ISO 17100 certification is the foundation of medical translation quality assurance. The standard sets the requirements for translator qualifications, independent revision, project management, and traceability that every serious LSP serving regulated medical clients should meet. AdHoc holds current certification to ISO 17100 and to ISO 18587 (the standard for full human post-editing of machine translation output), and both certificates are available on our site for any buyer who wants to verify them.
What buyers ask us next, once certification is established, is how the rest of the workflow fits together. ISO 17100 by design does not extend to MTPE (that is ISO 18587), to linguistic validation for patient-facing instruments (ISPOR Good Practices is the working reference), or to regulator-specific submission requirements like EMA Quality Review of Documents templates. Six months from now, an EMA reviewer opening your translated clinical trial protocol will want to see evidence of the full QA chain that produced it, not just the certificate at the top.
This article walks through that chain: what ISO 17100 commits an agency to, where back-translation belongs, when in-country review and cognitive debriefing earn their place, and how the layers combine by document type. There is a decision matrix in the middle, a vendor checklist near the end you can lift into your next RFP, and a sources footer for anyone tracing the regulatory and methodological claims back to primary references.
TL;DR
|
What is medical translation quality assurance?
Medical translation quality assurance is the documented set of process controls that produce translations of medical content defensible when an auditor or regulator asks how the work was done. It is process discipline, captured in records, that travels with the file long after delivery.
Quality assurance and quality control are related but separate. QA lives in the process: who is qualified to translate, who reviews, what records get kept. QC lives in the output: terminology checks, format compliance, and the linguistic verification that runs at the end. An agency running translation memory match checks at delivery is doing QC. An agency requiring every medical translator to demonstrate clinical subject matter experience before assignment is doing QA. Competent agencies do both. The QA half is what carries the audit trail.
The audience for that audit trail is what matters for medical content. A regulator asking whether a translated patient information leaflet was produced under controlled conditions wants process evidence. So does a notified body reviewing your medical device IFU. Matheson and Kollmer, writing in the Medical Writing journal of the European Medical Writers Association (March 2024), frame the same point in academic terms: QA in medical translation is process discipline rather than a property of the final text. The discipline is what an inspector reads six months after delivery.
Most medical translation sits at this intersection: an output that reads cleanly in-market and a process that withstands the audit.
What ISO 17100 certification tells you, and what comes next
ISO 17100 certification tells you the agency operates a quality management system that meets the international standard for translation services. The certificate is issued after a two-stage audit by an accredited certification body, is valid for three years, and is maintained through annual surveillance audits. For buyers evaluating LSPs, that is a meaningful signal: it shows the agency runs the documented procedures, translator qualifications, and independent revision process the standard requires.
What the certification does not do is audit every project on your account. The certification body audits the system; the system is what produces your project. So the natural next questions, once you have filtered for certified suppliers, are how the standard’s requirements play out on your specific medical content, and what the agency adds on top of ISO 17100 for the regulatory and methodology layers the standard does not extend to. The rest of this article walks through that frame.
What ISO 17100 requires (the standard in plain language)
ISO 17100:2015 specifies four core requirements. Qualified translators and revisers. A mandatory two-step process: translation by one professional, revision by an independent second. Documented project management. Continuous improvement procedures. Each one maps to a question you can ask a vendor. AdHoc holds current ISO 17100 certification (linked above), and what follows is what the certification commits us to.
Qualified translators and revisers (and what counts)
ISO 17100 lists three pathways for translator qualification. A recognised degree in translation, linguistics or language studies (or an equivalent degree with significant translation training) from a higher-education institution. A recognised degree in any other field from a higher-education institution plus two years of full-time professional translation experience. Five years of full-time professional translation experience without the formal qualification. Revisers must meet the same standard plus translation or revision experience in the relevant subject area.
These pathways are the floor for medical work, not the ceiling. The agency should layer specific medical subject matter experience on top. A linguist translating a clinical trial protocol needs more than a translation degree. They need to know the difference between an exclusion criterion and a withdrawal criterion (the words look similar; the meanings do not overlap). Some agencies will share anonymised translator profiles on request, which is reasonable to ask for during vendor evaluation.
The four-eyes principle
Every ISO 17100 project requires translation by one qualified professional and revision by a second qualified professional working independently. The reviser checks the target text against the source for accuracy, terminology, register, and compliance with project specifications.
