P31: Estimation Of Causal Effect In Integrating Randomized Clinical Trial And Observational Data
Poster Presenter
Yafei Zhang
PostDoc Fellow
Merck & Co., Inc. United States
Objectives
To bring together the breath and strength of real-world data (RWD) and randomized clinical trial (RCT) data, so that we can maximize the utility of RWD and answer broader questions.
Method
A three-step statistical framework was propose to corroborate findings from both RCT and RWD. The causal estimation is validated via the matched observational study with the target RCT by targeted maximum likelihood estimation.
Results
Simulation results suggest that the heterogeneity of patient population from RCT and RWD can leads to varying treatment estimation and the proposed approach may be able to mitigate such difference in the integrative analysis.
Conclusion
Directly combining RCT and observational study findings in a meta-analysis approach does not have a clear causal interpretation as the patient populations and study conditional of an RCT and an observational study are often not the same. The proposed approach allows us to validate the causal effects in an observational study by matching the study with RCT. The subsequent meta-analysis can only be feasible and interpretable if the observational study can produce consistent estimates of causal effects in a similar target population. We present a simulation study to demonstrate the difference in the estimates of causal effect with and without matching of RWD. And, the causal estimation of matched RWD can be closer to RCT true value than the reduced RWD. Moreover, the proposed method can identify the presence of unmeasured covariates by comparing causal estimation with RCT.
