Informed consent authoring that traces every sentence to the protocol and writes it in plain words
Consent forms traced to the protocol and written in plain words, amendments out in a week.
See one case, screen by screen ↓Software that writes the master informed consent form and its country, site and special versions from the protocol and the Investigator's Brochure, and keeps them current through every amendment. Informed Consent Authoring adds seven agents that draft, trace every statement to its source, simplify the language and check required elements, for a medical writer to decide and sign.
How to choose AI solutions for life sciences →
Updated
Every trial needs a master informed consent form written from the protocol and the Investigator's Brochure, then a country, site and special version for each country, each approved by its own ethics committee. Medical writers do this in Word, by hand, and do it again for every amendment: a redline of the protocol, a search for "12 visits" across a dozen forms, and the hope that every instance was found.
Nobody can say quickly which sentences in which country form must change, so country study managers wait for the master and then re-read the whole form. Protocol language leaks in — "hepatic transaminases", "pharmacokinetic sampling" — and committees send the forms back. A missed required element, such as the EU trial number or when the doctor may take someone out of the study, is a committee query and weeks lost in that country.
Estimated split for one protocol amendment across the master and its country forms, by hand and with the solution.
Seven agents read the protocol and brochure, write, trace, simplify and carry every change into each country form; a medical writer decides.
VLT-302 is a Phase 3 study of Veltrimab 150 mg every 2 weeks against placebo in moderate-to-severe plaque psoriasis — 640 participants, 112 sites, 14 countries. Amendment 3 and IB edition 7 arrived on 28 September. Here is what happens next, screen by screen, in the working solution.
Lena opens the consent family: the two sources, Protocol Amendment 3 and IB edition 7, then the master consent, then 11 country, site and special forms. Amendment 3 brings liver tests at Weeks 2, 4 and 8, safety follow-up from 12 to 16 weeks, and a new optional genetic sample. The ripple turns every form amber, with its old reading grade struck through and the grade it would have if the change went in as the protocol words it.
“Protocol VLT-302 Amendment 3 arrived 28 Sep — 12 consent forms need an update”
In the Germany form, 13 sentences change and the grade would go from 10.6 to 11.3 as written. Each old sentence is struck through above its new wording and the source it rests on: “You will have 13 study visits over up to 72 weeks” from the protocol, the new liver risk from the brochure, the EU trial number from the EU rules. Beside it, every committee's clock, sorted by days left: the US central IRB on Oct 21, Germany on Oct 23, Japan on Nov 6.
“The country adapter writes these as tracked changes once the master v4.0 is signed · Priya Raman accepts them for Germany”
The protocol reader reads Amendment 3 — 112 pages, the Schedule of Activities and 3 changes — in 4.2 s. The risk extractor takes 14 reactions with their frequencies and 1 new liver risk from IB edition 7. The consent composer writes 15 sections from template v6.2, key information first, and the form fills in as it goes.
“Draft ready — 10 rewrites proposed, 1 statement differs from the protocol, 2 required elements missing”
The traceability checker links 55 statements and flags one. H4 says “12 visits at the study site over approximately 68 weeks”; the amended Schedule of Activities, §1.3 on page 14, lists 13 on-site visits with Weeks 60 and 68 in place of Week 64. The passage sits next to the sentence with the corrected wording at grade 5.9. Lena applies the correction; she could also keep the original and explain why.
“Differs from the protocol. The amended Schedule of Activities has 13 visits (Week 64 replaced by Weeks 60 and 68) and a maximum of 72 weeks.”
The coach proposes a plain rewrite for every sentence above grade 8 and keeps its citation. “Pharmacokinetic sampling will be performed to characterize serum concentrations of the investigational product” (grade 17.8) becomes “We will take a small amount of extra blood to measure how much Veltrimab is in your body” (7.3). Lena can accept, edit or keep the original for each one; here she accepts all ten.
“10 rewrites accepted — document grade 8.5 → 6.8”
The elements checker reads the master against 21 CFR 50.25(a), (b) and (c), the key information section and EU CTR Article 29, and lists the statements that meet each one. Two are missing: the EU trial number, and when the study doctor may stop a person's participation under 50.25(b)(2). For each it drafts one plain sentence with its source, and Lena adds them — 22 of 22 present.
“The study doctor may take you out of the study if staying in could harm you or if you cannot follow the study plan.”
The participant view shows the key information page as a person would read it, and how long the whole form takes to read: about 17 minutes. The coach writes a five-question comprehension quiz for the consent interview, each question tied to a statement: “How many study visits will you have?” Lena can send it to the patient panel.
Release is blocked while a statement differs from its source or a required element is missing. Now 57 of 57 statements are traced and the grade is 6.8, so Lena signs as author with the meaning “Author — I confirm this text matches its sources”. Dr. Maya Chen signs for medical accuracy and Omar Haddad for privacy and legal, and master v4.0 is released.
“Every statement traced, every required element present, reading grade within target.”
