Document Type
Article
Publication Date
3-1-2023
Original Citation
Casiraghi E,
Wong R,
Hall M,
Coleman B,
Notaro M,
Evans M,
Tronieri J,
Blau H,
Laraway B,
Callahan T,
Chan L,
Bramante C,
Buse J,
Moffitt R,
Stürmer T,
Johnson S,
Raymond Shao Y,
Reese J,
Robinson P,
Paccanaro A,
Valentini G,
Huling J,
Wilkins K.
A method for comparing multiple imputation techniques: A case study on the U.S. national COVID cohort collaborative. J Biomed Inform. 2023;139:104295.
Keywords
JGM, Humans, COVID-19, Algorithms, Research Design, Bias, Probability
JAX Source
J Biomed Inform. 2023;139:104295.
ISSN
1532-0480
PMID
36716983
DOI
https://doi.org/10.1016/j.jbi.2023.104295
Abstract
Healthcare datasets obtained from Electronic Health Records have proven to be extremely useful for assessing associations between patients’ predictors and outcomes of interest. However, these datasets often suffer from missing values in a high proportion of cases, whose removal may introduce severe bias. Several multiple imputation algorithms have been proposed to attempt to recover the missing information under an assumed missingness mechanism. Each algorithm presents strengths and weaknesses, and there is currently no consensus on which multiple imputation algorithm works best in a given scenario. Furthermore, the selection of each algorithm’s pa- rameters and data-related modeling choices are also both crucial and challenging.
Comments
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/).