by Luli Zou
figures by Olivia Foster
Who benefits from medical research? We would like to think that everyone, regardless of background, has access to new discoveries and therapies. In reality, much of our knowledge about disease and treatment originates from research that does not reflect the diversity of all individuals with the disease. This imbalance stems from a variety of factors, including the same systemic prejudices and biases that have historically pervaded the U.S. and continue to affect broader questions of inequality to this day.
Underrepresentation and Bias in Scientific Research
Ever since the sequencing of the human genome in 2001, researchers have identified thousands of genetic variants, or places in the genome that vary from person to person, that are associated with increased disease risk. Genetic testing to identify what variants an individual possesses is becoming increasingly common, but the results must be interpreted with caution.
For example, in 2016, researchers showed that genetic variants that increase the risk of cardiomyopathy, a type of heart disease, in white individuals are in fact benign for black individuals. They found multiple records of black individuals who were misdiagnosed between 2005 and 2007 – nearly ten years prior to the study – as a result of this oversight. Furthermore, the researchers showed that if the original studies conducted in white populations had included even small numbers of black individuals, the misdiagnoses could have been prevented.
Although this is just one example, it exemplifies a much broader truth, namely that homogeneous studies have sometimes led to results that are not generalizable across populations. Unfortunately, this bias disproportionately affects minority individuals – specifically, people of nonwhite, non-European ancestry.
Many common chronic diseases in the U.S. such as asthma, diabetes, and heart disease affect minority populations at a higher rate than white individuals. Yet, only 19% of individuals studied in major genome studies are of non-European ancestry, and an even smaller percentage are of non-European, non-Asian ancestry (Figure 1).
The cardiomyopathy example is not the only result that does not generalize across ancestry groups. For example, one study found that many genetic variants associated with type 2 diabetes risk in European ancestry populations tend to be less associated with risk in African ancestry populations. Another study found that some genetic variants associated with chronic kidney disease risk in European ancestry populations did not replicate in American Indian populations.
Trust Violations and Other Barriers to Participation
Why haven’t non-European ancestry populations been included proportionally in scientific research? One reason is that a racially homogenous population can be desirable in scientific research, as they can make it easier to interpret a study’s findings. For example, if a study is conducted with a racially diverse group of individuals, researchers may not be certain that an increase in disease risk is due to the genetic variant they are interested in rather than, say, a different variant that happens to be more common in individuals of a certain ancestry.
Another major reason is that minority populations in the U.S. tend to avoid participating in scientific research, in part due to mistrust arising from a history of medical injustices. Infamous examples include the Tuskegee Syphilis Experiment (1932-1972), in which African-American men were given syphilis and purposefully left untreated, even after penicillin was discovered to be a cure; Henrietta Lacks, an African-American woman whose cancerous cells were taken without consent and are now commonly used in biological research; and the involuntary sterilization of over 3,000 American Indian women between 1973 and 1976 under beliefs of racial inferiority.
In addition to issues of trust, minority individuals face many other barriers to participation in medical research. One study suggests that physicians believe that African Americans are only two thirds as likely to adhere to their treatments as white patients. This bias may lead to physicians being reluctant to refer African American patients to clinical trials, because they do not believe the patients will adhere to the protocol. Minorities have also historically had poor access to medical care and are less likely to have health insurance, which is required for some clinical trials. As a result, minority individuals are more likely to have a less familiar relationship with their physicians, who in turn may be less likely to enroll them in a clinical trial.
The All of Us Research Program: Closing the diversity gap
Earlier this year, enrollment for the All of Us research program officially opened. Funded by the National Institutes of Health, All of Us seeks to address underrepresentation in biomedical research by recruiting a diverse population of at least 1 million individuals over the next five years in order to conduct statistically well-powered studies. All of Us defines diversity using many factors, including race and ethnicity, age, sex, gender identity, sexual orientation, disability status, access to care, income, educational attainment, and geographic factors. The goal is to include as many people from different walks of life as possible – especially those that have been historically underrepresented in biomedical research.
To deliver on this promise, All of Us is building a network of community organizations that will facilitate outreach, engagement, recruitment, and retention in order to overcome barriers to participation. Current partners include FiftyForward, which will conduct outreach to urban and rural, economically disadvantaged and older adult populations; the National Alliance for Hispanic Health, which will work with Hispanic communities; the Delta Research and Educational Foundation, which will engage the African-American community; and the San Francisco General Hospital Foundation, which will focus on sexual and gender minorities.
Although the priority is diversity, any adult in the U.S. who can give legal consent is eligible to participate. Over the course of the study, participants may donate blood or saliva samples, which will be used to sequence their DNA and identify genetic variants (Figure 2). Data on diverse individuals will lead to a better understanding of how the effects of genetic variants differ within minority populations and other subgroups. Individuals may also allow access to their electronic health records. These records contain participants’ medical history, diagnoses, medications, treatment plans, immunization dates, radiology images, allergies, and laboratory and test results. Researchers will be able to take advantage of this standardized resource to build models of disease occurrence. Participants will contribute their health data for at least 10 years if they remain in the study for the full duration.
With access to an unprecedented amount of information from individuals of diverse backgrounds, scientists may be able to pinpoint genetic risk factors for many diseases that are specific to diverse populations. Everyone, regardless of background, should be able to reap the benefits of scientific discovery. All of Us is a major step towards this goal.
Luli Zou is a first year Ph.D. student in the Biostatistics program at Harvard University.
Olivia Foster Rhoades is a fourth-year PhD student in the Biological and Biomedical SCiences program at Harvard & is pursuing a concentration in STS at the Harvard Kennedy School. You can find her on Twitter as @OKFoster.