Race & Ethnicity-based Epidemiology

Definitions

Race and ethnicity are often used interchangeably in the literature, often incorrectly. It is important to acknowledge that often the information that is presented to learners may be limited by the language and design of the literature on which they are based. It can be powerful to acknowledge this limitation with learners to help contextualize the landscape of research related to disease risk that attempts to dissect biosocial elements. It could also be beneficial to discuss these limitations by breaking down the difference between race, ethnicity, and ancestry and how they show up in our understanding of disease.

Race: The social construction and categorization of people based on perceived shared physical traits (APA Dictionary of Psychology).

Ethnicity: Characterization of people based on having a shared culture (e.g., language, food, music, dress, values, and beliefs) related to common ancestry and shared history (APA Dictionary of Psychology).

Ancestry: Characterization of people based on geography, genealogy, or genetics and subdivided as such. As the name suggests, geographic ancestry refers to the geographic region from which a person originates, and genealogic and genetic ancestries refer to ancestral pedigree and genetic character (Lu et al. 2022).

KEY POINT 1

Ancestry and genetic admixture (referring to genetic exchange between people from different ancestry that contributes to disease risk) are more useful in characterizing and systematizing disease risks than social constructs like race and ethnicity.

BUT (Swilley-Martinez et al., 2023):

Race and ethnicity are social determinants of health; the discussion should help contextualize systems barriers to quality care that disproportionately affect certain racial and ethnic groups.
Ignoring race and ethnicity altogether erases the disparities that are present in our system.
- Many studies in epidemiology attempt to remove race as a confounding factor but this erases subgroup differences in risk and response to exposure.

KEY POINT 2

At times, lecture content may be constrained/limited by the quality of data that is available in the literature. It is important to include this in a brief preamble prior to discussing this data in order to provide this crucial context. Guidance on how to give this preamble is provided below.

For example, many studies collapse racial categories into White, Black, Hispanic, and Other, which is problematic because the use of these categories severely limits the resolution of claims that can be made related to how different racial groups experience the healthcare system (Swilley-Martinez et al., 2023; Flanagin et al., 2021). Since this has been the prevailing format for how epidemiology is reported in some of the literature, there may not be sufficient data to break down the relative risk of subgroups, nor to provide context for why these disparities in risk exist.

General Approach to Common Feedback

"Population X is not a monolith."

The primary ask with this feedback is to be as specific as possible with the identity groups that you are referencing. There is a tendency to categorize and generalize large swaths of people in systems that aim to understand the interaction between these groups and disease risk or system barriers. This simplifies communities that often have as much diversity within groups as without them. Although systematization is necessary, especially in epidemiology, being as specific as possible means that the information is most accurate and useful.

Exercise

The following questions can help determine an approach to this feedback:

What is the goal of reporting this risk in this population? What is the actionable takeaway from the information?
What are the factors that may contribute to the disparity in risk with this particular group?
Is the grouping/categorization of the population appropriate? Would there be additional information revealed by being more granular?
Are there limitations to the data presented?

If there is time and you feel that the information is relevant, you can explore any one of these questions with learners in more depth. However, in most cases, reflecting on these questions and making brief references to what you have found can provide crucial context.

**If the data you are presenting with is limited in its specificity in the literature, or if the specific groupings are intentional, please disclose that upfront using the preamble template at the bottom of this page**

Evidence? Lack of contextualization?

Fundamentally, learners are just establishing their skills of critical appraisal, and a question about the strength of the evidence behind a statement is natural. However, claims about social groups or identities are more likely to be flagged by learners. This is because:

There are several cases where medicine has got these claims wrong and thus requires a more diligent appraisal to determine the validity of the claim being made (examples below).
Learners are increasingly more aware of these cases and may look to further validate claims about social groups and identities.

Important: Learners are vulnerable to amplifying implicit medical bias due to their relative lack of reference points early in their career. This means that claims made about a population (i.e. population X is more likely to get disease A) may bias their ability to build a wide differential that takes into account absolutes rather than risks. This is why providing additional context and evidence can contextualize the information appropriately for learners.

Additional Rationale & Theory

The key fact here is the learners are incredibly vulnerable to heuristics, particularly availability and representative heuristics. Ironically, this is a generalization about learners but one that is practically useful. This means that presenting information that makes a claim about a group of people can have an outsized effect on their ability to translate that knowledge to judge the relative risk of an individual from that group later in their career. This vulnerability puts an enormous responsibility on our teachers to be sure that factual claims about specific population groups are evidence-based and up-to-date, to minimize the risk of perpetuating stereotypes and outdated information, which can be amplified at this stage of learning medicine. Providing specific evidence that backs up these claims, even if as a supplementary resource, can both mediate this risk and provide a better learning experience.

