Full Text Screening in Elicit Systematic Reviews

For a rigorous, PRISMA-compliant literature review, abstract screening is only the first step. Researchers often encounter a significant pain point when a paper's full text reveals critical details—such as specific population definitions, study designs, or outcome measures—that disqualify a paper previously included based on its abstract alone.

Elicit addresses this challenge by introducing Full-Text Screening as a distinct, automated step in the Systematic Review workflow, allowing you to catch these mismatches early before committing to data extraction. This feature enables more rigorous and closer to PRISMA-compliant Systematic Literature Reviews (SLRs) within Elicit.

Full-text screening is an intermediary step that occurs between abstract screening and data extraction in your Systematic Review.

1. Start a Systematic Review: Begin a new review on the Elicit homepage, starting with the Gather Papers stage (Search/Upload).

2. Complete Abstract Screening: Follow the guided flow to complete your initial abstract screening.

3. Define Full-Text Screening: Once abstract screening is complete, click "Define full-text screening" at the top of the interface to move to this optional next stage.

4. Define Criteria: You can define your full eligibility criteria, using completely different criteria or more specific criteria than you used for abstract screening. Elicit will apply these criteria across the full text of the included papers.

5. Review Results: Elicit will use the full text of the papers to generate screening results and reasoning in the "Evaluate full-text screening" stage.

6. Proceed to Extraction: Once the full-text screening is complete, click "Define extraction" to begin the final data extraction process.

The Full-Text Screening step ensures a more accurate and efficient review by allowing you to apply criteria that genuinely require the complete paper.

This step is essential for criteria such as:

• Specific population definitions (e.g., distinguishing between adults and adolescents).

• Detailed intervention descriptions.

• Specific outcome measures.

• Fine-grained study design elements that authors may not include in the abstract.

By screening on these complex criteria before committing to extraction, you can catch critical mismatches early instead of discovering them late in your review. This capability complements our recent work on full-text analysis, including automatic retrieval and figure extraction.

The screening results overview will transparently show a breakdown of papers that were included but were ineligible because full texts were unavailable. You can, of course, decide to include these papers based only on their abstracts if your review methodology allows.

The most time-consuming part of a systematic review is often tracking down thousands of full-text PDFs across various publisher sites. Elicit's system works to automatically retrieve papers so you can focus on the research.

• Automatic Retrieval: Elicit attempts to find all the full texts for papers screened in from the abstract stage.

• Using the Elicit Browser Extension: Our Chrome extension uses your existing institutional access/logins to automatically pull full texts directly into your review, retrieving PDFs from journals directly.

  • Even without institutional access, the extension retrieves roughly 30% more papers than Elicit alone.

• Manual Upload: For any remaining papers, you can still upload the PDFs manually during Full Text Screening to enrich the abstract-only results.

Full-Text Screening is available for users on the Pro, Teams, and Enterprise plans.