Seminar by Lifeng Lin

Statistics and Biostatistics seminar series

Lifeng Lin
University of Arizona

Room: M3 3127

Monitoring Living Evidence for Research Synthesis

A living systematic review (LSR) is a continuously updated approach to synthesizing evidence in real time. Trial sequential analysis (TSA) plays a crucial role in LSRs and cumulative meta-analyses by assessing the conclusiveness of evidence. It utilizes sequential monitoring boundaries to evaluate intervention effectiveness and futility boundaries to determine whether an intervention differs significantly from or has a similar effect to the control. Despite its growing popularity, TSA’s reproducibility is often compromised by insufficiently detailed reporting of underlying assumptions.

Additionally, existing TSA methods rely heavily on interim analyses from randomized controlled trials, which typically involve more homogeneous participant groups than those in meta-analyses. In this presentation, we introduce novel methods designed to prevent the premature termination of LSRs, thereby enhancing the robustness of evidence synthesis. Our numerical studies demonstrate that these new methods offer greater reliability than existing approaches.