Accounting for bias due to outcome data missing not at random: comparison and illustration of two approaches to probabilistic bias analysis: a simulation study
The Monte Carlo bias analysis we describe is easy to implement in standard software and, in the setting we explored, is a viable alternative to a Bayesian bias analysis. We caution careful consideration of choice of auxiliary variables when applying imputation where data may be MNAR.
Authors
- Louise A C Millard (ORCID: https://orcid.org/0000-0003-4787-8411)
- Rachael A. Hughes (ORCID: https://orcid.org/0000-0003-0766-1410)
- George Davey Smith (ORCID: https://orcid.org/0000-0002-1407-8314)
- Emily Kawabata
- Kate M. Tilling (ORCID: https://orcid.org/0000-0002-1010-8926)
- Chin Yang Shapland (ORCID: https://orcid.org/0000-0002-5797-1241)
- Gemma Clayton (ORCID: https://orcid.org/0000-0002-9525-2758)
- Alice R Carter (ORCID: https://orcid.org/0000-0003-2817-4195)
- Maria Carolina Borges (ORCID: https://orcid.org/0000-0001-7785-4547)
- Debbie A. Lawlor (ORCID: https://orcid.org/0000-0002-6793-2262)
- Gareth J Griffith (ORCID: https://orcid.org/0000-0003-0481-3175)
- Tim P. Morris (ORCID: https://orcid.org/0000-0001-5850-3610)
- Alba Fernández‐Sanlés (ORCID: https://orcid.org/0000-0002-3587-8177)
- Daniel Major‐Smith (ORCID: https://orcid.org/0000-0001-6467-2023)
Institutions
- At Bristol (GB)
- University of Bristol (GB)
- MRC Clinical Trials Unit at UCL (GB)
- MRC Unit for Lifelong Health and Ageing (GB)
- MRC Epidemiology Unit (GB)
- University College London (GB)
Publication Details
- Journal
- BMC Medical Research Methodology
- Published
- 2024-11-13
- DOI
- https://doi.org/10.1186/s12874-024-02382-4
- Citations
- 7
- Primary Topic
- Statistical Methods and Bayesian Inference
- Type
- article
- Field-Weighted Citation Impact
- 4.96
Funders
- John Templeton Foundation
- Wellcome Trust
- UK Research and Innovation
- National Institute for Health and Care Research
- British Heart Foundation
- University of Bristol
- Medical Research Council