Causal Inference Approaches for Profiling | Professor Sharon-Lise Normand

Presenting at the STS Annual Meeting 2025

In her talk, “Improving Risk Adjustment in the Assessment of Congenital Heart Center Surgical Quality," Dr. Normand explained how modern causal inference techniques can offer a more accurate assessment of center performance.

SLN presenting STS CHSD research at STS Annual Meeting 2025

Modern Analytics to Improve Quality & Outcome Assessments Following Congenital Heart Surgery.

5R01HL162893-02

06/2022-05/2026

Annually more than 40,000 children undergo surgery for congenital heart defects (CHD) in the US and while patient outcomes have improved over the past 30 years, morbidity and mortality remain high for some children, and significant variation exists across the more than 100 US congenital heart centers. Using a registry containing detailed clinical data from virtually all U.S. congenital heart centers, we will develop machine learning and causal inference approaches to best utilize the diversity of the data to improve CHD surgical quality assessments. We expect the proposed research will transform our capabilities for assessing CHD surgical quality and drive national improvements in outcomes for children and families impacted by CHD.

2024

Yi Liu, Alexander W Levis, Sharon-Lise Normand, and Larry Han. 2024. “Multi-Source Conformal Inference Under Distribution Shift.”. Proceedings of Machine Learning Research, 235, Pp. 31344-82

Presentations

2024 American Causal Inference Conference, May 2024 in Seattle, WA. Larry Han oral presentation, “Data fusion for enhanced generalizability/transportability, efficiency, and heterogeneous treatment effect estimation”; Sharon-Lise Normand presented a plenary talk, “Causal Inference in the Trenches: Challenges and Opportunities.” https://sci-info.org/wp-content/uploads/2024/05/event_202312_agenda_pdf_aoggo.pdf

2nd Global Symposium of Research Methodology Innovation in Trauma and Emergency Care, May 2024 in Columbus, OH. Presented keynote address, “Modern Causal Inference Approaches for Health Services.” https://nationwidechildrens.cloud-cme.com/course/courseoverview?P=0&EID=30692

2024 Applied Statistics Symposium, June 2024 in Nashville, TN. Yi Liu presented, “Multi-source conformal inference under distribution shift.” https://symposium2024.icsa.org/detailed-agenda/

2024 International Conference on Machine Learning, July 2024 in Vienna, Austria. Larry Han poster presentation, “Multi-Source Conformal Inference Under Distribution Shift” https://icml.cc/virtual/2024/poster/32978

June 2024, Sharon-Lise Normand presented a Biostatistics Seminar at NYU Grossman School of Medicine, “Modern Causal Inference Approaches for Assessing Healthcare Quality.”

2024 Joint Statistical Meeting, August 2024 in Portland, OR. Larry Han presented, “Multi-Source Conformal Inference Under Distribution Shift.” https://ww3.aievolution.com/JSMAnnual2024/Events/viewEv?ev=3829

Sep 2024, Larry Han presented a Biostatistics Seminar at NYU Grossman School of Medicine, "Multi-Source Conformal Inference Under Distribution Shift."

2025 STS Annual Meeting, January 2025 in Los Angeles, CA. Meena Nathan presented, “Richard E. Clark Memorial Paper for Congenital Heart Surgery: Understanding Mortality Following Congenital Heart Surgery: What Do Procedure-Specific Factors Add?” Sharon-Lise Normand presented, “Improving Risk Adjustment in the Assessment of Congenital Heart Center Surgical Quality.” https://sts2025.eventscribe.net/searchGlobal.asp

April 2025, Sharon-Lise Normand presented, "Causal Inference Approaches to Assessing Healthcare Providers," at the European Causal Inference Conference in Ghent, Belgium.

May 2025, Sharon-Lise Normand presenting, "Modern Causal Inference Approaches to Profiling Healthcare Providers" at the at the 2025 Lifetime Data Science (LiDS) Conference in Brooklyn, NY.