01.Association15.Sensitivity analysis - sporedata/researchdesigneR GitHub Wiki
- In situations where researchers believe that not all the key confounders have been measured and that therefore they need to conduct a sensitivity analysis to evaluate the impact of potentially unmeasured confounders on the final conclusion of the study.
Sensitivity analysis is useful in assessing how robust an association is to potential unmeasured or uncontrolled confounding.
The E-value is the minimum strength of association, on the risk ratio scale, that an unmeasured confounder would need to have with both the treatment and outcome, conditional on the measured covariates, to fully explain away a specific treatment–outcome association [1].
The E-value allows an investigator to make statements of the following form: “The observed risk ratio of 3.9 could be explained away by an unmeasured confounder that was associated with both the treatment and the outcome by a risk ratio of 7.2-fold each, above and beyond the measured confounders, but weaker confounding could not do so; the confidence interval could be moved to include the null by an unmeasured confounder that was associated with both the treatment and the outcome by a risk ratio of 3.0-fold each, above and beyond the measured confounders, but weaker confounding could not do so.” [1]
[1] VanderWeele, T. J., & Ding, P. Sensitivity analysis in observational research: introducing the E-value. Annals of internal medicine. 2017 Aug 15;167(4):268-274.