The objective of this project is to write a detailed functional specification to perform inverse probability of treatment weighting (IPTW) estimation in the Sentinel Distributed Database.
This enhancement will enable IPTW for two treatment effect estimators: average treatment effect on the treated (ATT) and average treatment effect (ATE).
This specification will include options for IPTW development, such as stabilized weights vs. non-stabilized weights, trimming the non-overlapping area of treated and controls, weights for comparing two treatments or multiple treatments, and weights for summary level vs. subject level data. This specification will also include options for IPTW diagnosis checking, including summary statistics and graphical comparisons for distribution of weights, distribution of propensity scores, and balance of baseline covariates in unweighted and weighted samples.
Rongmei Zhang, PhD; Office of Biostatistics, Center for Drug Evaluation and Research, FDA, Silver Spring, MD
Rima Izem, PhD; Office of Biostatistics, Center for Drug Evaluation and Research, FDA, Silver Spring, MD