Revision is not proofreading. A proofreader works monolingually on the target text and catches typos and surface problems. A reviser works bilingually against the source and catches meaning errors and inconsistencies with terminology resources. Different jobs, different people, different skills. An agency offering “translation plus proofreading” is delivering something below ISO 17100, however the proposal is worded.
If you take one question from this article into your vendor evaluation, take this one: does the standard workflow include independent bilingual revision, or only proofreading? The answer often surfaces vendors using the wrong terminology in their own proposals.
Project management and traceability
ISO 17100 requires documented project records: who translated, who revised, what reference materials were used, what queries were raised and how they were resolved. The standard does not specify the technology, only that the records exist and can be produced on request.
Scattered email threads and shared drives do not meet the bar in practice. A translation management system that captures the full project trail in one place does. The traceability matters beyond regulatory housekeeping. It is what lets an account team honour your terminology consistently across projects, onboard a new project manager without losing context, and produce a project record when an auditor asks. AdHoc’s SmartDesk holds this audit trail for client accounts, with project history, queries, and resolutions in one searchable place.
What ISO 17100 does not cover
Three areas sit outside the standard by design. Machine translation post-editing has its own standard (ISO 18587, which the scope of ISO 17100 explicitly excludes). Linguistic validation for patient-facing instruments follows a separate methodology, with ISPOR Good Practices (Wild et al., 2005) as the working reference. Regulator-specific submission requirements like EMA Quality Review of Documents (QRD) templates are a regulatory matter rather than a translation-process one. Each is covered in the sections that follow.
| Standard | What it covers | When it applies in medical translation |
|---|---|---|
| ISO 17100:2015 | Human translation workflow: translator and reviser qualifications, two-step process, project management | Almost every medical translation project involving human translators |
| ISO 18587:2017 | Full human post-editing of machine translation output: post-editor qualifications and process | Wherever MTPE is part of the workflow |
| ISO 9001 | General quality management system applicable to any organisation | Agency-wide; relevant as evidence of operational maturity |
| ISO 13485 | Quality management system for medical device manufacturers and their suppliers | Translation agencies serving medical device clients under MDR or IVDR |
ISO 18587 and machine translation post-editing in medical work
ISO 18587:2017 is the standard for full human post-editing of machine translation output. The standard specifies the process and competence requirements for full post-editing, where the goal is output indistinguishable in quality from human translation (light post-editing, which aims only at understandability, is mentioned but is not the focus). AdHoc holds current ISO 18587 certification (linked above) alongside ISO 17100, which means MTPE in our workflows runs under documented process controls.
The standard requires post-editors to hold the same qualification level as ISO 17100 translators (the same three pathways, with “translation or post-editing” experience counting toward the experience requirements) plus specific training in post-editing methodology. That is the floor. The harder question, which the standard deliberately leaves to the buyer and the agency, is when MTPE is appropriate for a given piece of medical content.
A framing that holds up in practice: MTPE works where missing a subtle nuance is operationally inconvenient. MTPE fails where missing a nuance becomes a regulatory rejection or a patient safety issue.
| Content type | MTPE position |
|---|---|
| Internal SOPs and process documentation | Acceptable with full post-editing under ISO 18587 |
| Technical knowledge bases for internal use | Acceptable |
| General marketing materials (non-brand-critical) | Acceptable with post-editing and downstream review |
| Regulatory submissions to EMA, FDA, or notified bodies | Not acceptable |
| Patient information leaflets (PIL) | Not acceptable |
| Informed consent forms (ICF) | Not acceptable |
| Medical device IFUs (patient-facing) | Not acceptable |
| Clinical trial protocols | Not acceptable |
| Pharmacovigilance reports | Not acceptable |
The question for your vendor is not whether they use machine translation. It is under what conditions, with what post-editor qualifications, and with what process documentation. The written policy on when MTPE applies to your medical content is the thing to ask for.
Back-translation: when it is required and when it is not
Back-translation is the process of translating a translated text back into the source language by a separate linguist who has not seen the original. The aim is to verify meaning equivalence between source and forward translation. It is a verification technique with a specific use case rather than a default to apply across an entire translation programme.
The cost of applying it everywhere is real. Back-translation roughly doubles the linguist cost for affected documents because the volume of translation work doubles. It adds material time to the workflow, with the exact impact varying by document length, language pair, and the complexity of the reconciliation. Applying it where it does not belong inflates budgets without improving quality. Applying it where it does belong is the difference between a clean regulatory submission and a back-and-forth with the regulator.
