BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260526T085331EDT-93084iwjDV@132.216.98.100
DTSTAMP:20260526T125331Z
DESCRIPTION:Xu Shi\, PhD\n\nAssistant Professor\, Department of Biostatisti
 cs\n	University of Michigan\n\nWHEN: Wednesday\, February 14\, 2024\, from 
 3:30 to 4:30 p.m.\n\nWHERE: hybrid | 2001 McGill College Avenue\, room 114
 0\; Zoom\n\nNOTE: Dr. Shi will be presenting from Michigan\n\nAbstract\n\n
 The test-negative design (TND) has become a standard approach to evaluate 
 vaccine effectiveness. Despite TND's potential to reduce unobserved differ
 ences in healthcare-seeking behavior (HSB) between vaccinated and unvaccin
 ated subjects\, it remains subject to various potential biases. First\, re
 sidual confounding bias may remain due to unobserved HSB\, occupation as a
  healthcare worker\, or previous infection history. Second\, because selec
 tion into the TND sample is a common consequence of infection and HSB\, co
 llider stratification bias may exist when conditioning the analysis on tes
 ting\, which further induces confounding by latent HSB. Third\, generaliza
 bility of the results to the general population is not guaranteed. In this
  talk\, we present a novel approach to identify and estimate vaccine effec
 tiveness in the general population by carefully leveraging a pair of negat
 ive control exposure and outcome variables to account for potential hidden
  bias in TND studies. We illustrate our proposed method with extensive sim
 ulation and an application to COVID-19 vaccine effectiveness using data fr
 om the University of Michigan Health System.\n\nSpeaker bio\n\nXu Shi is a
 n Assistant Professor in the Department of Biostatistics at the University
  of Michigan. She is interested in developing statistical methods for elec
 tronic health records and claims data\, focusing on causal inference\, dat
 a harmonization across healthcare systems and comparative effectiveness an
 d safety research. \n
DTSTART:20240214T203000Z
DTEND:20240214T213000Z
SUMMARY:Double Negative Control Inference in Test-Negative Design Studies o
 f Vaccine Effectiveness
URL:https://www.mcgill.ca/epi-biostat-occh/channels/event/double-negative-c
 ontrol-inference-test-negative-design-studies-vaccine-effectiveness-353630
END:VEVENT
END:VCALENDAR