Comparison of methods to handle missing values in a continuous index test in a diagnostic accuracy study – a simulation study
Most methods perform well if the proportion of missing values is small. Given a higher proportion of missing values and MCAR, we would recommend to conduct a CCA and standard MI methods for a small and large sample size, respectively. In the absence of better alternatives we recommend to conduct a CCA and to discuss its limitations, if the sample size is small, and missing values are M(N)AR. Standard MI methods and the augmented inverse probability approach may be a good alternative, if the sample size and/or correlation increases. All methods are biased under MNAR and a low correlation.
Authors
- Johannes B. Reitsma (ORCID: https://orcid.org/0000-0003-4026-4345)
- Antonia Zapf (ORCID: https://orcid.org/0000-0002-8467-0508)
- Katharina Stahlmann (ORCID: https://orcid.org/0000-0002-4704-310X)
- Nandini Dendukuri (ORCID: https://orcid.org/0000-0002-2330-0976)
- Bas E Kellerhuis (ORCID: https://orcid.org/0009-0006-9996-1419)
Institutions
- Universität Hamburg (DE)
- Utrecht University (NL)
- University Medical Center Utrecht (NL)
- University of Oxford (GB)
- University Medical Center Hamburg-Eppendorf (DE)
- McGill University (CA)
Publication Details
- Journal
- BMC Medical Research Methodology
- Published
- 2025-05-27
- DOI
- https://doi.org/10.1186/s12874-025-02594-2
- Citations
- 1
- Primary Topic
- Statistical Methods and Bayesian Inference
- Type
- article
- Field-Weighted Citation Impact
- 2.63
Funders
- Deutsche Forschungsgemeinschaft