“Informed consent” is often used as shorthand for a signed form. That shorthand causes real problems for research teams and IRB/REC reviewers alike, because the form and the consent it documents are not the same thing. A signature on a document that nobody understood, or that was obtained under pressure, does not constitute informed consent — no matter how complete the paperwork looks. This guide covers what informed consent actually requires as a matter of research ethics and US federal regulation, why the form is only the documentation layer, which populations need additional protections, and when the requirement can be waived or altered.
What informed consent actually requires
Informed consent is a process, not an event. Federal regulation and research-ethics guidance converge on the same substantive requirements, whether the eventual documentation is a signed form, a verbal script, or an electronic acknowledgment:
- Voluntary participation. The decision to enroll must be free of coercion or undue influence — including subtle pressure from a treating clinician, a supervisor, or a financial incentive large enough to distort judgment. Under 45 CFR 46.116(a)(1), an investigator must seek consent only under circumstances that give the prospective participant “sufficient opportunity to discuss and consider whether or not to participate” and that minimize the possibility of coercion.
- Disclosure. The participant must receive the information “a reasonable person would want” to decide whether to take part — the study’s purpose, procedures, and expected duration; which procedures are experimental; reasonably foreseeable risks and discomforts; expected benefits (to the participant or to others); alternatives to participating; and how confidentiality will be protected.
- Comprehension. Disclosure alone isn’t sufficient — the information must actually be understood. Regulatory language requires it be conveyed “in language understandable to the subject,” which in practice means avoiding jargon, allowing time for questions, and — for non-English speakers or participants with low literacy — using a qualified interpreter or a simplified consent document rather than assuming a signature implies comprehension.
- Capacity to consent. The participant (or their legally authorized representative) must have the legal and cognitive capacity to make the decision. Where capacity is uncertain, this becomes a formal assessment question, not an assumption — see special populations below.
- An ongoing right to withdraw. Consent obtained at enrollment doesn’t exhaust the requirement. Participants must be able to withdraw at any point without penalty, and — for research involving more than minimal risk — must be told what withdrawal means procedurally (for example, whether already-collected data will be retained).
These map closely onto CASRAI’s own Informed consent dictionary entry, which defines it as “the voluntary agreement of a prospective research participant… given after disclosure of the information required by the applicable regulation and free from coercion or undue influence” and treats it explicitly as an ongoing dialogue rather than a single signing event.
The regulatory basis: the Belmont Report and the Common Rule
In the United States, the substantive requirements above trace directly to two connected sources.
The Belmont Report (1979), issued by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, set out three ethical principles for human-subjects research: respect for persons (treating individuals as autonomous agents, with extra protection for those with diminished autonomy), beneficence (maximizing benefit and minimizing harm), and justice (fair distribution of the burdens and benefits of research). The report ties these principles to three concrete applications — informed consent, assessment of risks and benefits, and selection of subjects. Informed consent is the direct application of respect for persons: it operationalizes autonomy into a specific procedural requirement.
The Belmont principles were subsequently codified into federal regulation as the Common Rule, 45 CFR Part 46, which governs federally funded and federally regulated human-subjects research in the US and is the regulatory basis most US IRBs operate under regardless of funding source. §46.116 sets the general requirements described above and lists the basic elements every consent process must cover (purpose and procedures, risks, benefits, alternatives, confidentiality, compensation/treatment for research-related injury where relevant, a statement that participation is voluntary, and — since the 2018 revisions — a statement about future use of identifiable data or biospecimens where applicable). Outside the US, the analogous substantive requirements are expressed through the Declaration of Helsinki and ICH Good Clinical Practice guidelines; the specific citations differ but the underlying ethical content — voluntariness, disclosure, comprehension, ongoing withdrawal rights — is consistent across frameworks.
The consent form documents consent — it isn’t consent itself
This is the most common misconception the search terms behind this guide (“informed consent form,” “informed consent example”) point to, and it’s worth stating precisely: the Common Rule regulates the informed-consent process under §46.116, and regulates documentation of that process separately, under §46.117. They are two different sections for a reason.
A compliant consent form typically restates the §46.116(b) elements in writing and provides a place for a signature or equivalent record. But a signed form with no evidence that the underlying disclosure-comprehension-voluntariness process actually happened does not satisfy §46.116 — an IRB or REC reviewing a consent form is checking that the document accurately discloses what §46.116 requires, not certifying that consent was validly obtained in every individual case. That distinction is also why CASRAI’s ethics approval statement dictionary entry treats an ethics-committee approval statement and a declaration that informed consent was obtained as two separate, commonly-paired disclosures in a manuscript, rather than one substituting for the other.
What a compliant process looks like in practice (illustrative pattern, not a specific documented case): a study coordinator reviews the consent document with the prospective participant in a private setting, explains procedures and risks in plain language, checks understanding by inviting questions rather than only asking “do you have any questions,” and only proceeds to signature once comprehension is reasonably confirmed and no pressure to consent is apparent. What fails the requirement, even with a fully executed form on file: a signature collected in a group setting with no individualized discussion, or enrollment obtained after a participant expressed hesitation and was not given a genuine opportunity to decline.
Special populations requiring additional protections
The Common Rule adds protections beyond §46.116 for participants whose capacity to consent, or freedom to decline, is not straightforward:
- Children/minors — Subpart D. Children generally cannot give legally binding consent. Instead, 45 CFR 46.408 requires both parental or guardian permission and, where the IRB determines the children are capable of it, the child’s own assent — their affirmative agreement to participate. Assent is distinct from mere non-objection: a child’s silence or failure to object is not assent.
