Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 461 - Design and Analytic Approaches to Address Unmeasured Confounding
Type: Contributed
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #313629
Title: Operationalizing a Definition of Statistical Convergence in the Context of Meta-Analysis
Author(s): Hayley Belli* and Ellen Caniglia and Scott Braithwaite and Andrea Troxel
Companies: New York University School of Medicine and New York University School of Medicine and New York University School of Medicine and New York University School of Medicine
Keywords: meta-analysis; convergence; reproducible research

Meta-analysis is a common statistical approach for synthesizing data across studies to evaluate the validity of a hypothesis. However, the question of how many studies to include or when researchers may need to add additional studies to a meta-analysis is a common problem. In this research, we propose statistical criteria for evaluating whether the result from a meta-analysis converges. We first surveyed the literature for randomized controlled trials that assessed as their primary objective reduction in low-density lipoprotein (LDL) cholesterol using either low-intensity statins, high-intensity statins, or ezetimibe. For each of these three groups, we extracted independent study effect sizes and applied the two-step convergence criteria assuming the chronological order in which the studies appeared in the literature. We also performed a sensitivity analysis by generating random permutations of the order of the studies and applying the same convergence criteria across these different scenarios. In the era of reproducible research, this work provides first-steps towards statistically assessing convergence in the synthesis of independently conducted randomized controlled trials.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program