|
Abstract:
|
Phase-three clinical trial results are typically complex, with multiple efficacy endpoints, multiple analyses for each endpoint, and multiple safety measures (i.e., adverse events). While these data typically are not analyzed after the regulatory filing, we contend that secondary trial analyses can identify populations with enhanced therapeutic effects, biomarkers for use in future studies, and safety effects for creation of risk minimization action plans. Retrospective analysis of postmarketing databases is a common approach to understanding clinical outcomes and therapy cost-benefit. Because these data are not from prospective, randomized, blinded designs, causal inference methods are required for valid inference. This presentation covers issues in the analysis of secondary clinical trial and retrospective postmarketing data by way of case studies in these settings.
|
