In many cultures, certain terms relating to health can carry stigma or social taboos. These require careful handling when developing health information. However, with the one-size-fits-all practice around the cultural adaptation of Clinical Outcome Assessments, items are developed with little thought about cultural sensitivities. Item development of COA needs to be informed by local knowledge around stigma and taboo. This should be built into the process from the start. Instead, Western concepts that do not have perceived sensitivities attached to them in the source culture are then forced through in translation. These, then, pick up shades of culturally-bound stigma in the translation process.

This is despite activities such as Linguistic Validation. By the time this happens, sometimes years after the development process has ended, it is often too late. The outcome is that the target version, if done well, is indeed the lexical equivalent of the source version. However, in the translation process, concepts like ‘cancer’ picks up additional ‘cultural coating’. It is perceived with different hues of perception. This influences understanding. This is rarely considered during the planning and development of a COA. Similarly, it is often not adequately detected or addressed during linguistic validation.

When COAs include direct translations of culturally sensitive terms, clinical study participants may misunderstand the question, refuse to answer or provide an altered response – providing false information that minimises their risk of heightened stigma. This leads to incomplete data and underreporting of critical health conditions. In turn, it can lead to skewed data and compromised clinical research validity. Taboo topics often involve mental illness, sexual and reproductive health, cancer and terminal illness.

Mental Illness

Mental health conditions are highly stigmatised in many societies, often associated with danger to other people. This can lead to reluctance to disclose symptoms. For example, the Chinese word for mental illness – jingshénbìng – is strongly associated with psychosis and insanity. This ‘cultural coating’ – the added meaning - can discourage people from acknowledging symptoms of depression or anxiety.

Cultures soften the taboo through euphemism. For example, in Japanese, Russian and Arabic, ‘mental illness’ may be expressed as ‘illness of the heart or mind’, ‘nervous disorder’, ‘psychological disorder’, ‘low energy’ to neutralise the association with institutionalisation or being ‘crazy’ - a negative term in any language.

For COA development, care needs to be taken to use neutral wording that gets rid of any stigma. When the COA reaches the Linguistic Validation process, document throughput can feel like an assembly line. When dealing with these topics, perhaps the conveyor belt can be slowed down a little and extra care given to topics that are potentially taboo. A thorough Translatability Assessment should signpost this prior to the translation stage.

In addition, COAs could use functional descriptions instead of diagnostic labels. So, instead of a question like:

“Do you have depression?”

Developers could consider the function description, such as:

“In the past two weeks, have you felt sadness that does not go away or loss of interests in activities that you usually do?”

Of course, at the iterative development stage – not solely the Linguistic Validation phase – pilot testing needs to be done to identify items that may discourage disclosure. These issues ideally should be identified at the item planning stage.

Sexual and Reproductive Health

Stigma around sexual and reproductive health exists globally, from mild embarrassment to fear of severe social sanction. This applies to sexually transmitted infections, infertility and erectile dysfunction. Many languages carry additional cultural baggage in their translated equivalents. East Asian languages often attach moral judgement to STIs. In Hindi, STIs are called the ‘hidden disease’, which says everything about local cultural attitudes. Questions around this topic can make patients feel embarrassed, judged and fearful of providing truthful answers if they suspect there could be a breach of confidentiality.

In COA development, there needs to be an effort from the beginning to tackle these issues head-on. For example, the developer could use indirect phrasing, focusing on the symptom, such as ‘Have you experienced symptoms such as unusual discharge?’, rather than referring explicitly to the name of the illness, which has high potential of carrying a culturally loaded meaning. Be mindful of using moralising language in translation. Give patients assurances about confidentiality, by making this explicit, e.g. at the beginning of the COA have a space where participant and clinical investigator can co-sign a confidentiality statement.

Cancer and Terminal Illness

In many cultures, the mere mention of cancer evokes dread and fear. There are also cultural attitudes towards terminal illness due to beliefs about fate and spiritual consequences. In addition, some cultures have a practice of withholding bad news diagnoses from people: in many societies a person with cancer may not even know it, but their diagnosing doctor does and, often, relatives. Instead, euphemisms are often used, with ‘cancer’ being described as “the bad disease” (Chinese), “the serious disease” (Japanese), “growth” (in many languages), “the long illness” (Arabic). Therefore, the naked use of ‘cancer’ in translation may cause discomfort, for example, if encountering it in a COA.

COA developers should consider neutral language. Instead of ‘cancer’, they could use:

“How long have you been diagnosed with your long-term illness?”

This allows respondents to acknowledge their condition without distress. Open-ended responses could also let patients describe their illness in their own words.

Final Thoughts

When COAs overlook cultural taboos, participants may provide altered responses, skip questions, or withdraw altogether. This affects data accuracy and limits the effectiveness of research.

COA developers must identify culturally sensitive terms early. Words carrying stigma should be replaced with neutral descriptions. Cross-cultural considerations should be part of development, not an afterthought. Capturing these issues only during Linguistic Validation may be too late.

I often ask: Is a one-size-fits-all approach to COA appropriate for international deployment?

A Final, Final Thought: Do NOT Use Inexperienced Interviewers

During cognitive debriefing, inexperienced interviewers may reinforce stigma, creating barriers to data collection. Earlier in my career, a translator-interviewer (location undisclosed) conducting epilepsy interviews requested extra space between themselves and participants, for fear of “being attacked.” Around the same time, another interviewer refused to ask at least 15 questions on sexual activity when working with endometriosis patients.

These experiences taught me harsh lessons. A lack of sensitivity or discomfort with taboo topics can make patients feel judged and ashamed. They may clam up, provide defensive responses, or withhold information. An inexperienced interviewer can make these issues worse.

When inexperienced interviewers are who may not appropriately trained for serious health research, this is disservice to patients everywhere, even if it is just for an activity like cognitive debriefing.

Thank you for reading,

Mark Gibson

Leeds, United Kingdom, March 2025