As we have stated in previous articles, the purpose of the linguistic validation process is to make sure that translated COAs remain conceptually equivalent to the source text. This means the job of Linguistic Validation is to provide evidence that:

·       Patients who speak different languages interpret the items in the same way

·       Cultural differences in language and expression do not distort meaning

·       The assessment is reliable and valid across populations.

Besides the fact that, in previous articles, we have argued that there is always an element of cultural distortion – or ‘coating’ – when a word is translated from one language to another and that there is an extent to which items are interpreted differently across cultures and languages. In addition, in a strict sense, whether an assessment is reliable or valid across populations is surely beyond the scope of Linguistic Validation, at least not in the way that it is typically carried out.

Nevertheless, this is what Linguistic Validation is supposed to achieve. It might be heresy to suggest otherwise. However, if Linguistic Validation is treated like a checkbox activity, one more bureaucratic obstacle to manoeuvre around, then it cannot do anything more than make a statement along the lines of:

 “An interviewer (whose interviewing credentials are not clear) in X locale, speaking Y language interviewed 5 people that the interviewer claims are native speakers, have the requisite illness (does anyone even check this?) who say they understand each item.”

It can only ‘validate’ to a relatively limited extent that patients in the target setting interpret items in the same way. It sheds only some light on distortion of meaning or ‘cultural coating’ of items. It cannot make statements about the reliability and the validity of the assessment across diverse cultural and linguistic populations.

What is the purpose of Linguistic Validation? Is it, as its name suggests, supposed to ‘validate’, like a rubber stamp, that a translation is sound in the target language? Or is it an opportunity for research into the Patient Voice, in linguistic and cultural settings that may never have been considered during the development phase?

We are firmly in the latter camp: Linguistic Validation is a research activity.  We believe that it should be more than a technical translation exercise. It is an opportunity to engage patients and explore how they truly experience and express their health concerns. When cognitive debriefing is supplemented by research methods from other disciplines, this opens up even more possibilities for elevating the Patient Voice.

Signs that it is just a Checkbox Activity

You can tell when cognitive debriefing has become a regulatory formality rather than meaningful research when:

·       Interviewers ask closed questions that may only elicit yes/no responses, without deeper questions or probing

·       Participants are not encouraged to discuss their own experiences

·       The focus is on whether an item is ‘correct’ in the sense of being understood and not whether it makes sense to patients in their own lives. Something being understood and something having meaning to people in their real lives are two very different things.

·       Sponsors are looking for no comments being raised and not on seeing where adjustments or refinements can be made to the COA based on patient feedback

·       The process is treated as a single-step validation activity, rather than an iterative research effort, although part of the problem is the late stage at which Linguistic Validation takes place. Not much can be done if serious issues are discovered. These issues are more than words being understood but how a COA is used and perceived holistically, or what kind of cognitive burden the COA places on someone from another culture due to local conventions of how information should be presented.

In contrast, a genuine research approach would aim to explore:

·       Whether the COA reflects how patients in the target locale describe their symptoms

·       How the wording of the items in the target translation aligns with patient experience, and not just focus on translation accuracy

·       Does the target language reflect the patient’s journey in the target linguistic and cultural setting?

How can Cognitive Debriefing be transformed into Meaningful Research?

The focus can be shifted from language-based to experience. In addition to simply confirming linguistic accuracy or fidelity to the source, interviewers could make more effort into finding out:

·       How a symptom can be described in the participants’ own words and what it means to them in their own lives

·       Whether the item reflects how they experience the illness or condition in question

·       If they think anything is missing from the questionnaire.

This way, cognitive debriefing, with very simple supplementary methods, fulfils the standard linguistic validation requirements, but also becomes a tool for gathering real-world patient insights.

Patients should also feel encouraged to expand on their responses, rather than confirming that something is clear and easy to understand. Debriefing should include follow-up questions that are focused on the ‘how’ and the ‘why’ participants interpreted an item in a certain way. In addition, asking participants for alternative wording could be an avenue of exploration of how a real person with the illness in question naturally talks about their condition.

Participants should not be passive but seen as co-researchers and partners who are helping to refine COAs.

The regulatory narrative could be reframed, where bodies such as FDA and EMA could recognise that:

·       Supplementary patient insights gathered from cognitive debriefing, beyond the responses to the standard questions, can improve COA validity

·       Project reports should highlight patient-driven suggestions for change to COA and not just confirmation that something is understood or not.

Cognitive debriefing is meant to test if COA content is understandable to patients in target locales. However, when treated as little more than a bureaucratic formality, it loses its research value.

Cognitive debriefing, as the most intensive part of the Linguistic Validation process, could do more than only check for linguistic equivalence. It could:

·       Explore how patients describe their symptoms and their experiences of living with a specific illness

·       Supplement the standard cognitive debriefing process with methods of investigation from related disciplines, as well as open-ended questions that allow for further probing. Cognitive debriefing should not be a script, but more of a guide

·       Go beyond translation equivalence to focus on relevance to what the item means to the patient’s own experience.

·       Ideally, cognitive debriefing should be done in the spirit of iterative design, where feedback leads to refinements in the target version COA, rather than problems being minimised or ‘disappeared’ altogether. This happens.

To conclude, Linguistic Validation should not be just about checking for equivalence. It should be about ensuring that the patient experience is captured and amplified as part of wider patient voice research. The breadth of diversity, cultural and geographical expanse afforded by the nature of multi-country, multilingual Linguistic Validation projects is too valuable an opportunity for it not to be harnessed.

Thank you for reading.

Mark Gibson

Leeds, United Kingdom, March 2025