Posted on February 1, 2019 at 9:00 AM
Through special arrangement with Taylor & Francis, AJOB posts its editorials on bioethics.net. This essay and the articles it references are also available on the publisher website.
by Thomas H. Murray
At barely 21 years of age, there are already calls for genetic exceptionalism to be retired. A little background on the circumstances of its birth might be helpful.
The labor was prolonged and arduous. It began, I suppose, with the work of the Task Force on Genetic Information and Insurance, an offshoot of the ELSI (Ethical, Legal and Social Implications) Working Group for the Human Genome Project. Our mission was to “anticipate ethical, legal, and social policy issues raised by the impact of genetic information on the public’s access to health care financing, life and disability insurance and to develop suggestions that could be used to forestall potential problems before they occur”. We were instructed to work with the insurance industry, regulators, consumers, professional groups, and researchers to develop a clear picture of the looming collision between the insurers’ business model that relied on predicting—and limiting liability for—risk, and the anticipated avalanche of genetic information that could be used to assess individuals’ likelihood of disease, disability and early death. We were also supposed to recommend policy options for dealing with the impending pile-up.
Insurers’ conception of distributive justice went by the name “actuarial fairness,” roughly the idea that justice consists in tailoring coverage and cost of a policy according to the likelihood that insurers would have to pay for a covered event. It’s not a crazy idea. A careless shop owner with a history of many small fires should have to pay more for fire insurance than a fastidious one. A 90-year-old, according to actuarial fairness, should pay more for the same amount of life insurance than a 20-year-old. Applying actuarial fairness to health insurance, however, feels heartless and perverse: The more likely one is to need coverage for a particular illness, the less willing companies are to cover it and the more you must pay if they do. The current debate over preexisting conditions touches this very nerve.
By 1991, when the task force began its work, the prospect that massive numbers of people might soon learn through genetic testing that they were predisposed to develop one or more serious maladies caused widespread anxiety. Insurers argued they should be allowed to treat evidence of genetic predispositions to disease just as they’d done with preexisting diseases.
The insurance task force was an ad hoc body of a working group (ELSI Research Program) of a joint scientific collaboration between the National Institutes of Health (NIH) and the Department of Energy—we had, that is, minimal institutional clout. During our tenure, there was a major undertaking at the White House, led by the wife of the then President, to thoroughly overhaul the American health care system. We were kept at arm’s length from that effort, which was shrouded in secrecy, but it seemed likely at the time that its work could have massive implications for how health insurance would operate in the future. In other words, we could only guess what the world of health insurance might be like when our report was published.
Nevertheless, we tried to fulfill our charge beginning with health insurance, hoping to get to life and disability income insurances once we’d completed our work on health. As an initiative of the Human Genome Project, our mission was to assess the impact of genetic information. Accordingly, we set out to show how genetic information was different from other kinds of health-related information.
We failed. Or, to put it more precisely, we looked one by one at the arguments offered for the uniqueness of genetic information in the context of health insurance. None of them, in the end, was convincing. Sometime later on I began using the term “genetic exceptionalism” to denote the claim that genetic information was so distinct in concept, practical implications, and moral import that it deserved to be singled out from other types of health-related information. At that time, influential authors were describing genetic information as “uniquely powerful and uniquely personal”. In an essay that, I believe, introduced the term in print, I bundled the various claims into three “concerns”: the argument from genetic prophecy, the concern for kin, and the fear of discrimination. All three concerns were real and relevant, but I argued that none of them justified the ominous warnings that our DNA was a “coded, probabilistic future diary” that others would read and misuse without our knowledge or permission.
Our recommendations strove to put genetic information in context. For example, the very first recommendation of our report read: “Information about past, present or future health status, including genetic information, should not be used to deny health care coverage or services to anyone”. We had come to see genetic information as a subset of all health-related information. It had its particular qualities that deserved to be understood and accommodated, but the similarities were even more important. We concluded that it was wiser not to regard genetic information as wholly distinctive in kind. I thought I detected some uncertainty even among the advocates for treating genetic information as different. In the article describing their commendable effort to draft a Genetic Privacy Act, Annas and colleagues included this very important conditional: “To the extent that we accord special status to our genes and what they reveal, genetic information is uniquely powerful and uniquely personal, and thus merits unique privacy protection”.
“To the extent that…” implicitly acknowledges the possibility that we could do otherwise: that genetic information may not be “uniquely powerful and uniquely personal”; that we are not obliged to accord it special status or unique privacy protection. What if instead we turned down the rhetorical temperature and domesticated genetic information—that is, welcomed it into the commodious abode of all health relevant information about ourselves? That, I confess, was my intention in naming and critiquing genetic exceptionalism.
