Screen papers and extract data for a systematic review
An agent screens abstracts against inclusion criteria and pulls study details into a table, with reviewers checking every call.
The use case
Systematic reviews require screening thousands of abstracts and extracting the same fields from each included paper, which takes research teams months. Elicit searches academic literature, screens papers against criteria you write, and extracts details such as sample size, methods and outcomes into a table with quotes from the paper. Reviewers can then work through the table instead of every PDF. Review standards still require people: at least one human checks each inclusion decision and each extracted value against the source quote before analysis.
Example workflow
- Write inclusion and exclusion criteria
- Search and screen papers
- Check each screening decision
- Extract fields with supporting quotes
- Verify extracted values before analysis
Tools
- Elicit
Risks and guardrails
- Missed papers bias the review; keep a second search method
- Extracted numbers can be wrong; check against quotes
- Report AI use in methods per journal rules
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- Elicit: elicit.com
- Elicit (official page): github.com/elicit/api-examples/blob/main/integrations/mcp/README.md (checked 2026-11-19)