BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260224T071415EST-39790kHJj0@132.216.98.100
DTSTAMP:20260224T121415Z
DESCRIPTION:Alexander Levis PhD\n\nAssistant Professor of Biostatistics| Un
 iversity of Pennsylvania\n\nWHEN: Wednesday\, February 25\, 2026\, from 3:
 30 to 4:30 p.m.\n	WHERE: Hybrid | 2001 91şÚÁĎÍř College Avenue\, Rm 1140\; Zo
 om\n	NOTE: Alexander Levis will be presenting in-person at SPGH \n\nAbstrac
 t\n\nBiomedical and public health researchers increasingly turn to large o
 bservational data\, such as those deriving from electronic health records 
 (EHR)\, to quantify the effects of medical treatments or other exposures. 
 Target trial emulation has emerged as a promising framework for conducting
  causal inference from these data sources. The approach aims to mitigate b
 ias in part by clarifying a target causal estimand and\, critically\, a co
 mmon notion of time zero. Notwithstanding the virtues of this conceptual f
 ramework\, current practice is often limited by reliance on strong paramet
 ric models\, implicit and often implausible structural assumptions\, and—p
 erhaps due to a lack of a universally accepted mathematical formalism—an i
 nability to formally handle commonly faced challenges (e.g.\, selection du
 e to missing data).In this talk\, I will present two recent projects that 
 strive to make progress towards remedying these issues\, focusing on estim
 ation of causal effects in the face of (1) missing eligibility status\, an
 d (2) possibly calendar time-varying effects. Our development begins by ma
 king explicit the structure of the observational (e.g.\, EHR-based) data u
 sed with target trial emulation\, the target causal estimands\, and assump
 tions under which the latter are identified from the observed data distrib
 ution.The ensuing proposed estimators incorporate machine learning\, and c
 an achieve asymptotic normality and semiparametric efficiency under relati
 vely weak assumptions. Finally\, we demonstrate practical use of our metho
 ds by analyzing an EHR-based study of long-term weight and glycemic outcom
 es\, comparing bariatric surgical intervention to no surgery among patient
 s with severe obesity.\n\nSpeaker Bio\n\nAlex Levis is an Assistant Profes
 sor of Biostatistics in the Center for Causal Inference and the Department
  of Biostatistics\, Epidemiology and Informatics at the University of Penn
 sylvania. Before joining Penn\, he obtained his PhD in Biostatistics at Ha
 rvard University in 2022\, and subsequently was a postdoctoral researcher 
 in the Department of Statistics & Data Science at Carnegie Mellon Universi
 ty. Alex's expertise lies in statistical methods development related to ca
 usal inference\, missing data\, non- and semi-parametrics\, machine learni
 ng\, and the use of administrative and electronic health record data for c
 omparative effectiveness research. He is interested in a broad range of ap
 plication areas spanning the biomedical sciences\, focusing recently on pr
 oblems in surgery\, experimental neuroscience\, and mental health. https:/
 /www.awlevis.com/ \n
DTSTART:20260225T203000Z
DTEND:20260225T213000Z
SUMMARY:Challenges and Flexible Solutions for Target Trial Emulation in Ele
 ctronic Health Record Data
URL:/spgh/channels/event/challenges-and-flexible-solut
 ions-target-trial-emulation-electronic-health-record-data-371376
END:VEVENT
END:VCALENDAR