Full Text Screening
Screen the full texts of papers that passed abstract screening, verify each decision against the PDF, and confirm the stage to move to data extraction.
What is Full Text Screening?
The second and final screening pass. Paperguide AI reads the full PDF of every paper that survived abstract screening, evaluates it against your full text criteria, and links each decision to the exact statements it relied on. A person then verifies that work before the review moves on.
Before you start: Your full text screening criteria must be defined in the Protocol, and Title & Abstract Screening must be confirmed. Assign your verifier before starting, not after.
How this stage works
- A Full Access member starts AI screening. AI reads the full text of every paper with a PDF and evaluates it against your full text criteria.
- Papers without a PDF are marked ineligible. They are held out of the run until a PDF is attached.
- A verifier reviews the decisions. They check each one against the cited statements in the PDF and override anything that is wrong.
- The verifier confirms the papers. Only then does data extraction unlock.
How this differs from abstract screening
The workflow is the same. What changes is what AI reads and how firm the decision has to be.
| Title & Abstract Screening | Full Text Screening | |
|---|---|---|
| What AI reads | Title and abstract only | The full PDF |
| Decision | Include, Exclude, or Maybe | Include or Exclude |
| Evidence | Sentences from the abstract | Statements from the full text, with page or section location |
| PDF required | No | Yes |
Abstract screening is deliberately sensitive: it keeps anything it cannot rule out. Full text screening is confirmatory. The full paper should contain everything needed to reach a final answer, so there is no Maybe at this stage. Every paper ends up either included or excluded.
This is also where criteria that depend on detail get settled: methodology, participant characteristics, intervention specifics, and outcome measurement are often reported only in the body of a paper, never in its abstract.
What gets screened
Every paper marked Included or Maybe in abstract screening carries forward. Excluded papers stop at the previous stage.
Each paper needs a full text PDF. Papers without one are marked ineligible and held out of screening until a PDF is attached. They are not excluded on their merits, only set aside because there was nothing to read.
Before you start
Set up your full text criteria
Full text criteria are separate from your abstract criteria and usually stricter, because the full paper supports judgments an abstract cannot.
Criterion order determines exclusion reasons. When a paper fails, the first criterion it fails, in the order you listed them, becomes its exclusion reason.
Add missing PDFs
Papers usually lack a PDF for one reason: the full text sits behind a paywall or requires institutional authentication.
Two ways to attach one:
- Use the Paperguide browser extension to fetch the PDF while you browse, including paywalled papers through your organization's proxy
- Upload the file manually from your device
You do not have to resolve every missing PDF before starting. Papers can be screened later as their PDFs arrive. But every gap you leave is a second pass you will have to make.
Assign a verifier
Assign the person responsible for verifying this stage before screening starts. Assignment determines who can act on the results and who gets notified when the run finishes.
| If the stage is | Who verifies it |
|---|---|
| Assigned to a member | That member. They receive the notification when screening finishes. |
| Not assigned to anyone | Any Full Access or Review member. All of them are notified and can work through the results together. |
View members can never verify, assigned or not.
Running AI screening
Only Full Access members can start AI screening. Review members verify results but cannot trigger the run.
When the run starts, Paperguide indexes the attached PDFs and screens each paper as its indexing completes. Full text screening therefore takes longer than abstract screening: a full paper carries far more text than an abstract, and it has to be processed before it can be read.
Results appear in the screening workspace as the run progresses, so you can begin reviewing before it finishes. Paperguide also sends an email on completion: to the assigned verifier, or to every Full Access and Review member if the stage is unassigned.
How AI reaches a decision
For each criterion, AI locates the relevant material in the full text and judges whether the criterion is satisfied. Each one returns:
- A verdict, yes or no
- The cited evidence, the specific statements in the paper supporting that verdict
- Location, the page or section the evidence came from
The overall decision follows directly from the criterion verdicts:
| If | Decision |
|---|---|
| Every criterion is satisfied | Include |
| Any criterion fails | Exclude, with that criterion as the exclusion reason |
Read the criterion verdicts before changing an overall decision. When you disagree with a result, the cause is almost always one criterion judged differently than you would judge it.
Vague criteria produce vague evidence. AI can only find what your criterion instructions tell it to look for, so "appropriate methodology" will return weaker results than a criterion naming the study designs you accept.
Working through the results
The screening workspace puts the paper list on one side and the decision detail on the other. Select a paper to see every criterion verdict, its cited evidence, and where in the paper that evidence came from.
The PDF viewer
A built-in PDF viewer opens beside the decision panel, so the paper and the reasoning about it stay on the same screen.
Select a citation and the viewer highlights the supporting statements in the full text: the exact sentences AI used to reach that verdict, in place, with the surrounding paragraph around them. You can also move to any page in the PDF yourself while reviewing.
This is what makes verification real. A verdict on its own asks you to trust a summary. A verdict with its supporting statements highlighted in the source asks you to check it, and checking takes seconds rather than a hunt through the paper.
Managing the queue
Papers are grouped into decision tabs, and you can filter by decision or verification status and sort your queue to work through a large set without losing your place.
Verifying and overriding
The AI decision is a recommendation. A confirmed decision is one a person has reviewed and accepted.
To verify a paper, work through the criteria, paying most attention to any marked no, since a single failed criterion determines the outcome. Select the citations to highlight the supporting statements and read them in context. If you agree, confirm. If you disagree, override to Include or Exclude, selecting the failed criterion when you exclude.
To confirm, either confirm papers individually as you go, or select Confirm All once you have worked through the set. Confirming records the AI decision as approved, or your override as the final decision. Both are kept, so the review holds a record of what AI decided and what a person changed.
Full Access members can review and override decisions after verification is complete, including decisions made by another member.
Screening papers whose PDFs arrive later
You can add a PDF to an ineligible paper at any point, including after AI screening has run and after verification is complete.
Once a PDF is attached, Start screening becomes available for the newly ready papers. Selecting it screens only those papers, leaving completed work untouched. A verifier then reviews and confirms them the same way as the rest.
This means a missing PDF never has to block your review permanently. If you can obtain the paper later, it can still be assessed.
<Callout kind="info" collapsed="false"> Work through your ineligible papers before treating the stage as finished. A paper you could have retrieved is a paper your review should have assessed, and reviewers will ask. </Callout>
Confirming the stage
Once every screened paper has been verified, confirm the stage.
You then see the final counts: Included, Excluded, and Ineligible because no PDF was available.
Included papers move forward to data extraction. Excluded papers stop here and stay recorded in the review flow. Ineligible papers remain held until a PDF is added.
Confirming unlocks Data Extraction.
How this feeds PRISMA
This stage populates the eligibility section of your PRISMA diagram:
- Papers with a PDF are counted as assessed for eligibility
- Papers excluded here are recorded as full text exclusions, with the criterion each one failed
- Ineligible papers are counted as not retrieved
Missing PDFs are therefore not only a workflow problem but part of your review's reported accounting. Resolve them early if you want your totals to settle before you write up.
If your protocol or papers change
If you change your full text criteria in the protocol, screening may need to run again so decisions stay consistent with the criteria you are now using. How much needs rescreening depends on what changed.
If you add papers to the collection, they pass through abstract screening first, then arrive here. Only the new papers are screened.
Exporting screening data
You can export your full text screening results, including decisions, exclusion reasons, and the cited evidence behind them. Useful for sharing progress with collaborators outside Paperguide and for keeping a screening record alongside your submission.