iiinova — where your digital self keeps evolving

Alex Levis

@awlevis.bsky.social

Asst Prof of Biostats @ University of Pennsylvania Center for Causal Inference www.awlevis.com interests: causal inference, distribution shift, machine learning, non/semiparametrics, w/ applications in EHR data & beyond

Posts Replies Media

Alex Levis @awlevis.bsky.social · 10/12/2025

Such a rewarding project with @gabeloewinger.bsky.social and the team! Participant blinding is a SUPER old idea: as late as 1784!! Yet (un)blinding seems (to me) still poorly understood, e.g., compared to confounding, selection bias. I hope our work clarifies some issues, esp. in mental health RCTs

060

Reposted by Alex Levis

Alec McClean @alecmcclean.bsky.social · 13/12/2024

New-ish paper alert! arxiv.org/abs/2410.13522 We tackle the challenge of comparing multiple treatments when some subjects have zero prob. of receiving certain treatments. Eg, provider profiling: comparing hospitals (the “treatments”) for patient outcomes. Positivity violations are everywhere.

arxiv.org

Fair comparisons of causal parameters with many treatments and positivity violations

Comparing outcomes across treatments is essential in medicine and public policy. To do so, researchers typically estimate a set of parameters, possibly counterfactual, with each targeting a different ...

1285