The country adapter carries master v4.0 into every country, site and special form as tracked changes and the elements checker runs each country's rules: Germany gets 13 tracked changes and 4 country rules checked, including the participant insurer and policy number. The new optional genetic research form is drafted from protocol §8.9. Every ethics package — four forms for the United States, four for Germany — reads “Ready to send”.
The country adapter never changes text the master did not change; every change is tracked.
Back on the consent family, every form reads “changes ready”: 12 of 12 updated, the master released at 6.8 and Germany at 6.9 instead of 11.3. Each committee's clock now reads “Changes written”. Priya accepts the German tracked changes with her own signature before the package goes to the ethics committee — people submit, not the agents.
The same amendment, seen by the five people who carry it — what it looked like before, and what it looks like now.
Seven agents read the protocol and brochure, write, trace, simplify and carry every change into each country form; a medical writer decides.
Reads the protocol and each amendment — synopsis, Schedule of Activities, design, procedures — and lists what changed, each fact with its section and page.
Pulls risks from the Investigator's Brochure with their frequencies against placebo and turns percentages into “about N in 100 people”.
Writes the consent form section by section from the sponsor template and the extracted facts — key information first — and keeps the source link on every statement.
Checks every statement against the passage it cites and flags any number, visit, dose, duration or frequency that differs, with the reason and a corrected sentence.
Scores every sentence with Flesch-Kincaid and Gunning Fog and proposes rewrites at or below the target grade that keep every fact of the cited source.
Checks 21 CFR 50.25, the key information rule, EU CTR Article 29 and each country's rules, lists the statements that meet each one and drafts a plain sentence for anything missing.
Carries master changes into every country, site and special form as tracked changes, keeps country-specific wording and prepares the text for certified translation and back-translation.
Accepts each change. The agents propose; a named person decides.
13 screens from the working solution, on its sample data. Pick one to see it large.
What the agents check against and who signs are settings, set once by the study team.
| Setting | Default | Choose from |
|---|---|---|
| Target reading grade for consent forms | Grade 8 | Step up or down, one grade at a time |
| Target for assent forms (ages 12–17) | Grade 5 | Step up or down, one grade at a time |
| Score with | Flesch-Kincaid | Flesch-Kincaid · Gunning Fog · Both |
| Never rewrite required wording | On | On · Off |
| Explain frequencies as “about N in 100 people” | On | On · Off |
| Medical accuracy signer | Dr. Maya Chen | Dr. Maya Chen · Omar Haddad |
| Patient panel review of the key information page and quiz | Dana Okafor | Dana Okafor · Not required |
| Sponsor master template | Master ICF template v6.2 | Master ICF template v6.2 · v6.1 |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample study · “target” = the design goal, measured in the live solution · “estimated” = our estimate · “typical” = published figures (J Clin Transl Sci 2024 review of 5,239 trial consent forms — 91% above grade 8, mean Flesch-Kincaid grade 10.99; Tufts CSDD 2024 amendment benchmark, Ther Innov Regul Sci — 260 days from the need for an amendment to the last ethics approval). Regulations named: 21 CFR 50.25, 45 CFR 46.116, EU CTR Art. 29. People, companies, studies and medicines named on this page are fictional — characters and sample data in the working solution.
Software that writes the master informed consent form and its country, site and special versions from the protocol and the Investigator's Brochure, and keeps them current through every amendment. Informed Consent Authoring adds seven agents that draft, trace every statement to its source, simplify the language and check required elements, for a medical writer to decide and sign.
The protocol reader lists what the amendment changed and maps it to every section of every form. The consent family shows which sentences in which form must change, the reading grade each form would have and each ethics committee's clock. Once the master is released, the country adapter writes the changes into each form as tracked changes.
The plain-language coach scores every sentence with Flesch-Kincaid and Gunning Fog against your target — grade 8 for consent forms and grade 5 for assent forms by default — and proposes a rewrite for each sentence above it that keeps every number and risk of the cited source. A writer accepts, edits or keeps each one.
Yes. The elements checker reads each form against 21 CFR 50.25(a), (b) and (c), the key information section, EU CTR Article 29 for EU versions and your country rules library, lists the statements that meet each element and drafts a plain sentence for anything missing.
No. Required wording such as the ClinicalTrials.gov statement is locked and the coach cannot rewrite it. The country adapter never changes text the master did not change, and every change it makes is tracked.
Yes. Agents propose wording; a named writer accepts each change. A statement that differs from its source blocks release until a person decides. The author, the medical monitor and privacy and legal counsel sign the master, each country study manager accepts their form, and people — not agents — submit to ethics committees.
Yes. The reading targets, the scoring method, the wording lock, natural frequencies, the signer for each role, patient panel review, the sponsor template version and the country rules library are all settings.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your master ICF template, country rules and a current protocol with its brochure — and it goes live once every quality gate has passed. We will walk you through it on one of your own studies first.
A checked first-draft protocol in days, in your approved language.
1-min demoeTMF Quality CheckAn inspection-ready trial master file every day, each site scored on what is due today.
1-min demoClinical Study Report AuthoringResults drafted within days of database lock, linked to tables.
1-min demoWe'll run Informed Consent Authoring on one of your own protocols, its brochure and your master template.