EUVIMEDEUVIMED entdecken
MetaanalyseOpen Access · CC BY 4.0

Multiple imputation of incomplete multilevel data using Heckman selection models

Johanna Muñoz iD, Orestis Efthimiou iD, Vincent Audigier iD, Valentijn M T de Jong iD, Thomas P A Debray iD

Statistics in Medicine · 10.02.2024 · 43 (3) · 514-533

DOI: 10.1002/sim.9965 ↗

Kurz gefasst

Metaanalyse in der Zeitschrift Statistics in Medicine, online erschienen am 11.12.2023. Verfasst von Johanna Muñoz und 4 weiteren Autor:innen. Beteiligte Einrichtung: Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands. Hauptthemen laut MEDLINE: Biomedical Research. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Horizon 2020 Framework Programme.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Missing data is a common problem in medical research, and is commonly addressed using multiple imputation. Although traditional imputation methods allow for valid statistical inference when data are missing at random (MAR), their implementation is problematic when the presence of missingness depends on unobserved variables, that is, the data are missing not at random (MNAR). Unfortunately, this MNAR situation is rather common, in observational studies, registries and other sources of real‐world data. While several imputation methods have been proposed for addressing individual studies when data are MNAR, their application and validity in large datasets with multilevel structure remains unclear. We therefore explored the consequence of MNAR data in hierarchical data in‐depth, and proposed a novel multilevel imputation method for common missing patterns in clustered datasets. This method is based on the principles of Heckman selection models and adopts a two‐stage meta‐analysis approach to impute binary and continuous variables that may be outcomes or predictors and that are systematically or sporadically missing. After evaluating the proposed imputation model in simulated scenarios, we illustrate it use in a cross‐sectional community survey to estimate the prevalence of malaria parasitemia in children aged 2‐10 years in five regions in Uganda.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Biomedical ResearchChildHumansCross-Sectional StudiesUganda

Stichwörter der Autor:innen

Heckman modelIPDMAmissing not at randommultiple imputationselection models

Autor:innen und Institutionen

  • Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
  • Institute of Primary Health Care (BIHAM), University of Bern, Bern, Switzerland.
  • Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland.
  • Conservatoire national des arts et métiers (CNAM), Laboratoire CEDRIC-MSDMA, Paris, France.
  • Data Analytics and Methods Task Force, European Medicines Agency, Amsterdam, The Netherlands.
  • Smart Data Analysis and Statistics, Utrecht, The Netherlands.

ORCID-Profile: Johanna Muñoz, Orestis Efthimiou, Vincent Audigier, Valentijn M T de Jong, Thomas P A Debray

Förderung

  • Horizon 2020 Framework Programme

Zitieren

APA

Muñoz, J., Efthimiou, O., Audigier, V., de Jong, V. M. T., & Debray, T. P. A. (2024). Multiple imputation of incomplete multilevel data using Heckman selection models. Statistics in Medicine, 43(3), 514-533. https://doi.org/10.1002/sim.9965

Vancouver

Muñoz J, Efthimiou O, Audigier V, de Jong VMT, Debray TPA. Multiple imputation of incomplete multilevel data using Heckman selection models. Statistics in Medicine. 2024;43(3):514-533. doi:10.1002/sim.9965

BibTeX
@article{Muoz2024,
  author = {Muñoz, Johanna and Efthimiou, Orestis and Audigier, Vincent and de Jong, Valentijn M T and Debray, Thomas P A},
  title = {Multiple imputation of incomplete multilevel data using Heckman selection models},
  journal = {Statistics in Medicine},
  year = {2024},
  volume = {43},
  number = {3},
  pages = {514--533},
  doi = {10.1002/sim.9965}
}
Kostenlos starten →