The Role of Measured Covariates in Assessing Sensitivity to Unmeasured Confounding
Observational Studies · 2026
DOI: 10.1353/obs.00008 ↗Kurz gefasst
Fachartikel in der Zeitschrift Observational Studies, 2026. Verfasst von Abhinandan Dalal und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Statistics and Data ScienceUniversity of PennsylvaniaPhiladelphia, PA, USA. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Sensitivity analysis is widely used to assess the robustness of causal conclusions in observational studies, yet its interaction with the structure of measured covariates is often overlooked. When latent confounders cannot be directly adjusted for and are instead controlled using proxy variables, strong associations between exposure and measured proxies can amplify sensitivity to residual confounding. We formalize this phenomenon in linear regression settings by showing that a simple ratio involving the exposure model coefficient and residual exposure variance provides an observable measure of this increased sensitivity. Applying our framework to smoking and lung cancer, we document how growing socioeconomic stratification in smoking behavior over time leads to heightened sensitivity to unmeasured confounding in more recent data. These results highlight the importance of multicollinearity when interpreting sensitivity analyses based on proxy adjustment.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Statistics and Data ScienceUniversity of PennsylvaniaPhiladelphia, PA, USA.
- Department of BiostatisticsNew York UniversityNew York, NY, USA.
Zitieren
APA
Dalal, A., Horng, I., Feng, Y., & Small, D. S. (2026). The Role of Measured Covariates in Assessing Sensitivity to Unmeasured Confounding. Observational Studies. https://doi.org/10.1353/obs.00008
Vancouver
Dalal A, Horng I, Feng Y, Small DS. The Role of Measured Covariates in Assessing Sensitivity to Unmeasured Confounding. Observational Studies. 2026. doi:10.1353/obs.00008
BibTeX
@article{Dalal2026,
author = {Dalal, Abhinandan and Horng, Iris and Feng, Yang and Small, Dylan S},
title = {The Role of Measured Covariates in Assessing Sensitivity to Unmeasured Confounding},
journal = {Observational Studies},
year = {2026},
doi = {10.1353/obs.00008}
}
EUVIMED