Key takeaways
- Linguistic validation confirms the conceptual equivalence and content validity of a translated patient-reported outcome measure, usually by testing it with patients through cognitive interviews.
- The ISPOR Principles of Good Practice set out ten steps, from preparation and two independent forward translations to cognitive debriefing, proofreading and a final report.
- ISPOR recommends cognitive debriefing with a group of 5 to 8 respondents in the target country; the PRO Consortium sets a minimum of 5 participants per language and country.
- FDA guidance on PRO measures asks sponsors to show that measurement properties are adequately similar across all language versions used in a trial.
- Permission from the instrument's copyright holder is normally needed before any translation work starts.
This article is for clinical operations teams, outcomes researchers and sponsors who use questionnaires completed by patients in international studies. It explains what linguistic validation is, the recognised steps, how many patients are involved, and what regulators look for.
A patient-reported outcome (PRO) is a measurement that comes directly from the patient, such as a pain score, a symptom diary or a quality-of-life questionnaire. PROs belong to the wider family of clinical outcome assessments (COAs), which also includes clinician-reported and observer-reported measures.
What is linguistic validation?
Linguistic validation is a structured translation and testing process that shows a translated questionnaire means the same thing to patients as the original. The PRO Consortium, in its published consensus on its translation process, defines it as follows:
"The process of assessing and confirming the conceptual equivalence and content validity of translations of patient-reported outcome (PRO) measures. Usually, linguistic validation refers to a process whereby translated text is actively tested with patients in the target population and target language group through cognitive interviews." (Eremenco et al., PRO Consortium translation process, Journal of Patient-Reported Outcomes)
Conceptual equivalence means the item asks about the same idea, even if the words differ. A question about feeling "blue" cannot be translated literally into most languages; the translation has to capture low mood in an idiom that patients in that country actually use.
Why do regulators expect validated translations?
Regulators expect validated translations because pooled trial data are only meaningful if every language version measures the same thing. The US Food and Drug Administration's guidance on patient-reported outcome measures states:
"Regardless of whether the instrument was developed concurrently in multiple cultures or languages or whether a fully developed instrument was adapted or translated to new cultures or languages, we recommend that sponsors provide evidence that the content validity and other measurement properties are adequately similar between all versions used in the clinical trial." The same guidance adds that FDA "will review the process used to translate and culturally adapt the instrument".
Its appendix lists what a submission should contain under "Language Translation and Cultural Adaptation", including the process used, a description of patient testing, copies of translated versions and evidence that measurement properties are comparable. A straightforward translation without this documentation leaves a gap in the dossier.
What are the steps of linguistic validation?
The most widely cited method is the ten-step process in the ISPOR Task Force report, Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes (PRO) Measures (Wild et al., Value in Health, 2005). The full report lists these steps:
- Preparation: obtaining permission, understanding the concepts behind each item and recruiting in-country experts.
- Forward translation: at least two independent translations into the target language.
- Reconciliation: merging the forward translations into a single version.
- Back translation: translating the reconciled version back into the source language.
- Back translation review: comparing the back translation with the original and resolving discrepancies.
- Harmonisation: comparing language versions with each other to keep the approach consistent.
- Cognitive debriefing: testing the translation with patients from the target population.
- Review of cognitive debriefing results and finalisation.
- Proofreading.
- Final report documenting every decision.
The ISPOR report asks forward translators to produce "colloquial translations that will be easily understood by the general lay population". That instruction rules out technical or literary wording, even when it is accurate.
How many patients are needed for cognitive debriefing?
Cognitive debriefing usually involves a small group of patients per language. The ISPOR report recommends testing on "a group of 5 to 8 respondents in the target country", who should be native speakers representative of the target population by sex, age, education and diagnosis.
The PRO Consortium's process sets a minimum of 5 participants per language and country, matched to the target population "for as many criteria as reasonably practical", with interviews in person where possible. In each interview, the patient completes the questionnaire and then explains in their own words what each item means to them, which reveals misunderstandings that translators cannot see on paper.
Universal or country-specific translations?
A sponsor must decide whether one translation will serve every country that shares a language, or whether each country gets its own version. The PRO Consortium describes both approaches.
| Approach | How it works | When it suits |
|---|---|---|
| Universal translation | One version is developed for all countries speaking the same language, focusing on shared vocabulary | Trials needing pooled data and a single version per language |
| Country-specific translation | Separate versions, for example for Spanish in Spain, Mexico, Chile and the US | Where vocabulary for symptoms or daily activities differs between countries |
| "Mother" version with adaptations | One country's version goes through the full process, then is adapted for others, for example German for Austria and Switzerland | Multi-country studies wanting local fit without repeating every step |
Who should carry out linguistic validation?
Linguistic validation needs a team, not a single translator. The ISPOR report describes a project manager, a key in-country consultant from a "medical/health/psychology/social science background", professional forward translators who are native speakers of the target language, and back translators who are native speakers of the source language.
Back translators "should have no prior knowledge of the measure" and should not see the source text, so that their version is an honest test. Our article on back translation explains this step in more detail.
Common mistakes in linguistic validation
- Skipping permission. ISPOR's first critical component is to obtain permission from the developer, who may also require a particular method or review.
- Using one translator for everything. Independent forward and back translations are the core safeguard.
- Testing with the wrong people. Healthy volunteers or bilingual staff do not reveal how patients with the condition read the items.
- Recording decisions poorly. Without an item-by-item report, the sponsor cannot show regulators how equivalence was reached.
- Treating electronic versions as identical. Moving a questionnaire to a tablet or app can change layout and response scales, so check the screens too.
Linguistic validation is one part of clinical trial translation. See our guide to clinical trial document translation for consent forms and labels, and our medical and pharmaceutical translation guide for the wider picture.
How TranslationWorks can help with PRO translation
TranslationWorks is a UK-based translation agency that provides human translation of clinical and patient-facing documents by native-speaking professional linguists in more than 150 languages. For questionnaires that need forward translation, back translation or tailored review steps, our custom language services can be scoped around the method your sponsor or instrument developer requires.
TranslationWorks holds ISO 17100, ISO 9001 and ISO 27001 registration and Cyber Essentials certification. Tell us the instrument, languages and required method, and we will prepare a free quote within 24 hours, confirming which steps we can deliver.
