Bounds on Direct Effects in the Presence of Confounded Intermediate Variables
- 19 December 2007
- journal article
- research article
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 64 (3) , 695-701
Abstract
Summary This article considers the problem of estimating the average controlled direct effect (ACDE) of a treatment on an outcome, in the presence of unmeasured confounders between an intermediate variable and the outcome. Such confounders render the direct effect unidentifiable even in cases where the total effect is unconfounded (hence identifiable). Kaufman et al. (2005, Statistics in Medicine24, 1683–1702) applied a linear programming software to find the minimum and maximum possible values of the ACDE for specific numerical data. In this article, we apply the symbolic Balke–Pearl (1997, Journal of the American Statistical Association92, 1171–1176) linear programming method to derive closed‐form formulas for the upper and lower bounds on the ACDE under various assumptions of monotonicity. These universal bounds enable clinical experimenters to assess the direct effect of treatment from observed data with minimum computational effort, and they further shed light on the sign of the direct effect and the accuracy of the assessments.Keywords
This publication has 14 references indexed in Scilit:
- Estimation of Direct Causal EffectsEpidemiology, 2006
- Counterfactual Links to the Proportion of Treatment Effect Explained by a Surrogate MarkerBiometrics, 2005
- Improved estimation of controlled direct effects in the presence of unmeasured confounding of intermediate variablesStatistics in Medicine, 2005
- Direct and Indirect Causal Effects via Potential Outcomes*Scandinavian Journal of Statistics, 2004
- Bounds on Treatment Effects From Studies With Imperfect ComplianceJournal of the American Statistical Association, 1997
- Bounds on Treatment Effects from Studies with Imperfect ComplianceJournal of the American Statistical Association, 1997
- Identifiability and Exchangeability for Direct and Indirect EffectsEpidemiology, 1992
- Statistical validation of intermediate endpoints for chronic diseasesStatistics in Medicine, 1992
- Identifiability, Exchangeability, and Epidemiological ConfoundingInternational Journal of Epidemiology, 1986
- The Lipid Research Clinics Coronary Primary Prevention Trial results. I. Reduction in incidence of coronary heart diseaseJAMA, 1984