Exculpatory Language in Informed Consent

An informed consent document can be accurate, well-organized, and still be non-compliant if a single sentence asks a participant to give up a legal right or shields the sponsor or investigator from liability. That specific drafting problem is called exculpatory language, and both the Common Rule and FDA regulations prohibit it outright, in every consent form, regardless of risk level, funding source, or whether the study is otherwise fully compliant.

This guide explains what exculpatory language is, why it is prohibited even when the underlying facts it describes are true, how to identify it during IRB review, and how it differs from other consent-related provisions it is commonly confused with, including waiver of consent, liability-limiting research-injury disclosures, and language governing a legally authorized representative’s role.

What is exculpatory language?

Exculpatory language is any wording in an informed consent document, written or oral, through which a research subject (or their legally authorized representative) is made to waive, or made to appear to waive, any of the subject’s legal rights — or through which the investigator, sponsor, institution, or its agents are released, or appear to be released, from liability for negligence.

The prohibition is codified in two parallel places:

45 CFR 46.116(a)(6) (the Common Rule, HHS and the other Common Rule departments/agencies): “No informed consent may include any exculpatory language through which the subject or the legally authorized representative is made to waive or appear to waive any of the subject’s legal rights, or releases or appears to release the investigator, the sponsor, the institution, or its agents from liability for negligence.”
21 CFR 50.20 (FDA-regulated clinical investigations): the same prohibition, worded almost identically, appears as the closing requirement of FDA’s general informed consent regulation.

Because the two provisions are near-identical in substance, a study that is both Common-Rule-covered and FDA-regulated (for example, an NIH-funded IND trial) does not need to satisfy two different exculpatory-language standards — complying with one satisfies both in practice, though each remains a separately codified regulation.

Why the prohibition exists, even for true statements

The rule is not about accuracy. A sentence can be entirely factual and still be exculpatory, because the prohibition targets the legal effect of the language, not its truthfulness. Informed consent exists so that a prospective subject can make a voluntary, informed decision about whether to participate — not so that participation becomes conditional on surrendering rights the subject would otherwise retain if something goes wrong. HHS’s Office for Human Research Protections (OHRP) has explained this distinction since a 1996 guidance letter on exculpatory language, and both OHRP and the 2011 Federal Register draft guidance on the same topic use the same core test: does the language cause the subject to give up a legal right or claim, or does it merely describe what will or will not be provided?

Worked examples: acceptable vs. exculpatory phrasing

Because the line between a permitted disclosure and a prohibited waiver can be subtle, OHRP’s guidance walks through specific phrasing. The general pattern that IRBs use in practice:

Likely exculpatory (prohibited)

“By signing this form, you release the investigator, the sponsor, and the institution from any liability for negligence.”
“I agree not to sue or file a claim against [institution] as a result of my participation.”
“The investigators are not responsible for any complications resulting from this research.”
“I waive my legal rights to compensation by agreeing to participate.”

Each of these asks the subject to give up something — a claim, a right to sue, a right to seek compensation — as a condition of participating, or states that the institution bears no responsibility at all, even for its own negligence.

Acceptable (a factual disclosure, not a waiver)

“[Institution] does not provide compensation for a research-related injury beyond the immediate treatment of that injury.” — this describes a real limitation on what the sponsor will pay for; it does not ask the subject to release anyone from liability for negligence, and it does not affect the subject’s ability to pursue a legal claim.
“If you are injured as a result of participating in this study, treatment will be available. Your insurance company may be billed for this treatment.” — a factual description of the injury-treatment plan.
“You are not giving up any of your legal rights by signing this form.” — an affirmative statement that no rights are being waived (some IRBs require or recommend a sentence like this precisely to avoid any ambiguity).

The distinguishing test is whether the sentence is a disclosure (what will or won’t happen, what is or isn’t provided) or a release (what the subject gives up, or what the institution is freed from). Compensation-for-injury language, required as a disclosed element of consent under 45 CFR 46.116(b)(6)/(c)(7) and 21 CFR 50.25(a)(6), is a good example of language that sits right next to the exculpatory-language line: it is required, and it can truthfully say compensation is limited or unavailable, but it cannot be phrased as the subject releasing the sponsor from liability for negligence.

How exculpatory language differs from related consent concepts

Exculpatory language is frequently confused with several adjacent, but legally distinct, consent-related mechanisms. Getting the distinctions right matters for IRB reviewers and study teams drafting or revising consent documents.

