Translating Informed Consent Forms Without Losing Validity

The Informed Consent Form (ICF — TCLE in Brazilian Portuguese) is the one document in a clinical study that the participant actually reads, signs, and takes home. A poorly translated protocol delays a submission; a poorly translated ICF compromises the consent itself — which is why Research Ethics Committees (CEP) and CONEP treat the translation of this document with a stricter standard than the rest of the dossier.

This guide covers what Brazilian law requires, the specific rules for vulnerable populations and non-literate participants, the most common mistakes in ICFs translated from international templates, and what must always be preserved — even when the translation needs to rewrite a sentence from scratch.

What Resolution CNS 466/2012 and Law 14.874/2024 require

Resolution CNS 466/2012, from Brazil’s National Health Council, sets the ethical guidelines for research with human subjects in Brazil and is the regulatory baseline against which every national ICF is assessed — including ones translated from the sponsor’s international standard template. It requires the form to be written in accessible language, to reflect voluntariness, and to clearly disclose risks, benefits, and the right to withdraw consent at any time, without penalty to the participant.

The more recent Law 14.874/2024 formalized a terminology shift that ethics committees had already been adopting: the term “research subject” was replaced by “research participant”throughout Brazilian legislation. It’s not just a style preference — it’s the official term now, and an ICF translation that still reads “subject” (a literal carryover from the English source) immediately signals that the text wasn’t adapted to the current Brazilian regulatory framework.

Vulnerable populations: when the ICF isn’t enough on its own

Not every study ends with a single form signed by the participant themselves. Resolution CNS 466/2012 requires additional safeguards for groups whose age or condition reduces their decision-making capacity — and the translation needs to reflect which document applies in each case, not just translate the standard ICF:

• Children and adolescents. Beyond the ICF signed by the legal guardian, studies involving minors typically require a separate assent form— written in language and at a reading level appropriate to the participant’s age group, not an abridged copy of the adult form.
• Participants with reduced consent capacity.When the study condition involves cognitive impairment, the legal guardian’s consent doesn’t waive the need to explain the study to the participant themselves, to the extent of their understanding — which may also require a second, simpler version of the document.

Translating only the adult ICF and missing that the protocol also calls for a separate assent form is one of the most common gaps in packages from international sponsors, where a single “informed consent form” sometimes covers both roles — and adapting it to the Brazilian model means splitting that into two documents.

Non-literate participants: the impartial witness requirement

Another point that a translation of an international ICF rarely anticipates: when the participant (or their guardian) cannot read, a signature alone is not enough. Resolution CNS 466/2012 requires, in that case, that the ICF be read aloud and signed by an impartial witness— someone with no ties to the research team — who attests that the content was understood. The translated text therefore needs to work just as well read aloud as it does read silently, which reinforces why an ICF translation can’t sound like a technical journal article.

The risks of a literal or automated translation

International ICFs arrive in Brazil as part of a multicenter study’s regulatory package, usually written with the source country’s legal system and medical vocabulary in mind. Translating word for word — or relying on a generic AI with no regulatory context — tends to create three recurring problems:

• Mismatched reading level. The ICF needs to be understandable to a lay participant, not a physician. A translation that mirrors the sentence structure of technical English tends to sound overly formal and hard to follow — something the CEP can send back for a rewrite.
• Inconsistency with the protocol. If the drug name, route of administration, or an adverse event appears with different terms in the ICF and the protocol, the committee flags the discrepancy and pauses its review until it’s fixed.
• Outdated terminology.“Sujeito de pesquisa”, untranslated anglicisms, or adverse-event vocabulary outside the Portuguese MedDRA standard — any of these signal a rushed translation.

A real example of how this shows up, sentence by sentence:

The first version is a word-for-word translation of the English source into Portuguese that mirrors clinical-label vocabulary rather than everyday speech. The second says exactly the same thing in the language Resolution CNS 466/2012 requires: accessible to a lay participant.

What can never change in translation

Rewriting the language to be accessible doesn’t mean taking liberties with the content. Some elements of the ICF are non-negotiable — they must appear in the translation exactly as in the original, without paraphrasing:

Back-translation: when it’s worth it

Back-translation — translating the Portuguese ICF back into the source language and comparing it against the original — isn’t a universal requirement, but it’s widely used by international sponsors and requested by ethics committees in multicenter studies as evidence that meaning was preserved. It works best as an audit step after the final translation, done by a different professional than the original translator — the same principle behind the two-pass review (translator + reviewer) that any serious regulatory translation should have. Back-translation, however, never replaces the reading-level review described above: it confirms meaning was preserved, not that the text became accessible.

Checklist before submitting to the CEP

How this works in practice

Every ICF processed by SealScribe goes through a glossary built from official sources (Resolution CNS 466/2012, Law 14.874/2024, ANVISA’s Controlled Vocabulary, Portuguese MedDRA), with automatic blocking of outdated terms like “sujeito de pesquisa”, and review before delivery — the same two-pass principle described above. When the document contains examples with fictitious names or contact details, anonymization kicks in before anything is sent to the cloud, without altering a single clinical data point — and it comes included with every order at no extra cost, not as a separate add-on.