A systematic analysis of meta-analyses: Development of the Meta-Analysis Diversity Index (MADI)

Summary: Abstract Body: There are few objective measures of efficacy of a meta-analysis; thus, a systematic review study was conducted to quantify the effect of varying diversity of included studies in meta-analyses published in dermatologic journals. Until December 2024, a comprehensive literature review was conducted, and 123 studies were revised. The 51 included meta-analyses synthesized 943 studies and 28,049,093 participants. Meta-analyses with one predominant study had pooled event rates within the same confidence intervals of that study. A linear regression model was used comparing predominant study event rate to the overall pooled event rate, yielding a positive association (R2 = 0.9793, p= 1.28554E-10). This led to the development of the MADI, which can be calculated using the equation . A MADI score greater than 1.3 predicts greater diversity within the included meta-analyses, indicating novel study. Meta-analyses with less diverse included studies, or one predominant study do not provide new information. Rather, authors should utilize the MADI score and instead perform a replication study with a larger cohort to ensure reproducibility and variability of the predominant study. Pritika Parmar¹, Parker Juels¹, Robert Dellavalle³, Hensin Tsao² 1. University of Colorado Anschutz Medical Campus School of Medicine, Aurora, CO, United States. 2. Dermatology, Harvard Medical School, Boston, MA, United States. 3. Dermatology, University of Minnesota Medical School, Minneapolis, MN, United States. Clinical Research: Epidemiology and Observational Research