Document is current

Article History

Received: 11 April 2024

Accepted: 12 May 2025

First Online: 17 May 2025

Declarations

:

: The authors declare no competing interests.

: Supplementary material is available online.

: The MIMIC dataset, version 1.4, utilized in this study is publicly available and can be accessed online6.

: The simulated dataset, which emulates the missing values observed in the real-world Geisinger data, will be made available upon request, request can be made to Jiang Li (jli@geisinger.edu). The simulated data is accessible via the GitHub repository at .

: Due to privacy and confidentiality concerns, we are unable to publicly share the Geisinger real-world dataset. The data was collected from a large integrated healthcare system encompassing multiple hospitals in the United States, necessitating strict adherence to privacy regulations and ethical considerations. However, data can be shared upon execution of a data-sharing agreement; interested parties can contact Jiang Li (jli@geisinger.edu) to request access.

: Due to privacy and confidentiality concerns, we are also unable to publicly share the Penn State Health real-world dataset. However, data can be shared upon execution of a data-sharing agreement; interested parties can contact Vida Abedi or Wenke Hwang (vabedi@pennstatehealth.psu.edu or whwang@pennstatehealth.psu.edu) to request access.

: To promote transparency and enable reproducibility, the code used in this research and the Pympute package is released as open-source resources on GitHub () (upon publication). Interested researchers can access the codebase, utilize Pympute for their imputation tasks, and replicate the methodology employed in this study.