|
Abstract:
|
It is a common interest in medicine to determine whether a facility meets a benchmark created from an aggregate reference population, after accounting for distributional differences in multiple covariates. Due to the difficulties and expenses of collecting individual-level data, however, it is often the case that only marginal distributions of the covariates are available, making covariate-adjusted comparison of the studied facility to the benchmark difficult.
We propose and evaluate a novel approach for conducting these covariate-adjusted comparisons when only marginal covariate distributions of the studied facility are known. Our approach combines and extends methods which are traditionally unrelated both to each other and to the medical problem of interest. These methods include Iterative Proportional Fit, which estimates the cells of a contingency table when only marginal sums are known, and Synthetic Control Methods, which extract information from a dataset when not all collected data points are equally representative of the object of study.
|
