How to Use Claude as a Medical Writer: What It Can Draft

June 11, 2026

Skill · .md

The skill behind this guide: Nora - Medical Writer AI Skill. Asks for audience, document type and jurisdiction first, and marks missing data instead of inventing it - $14.99, one payment, yours permanently.

View the Nora skill →

A model can take the first-draft burden off structuring a document, recasting the same trial data for a different reader, and turning clinical language into plain language. It cannot be trusted with a reference, it does not know your approved label, and it has no idea which of your inputs you were allowed to paste. Nothing drafted this way goes to a patient, a regulator or a journal without qualified medical, scientific and regulatory review. This is not medical advice.

Why is citation fabrication the first thing to deal with?

Because in this field a fake reference does not just embarrass you. A model will produce a real-sounding author list, a plausible journal, a volume, a page range and a well-formed DOI for a paper that does not exist - and the output carries no signal that it invented any of it. No hedge, no missing field, no lower confidence.

It does this most often precisely where you most want the support: the specific claim at the hinge of your argument, in the place where the real literature is thinnest. If a well-covered topic is being cited, the model has seen genuine papers and is more likely to name one. If your sentence needs a study nobody ran, you get a study nobody ran.

The adjacent failures are worse in medical writing because they survive a quick look: a real paper with invented page numbers; a real paper whose finding was in a different population, at a different dose, or over a different follow-up than the sentence you attached it to; a resolving DOI pointing at an unrelated article. Every reference gets resolved against the actual source - DOI, PubMed, the journal - before it goes anywhere.

The method that works: never ask for sources

Reverse the order. Have the model mark the claims that need citing and tell you what to search for. Find the sources yourself. Then paste the real source text back in and have it check your draft against what the source actually says - which is the step that catches the more common problem of a real reference stretched past its evidence.

Prompt 1 - mark the claims, supply no references

You are reviewing a draft [DOCUMENT TYPE: manuscript discussion / background section / plain-language summary] in [THERAPEUTIC AREA]. Do NOT supply any citation, author name, year, journal or DOI. If you are tempted to name a source, name a search instead. Mark every sentence making a factual, clinical or quantitative claim that requires a source. For each, output: - the sentence, quoted exactly - the evidence level that would be needed to support it as written (RCT, meta-analysis, observational, guideline, label/SmPC, preclinical) - three PubMed search strings, in decreasing breadth - whether the claim, as phrased, is stronger than the evidence type it is likely to have Separately, list every sentence that states or implies efficacy, comparative efficacy, or safety. Those go to regulatory review. DRAFT: [PASTE]

Prompt 2 - check the paraphrase against the real source

Below is a passage from my draft, then the actual text of the source I cited for it. I have read the source. Check my use of it. For each claim in my passage, state whether the source text: (a) states it directly, (b) implies it but does not state it, (c) states something narrower - different population, dose, endpoint, follow-up, comparator or magnitude, (d) does not support it at all. Quote the exact line you are judging against. If you cannot find a supporting line, answer (d). Do not fill any gap from your own knowledge and do not tell me what the wider literature says. Flag separately any number in my passage that does not appear verbatim in the source. MY PASSAGE: [PASTE] SOURCE TEXT: [PASTE THE RELEVANT PAGES]

Skill spotlight

Nora - Medical Writer AI Skill

$14.99this skill vs $100hiring a freelance medical writer/hr

Asks for audience, document type and jurisdiction before it writes a line, then works only from the data you give it - marking anything absent as [DATA TO BE PROVIDED] rather than filling the gap. Every output comes back labelled as a draft for qualified review.

View Nora - Medical Writer AI Skill →

What can you actually put into it?

This is a real decision with real rules, and it is decided before you paste, not after. Two separate problems get confused.

Patient data. Under HIPAA, a vendor processing protected health information on your behalf needs a Business Associate Agreement. Consumer chat tiers generally do not come with one, so “I will just paste the case summary” is a decision about your organisation's compliance posture, not a convenience. De-identification also has defined standards - Safe Harbor and Expert Determination - and removing the name is not one of them. Dates, a rare diagnosis, a small site population, an unusual combination of ages and locations: any of these can re-identify a person. Under GDPR, health data is a special category with its own basis, minimisation and retention requirements, and where the processing physically happens matters.

Unpublished trial data. Different rulebook entirely. This is usually confidential to the sponsor under contract, and a confidentiality agreement does not care whether the recipient is a person or a tool. Topline results before disclosure can also be material non-public information for a listed sponsor.

The practical answer: ask what tier and what agreement your organisation has in place, and get the answer before you draft anything real. Learn the workflow on synthetic or already-published data - it teaches you exactly the same thing.

What does the model not know about regulated content?

It does not know your label. It writes from the published literature and from how drug copy generally sounds, which is why it drifts into efficacy phrasing that goes beyond an approved indication without any sense that it has crossed a line.

