Research & PolicyFree
IRB Protocol Narrative From Study Notes
A protocol narrative written against the approval criteria your IRB applies, with every risk paired to a protection and every unanswered question flagged.
River's IRB protocol narrative takes what you know about your study and writes the sections a review board reads. Those are background and rationale, objectives, design, population and recruitment, consent, procedures, risks and how each is minimized, benefits, privacy and data security, monitoring, and the analysis plan. Paste a study summary, a grant aims page or rough notes. The narrative follows your institution's template if you include its headings, and every question your description leaves open is listed for you to answer.
Unlike the protocol templates that rank for this search, this writes toward the findings a board must make before it can approve, rather than toward a list of headings. A template asks you to describe risks. A reviewer asks whether each risk is minimized and reasonable against the benefits, whether selection is equitable, and whether privacy is protected. So every risk is paired with its protection, and anything the design does not yet answer is flagged rather than filled with confident prose.
This is for investigators, research coordinators, fellows and residents writing a first or revised submission, and for study teams answering a board's requested changes. Write the narrative before the consent form, because the form has to match it. When the study needs a regulatory file, the trial site regulatory binder keeps it. A quality improvement project that might not be research is worth checking against your board's determination process before you write anything. If the study is also going to NIH, a mock NIH review of your aims page checks the grant side.
Write to the approval criteria, not to the headings
Under the Common Rule, a board can approve research only after it determines that specific criteria are met. Risks are minimized and reasonable in relation to anticipated benefits, selection of subjects is equitable, consent will be sought and documented, data are monitored where appropriate, and privacy and confidentiality are protected. Reviewers read your narrative looking for the evidence behind each finding. A narrative where each criterion has an obvious answer is faster to review than a longer one that leaves the board to hunt.
Consent has a newer rule worth designing for. The revised Common Rule requires consent to begin with a concise and focused presentation of the key information most likely to help someone decide whether to take part. That means the narrative should already say, in plain terms, what participation involves, how long it lasts, the main risks and any benefit. If you cannot state those in a short paragraph, the consent form will not manage it either, and the board will notice.
Monitoring should match risk, not habit. The NIH policy on data and safety monitoring requires monitoring for every clinical trial it supports, and an independent board for multi-site trials of interventions that carry potential risk. A single-site observational study with a blood draw usually needs a clear plan for the investigator to review adverse events, not a committee. Say who reviews what, how often, and what would pause the study. Boards push back on plans that are vague, not on plans that are small.
How it works
Describe the study
Paste a summary, an aims page or rough notes, plus your institution's template headings if you have them.
Pick the review level
Exempt, expedited or full board, or ask River to suggest which fits, for your board to confirm.
Get the narrative
Every section drafted, each risk paired with its protection, and open questions listed rather than guessed.
Answer and revise
Answer the open questions in chat, or paste the board's requested changes and revise against them.
What you get
- Background, objectives, design, procedures and analysis plan written from your study summary or aims page
- Each risk paired with how it is minimized, ordered by likelihood and seriousness for this study
- Recruitment and consent described step by step, including who approaches participants and when
- A key information paragraph drafted for the consent form, consistent with the narrative's own description
- Privacy, data security and monitoring plans sized to the actual risk of this particular study
- Every question your description leaves open listed for you, so nothing is guessed on your behalf
Common questions
Will it decide whether my study is exempt?
No. Only your institution can make that determination, and most boards want exempt studies submitted to them rather than self-declared. The narrative says which category your description seems to fit and why, so the person deciding has the reasoning in front of them. If the answer turns on a detail you did not give, that detail is flagged as an open question.
Can I paste our institution's template?
Yes, and you should. Paste the section headings, or the whole blank template, along with your study notes, and the narrative follows that order and those labels. Boards are used to reading their own form, and a narrative that matches it field by field is easier to review than one that makes the reviewer translate between formats.
Our study only uses existing records. Is this still useful?
Yes, and the questions change. The narrative focuses on which records, the date range, who extracts them, whether identifiers are kept, how the data are stored and destroyed, and whether a waiver of consent or authorization is requested and why. For studies built on registry data, the clinical registry pack keeps the data definitions in one place.
Will it invent background references?
No. Background and rationale are written from what you supplied, and any statement that needs a source is marked for one rather than filled with a reference that may not exist. If you paste your own references or an aims page with citations, those are used as given. Checking that each source says what the narrative claims stays with the investigator.
How does it handle vulnerable participants?
It asks what the regulations ask: whether children, prisoners, pregnant participants or adults who may lack capacity to consent are involved, and if so, which additional protections apply. Including any of them changes the consent process and sometimes the review level, so the narrative names the protections and flags where your board's local policy has to supply the detail.
Is the protocol the same as the consent form?
No, but they must agree. The protocol describes the study for the board, and the consent form describes it for participants. Write the protocol first, then the form, and check every procedure, risk and time commitment against each other. The key information paragraph drafted here is a starting point for the form, not a finished consent document.
Sources
Built by
Chandler Supple
Co-founder & CEO, River
Chandler has spent the last six years working in AI, building tools for millions of users and Fortune 100 companies. In 2025, he co-founded River to build a new kind of operating system for AI agents.
IRB Protocol Narrative From Study Notes
Describe what you need, add any files you have, and your workspace opens with the work already underway.