Truace
Subscribe

Unmasking bias in the evidence ecosystem: a panoramic analysis of 311,751 meta-analyses using an artificial intelligence agent-based approach

Traditional secondary meta-analysis workflows are highly labour-intensive, time-consuming, and difficult to update in real time. Currently, there is a lack of comprehensive artificial intelligence frameworks capable of automating the entire meta-analysis workflow, including literature screening, data extraction, and quality assessment. Furthermore, a large-scale structured database for systematically analysing the global landscape of published meta-analyses remains unavailable. In this viewpoint, we aimed to eva…

By the Truvace desk1 min read1 sourcePeer-reviewedTRV-2026-0983· cite & verify
Unmasking bias in the evidence ecosystem: a panoramic analysis of 311,751 meta-analyses using an artificial intelligence agent-based approach
N-gram analysis results – top 20 recurrent 4-g for the 137 descriptions of "privacy" in AI governance guidelines by Authors of the study: Nicholas Kluge Corrêa Camila Galvão James William Santos Carolina Del Pino Edson Pontes Pinto Camila Barbosa Diogo Massmann Rodrigo Mambrini Luiza Galvão Edmund Terem Nythamar de Oliveira. CC BY 4.0 · https://creativecommons.org/licenses/by/4.0

In brief

Traditional secondary meta-analysis workflows are highly labour-intensive, time-consuming, and difficult to update in real time. Currently, there is a lack of comprehensive artificial intelligence frameworks capable of automating the entire meta-analysis workflow, including literature screening, data extraction, and quality assessment.

In this viewpoint, we aimed to evaluate the feasibility of large language models in automating meta-analysis workflows and develop the Meta-Analysis Screening, Transformation and Evaluation Review Agent (MASTER) agent; establish a large-scale Unified Meta-Analysis Repository (UMAR) and perform an exploratory panoramic analysis of the current evidence ecosystem; and develop an Agent-based Secondary Meta-analysis Platform (ASAP), integrating these capabilities. Here, we provide an initial exploration of the technical feasibility and scalability of artificial intelligence-driven automated meta-analysis.

Main points

  1. Traditional secondary meta-analysis workflows are highly labour-intensive, time-consuming, and difficult to update in real time.
  2. Currently, there is a lack of comprehensive artificial intelligence frameworks capable of automating the entire meta-analysis workflow, including literature screening, data extraction, and quality assessment.
  3. Furthermore, a large-scale structured database for systematically analysing the global landscape of published meta-analyses remains unavailable.

The problem

Currently, there is a lack of comprehensive artificial intelligence frameworks capable of automating the entire meta-analysis workflow, including literature screening, data extraction, and quality assessment.

The rundown

Furthermore, a large-scale structured database for systematically analysing the global landscape of published meta-analyses remains unavailable. In this viewpoint, we aimed to evaluate the feasibility of large language models in automating meta-analysis workflows and develop the Meta-Analysis Screening, Transformation and Evaluation Review Agent (MASTER) agent; establish a large-scale Unified Meta-Analysis Repository (UMAR) and perform an exploratory panoramic analysis of the current evidence ecosystem; and develop an Agent-based Secondary Meta-analysis Platform (ASAP), integrating these capabilities.

Sources

  1. Peer-reviewedJournal of Global Health2026-09-04

The debate