When back-translation is the right step
The clearest case is FDA-regulated clinical outcome assessment (COA) and patient-reported outcome (PRO) instruments. FDA expects documented linguistic validation for instruments used in registration trials, though the agency does not strictly mandate a single methodology. Its guidance language is “fit-for-purpose” with evidence of conceptual equivalence. In practice, the ISPOR Good Practices guidance (Wild et al., 2005) is the working reference most sponsors and contract research organisations follow, and back-translation sits inside it as one of ten steps: Preparation, Forward Translation, Reconciliation, Back Translation, Back Translation Review, Harmonization, Cognitive Debriefing, Review of Cognitive Debriefing Results and Finalization, Proofreading, and Final Report.
Informed consent forms come next. Many institutional review boards require back-translation of the ICF as part of approval, though the requirement is institution-specific. Confirm with the IRB before scoping. Some accept certified translation with independent review in place of back-translation; some do not.
Some pharmacovigilance content rounds out the list. Individual case safety reports and case narratives are back-translated as a verification step before regulatory submission in some contexts. The requirement varies by regulator and case type.
When back-translation is unhelpful
Outside those contexts, back-translation is the wrong tool for the job. Marketing content needs cultural and brand-voice review; verifying that the literal meaning made it across does not tell you whether the German version reads like German marketing. Internal SOPs need terminology consistency. Technical documentation needs subject-matter review by someone who would notice if a procedure had been described in a way that would not work in practice.
For these content types, in-country review by a target-market reader does what back-translation cannot. It catches what readers will and will not understand, and whether the language fits the market. Specifying back-translation as a default across a translation programme inflates cost without improving outcomes. Specifying it by document type, with a written rationale for each inclusion and exclusion, is the more useful position.
The methodology and the common pitfalls
Back-translation runs in three core steps. Forward translation produces the target text from the source. An independent back-translator, working without access to the original, translates the target back into the source language. A reconciliation review compares the back-translation to the source, flags meaning discrepancies, and resolves each one through revisions to the forward translation. The resolution is documented; that documentation is the audit trail.
Three failure modes show up regularly enough to call out during vendor evaluation. The same linguist producing both forward and back translations, which defeats the verification purpose. Producing the back-translation but skipping reconciliation, which turns the exercise into paperwork. Running reconciliation but failing to document discrepancies and resolutions, which means the audit trail does not exist when an inspector asks for it. Each one is a question worth asking: who does the back-translation, how is reconciliation handled, what does the documentation look like.
In-country review and cognitive debriefing
In-country review is the validation step where a native-language reviewer in the target market reads the translation for clarity, cultural fit, and regulatory acceptability. Cognitive debriefing extends in-country review by testing comprehension with patient-representative readers. It applies to patient-facing instruments where reader understanding determines whether the instrument works as designed. Both sit outside ISO 17100. The standard covers the linguistic process; in-country review and cognitive debriefing cover what happens when the translated text meets its actual reader in its actual market.
Who runs in-country review
An in-country reviewer is a native speaker of the target language working in or familiar with the target market. They are independent of the translator and the reviser. Their job is to catch what the ISO 17100 process is not designed to catch: whether the translation reads naturally to someone in this market, whether it carries the right register, whether it matches how regulators in this market expect this kind of content to read.
In-country reviewers are often client-side employees: a regional regulatory affairs lead, an in-market medical adviser, a country manager with content review in their remit. The reviewer with the most reliable judgment for your content frequently sits inside your own organisation. A useful translation agency supports that workflow rather than insisting on its own reviewer pool. AdHoc’s SmartEdit lets in-country reviewers edit content directly in the final layout in a browser, with no InDesign licence required and no PDF mark-up cycle. Edits flow back into the translation memory, so the next project benefits from this one.
When cognitive debriefing applies
Cognitive debriefing is comprehension testing with patient-representative readers. It applies to instruments where reader understanding determines regulatory acceptance or patient outcomes: informed consent forms in clinical trials, patient information leaflets for medicinal products, patient-reported outcome instruments used in registration trials.
The ISPOR methodology pulls a small sample of in-target readers (5 to 8 per language, with WHO guidance recommending a minimum of 10) and asks them to read the translation, then asks structured questions that surface points where comprehension breaks down. The output feeds back into a revised translation. You will see the process called cognitive debriefing in some contexts, linguistic validation in others, patient acceptability testing in a third; the underlying logic stays the same.
