Biostatistics Dept Seminar: Addressing Unmeasured Confounding in Observational Studies: Insights from Negative Control Methods and Proximal Causal Inference

Biostatistics Department Seminar 

Title: Addressing Unmeasured Confounding in Observational Studies: Insights from Negative Control Methods and Proximal Causal Inference 

Abstract: Unmeasured confounding is a major source of skepticism about the validity of findings in observational studies. To address this issue, negative control methods and proximal causal inference have gained increasing popularity. These methods detect and account for unmeasured confounding bias by leveraging variables that do not have a direct causal relationship with either the exposure or the outcome of interest. They are particularly advantageous in studies based on electronic health records (EHRs), where unmeasured confounding is common while candidate negative controls are abundant. In this talk, I will first provide a brief introduction to the concepts of negative controls and proximal causal inference, along with an overview of the current literature. I will then illustrate how these methods can be applied to address selection and confounding bias in test-negative design studies evaluating vaccine effectiveness using EHR data. I will conclude with a discussion on future research directions.

Speaker

Xu Shi is an associate professor in Biostatistics at the University of Michigan, School of Public Health.

Zoom Registration

If you would like to join via Zoom, please register here.

2024-2025 Monday Seminar Series

All seminars are held at 12:05 PM via Zoom and onsite. View all seminar information here.