|
Abstract:
|
Evidence-based clinical decision making involves synthesizing available evidence from multiple resources, including randomized controlled trials (RCT) and real-world data (RWD). RCTs are the gold standard for assessing treatment efficacy within a well-defined cohort. However, given the specific eligibility criteria for trial enrollment, coupled with the rapidly changing nature of multidisciplinary cancer care, it is uncertain if the efficacy observed in RCTs can be generalized to broader clinical practice. While these parallel data sources are all informative, properly integrating RWD with RCT will provide insights to understand the observed discrepancy of treatment efficacy and to identify subgroups of patients with different treatment efficacies. We will re-analyze a randomized controlled trial: NSABP trial was conducted during 1970-1980 to compare the local therapies for treatment of early stage breast cancer. Recently a growing body of literature based on RWD reported somewhat different findings from what obtained from the NSABP trial. We have performed comprehensive analyses including resampling and simulation studies to better understand the reasons behind the discrepancy.
|