Not every translation agency offers cognitive debriefing directly. Many partner with specialist research vendors who handle participant recruiting and structured interviewing. If your translation requires it, ask explicitly: who runs the debriefing, what protocol do they follow, how does the output integrate back into the translation record.
How to combine ISO 17100, back-translation, and in-country review by document type
The right QA combination depends on what you are translating and who is going to read it. A vendor applying the same stack to every job is not matching method to risk. The matrix below covers the document types that account for most regulated medical translation work, with the QA layers each typically needs.
Treat the matrix as a starting point rather than a finished decision. Specific regulators, protocols, and IRBs can add requirements that override the general pattern. Use it to frame your RFP conversation and to spot vendors who default to the same QA stack regardless of document. Confirm requirements with the specific regulator before locking scope.
Document type | ISO 17100 | MTPE under ISO 18587 | Back-translation | In-country review | Cognitive debriefing |
|---|---|---|---|---|---|
| Clinical trial protocol | Required | Not appropriate | Optional | Recommended | Not applicable |
| Informed consent form (ICF) | Required | Not appropriate | Often required by IRB | Required | Recommended |
| Patient information leaflet (PIL) | Required | Not appropriate | Not typically | Required | Recommended |
| Investigator’s brochure | Required | Not appropriate | Optional | Optional | Not applicable |
| Medical device IFU (high-risk, patient-facing) | Required | Not appropriate | Optional | Required | Recommended |
| Medical device IFU (HCP-facing) | Required | Not appropriate | Not typically | Recommended | Not applicable |
| COA and PRO instruments | Required | Not appropriate | Standard ISPOR step | Required | Required |
| Pharmacovigilance reports (ICSRs) | Required | Not appropriate | Context-dependent | Optional | Not applicable |
| Medical marketing and HCP content | Required | Possible in some contexts | Not typically | Recommended | Not applicable |
| Regulatory submission documentation | Required | Not appropriate | Context-dependent | Recommended | Not applicable |
| Internal SOPs and technical documentation | Required (MTPE acceptable) | Acceptable | Not typically | Recommended for cross-region use | Not applicable |
Two patterns are worth pulling out. ISO 17100 applies almost universally; the agency’s process sits underneath everything else, and what changes from document to document is what you add on top. And the most demanding rows (ICF, COA/PRO instruments, high-risk patient-facing IFU) take multiple QA layers because the consequence of error lands on the patient. If a vendor proposes the same QA stack for a clinical trial protocol and a PRO instrument used in the same trial, treat it as a flag. The protocol needs solid ISO 17100 with optional back-translation. The PRO instrument needs the full stack including cognitive debriefing. A supplier with real medical experience can articulate the distinction without prompting. The HCP content and medical marketing row at the bottom of the matrix sits under a separate regulatory regime – the EFPIA, ABPI, and MedTech Europe codes rather than MDR or marketing authorisation rules – and is covered in HCP content translation.
How EU MDR, IVDR, FDA, and EMA QRD requirements shape translation QA
Your QA stack is partly a regulatory decision. EU MDR and IVDR require IFU and labelling translations to be controlled and traceable. EMA QRD templates constrain what you translate and how. FDA expects evidence of linguistic validation for patient-facing trial instruments. The regulator your content is heading to should be an input on your translation supplier, not an afterthought.
EU MDR and IVDR
Regulation (EU) 2017/745 (MDR) Article 10(11) and Annex I Section 23 require manufacturers to ensure that medical devices are accompanied by labelling, IFU, and certain technical documentation in the official Union language(s) determined by the Member State where the device is made available. Regulation (EU) 2017/746 (IVDR) Article 10(10) and Annex I Section 20 set equivalent requirements for in vitro diagnostic devices. The European Commission maintains a Member State language requirements table (currently Revision 3, August 2025). For the deeper operational read on hardware-device translation under MDR (IFU, label, implant card, SSCP), see MDR translation requirements. For SaMD and software-specific artefacts under MDR Rule 11 and the EU AI Act, see medical software translation.
The translation implications run two ways. Traceability: the manufacturer must demonstrate, when asked, that the translation was produced under a controlled process; ISO 17100 supplies the documented procedures. Consistency: a notified body comparing the German IFU to the French IFU expects equivalence; the translation memory and terminology management layer that sits under any serious medical translation supplier is what produces it. AdHoc maintains client-specific translation memories and terminology resources that compound across projects.