Yes, genetic information has certain properties that warrant close attention. But nothing that launches it into some unique universe of moral, legal, and policy concerns. Sounding the alarm about “genetic exceptionalism” was intended to refocus us on the particulars, on the multiple factors that shaped our understanding—and misunderstandings—of genetics; how it would be used in the clinic; the social, political, and economic forces that would affect how it would be employed to our benefit or our detriment—in a word, on the context.
Not surprisingly, I applaud the intent of “Genomic Contextualism: Shifting the Rhetoric of Genetic Exceptionalism.” Garrison, Brothers, Goldenberg, and Lynch (2019) argue for a “fundamental duality” such that “genomic tests both share characteristics with other types of medical tests and represent a combination of features that make them distinct” (1). In place of a “fundamental duality” I might prefer the more modest notion that information about us derived from genomic tests will be like other sorts of information about us in some ways and different in other ways. That doesn’t sound particularly profound. But it reminds us that we can’t escape the hard work of exploring the many dimensions of context. There will be times when the genetic link to forbears or descendants will be critical, as Botkin notes regarding finding a BRCA1 mutation in a newborn whose mother may be at increased risk for breast and ovarian cancer. The link to other generations, however, isn’t as important when the question is whether an insurer can deny that person coverage for breast cancer treatment. Whether similarities or differences matter more depends on the context.
The many excellent commentaries accompanying the article display a commendable attention to context in both substance and method. The particular frameworks we use influence what we look for and, therefore, what we are capable of seeing. “Genetic contextualism” employs two frameworks, one broad, the other focused on ethical analysis. From rhetoric it draws on stasis theory, which directs our attention to four dimensions: facts, understandings, values, and actions. If we construe these dimensions broadly enough, it’s difficult to see what could be left out! The history of eugenics and “racial hygiene”? Capitalism and the idea of insurance as a sharing of risk? Whether health care insurance has a moral mission that distinguishes it from life insurance? It’s also important to remember the four dimensions aren’t temporally linear, as Condit observes. Facts that at first seemed insignificant may be recognized later as massively important, once understandings come into focus, values at stake are identified, and the universe of possible actions is refined.
The other framework the article uses is “principlism,” what it characterizes as “the well-defined principles of biomedical ethics.” I can do no better than quote from John Arras’s “Methods in Bioethics: The Way We Reason Now”: “Interpreted (correctly) as a framework for moral deliberation and justification, as opposed to an algorithm susceptible to more or less mechanical application, principlism has made major contributions to the field. On the other hand, many enthusiasts of principlism did tend to naively apply the ‘Georgetown mantra’ in an unreflective and ultimately indefensible manner”. John traces the evolution of Beauchamp and Childress’s account of principlism as it evolved ultimately to embrace reflective equilibrium in what he dubs “a never-ending dialectic of justification.”
To the extent that we treat stasis theory or principlism as a box-ticking exercise or checklist, I believe that we undermine whatever is most helpful about them. Then again, we can use them as starting points, reminders that we had best pay attention to critical aspects of whatever problem we confront. To put it another way, they urge us to be attentive to context in multiple dimensions. Many years ago I suggested a bioethics analogue to the three laws of real estate: location, location, location. For bioethics the laws should be: context, context, context.
This is not easy. It requires wide reaching imagination, sensitivity to different perceptions and interests, and never-ending modesty about the limitations of our wisdom, along with the possibility that as we learn more, our judgments will have to be revised, and, perhaps, apologies will have to be proffered.
And so, despite its relative youth, I agree with the authors of the lead article and commentaries that genetic exceptionalism should be retired. It served a purpose in its time as a counter against genetic determinism and the exaggerated fears it engendered. But it’s been pulled in so many, even contradictory, directions in recent years that its meaning and usefulness have dissipated. “Contextualism” is an improvement, to be sure. But attending carefully to context should apply to everything we study, not only HIV, genetics, and genomics.
The phrase “genetic exceptionalism” was meant as a counterweight to the idea that our genome was a mysterious, opaque, and ominous “future diary” in which our life was already recorded. As an alternative, I asked, why not regard our genome as a list of the obstacles we are likely to encounter and perhaps a somewhat better prediction of how long we have to do what matters to us, to be with the people we love, and to accomplish the tasks we have set for ourselves? Our genes no more dictate what is significant about our lives than the covers and pages of a blank diary dictate the content of what is written within. Our genes might be regarded metaphorically as the physical, but blank, volume in which we will create our diary. Some volumes have fewer pages in which to write, some more. Certain pages, often toward the back of the volume, may be more difficult to write on. And some leaves may require great skill and effort to open at all. But the physical volume is not the content of the diary. The content we must write ourselves.
With many fewer pages remaining in my own diary than when I first wrote this, I am grateful for the (mostly) younger voices taking on the challenge of how to place genomic information in context.