Regression-based proximal causal inference
Open Access
- 25 September 2024
- journal article
- research article
- Published by Oxford University Press (OUP) in American Journal of Epidemiology
- Vol. 194 (7) , 2030-2036
Abstract
Negative controls are increasingly used to evaluate the presence of potential unmeasured confounding in observational studies. Beyond the use of negative controls to detect the presence of residual confounding, proximal causal inference (PCI) was recently proposed to de-bias confounded causal effect estimates by leveraging a pair of treatment and outcome negative control or confounding proxy variables. Formal methods for statistical inference have been developed for PCI; however, these methods can be challenging to implement because they involve solving complex integral equations that are typically ill-posed. We develop a regression-based PCI approach, employing 2-stage generalized linear regression models (GLMs) to implement PCI, which obviates the need to solve difficult integral equations. The proposed approach has merit in that (1) it is applicable to continuous, count, and binary outcomes cases, making it relevant to a wide range of real-world applications; and (2) it is easy to implement using off-the-shelf software for GLMs. We establish the statistical properties of regression-based PCI and illustrate their performance in both synthetic and real-world empirical applications. This article is part of a Special Collection on Methods in Social Epidemiology.Keywords
All Related Versions
Funding Information
- National Institutes of Health (R01AG065276, R01AI127271)
This publication has 21 references indexed in Scilit:
- Using negative controls to adjust for unmeasured confounding bias in time series studiesNature Reviews Methods Primers, 2023
- Double Negative Control Inference in Test-Negative Design Studies of Vaccine EffectivenessJournal of the American Statistical Association, 2023
- Semiparametric Proximal Causal InferenceJournal of the American Statistical Association, 2023
- INTRODUCING PROXIMAL CAUSAL INFERENCE FOR EPIDEMIOLOGISTSAmerican Journal of Epidemiology, 2023
- Proximal causal inference without uniqueness assumptionsStatistics & Probability Letters, 2023
- Current Challenges With the Use of Test-Negative Designs for Modeling COVID-19 Vaccination and OutcomesAmerican Journal of Epidemiology, 2022
- A Selective Review of Negative Control Methods in EpidemiologyCurrent Epidemiology Reports, 2020
- Estimating the causal effect of zidovudine on CD4 count with a marginal structural model for repeated measuresStatistics in Medicine, 2002
- Causal Inference from Complex Longitudinal Data1997
- A Controlled Trial to Improve Care for Seriously III Hospitalized PatientsJAMA, 1995