Collecting demographic data allows us to understand our participant populations in a structured way, so we can analyze variability in the disease experience based on factors such as age, gender, and ethnicity. Analyzing differences in the pattern of ill health and disease in specified populations is the driving force for shaping public health policies2, and robust demographic data is thus essential for visibility of diversity. To improve approval timelines, access to innovative therapies, and screening and treatment guidelines, it is necessary to critically assess the equity of healthcare delivery across ethnicity, race, and other socioeconomic factors3.
What are the obstacles to collecting ethnicity data in multi-country studies?
To put it simply, it is impossible to have an international standard for collecting ethnicity or ethnic group data.
For instance, the classifications of “American Indian” and “Alaska Native” in the US, or subclassifications of “white British” and “white Irish” in the UK, will be much less relevant in most other countries4. Alternatively, a classification used in one country (e.g., “Gypsy” in the UK) may be inappropriate in another, where a different term may be used to capture people who identify in a similar way. While the collection of ethnicity data is well established in certain countries, others, such as France and Germany, actively avoid collecting data on race or ethnicity of their citizens.
Naturally, these differences in classifications between countries can complicate various stages in the development of international real-world studies, in particular:
1. Localization
Clients may choose to extend ethnicity lists to cover all bases for a particular country, which—if certain categorizations have no direct equivalent in one country compared with another—can confound the translation process. This creates the need for cultural adaptation and localization, as unique ethnicity lists may be required for each country.
2. Data capture
To facilitate the comparison of data across a patient population, questions in surveys tend to have the same answer options, regardless of the language. When it comes to ethnicity lists, investigators may end up with different datasets depending on the country.
What is the solution?
There are ways to handle issues in collecting ethnicity data, to ensure generation of real-world evidence that paints a full and meaningful picture of disease impact:
- Research and plan studies carefully to minimize issues with data capture further down the line. Ethnicity lists are likely to be study-specific, taking into account the countries involved and their associated national regulations, while capturing the main disease-relevant population groups in accordance with clinical trials or prior research.
- Involve localization experts, particularly those specialized in life science translation, who will be able to guide the localization process to meet national standards as well as study endpoints.
- Guarantee regulatory compliance by going through the appropriate ethical approval pathway and ensuring that ethnicity data can be collected in a particular country for that study.
At Vitaccess, we have experience in running real-world studies around the globe. To learn how our in-house experts can help you reach your patient-centered outcomes goals, contact us at info@vitaccess.com.
References
- Connelly R et al. Method Innov 2016;9:1–10.
- Bhopal R. J Epidemiol Community Health 2004;58(6):441–5.
- Ju-Young JS et al. International and global issues – differences in health systems, patient populations, and medical practice. In: Girman CJ and Ritchey ME eds. Pragmatic Randomized Controlled Trials. Academic Press;2021:257–72.
- GOV.UK. Data in government. 2022. Available at: https://dataingovernment.blog.gov.uk/2022/01/25/comparing-ethnicity-data-for-different-countries/. Accessed: Aug 2022.
By Fatemeh Amini and Anna Richards
