I love the spectacular symbiosis of my vegetable garden as harvest time approaches. Beanstalks spiral up cornstalks, their tendrils teasing nearby tomato…
I just finished revising the fourteenth edition of my college textbook, Human Genetics: Concepts and Applications. The first was published at the dawn of the human genome sequencing era, 1994. I’m accustomed to incorporating feedback from professors and updating content every 2 or 3 years, but this revision threw something new at me: the publisher asking all textbook authors to strive for DEI:
DIVERSITY: depicting various identities and differences
EQUITY: providing fair and equitable access and opportunity
INCLUSION: respecting and welcoming all individuals
Ironically, just as I finished the new edition, the American College of Medical Genetics and Genomics (ACMG) published a “points to consider” statement in Genetics in Medicine, “Clinical, technical, and environmental biases influencing equitable access to clinical genetics/genomics testing.”
The subtext: Textbooks shouldn’t use only or mostly photos of white people, and interpreting DNA test results shouldn’t be based on research done mostly on white people.
What I’ve Done Right
I hadn’t known DEI was a thing. So when my publisher hired a company to pepper the pages of my previous textbook edition with yellow flags trumpeting where I’d written something insensitive, I initially referred to them as “the PC police.” I thought I’d been doing a decent job.
I’d always written “people with” a condition, not diabetics or hemophiliacs, and certainly not cancer “victims.”
I used black or African-American, white or Caucasian, depending upon which name was trending when the book came up for revision (every 2-3 years).
I covered transgender identity in the “Matters of Sex” chapter long before it made headlines.
A decade ago I changed “mental retardation” to “intellectual disability” to align with updates in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM).
I’ve always covered links between ancestral groups and genetic disease risk, because that’s been a basic part of the field. And I’ve used examples from population groups that reflect the reality that we do not choose our partners at random, as theoretical calculations in population genetics assume. I’ve pointed out the limitation of basing conclusions, let alone medical advice, on data from the UK Biobank, with its majority of a half million genomes from white Europeans. That’s led to health information that might not apply to other groups, and is a situation that is, finally, changing.
What I’ve Done Wrong
The historical scars of eugenics linger in the background of many discussions of genetics and society. But the yellow flags in my textbook were more specific. They requested word-by-word changes that in some cases bordered on the absurd, but overall satisfy the DEI requirements.
Here are some examples.
I had no inkling that the word “normal” is offensive, and using it makes me “ableist.” The new edition instead has “typical” and “atypical,” never “normal” or “abnormal,” even if I’m writing about chromosomes, which don’t have feelings. “Birth defects” are now “congenital disorders.” “Homosexuality” has become “same-sex attraction,” which sounds stilted and just repeats what homosexuality means. In the “people with fill-in-the-blank” paradigm, someone “has infertility.”
The yellow flaggers didn’t notice cis and trans describing relative positions of gene variants on their chromosomes – cis if on the same one, trans if on either of the two chromosomes of a pair. Chemistry uses cis and trans too. These are standard scientific terms that predate application to sexual attraction among humans by decades. So the flaggers didn’t have much in the way of a science background.
But some gaffs are cringeworthy to me now, like the noun that I used for Sally Hemings and the people of African ancestry who traveled during the Great Migration from the rural south from 1910 to 1970. After reading the 1619 Project, I’ll always use “enslaved.”
The yellow DEI flags also accused me of “othering.” I hadn’t known that was a thing either.
Consider Menkes disease. My textbook refers to it as “kinky hair disease,” because, well, that’s what it’s been called, for decades. Apparently hair texture, which physicians note in diagnosing this and a few other rare conditions, is a slur. Another yellow flag helpfully clarified my insult where I’d written kinky hair again, in an end-of-chapter question:
“Tribbles are extraterrestrial mammals that long ago invaded a starship on the television program Star Trek. A gene called frizzled causes kinky hair in female tribbles who inherit just one allele. However, two mutant alleles must be inherited for a male tribble to have kinky hair.”
The yellow tag reads: “The use of kinky can be considered othering since it is primarily associated with Black hair. In this context, it is literally being used as a trait example for aliens.”
I was also chastised for an end-of-chapter question based on H.G. Wells’ The Time Machine:
“In the distant future on Earth, humanity has diverged. The Morlocks live underground and have dark skin, dark hair, and are aggressive. The above-ground Eloi are blond, fair-skinned, and meek.”