EMA QRD templates
The European Medicines Agency publishes Quality Review of Documents (QRD) templates for product information: SmPC, package leaflet, labelling. The templates are language-specific (all official EU languages plus Icelandic and Norwegian) and constrain wording for standard sections. The current centrally-authorised human-product template is version 10.4 (February 2024), with revision toward version 11 underway. Translation work on EMA submissions has to fit the template version current at submission time. For the deeper read on EMA QRD work – the 25-day linguistic review, Eudralink Day 5 submission, the centralised vs MRP/DCP language scope – see package leaflet translation.
Ask your supplier whether they work to current QRD templates, how new template versions propagate across active accounts, and whether they maintain EMA template alignment in their translation memory. Deviation from QRD wording on standard sections is a recurring cause of regulatory back-and-forth, often introduced by suppliers who do not check templates against the version current at submission time.
FDA expectations for clinical trial translation evidence
FDA expectations focus on the evidence trail for translation of patient-facing trial documents: ICFs, COA instruments, PRO instruments. For the deeper read on clinical trial translation across the EU CTR, the UK CTR 2025, and ICH E6(R3), see clinical trial translation. The agency expects documented linguistic validation for instruments used in registration trials but, importantly, does not strictly mandate a single methodology. Its guidance language is “fit-for-purpose” with evidence of conceptual equivalence. In practice, the ISPOR Good Practices guidance is the working reference most sponsors and CROs follow, supplemented by FDA’s own Patient-Focused Drug Development (PFDD) Guidance Series.
What this means in practice: your translation supplier should be able to produce a linguistic validation report on demand. The report documents who did each step, what discrepancies were found, and how they were resolved. The report is what an FDA reviewer asks for if questions arise.
A vendor evaluation checklist for medical translation buyers
The checklist below is built to lift directly into your RFP or vendor scorecard. Each item is a question, with a one-line note on what a credible answer looks like.
Certifications and standards
- Is the agency certified to ISO 17100? Good answer: Yes, with a current certificate available on request.
- Is the agency certified to ISO 18587 if MTPE is in scope? Good answer: Yes, with a written policy on when MTPE applies to medical content.
- Are sector-specific certifications held where relevant (for example, ISO 13485 for suppliers serving medical device clients under MDR)? Good answer: A clear yes or a clear no with reasoning.
Translator qualifications and pool depth
- How does the agency qualify a linguist as a medical translator beyond the three ISO 17100 pathways? Good answer: Documented criteria including subject matter experience and tested medical work.
- How many medical translators does the agency have for your language pairs? Good answer: Enough to absorb volume spikes without single-resource dependency.
- Are the same linguists assigned across projects to build account familiarity? Good answer: Yes, with documented backup plans for absence.
Process and traceability
- How is the four-eyes principle implemented? Good answer: A documented workflow with named roles (translator, reviser) and a clear distinction from proofreading.
- How are project records maintained, and how long are they retained? Good answer: A TMS or equivalent system with audit-trail visibility for the client.
- Can the agency produce a project record on demand for audit purposes? Good answer: Yes, with an example of the format available.
Regulatory expertise
- Does the agency work to current EMA QRD templates? Good answer: Yes, with a documented process for propagating template updates.
- Has the agency produced linguistic validation reports for FDA-regulated COA or PRO work? Good answer: Yes, with an example available under NDA.
- How does the agency stay current with MDR and IVDR labelling and IFU requirements? Good answer: A named process or person responsible.
Technology
- Does the agency provide a customer portal or TMS with project visibility and reporting? Good answer: Yes, with a demonstration available before contract.
- How is terminology managed across projects and languages? Good answer: Client-specific termbases and translation memories maintained centrally.
- Can in-country reviewers edit content in the final layout without InDesign? Good answer: Yes, with a browser-based review tool.
Confidentiality and data security
- What confidentiality controls are in place for unpublished clinical and regulatory content? Good answer: Documented NDAs and role-based access in the TMS.
- Where is data stored, and is it GDPR-compliant for EU operations? Good answer: EU-based storage or documented adequacy for transfers.
- Is there a data breach notification process? Good answer: Yes, with named contacts and timing commitments.
