The method uses a hierarchical outcome tree to group clinically related diagnostic codes and applies scan statistics to adjust for multiple testing of correlated hypotheses when screening thousands of potential adverse events for signal identification activities. The tree-based scan statistic (TreeScan Martin Kulldorff, Harvard Medical School, Boston, Massachusetts) is a statistical data-mining method, compatible with multiple study designs, that can be used to identify drug safety signals from health insurance claims and electronic health records databases ( 1, 2). Potential signals should be followed by pharmacoepidemiologic assessment where confounding control is tailored to the specific outcome(s) under investigation. The choice of PS depends on the level of concern about residual confounding versus loss of power. Unlike tailored covariates, empirical and predefined general covariates can be applied “out of the box” for signal identification. Inclusion of empirically selected covariates can provide better proxy coverage for confounders but can also decrease statistical power. Inclusion of covariates tailored to exposure did not appreciably affect screening results. For each pair, statistical alerting patterns were similar with alternative PSs (≤11 alerts in 7,996 outcomes scanned). For each pair, we evaluated 5 candidate PSs with different combinations of 1) predefined general covariates (comorbidity, frailty, utilization), 2) empirically selected (data-driven) covariates, and 3) covariates tailored to the drug pair.

We selected 4 pairs of medications with well-understood safety profiles.

However, it is unclear which variables to include in a PS for applied signal identification studies to simultaneously adjust for confounding across potential outcomes. Simulation has demonstrated the promise of TreeScan with a propensity score (PS)-matched cohort design. The tree-based scan statistic (TreeScan Martin Kulldorff, Harvard Medical School, Boston, Massachusetts) is a data-mining method that adjusts for multiple testing of correlated hypotheses when screening thousands of potential adverse events for signal identification.

Sas university edition propensity score