Reads the yellow tag, “Though this is in reference to a story, this example perpetuates the common literary trope of portraying darker skinned people as subhuman.”
The flaggers must not have seen the film or read the book of The Time Machine. The Morlocks were smarter than the Eloi, whom they ate.
I’ve abandoned classic terms from genetics now deemed upsetting, once they were pointed out to me.
Members of small indigenous groups may be “vulnerable,” not “vanishing.” Uncommon conditions are “rare,” not “orphan.” The term comes from the Orphan Drug Act of 1983, which encouraged drug development for rare conditions.
African-American, aboriginal American, Eskimo, and Native American have been replaced. “Tribe” is taboo, First People or First Nation okay, although precise names are best, such as Cherokee. And Chang and Eng weren’t famous Siamese twins. They were famous conjoined twins.
I’m sure I haven’t found all my gaffes.
The American College of Medical Genetics Weighs in on Bias in Genetic Testing
Historically, for economic reasons, tests to identify carriers of certain single-gene conditions focused on population groups at higher risk, like Ashkenazi Jewish people and Tay-Sachs disease, and Blacks and sickle cell disease. Vastly improved next generation DNA sequencing now is making genetic testing panethnic. So everyone can be tested for everything – theoretically, at least.
Now that the time has come for genetic testing to enter the medical mainstream, many clinicians will need to get up to speed. That’s what inspired the ACMG’s “points to consider” statement. The organization’s Social, Ethical, and Legal Issues committee joined the Diversity, Equity, and Inclusion committee to advise clinicians on avoiding bias in genetic testing, assuming that bias “is a pervasive part of the human condition.”
The committees identify three areas of bias in genetic testing that especially impact marginalized groups and underrepresented minorities: environmental, clinical, and technical.
ENVIRONMENTAL BIAS includes issues of housing, employment, health insurance, food insecurity, stress, and interactions with colleagues, peers, and medical institutions.
Specifically, clinicians should:
• Recognize practices that have led to harm, medical mistrust, stigmatization, and discrimination.
• Respect autonomy, dignity, and traditional beliefs that might affect uptake of genetic testing.
• Represent these groups in genetics research.
• Consider genetic testing equal to other lab tests, “to guarantee equitable access to care, timely diagnosis, and appropriate management and therapy.”
CLINICAL BIAS that contributes to inequities in genetic testing and test interpretation include access to genetic services, poor knowledge of genetics among clinicians and patients, diversity of the genetics workforce, and evaluating genetic risks when knowledge of family history and/or genetic ancestry is incomplete.
Specifically, clinicians should:
• Recognize family communication styles and cultural dynamics that might affect provision of the information needed to construct pedigrees, from which genetic counselors deduce risks and assess appropriate tests.
• Avoid use of racial and ethnic categories as proxies for genetic ancestry.
• Provide resources in different languages.
TECHNICAL BIAS arises when a gene variant, or collection of gene variants (mutations), indicate a health risk in one population group but not in another, or to a different degree, because genes interact with each other and with environmental influences. For years, data from the UK Biobank has guided interpretation of many mutations. A stark example is a heart condition, hypertrophic cardiomyopathy. A mutation associated with the illness in white Europeans does not have the same effect on people of African ancestry. So it’s important to back up genetic data with clinical diagnostic evidence.
Physicians with minimal knowledge of genetics have made mistakes when interpreting genetic test results. A particularly stunning one is mistaking carrier status (a heterozygote for a recessive condition) for having the associated disease- or not explaining the distinction clearly enough to avoid patients reaching that conclusion. Similarly, some physicians have failed to follow up prenatal screens on cell-free fetal DNA in the maternal bloodstream with the old-fashioned chromosome checks needed to confirm suspicions.
Specifically, clinicians should:
• Recognize that current population databases used to interpret some genetic test results may still be based on white Europeans, with many ancestral populations underrepresented.
• Until pangenome analysis becomes possible showing all variation, use ethnic-neutral gene testing panels of variants common in different ancestral populations.
• Form global consortia to share DNA data from diverse biogeographical regions.
Dena R. Matalon, co-lead-author, explained the value of the two committees joining forces to interrogate the genetic testing situation. “Our goal in working together was to raise awareness about how bias affects all of us, ways in which it does that we may not have realized, how bias affects underrepresented minorities, and what we can do to begin to mitigate it. We ultimately want to provide more equitable clinical genetic care for all, and hope this document helps us take a step in that direction.”