|
Abstract:
|
In observational studies evaluating the effect of treatment on adverse events among geriatric or severely ill patients, death among subjects complicates the analysis. This problem is often referred to as outcome “truncation” by death. Traditionally, researchers estimate the survivor average causal effect (SACE) using principal stratification to assess the effect of treatment among survivors. This approach does not account for the effect that either treatment regimen had on the occurrence of death, or the effect of either treatment among those who died. We propose an alternate, Bayesian method which imputes counterfactual mortality and adverse event outcomes for each participant under the alternate treatment. We also define a composite ordinal outcome for each patient, combining the occurrence of death and the adverse event of interest in an increasing scale of severity. This allows for direct treatment comparison on the whole sample after imputation, while evaluating the occurrence of death and the event outcome concurrently. We implement our procedure to analyze the incidence of heart failure among geriatric patients being treated for Type II diabetes with different drugs.
|