Acknowledging the Social Determinants of Health

This is essentially an extension of the point made above about providing the appropriate context for a claim. If the claim is that population X is more susceptible to a certain disease, this is historically where the information ends. However, this relative risk may be due to ancestry (i.e. genetics) which confer biological differences to give rise to this risk, or it may be due to social determinants that are largely shaped by their environment and the systems in place. Providing context as to why certain disparities exist is more informative and accurate.

Guidance on Language/Terms

KEY POINT 3

Accuracy requires specificity (refer to APA Language Guide)
*If referring to research that uses Statistics Canada's population groupings, refer here for more granular definitions.

African American/Black

African American refers to the ethnicity; Black is a racial group/category.
African American should not be used universally, as to avoid obscuring more specific references to ethnicity (i.e. Nigerian, Kenyan, Jamaican, Bahamian).

Arab, Middle Eastern, and North African (Arab/MENA)

Arab is an ethnicity and ancestry; therefore. some people of MENA descent may refer to themselves as Arab Americans/Arab Canadians but this is not universally true for the region.

Asian/Asian American (Canadian)

Asian refers to people of Asian ancestry from Asia; Asian Canadian refers to people of Asian ancestry from Canada. They are not synonymous.
Refer to specific regions where possible (i.e. South Asian, East Asian, and Southeast Asian).

BIPOC (Black, Indigenous, and People of Colour)

Use more specific terms when referring to racial or ethnic groups where possible.
If specificity is not possible, use "people/person of colour" or "communities of colour" as alternatives.
For a more information on the background of the term BIPOC, refer to APA Language Guide.

Indigenous

Indigenous is a recommended umbrella term to refer to the first peoples of Canada.
- You may also come across the terms First Nations, Métis, and Inuit (FNMI) populations; these are also acceptable.
Avoid use of Aboriginal or Indian unless referring to legislation that continues to use this outdated terminology.

Sample Preamble

Statement: The South Asian population is at higher risk of cardiovascular disease.

Potential Learner Feedback: The South Asian population is not a monolith; what specific subpopulations within the South Asian community are you referring to, and what are the potential reasons for their higher risk?

A preamble could have addressed feedback concerns and prevented the feedback in the first place.

Preamble: Epidemiological data provided today is (1) limited by the population groups chosen by the literature, and (2) limited in its resolution of smaller subgroups, as it has to balance clinical significance with data specificity to allow us to make actionable assertions about population risk. It may be warranted to advocate for more granular population subgrouping in research in the future. Appreciate that these are assertions about population risk, and learners should be aware of letting relative risk skew their broad differentials when met with an individual.

A preamble can set the foundation for the information to be shared. These can be followed up by additional context that can enriching.

Additional context: Studies that do look at more granular subgroups of South Asian populations find that their subgroups individually and independently have higher cardiovascular risk compared to other populations in Europe, Asia, and the Americas. Therefore, regardless of how granular we get, the statement about the higher risk of CVD in the South Asian population holds true.

Want to learn more about anti-racism in medicine? Check out the online Faculty Development module called "Anti-Black Racism in Medicine" by logging into your Instructor Homepage with your CCID.

Additional Resources on Systemic Racism in Research & Practice

References

Flanagin, A., Frey, T., Christiansen, S. L. & Committee, A. M. of S. (2021). Updated Guidance on the Reporting of Race and Ethnicity in Medical and Science Journals. JAMA, 326(7), 621–627. https://doi.org/10.1001/jama.2021.13304
Lu C, Ahmed R, Lamri A, Anand SS. Use of race, ethnicity, and ancestry data in health research. PLOS Glob Public Health. 2022 Sep 15;2(9):e0001060. doi: 10.1371/journal.pgph.0001060. PMID: 36962630; PMCID: PMC10022242.
Swilley-Martinez, M. E., Coles, S. A., Miller, V. E., Alam, I. Z., Fitch, K. V., Cruz, T. H., Hohl, B., Murray, R. & Ranapurwala, S. I. (2023). “We adjusted for race”: now what? A systematic review of utilization and reporting of race in American Journal of Epidemiology and Epidemiology, 2020–2021. Epidemiologic Reviews, 45(1), 15–31. https://doi.org/10.1093/epirev/mxad010
Samie, A. (2023). intersectionality. Encyclopedia Britannica. https://www.britannica.com/topic/intersectionality

Page updated

Report abuse