is widely accepted by methodologists as an appropriate way to deal with missing data. We aim to provide practical guidance on handling missing data for prediction modelling. We hereto propose a five-step approach, centred around single and multiple imputation: 1) explore the missing data patterns; 2) choose a method of imputation; 3) perform imputation; 4) assess diagnostics of the imputation; and 5) analyse the imputed datasets. We illustrate these 5 steps with the estimation and validation of the IMPACT prognostic model in 1375 patients from the CENTER-TBI database, included in 53 centers across 17 countries, with moderate or severe TBI in the prospective European CENTER-TBI study. Future prediction modelling studies in acute diseases may benefit from following the suggested 5 steps for optimal statistical analysis and interpretation, after maximal effort have been made to minimize missing data.
    Short Title Missing data in prediction research
    Date Added 1/26/2021, 8:29:03 PM
    Modified 1/26/2021, 8:30:17 PM

    Tags:

    • teaching-mds
    • imputation
    • prediction
    • missing
  • Review: A gentle introduction to imputation of missing values

    Type Journal Article
    Author Donders
    Author Geert J. M. G. van der Heijden
    Author Theo Stijnen
    Author Karel G. M. Moons
    Volume 59
    Pages 1087-1091
    Publication J Clin Epi
    Date 2006
    Extra Citation Key: don06rev tex.citeulike-article-id= 13265491 tex.posted-at= 2014-07-14 14:09:58 tex.priority= 0
    Date Added 7/7/2018, 1:38:33 PM
    Modified 11/8/2019, 8:01:59 AM

    Tags:

    • teaching-mds
    • missing-data
    • imputation

    Notes:

    • simple demonstration of failure of the add new category method (indicator variable)

  • Using the outcome for imputation of missing predictor values was preferred

    Type Journal Article
    Author Karel G. M. Moons
    Author Rogier A. R. T. Donders
    Author Theo Stijnen
    Author Frank E. Harrell
    URL http://dx.doi.org/10.1016/j.jclinepi.2006.01.009
    Volume 59
    Pages 1092-1101
    Publication J Clin Epi
    Date 2006
    Extra Citation Key: moo06usi tex.citeulike-article-id= 13265492 tex.citeulike-linkout-0= http://dx.doi.org/10.1016/j.jclinepi.2006.01.009 tex.posted-at= 2014-07-14 14:09:58 tex.priority= 0
    DOI 10.1016/j.jclinepi.2006.01.009
    Date Added 7/7/2018, 1:38:33 PM
    Modified 11/8/2019, 8:01:59 AM

    Tags:

    • teaching-mds
    • missing-data
    • graphics
    • imputation
    • response
    • aregimpute
    • mice

    Notes:

    • use of outcome variable; excellent graphical summaries of simulations

  • Imputation of missing values is superior to complete case analysis and the missing-indicator method in multivariable diagnostic research: A clinical example

    Type Journal Article
    Author Geert J. M. G. van der Heijden
    Author Donders
    Author Theo Stijnen
    Author Karel G. M. Moons
    URL http://dx.doi.org/10.1016/j.jclinepi.2006.01.015
    Volume 59
    Pages 1102-1109
    Publication J Clin Epi
    Date 2006
    Extra Citation Key: hei06imp tex.citeulike-article-id= 13265493 tex.citeulike-linkout-0= http://dx.doi.org/10.1016/j.jclinepi.2006.01.015 tex.posted-at= 2014-07-14 14:09:58 tex.priority= 0
    DOI 10.1016/j.jclinepi.2006.01.015
    Date Added 7/7/2018, 1:38:33 PM
    Modified 11/8/2019, 8:01:59 AM

    Tags:

    • teaching-mds
    • precision
    • missing-data
    • bias
    • imputation
    • single-imputation
    • complete-case-analysis
    • extra-categories
    • missingness-indicator

    Notes:

    • Invalidity of adding a new category or an indicator variable for missing values even with MCAR