In previous articles, we looked at the merits of concept elaboration and translatability assessments. These are two critical processes to lay the groundwork for Clinical Outcome Assessments to be accurate, sensitive and cross-culturally accurate. Each plays a unique role in the development of COAs. However, the real power comes when they are integrated thoughtfully into a single, continuous process – rather than treated as isolated or sequential tasks.

This article looks at how these two activities align, why they are stronger together and where there is unexplored middle ground.

Recap:

Concept Elaboration is a process of clarifying and defining what each item in a COA really intends to measure: what is this item for? Why is it included in this assessment? The process of concept elaboration involves deconstructing each item and unpacking the meaning of every word and phrase contained in that item. It involves understanding the context of the assessment and considering how a concept is experienced, expressed and interpreted across different settings.

Translatability Assessment is a pre-translation evaluation of how well a COA can be translated, culturally adapted and understood across different languages. It identifies risks such as idioms, metaphors, cultural mismatches, untranslatability and ambiguity before translation even begins. Ideally, a translatability assessment should flag in advance the kind of ‘cultural coating’ that an item may acquire during the translation process, or indeed, ‘cultural stripping’, where an item in the source language may lose nuances in a target language.

Why these two approaches should be integrated

First and foremost, both activities occur before translation. Usually, concept elaboration happens first and then translatability. The latter could be seen as an extension of the former.

Very often, these two activities are siloed by sponsors. Sometimes one happens, but not the other. When they are siloed, it feels like only half an insight has been achieved: fifty percent of the job done. Concept elaboration on its own might define the concepts clearly in English, but whether these concepts can feasibly transfer into a new cultural setting is not checked. Translatability used on its own might flag problems with language but lacks insight into the underlying intent of them, leading to conjecture and literal substitutions.

When they work together, they are like two halves of the same orange. One focuses on making sure what the COA says is conceptually sound. The other ensures how it is said is flexible enough to carry that meaning across languages.

The advantages of integrating both:

-            It means the COA is designed for global applicability from a pre-translation state (although it should really happen much earlier in the development phase)

-            Conceptual integrity and cultural appropriateness are baked in, not patched together as an after-thought.

-            Translation becomes smoother, faster and more accurate. There is no excuse for translator error if a potential pitfall had already been highlighted in the concepts and translatability phase.

What About the Development Phase?

There is no reason to delay concept elaboration and translatability until just prior to deployment of a COA in other cultures. This means that the items in the source version are developed, agreed upon, tested, validated, submitted and licensed. Critical mistakes such as untranslatability in target languages can only be patched over and not truly addressed because the source COA cannot be changed.

Concept elaboration and translatability could be incorporated at the earliest stage of item development. They force the development team into providing answers to the questions: “What does this item mean?” and “Can this be said meaningfully in other languages?”

The Middle Ground

There is also middle ground, like the field of Rumi, where the concept elaboration also incorporates a kind of translatability activity that is agnostic. By this, I mean scanning the item for any possible translatability issue without focusing on a particular language. This would be cross-cultural horizon scanning, flagging issues that could focus the mind of the in-country translatability expert.

This ‘agnostic translatability assessment’ would be a simple extension to the concept elaboration process. In fact, there should already be elements of agnostic translatability comments in all concept elaboration activities. This is the middle ground between the two activities, the bridge that seamlessly connects concept elaboration to translatability. However, it requires significant in-house linguistic knowledge (believe it or not, many companies in this space do not possess much of this), but it can also be AI-assisted.

This approach allows the COA to protect its core constructs while allowing for linguistic and cultural variation, seeing from a distance and preparing oneself for it.

Bridging concept elaboration and translatability with a middle ground of a language scanning / agnostic translatability exercise is not just a technical choice. It is a philosophical one as well. It is about translating with diversity in mind, and it ought to be incorporated into COA development from the beginning. This is designing for a global audience and not just the immediate cultural setting of where the COA was developed. Health is universal and its experience and expression are profoundly culturally felt.

Embedding clarity of concepts and the kind of cross-cultural foresight described in this article, developers have the best chance of hearing, as well as respecting, the voices of patients around the world. This is about solid science and good care.

Thank you for reading,

Mark Gibson

Leeds, United Kingdom, April 2025