In the US, prescription drug promotion sits under 21 CFR 202.1. Material must present a fair balance between risk and benefit information - a structural requirement about how thoroughly each is treated, not a matter of tone - and must not be false or misleading about side effects, contraindications or effectiveness, or omit material facts about consequences of use as recommended. Promotional materials are submitted to FDA on Form 2253 at first dissemination, and the Office of Prescription Drug Promotion issues untitled and warning letters when they are not compliant. A model asked for “engaging” copy will produce benefit-heavy text by default, because that is what most marketing copy looks like.

Structured documents have prescribed contents: ICH E3 for clinical study reports, the SmPC and package leaflet formats in the EU, and reporting guidelines such as CONSORT for randomised trials, PRISMA for systematic reviews and STROBE for observational studies. The characteristic failure is not a section that looks wrong - it is a section that reads perfectly and silently omits a required item.

For journal submission, ICMJE recommendations state that a chatbot cannot be listed as an author, because authorship requires responsibility for the accuracy and integrity of the work. Authors are asked to disclose AI-assisted technologies at submission, describe how they were used, and remain responsible for verifying everything an AI touched. The same discipline is covered from the academic side in our academic writer guide, and reference-checking in the citation manager guide.

What is it genuinely good at?

Recasting. The same dataset has to become a clinical study report section, a manuscript, a clinician brief and a patient leaflet, in four registers, and that translation work is real hours. A model given the numbers and told the audience does the structural and linguistic lift while you keep the judgement.

It is also good at first-pass structure for a document type you write rarely, at flagging where your own draft has drifted in register, and at producing the awkward questions a reviewer will ask before the reviewer asks them.

Prompt 3 - recast for a different reader without adding claims

Rewrite the passage below for [AUDIENCE: patients and carers / prescribing clinicians / an HTA committee / a lay press officer] in [JURISDICTION: US / EU / UK], target reading level [e.g. UK reading age 12]. Hard rules: - Introduce no fact, number, benefit, comparison or reassurance that is not already in my text. Anything missing goes in a GAPS list. - Do not convert an absolute number into a relative one, or the reverse, and do not recalculate anything. Copy figures verbatim. - Keep uncertainty intact. If I wrote that an effect was not statistically significant, it stays not significant. - Keep risk information as prominent as benefit information. - Where a required element of this document type is missing from my text, insert [DATA TO BE PROVIDED] - never a placeholder value. Return the rewrite, then GAPS, then a list of every sentence a regulatory reviewer should look at and why. PASSAGE: [PASTE]

Prompt 4 - pre-review pass before it reaches your reviewer

Act as a critical first reader of the draft below. This is [DOCUMENT TYPE] for [AUDIENCE] in [JURISDICTION], concerning [PRODUCT / INTERVENTION], indication as written in the label: [PASTE THE APPROVED INDICATION TEXT]. You are not approving anything. Produce a list for a qualified reviewer. Flag, with the sentence quoted: 1. Any claim that goes beyond the indication text I pasted. 2. Any comparative claim, explicit or implied by ordering or emphasis. 3. Any place where benefit is stated more fully than risk. 4. Any number that appears without its denominator, timeframe, population or uncertainty. 5. Any causal language attached to non-causal evidence. 6. Any required element of this document type that is absent - name the guideline or format you are checking against, and say plainly if you are not confident it applies in this jurisdiction. Do not rewrite anything. Do not reassure me. DRAFT: [PASTE]

Complementary prompt pack

AI Medical Documentation Writer

$9this pack vs $25a medical scribe service/hr

For the everyday half of the job - clinic letters, case summaries, specialist correspondence - drafted for clinician review before anything is sent. The same confidentiality question applies before you paste.

View AI Medical Documentation Writer →

How do you install the Nora skill?

The download is a ZIP with SKILL.md at the root of the archive - not inside a nested folder, which is the usual reason an upload fails. In the Claude desktop app, open Customize → Skills, upload the ZIP, and toggle it on. Claude loads it whenever your request matches what it covers.

Skills are available on Free, Pro, Max, Team and Enterprise. The real requirement is code execution: on Free, Pro and Max, enable Code execution and file creation under Settings → Capabilities. On Team it is on by default at the organisation level; on Enterprise an owner enables it in organisation settings. If the Skills menu is greyed out on your work account, that is an admin switch someone has turned off, not a plan limitation - and in a life sciences organisation it may have been turned off deliberately, which is worth asking about before you route around it.

In ChatGPT or Gemini there is no upload step: open SKILL.md, copy the contents, and paste them into custom instructions. You lose automatic triggering and keep the method.

Where does it fail?