Exculpatory language vs. waiver of informed consent

A waiver of informed consent is an IRB-approved decision, under a defined regulatory test (45 CFR 46.116(f) or the FDA’s parallel emergency and minimal-risk waiver provisions), that the consent process itself is not required for a specific study or specific research activities — typically because the research is minimal risk and could not practicably be carried out otherwise. Exculpatory language, by contrast, is a drafting defect within a consent form that IRBs must catch and reject during ordinary review of studies that do require consent. One is a decision about whether consent is needed at all; the other is a rule about what a consent form, once required, is not allowed to say. See CASRAI’s guide on Waiver of Informed Consent for the full waiver test.

Exculpatory language vs. the legally authorized representative’s role

When a subject cannot consent for themselves, a legally authorized representative (LAR) may provide surrogate consent on the subject’s behalf, under state or local law defining who qualifies. The exculpatory-language prohibition applies identically whether the person signing is the subject or their LAR — the regulatory text at 45 CFR 46.116(a)(6) explicitly names the “subject or the legally authorized representative” as the party who cannot be made to waive rights. An LAR’s authority to consent does not create any additional latitude for exculpatory language; if anything, IRBs scrutinize consent documents for populations that rely on LARs (e.g., cognitively impaired adults, unconscious emergency patients) especially closely, since these subjects are less able to push back on unclear or coercive phrasing themselves. See CASRAI’s guide on Legally Authorized Representative (LAR) in Research Consent for who qualifies as an LAR and how surrogate consent works.

Exculpatory language vs. required compensation-for-injury disclosure

As covered above, the required disclosure of whether and how research-related injury will be compensated or treated is a separate, affirmative consent element (45 CFR 46.116(b)(6)/(c)(7); 21 CFR 50.25(a)(6)) that consent forms covering more-than-minimal-risk research must include. It becomes an exculpatory-language problem only when it is phrased as a release of liability rather than a factual statement of what is or is not provided.

Catching exculpatory language during IRB review

Because exculpatory language is a drafting problem rather than a study-design problem, it is typically caught during initial or continuing consent-document review, not during the risk/benefit analysis of the protocol itself. IRB staff and reviewers commonly look for:

Release/waiver verbs aimed at the subject: “release,” “waive,” “hold harmless,” “agree not to sue,” “give up my right to.”
Blanket no-liability statements about the investigator, sponsor, or institution (“the researchers are not responsible for…”) rather than statements limited to describing what will or won’t be provided.
Sponsor-supplied consent template language, especially in industry-sponsored trials, since exculpatory clauses sometimes originate in commercial contract-style boilerplate that was not written with 45 CFR 46.116(a)(6)/21 CFR 50.20 in mind and needs to be edited out before IRB approval.
Assumption-of-risk framing that goes beyond disclosing risk (which is required) into asking the subject to accept legal responsibility for those risks in place of the investigator or sponsor.

Because the prohibition is absolute — it applies to minimal-risk and greater-than-minimal-risk research alike, and to both Common Rule and FDA-regulated studies — an IRB cannot approve a consent document containing exculpatory language regardless of how the rest of the protocol is designed. The standard remedy is a revise-and-resubmit consent document, not a protocol-level change.

Frequently asked questions

Is it exculpatory to say the study sponsor won’t pay for treatment of a research-related injury?

Not by itself. Stating that compensation or free treatment is not provided is a required factual disclosure under 45 CFR 46.116(b)(6)/(c)(7) and 21 CFR 50.25(a)(6). It becomes exculpatory only if it is phrased as the subject giving up the right to seek compensation through other means (for example, a lawsuit), rather than simply describing what the sponsor will and will not pay for.

Does the exculpatory-language prohibition apply to minimal-risk studies?

Yes. 45 CFR 46.116(a)(6) and 21 CFR 50.20 apply to any consent document, regardless of the study’s risk level. Some minimal-risk studies qualify for a waiver of the consent process entirely under 45 CFR 46.116(f), which is a different question from what a consent form may say when consent is required.

Who is responsible for catching exculpatory language before a study starts?

The IRB reviewing the protocol and consent document bears ultimate responsibility, but study teams and research administrators drafting or adapting consent templates — especially sponsor-supplied templates in industry-funded trials — should screen for exculpatory phrasing before submission to reduce review cycles.

Does exculpatory language apply to a legally authorized representative signing on a subject’s behalf?

Yes. 45 CFR 46.116(a)(6) explicitly names the legally authorized representative alongside the subject as a party who cannot be made to waive the subject’s legal rights.