Causal Identification Conditions for the Effect of Treatment in the Treated: Illustration Using the Northwest Germany Stroke Registry
While ATE estimates suggested harm from tPA, ATT estimates suggest a protective effect. Simulations show that when one-sided positivity violations exist, epidemiologists can leverage weaker identification conditions to consistently estimate the ATT, even when estimates of the ATE are biased.
Authors
- Alexander Breskin (ORCID: https://orcid.org/0000-0001-5568-2510)
- Tobias Kurth (ORCID: https://orcid.org/0000-0001-7169-2620)
- Catherine Wiener (ORCID: https://orcid.org/0000-0001-9625-0319)
- Michele Jönsson Funk (ORCID: https://orcid.org/0000-0002-3756-7540)
- Paul N. Zivich (ORCID: https://orcid.org/0000-0002-9932-1095)
- Stephen R. Cole (ORCID: https://orcid.org/0000-0003-2117-1311)
- Klaus Berger
Institutions
- University of North Carolina at Chapel Hill (US)
- Regeneron (United States) (US)
- University of Münster (DE)
- House of Representatives (NL)
- Charité - Universitätsmedizin Berlin (DE)
Publication Details
- Journal
- Epidemiology
- Published
- 2025-10-14
- DOI
- https://doi.org/10.1097/ede.0000000000001924
- Primary Topic
- Advanced Causal Inference Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00