- Individuals with cognitive or decisional impairment. Where capacity is diminished or fluctuating, consent may need to be obtained from a legally authorized representative, and some protocols require ongoing reassessment of the participant’s capacity rather than a one-time determination at enrollment. The Common Rule does not set out a single national standard for this population the way it does for children or prisoners; institutional IRB policy and applicable state law typically govern the specific procedure.
- Prisoners — Subpart C. Because incarceration itself can compromise the voluntariness of a decision to participate (including through the promise of favorable treatment), Subpart C imposes additional safeguards, including specific IRB composition requirements and constraints on the categories of research prisoners may be enrolled in.
- Pregnant women, human fetuses, and neonates — Subpart B. Additional safeguards apply given the potential for research risk to extend to a fetus or neonate who cannot themselves consent; where a research participant under Subpart B is also a child under Subpart D’s definition, both sets of protections apply.
Outside the federal human-subjects framework specifically, professional bodies impose parallel obligations — for example, the American Psychological Association’s Ethics Code (Standard 8.02, Informed Consent to Research) requires disclosure of purpose, procedures, the right to decline or withdraw, foreseeable risks and benefits, confidentiality limits, and who to contact with questions, closely mirroring the Common Rule elements. Readers arriving at this guide via “informed consent in counseling” or “informed consent psychology” searches are often looking for the clinical/therapeutic-consent obligation rather than the research-consent obligation covered here — related but governed by a different standard (for psychologists, primarily Standard 10.01, Informed Consent to Therapy, rather than Standard 8.02).
When consent can be waived or altered
The Common Rule allows an IRB to waive or alter informed consent, or waive only the documentation of it, under two separate provisions that are frequently conflated:
- Waiver or alteration of the consent process itself — §46.116(f). An IRB may waive some or all of the required elements of consent, or alter them, only where the research involves no more than minimal risk, the waiver or alteration will not adversely affect participants’ rights and welfare, the research could not practicably be carried out without the waiver or alteration, and — whenever appropriate — participants will be given additional pertinent information after participation.
- Waiver of documentation only — §46.117(c). This is the provision behind the “minimal risk, form would be the only identifying record” scenario. An IRB may waive the requirement for a signed consent form (while the underlying consent process still must occur) where either: (a) the signed form would be the only record linking the participant to the research, and the principal risk of the research is a confidentiality breach — in which case each participant is asked whether they want documentation linking them to the study, and their wishes govern; or (b) the research presents no more than minimal risk and involves no procedures for which written consent is normally required outside a research context. This is why some minimal-risk survey or interview studies use an information sheet plus verbal consent rather than a signature — the identifying document itself would be the primary privacy risk.
The distinction matters operationally: a waiver under §46.116(f) changes what participants are told or whether the consent process occurs at all; a waiver under §46.117(c) changes only whether a signature is collected — the disclosure-comprehension-voluntariness process under §46.116 still has to happen either way.
Where this fits in the broader approval process
Informed consent is one of the substantive criteria an IRB or REC evaluates when it reviews a protocol — alongside risk/benefit balance and equitable subject selection — but the consent process itself is typically reviewed and refined iteratively, not just approved once at the study’s outset. For the mechanics of which studies require IRB/REC review, which review pathway applies, and how long approval typically takes, see CASRAI’s IRB/REC approval process guide. For research that qualifies for exemption from ongoing IRB oversight (which does not mean consent requirements disappear — most exemption categories still require some form of disclosure), see the exempt human subjects research dictionary entry. For how a manuscript documents that both ethics approval and informed consent were obtained, see the ethics approval statement entry. The reviewing body itself is defined in the IRB (Institutional Review Board) dictionary entry.
Related consent models with their own CASRAI dictionary entries — useful where a single, static consent form doesn’t fit the research design — include broad consent (a single consent for storage and future secondary use of biospecimens/data, permitted under the 2018 Common Rule revisions as an alternative to study-specific re-consent) and dynamic consent (an ongoing, typically digital consent relationship allowing participants to review and adjust their preferences over time, rather than a single point-in-time decision). For research involving Indigenous communities and their data, Free, Prior and Informed Consent (FPIC) is a related but distinct standard, grounded in the UN Declaration on the Rights of Indigenous Peoples rather than the Common Rule.
Frequently asked questions
Is a signed consent form legally the same thing as informed consent?
No. The form documents that a consent process took place; it is not itself the consent. Under the Common Rule, the process is governed by §46.116 and documentation is governed separately by §46.117 — a signed form obtained without genuine disclosure, comprehension, and voluntariness does not satisfy the informed-consent requirement, even though it exists on file.
What are the core elements a valid informed consent process must include?
Voluntary participation free of coercion, disclosure of the information a reasonable person would want (purpose, procedures, risks, benefits, alternatives, confidentiality protections), genuine comprehension by the participant, capacity to consent, and an ongoing right to withdraw without penalty.
Can informed consent be obtained verbally instead of with a signed form?
Yes, where an IRB has waived documentation under §46.117(c) — typically for minimal-risk research where a signed form would be the only record linking the participant’s identity to the study, or where the research involves no procedures that would normally require written consent outside a research setting. The underlying consent process (disclosure, comprehension, voluntariness) still has to occur; only the signature requirement is waived.
What’s the difference between assent and consent for research involving children?
Consent is the legally binding agreement; children generally cannot give it. Under 45 CFR 46.408, research involving children instead requires the parent or guardian’s permission plus, where the IRB finds the child capable, the child’s own assent — their affirmative agreement, which is distinct from simply not objecting.
Does informed consent still apply if research qualifies for IRB exemption?
Usually, yes, in a modified form. Exemption removes the research from ongoing IRB review under most categories, but most exemption categories still expect some form of disclosure to participants; exemption is not a waiver of the ethical obligation to disclose and obtain agreement. See CASRAI’s exempt human subjects research entry for the specific categories.
