Real-world evidence, read before the reviewer does
Is your real-world evidence protocol ready for the review division?
Paste the outline. In one pass, RWE Readiness Checker tells you which of 29 recorded expectations the text actually evidences, which it contradicts, and which it never mentions. Every finding carries the rule behind it and the timestamp in the source.
Read from one recorded source, not invented. Every check carries its timestamp in the episode, so you can hear the sentence behind the finding. The legal basis is read from 29 held controls across 7 frameworks.
Free, no sign-up. The paste stays in your browser. The paid pack adds the repair wording and the review document.
Paste the outline
Text stays in this browser"Statistical analysis plan: to be finalised after database lock." A specimen finding, source 30:36.
No account, no upload, no model call. The text is read by a script in this page and is not sent to us.
The reading is deterministic. Every finding names the rule, quotes the line it read and says what would resolve it. Two people pasting the same outline get the same findings in the same order.
Saved checks in this browser
No checks saved in this browser yet. Saving keeps the outline and the reading in this browser only.
Five families, 29 checks, one recorded source
The rule set is read from the FDA Watch episode on real-world evidence in accelerated approvals, from the point where the panel turns to what the statute requires before approval to the point where it closes on data rights. Nothing is added from a template.
Planned before the data
The protocol, the analysis plan, the estimand and the confirmatory study are in motion before anyone sees a result.
The six checksFit for the question
Relevance, accuracy, completeness and traceability, with source verification and the submission standard named.
The seven checksDoes not hinge on one choice
Sensitivity and subgroup analyses fixed in advance, and a stated view of what would change the conclusion.
The four checksA fair comparison
Eligibility, endpoints, confounding, contemporaneous care and the care standard at the sites, matched to the trial.
The six checksThe right to use it, and to prove it
The submission use, the data use agreement, the source records, privacy and the chain through every hand.
The six checksMet, partly evidenced, missing, contradicted
Four states, one meaning each. A line that denies or defers an element is a finding, never a pass.
How the reading worksA reviewer reads the plan, then asks for the record
The episode describes a bar that rises with the role the evidence plays. Real-world data offered as the main comparison for an effectiveness claim is read harder than data used for context, and an external control is where the bar is highest. The panel is specific about what that means in practice: a protocol and analysis plan shared before the study, data that is relevant, accurate, complete and traceable, sensitivity and subgroup analyses fixed in advance, and the right to use the data in the submission.
Those are checkable statements. This product turns them into 29 checks, reads the outline you paste, and quotes the line it read. It is not a template and it does not write your protocol. It tells you what the text does not yet say.
What it will not do. It will not tell you that a study will be accepted. It does not read the datasets, the protocol appendices or the agreements. It reads the outline in front of it, and it says when a line is absent rather than inferring it from a heading.
What it is good at. Finding the gap that is cheapest to fix now: a plan deferred to after database lock, an external control matched only on age and sex, a data use agreement still in negotiation, a source record nobody has the right to inspect.
The specimen, already run
A draft outline for a single-arm cell therapy study with an external control. It reads like a real draft: the headings are there, the commitments are not. The check finds the contradictions first.
| Reading | Count | What it means |
|---|---|---|
| Met | 3 | The outline carries every element the check needs. |
| Partly evidenced | 15 | The subject is named, at least one element is missing. |
| Not in the outline | 5 | Nothing in the paste speaks to the check. |
| Contradicted | 6 | A line denies or defers the element. |
Readiness reading: 36/100. Thin. The outline does not yet evidence what a review division will look for, and several elements are missing or contradicted. See the specimen and its first findings.
Free to check. Pay for the repair wording
The check
USD 0
no account
- All 29 checks, all findings, in the browser
- The rule and the timestamp behind each finding
- The line it read, quoted with its section
- CSV download of the findings
Protocol repair pack
USD 199
one payment, yours to keep
- The 29 checks as a review document, PDF and XLSX
- A paste-ready repair clause for every check
- The evidence and owner columns a reviewer asks for
- The register CSV and the outline template
- Regenerated when the rule set changes
Team desk
USD 39
a month, cancellable
- Everything in the pack
- Saved checks in the browser, with a backup file
- Re-check a revised outline and compare readings
- The watch: an email when a rule or its source changes
- Annual invoicing on request, two months free
One click to checkout: card, Apple Pay or Google Pay, and your account is created from the email you give there, no sign-up form and no link to wait for.
The questions buyers ask first
What does RWE Readiness Checker actually read?
It reads the protocol outline or synopsis you paste: the headings and the sentences under them. It does not read the full protocol, the datasets or the agreements, and it says so when it cannot find an element.
Is this an FDA decision or a legal opinion?
No. It is a reading of your text against checks read from one published episode. A finding is not a decision by any agency, and a met check is not a finding that the evidence is adequate.
Does my protocol text leave my browser?
No. The free check runs as a script on the page. The text is not uploaded, stored on our servers or sent to a model. Saved checks stay in this browser unless you download the backup.
What is the difference between partly evidenced and contradicted?
Partly evidenced means the subject is named and at least one element is missing. Contradicted means a line denies or defers the element, for example a statistical analysis plan that is to be finalised after database lock. A contradiction is the stronger finding.
Where do the rules come from?
From the FDA Watch episode on real-world evidence in accelerated approvals, from 19:57 to 39:36, with the timestamp on every rule. The legal basis is read from the held frameworks named on the basis page.
Do I need an account?
Not for the free check. An account is created from the email at checkout when you buy the repair pack or the team desk.