MD3 RISK ESTIMATION BY BOOSTED DOUBLY ROBUST METHOD.

The choice of a covariate adjustment method may affect risk estimates in observational studies. Statistical arguments suggest that BPS reduces the bias in the risk estimates and that DR reduces bias and variance still further. We computed the ratio of the RRs (RRR) and 95 percent confidence limits in the treatment and comparison cohorts using all methods. [Extracted from the article]