Arithmetic on your data. It will re-derive a percentage, a confidence interval or an incidence rate from your text and get it wrong while sounding certain. Copy every figure by hand from the source table.
Coded terminology. MedDRA preferred terms, ICD and ATC codes come back plausible and frequently wrong. These are lookups, not language tasks.
Jurisdictional structure. A US prescribing information document and an EU SmPC are not the same document with different spelling. Asked for “prescribing information” it will blend both.
Fair balance. Left to its own defaults it writes benefit-forward copy. Balance has to be an explicit instruction and still has to be checked by someone qualified.
Agreeableness. Ask whether your draft is compliant and you will usually be told yes. Ask what a reviewer would object to instead.
Audit trail. Chat output has no version history, no reviewer sign-off and no controlled record. In an environment governed by 21 CFR Part 11 or EU Annex 11, the draft has to enter your validated document system, not live in a conversation.
Confidentiality. This one fails at the moment of pasting. There is no later fix.

Who is this for?

Medical writers, regulatory and clinical teams in pharma and CROs, medical communications agencies, and academic researchers writing studies up. It works in Claude, ChatGPT or any AI chat. For adjacent clinical documents - consent forms, clinical summaries, patient materials - the health & medical skills collection and the health & medical prompt packs cover more of the same ground, each as a focused assistant rather than a general chatbot. Everything here produces drafts for qualified review, and none of it is medical advice.

In summary:

Never take a reference from a model, settle the confidentiality question before you paste, and treat every output as a draft for qualified medical and regulatory review. For regulatory, clinical and patient-facing documents use Nora - Medical Writer AI Skill ($14.99); for day-to-day clinic letters and case summaries, the AI Medical Documentation Writer ($9) prompt pack covers the rest. Both work in Claude, ChatGPT and any AI chat, with a 30-day money-back guarantee.

Skill · .md · Works with Claude & ChatGPT

Nora - Medical Writer AI Skill

One file that fixes audience, document type and jurisdiction before a word is written, works only from the data you supply, and marks every gap as [DATA TO BE PROVIDED] instead of inventing one. Drafts for qualified review. No subscription. Yours permanently.

$14.99

Get the Nora skill →

KissMySkills is a marketplace of 853 AI skills, 158 prompt packs, 55 agents & free tools for Claude, ChatGPT & any AI chat.

Related Skill Guides

Frequently asked questions

Can I use Claude for medical writing?+

For first drafts, structure, and recasting the same data for a different audience, yes — with three conditions. Every reference must be resolved against the actual source before it goes anywhere. The confidentiality question about what you paste must be settled first. And nothing reaches a patient, a regulator or a journal without qualified medical, scientific and regulatory review. This is not medical advice.

Why does AI invent references in medical writing?+

A citation is a highly patterned string — authors, year, title, journal, volume, pages, DOI — and a model that continues patterns produces all of those features whether or not the paper exists. Nothing in the output marks an invented one. It happens most often exactly where you most want support: the specific claim at the hinge of your argument, where the real literature is thinnest. Adjacent failures include a real paper with invented page numbers, and a real finding from a different population, dose or follow-up than your sentence implies.

Can I put patient data or unpublished trial data into an AI chat?+

Treat these as two separate questions with two separate rulebooks. Under HIPAA, a vendor processing protected health information needs a Business Associate Agreement, and consumer chat tiers generally do not carry one. De-identification has defined standards — Safe Harbor and Expert Determination — and removing a name is not one of them; dates, rare diagnoses and small site populations can re-identify. Under GDPR, health data is special category with its own basis, minimisation and retention rules. Unpublished trial data is usually confidential to the sponsor under contract regardless of privacy law. Ask what agreement your organisation has in place before you paste anything real.

What does an AI get wrong about promotional and regulatory medical content?+

It does not know your approved label, so it writes efficacy language drawn from the literature rather than the indication. In the US, prescription drug promotion sits under 21 CFR 202.1, which requires fair balance between risk and benefit information and prohibits material that is false or misleading about side effects, contraindications or effectiveness; materials are submitted to FDA on Form 2253 at first dissemination. Structured documents have prescribed contents — ICH E3 for clinical study reports, SmPC and package leaflet formats in the EU, CONSORT, PRISMA and STROBE for reporting. The typical failure is a section that reads perfectly and quietly omits a required item.

Can I list Claude as an author on a paper?+

No. ICMJE recommendations state that a chatbot cannot be listed as an author, because authorship requires responsibility for the accuracy, integrity and originality of the work. Authors are asked to disclose AI-assisted technologies at submission, describe how they were used, and remain responsible for verifying everything the AI touched.

How do I install the Nora Medical Writer skill?+

The download is a ZIP with SKILL.md at the root of the archive, not inside a nested folder. In the Claude desktop app open Customize then Skills, upload the ZIP and toggle it on. Skills work on Free, Pro, Max, Team and Enterprise; the real requirement is code execution — enable Code execution and file creation under Settings then Capabilities on Free, Pro and Max. On Team it is on by default and on Enterprise an owner enables it, which is why the menu is sometimes greyed out on a work account. In ChatGPT or Gemini, paste the contents of SKILL.md into custom instructions instead.