Research consent forms: the elements, the IRB, and the clause you may not include

Written for the sponsor, coordinator or start-up preparing a consent document, and for the person being asked to sign one. It describes the federal element lists. It is not a substitute for your institutional review board, which reviews the actual document, or for regulatory counsel, which is not optional in this area.

Which rulebook applies to you

Two federal frameworks govern research consent in the United States, and confusing them is the most common structural error.

For applicable clinical trials the FDA-regulated form also carries a statement that a description of the trial will be available on a public registry. Where the study sits under both frameworks, drafting to the stricter of the two is the only version that clears review once.

The eight basic elements

These are required in every consent document unless the review board has specifically approved an alteration or waiver.

The key information section that comes first

The 2018 revision added a requirement that reshapes the document rather than adding to it. Consent must begin with a concise and focused presentation of the key information most likely to help a prospective participant understand why they might or might not want to take part, organised and presented in a way that facilitates comprehension.

The preamble to the rule identifies what typically belongs there: that participation is voluntary; the purpose, duration and procedures; the risks; the benefits; and the alternatives to taking part. In other words, the decision, ahead of the detail.

The problem it was written to solve is worth naming, because it explains how the section should be judged. Consent forms had become long, complex and burdened with competing purposes — serving legal review, sponsor requirements and institutional protection alongside the participant. A key information section that is itself three pages of defined terms has reproduced the problem it was meant to fix.

The clause you are not permitted to include

This is the provision that surprises people arriving from commercial drafting. No informed consent may include exculpatory language through which the participant is made to waive or appear to waive any legal rights, or which releases or appears to release the investigator, the sponsor, the institution or its agents from liability for negligence.

Note how wide "appears to release" is. It catches not only a release clause but language drifting towards one: a statement that the participant assumes all risks, a sentence saying the institution is not responsible for injury, a waiver of the right to seek compensation. A form may accurately state that no compensation programme exists, if that is the fact. It may not ask the participant to give up a claim.

The contrast with a commercial waiver is instructive. In fitness and recreation, releasing liability for ordinary negligence is the whole point of the document, enforceable in most states and void in a handful — the mechanics are set out in are liability waivers enforceable. In research, that same clause is not a risk to be weighed. It is prohibited, and a form containing it should not clear review.

Alternatives are the element most often thinned out

Of the eight basic elements, alternatives is the one that most often appears as a single sentence. It requires disclosure of appropriate alternative procedures or courses of treatment that might be advantageous to the participant, and the realistic scope of that includes approved therapies for the condition, other forms of treatment such as surgical options, diagnostic approaches, supportive care, and a description of the current recognised standard of care.

The reason it thins out is structural rather than dishonest: the people best placed to write the section have an interest in enrolment. It is worth treating as the element to check hardest on your own document, because it is also the one a participant relies on most heavily in deciding whether the study is a reasonable thing to do rather than a good study in the abstract.

Where non-clinical organisations get caught

The rules above are not confined to hospitals. Wellness companies, digital health products, supplement brands and app developers routinely run something they describe as a study — a trial cohort, an outcomes analysis, a pilot with volunteers — and the paperwork they reach for is a customer agreement or a testimonial release.

Whether an activity is research involving human subjects, and whether it is a clinical investigation of an FDA-regulated product, are definitional questions with regulatory consequences. They are not questions to resolve internally by deciding the activity feels like product development. If you intend to publish the results, make claims from them, or submit them to a regulator, get the determination made properly before you enrol anyone, because consent obtained under the wrong framework cannot be repaired retrospectively.

Privacy sits on top, not inside

Research consent authorises participation. It does not, by itself, authorise a covered entity to use or disclose protected health information for the research — that runs on its own authorisation with its own required elements, and special rules govern when it may be combined with the consent document. The general shape of those requirements is in HIPAA authorization or records release.

Where the research is embedded in care, both obligations apply at once. The participant is consenting to treatment under the ordinary disclosure standard — described in what informed consent requires — and separately to research participation under the element list. Collapsing the two into one document is possible in some settings and is a decision for the review board, not a drafting convenience.

What a good document looks like from outside

A page of key information a person could act on. Then the elements as labelled sections, so a reviewer can find each one without hunting. Alternatives written as though the reader might reasonably choose one. Contacts split three ways — the study, participants' rights, and injury — because a participant with a complaint should not have to route it through the investigator. Nothing anywhere that looks like a release.

That is a document that reads as a disclosure rather than a defence, which is what the regulations are structured to produce. The forms that fail review are almost always the ones drafted by working outwards from the institution's exposure instead of inwards from the participant's decision.