How AdHoc approaches medical translation quality
We have been translating for pharmaceutical, biotech, and MedTech clients for over thirty years. The methodology in this article maps onto how the work runs at our end. ISO 17100 and ISO 18587 are the floor (both certified, both auditable). Back-translation, in-country review, and cognitive debriefing get added by document type and regulator. The combination of people, process, and technology is what we lean on; the documented controls are what we can show.
5,500+ specialist linguists working into their native language, with medical translators assigned to client accounts and kept consistent across projects. Dedicated project managers who learn each client’s regulatory markets and document workflows. Centrally maintained terminology resources accessible to in-country reviewers. Offices across Northern and Western Europe, India, and the US.
If you would like to talk through how your medical translation programme is currently set up, or how a switch in supplier would look in practice, speak with our localisation team.
Frequently asked questions about medical translation quality assurance
What is the difference between ISO 17100 and ISO 18587?
ISO 17100:2015 covers human translation workflows: qualified translators, an independent reviser, documented project management, traceable procedures. ISO 18587:2017 covers full human post-editing of machine translation output: post-editor qualifications, the post-editing process, and conditions the output must meet. A translation agency working with both human translation and MTPE typically holds both certifications. ISO 17100 applies wherever a human translator produces the target text; ISO 18587 applies wherever MTPE is part of the workflow.
Is back-translation always required for clinical trial documents?
No. Back-translation is the standard methodological step in the ISPOR-endorsed linguistic validation sequence the FDA expects for patient-reported outcome (PRO) and clinical outcome assessment (COA) instruments. Many institutional review boards require it for informed consent forms, though the requirement varies by IRB. It is not typically required for clinical trial protocols, investigator’s brochures, or pharmacovigilance reports outside specific contexts. The question for any trial bundle is which documents need it. Confirm requirements with the specific regulator or IRB before scoping the work.
When can machine translation be used for medical content?
Machine translation with full post-editing under ISO 18587 can be appropriate for internal SOPs, technical knowledge bases, and reference content where a qualified post-editor reviews the output against the source. It is not appropriate for regulatory submissions, patient-facing leaflets, informed consent forms, or content where patient safety or regulatory acceptance is at stake. The post-editor must hold ISO 17100-level qualifications plus specific post-editing training. Ask to see the ISO 18587 certificate and the written policy on when MTPE applies.
How much does back-translation add to a translation budget?
Back-translation roughly doubles the linguist cost for the affected documents, because the volume of translation work doubles (you are paying for an additional independent translator to produce the back-translation, plus reconciliation work). It also adds time to the workflow, varying with document length, language pair, and the complexity of the reconciliation. Ask your supplier for an estimate against your specific document. The cost is one reason to apply back-translation selectively, against the document types that need it, rather than across an entire translation programme.
Who qualifies as an in-country reviewer?
An in-country reviewer is a native speaker of the target language working in or familiar with the target market. They are independent of the translator and the reviser. For patient-facing content the reviewer should have medical or regulatory background. For HCP-facing content, clinical familiarity matters. Many in-country reviewers are client-side employees: regional regulatory affairs leads, in-market medical advisers, country managers with content review in their remit. A useful translation supplier supports that workflow rather than insisting on its own reviewer pool.
Sources
- ISO 17100:2015 (Translation services. Requirements for translation services). International Organization for Standardization. iso.org.
- ISO 18587:2017 (Translation services. Post-editing of machine translation output. Requirements). International Organization for Standardization. iso.org.
- Regulation (EU) 2017/745 on medical devices (MDR), Article 10(11) and Annex I Section 23. EUR-Lex.
- Regulation (EU) 2017/746 on in vitro diagnostic medical devices (IVDR), Article 10(10) and Annex I Section 20. EUR-Lex.
- European Commission, MDR Language requirements for manufacturers, Revision 3, August 2025.
- European Medicines Agency, QRD product information template version 10.4 (February 2024). ema.europa.eu.
- Wild D, Grove A, Martin M, et al. Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures: report of the ISPOR Task Force for Translation and Cultural Adaptation. Value in Health. 2005;8(2):94-104.
- Matheson J, Kollmer M. Quality assurance in medical translation. Medical Writing (European Medical Writers Association). 2024;33(1):34-36. DOI: 10.56012/qjes2321.
- FDA Patient-Focused Drug Development (PFDD) Guidance Series. U.S. Food and Drug Administration.