EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Use of Inverse Probability Weighting in Cardiovascular Surgical Outcomes Research: Principles, Limitations, and Recommendations

Zhongjian Liu, Wei-Guo Ma iD, Lei Liu iD, Prashanth Vallabhajosyula, Haibo Wang

European Journal of Cardio-Thoracic Surgery · 2026

DOI: 10.1093/ejcts/ezag122 ↗

Kurz gefasst

Fachartikel in der Zeitschrift European Journal of Cardio-Thoracic Surgery, 2026. Verfasst von Zhongjian Liu und 4 weiteren Autor:innen. Beteiligte Einrichtung: Peking University First Hospital, Peking University, Beijing 100034, China. Hauptthemen laut MEDLINE: Cardiovascular Surgical Procedures, Outcome Assessment, Health Care. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 12 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Major Science and Technology Major Projects.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This statistical primer presents a comprehensive introduction to the use of inverse probability of treatment weighting (IPTW), a method for rebalancing group confounders at baseline in cardiovascular surgical outcomes research. We leveraged a retrospective study comparing the outcomes of 2 different surgical approaches to aortic arch pathology to illustrate the essential aspects of performing IPTW, including basic theories, application, extreme weights, underestimated variance, heterogeneity of treatment effect, and multicollinearity. The limitations of IPTW were discussed with examples of solutions. Ten practical recommendations are made on avoiding the pitfalls in the use of IPTW. Proper use of IPTW may effectively reduce confounding bias by balancing confounders between different groups at baseline, retain a relatively large sample size, and allow for integration with other statistical methods. However, it is crucial to be fully aware of its limitations and carefully select appropriate methods to avoid improper use and data misinterpretation in cardiovascular outcomes research.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cardiovascular Surgical ProceduresOutcome Assessment, Health CareHumansAorta, ThoracicProbabilityRetrospective StudiesPropensity Score

Stichwörter der Autor:innen

confounding factorinverse probability of treatment weightingoutcomes researchpropensity score

Autor:innen und Institutionen

  • Peking University First Hospital, Peking University, Beijing 100034, China.
  • Clinical Research Institute/Institute of Advanced Clinical Medicine, Key Laboratory of Epidemiology of Major Diseases, Peking University, Beijing 100191, China.
  • Key Laboratory of Epidemiology of Major Diseases (Peking University), Beijing 100191, China.
  • Division of Cardiac Surgery, Yale University School of Medicine, New Haven, CT 06519, United States.
  • Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Beijing 100029, China.
  • Division of Biostatistics, Washington University in St. Louis, St. Louis, MO 63110, United States.

ORCID-Profile: Wei-Guo Ma, Lei Liu

Förderung

  • National Major Science and Technology Major Projects

Zitieren

APA

Liu, Z., Ma, W. G., Liu, L., Vallabhajosyula, P., & Wang, H. (2026). Use of Inverse Probability Weighting in Cardiovascular Surgical Outcomes Research: Principles, Limitations, and Recommendations. European Journal of Cardio-Thoracic Surgery. https://doi.org/10.1093/ejcts/ezag122

Vancouver

Liu Z, Ma WG, Liu L, Vallabhajosyula P, Wang H. Use of Inverse Probability Weighting in Cardiovascular Surgical Outcomes Research: Principles, Limitations, and Recommendations. European Journal of Cardio-Thoracic Surgery. 2026. doi:10.1093/ejcts/ezag122

BibTeX
@article{Liu2026,
  author = {Liu, Zhongjian and Ma, Wei-Guo and Liu, Lei and Vallabhajosyula, Prashanth and Wang, Haibo},
  title = {Use of Inverse Probability Weighting in Cardiovascular Surgical Outcomes Research: Principles, Limitations, and Recommendations},
  journal = {European Journal of Cardio-Thoracic Surgery},
  year = {2026},
  doi = {10.1093/ejcts/ezag122}
}
Kostenlos starten →