© 1998 by Oxford University Press
Should the one speak for the many? Or the many for the one? This question haunts the hunt for medically relevant genes as researchers focus on small, culturally and/or geographically isolated populations.
Some bioethicists have raised the question of whether the community as a whole, not simply individuals, needs to consent to research that may affect it. And if so, is it the group or the individual that has the final say? One group of researchers has obtained just such a group consensus, which they think could be used as a model for future research.
The University of Oklahoma researchers, led by Morris W. Foster, Ph. D., associate professor in the university's Department of Anthropology, believe it is the first formally published example of a group consensus obtained for genetic research.
The agreement with the Apache Tribe of Oklahoma, published in the September issue of the American Journal of Human Genetics, attempts to address issues that have arisen as science has looked to these socially identifiable populations for genes: How can both communal and individual decision making be respected? How can researchers minimize the risks of stigmatizing entire categories of people? And who should profit from valuable information derived from a unique population?
Working System
In the case of the Apache, Foster's group worked with the systems the Apache already had in place to reach community decisions to set up a Tribal Internal Review Board, which negotiated the format for the study that was to take place. While community approval of the research project was needed, each individual had the option of not participating, and steps were taken to protect the privacy of those who chose not to do so.
In order to deal with the risks of stigmatizing the entire Apache population, the researchers agreed to publish the results without naming the tribe if the Apache felt the study would stigmatize them. Foster's group likens this to the publishing of pedigrees without naming the family; the genetic factors analyzed are not invalidated by not using a recognizable name.
Share Profits
On the question of profit, the tribe will receive a share of all royalties from any intellectual property that may result from the research. The tribe earmarked any such funds for the promotion of the health and education of tribal members.
"By actually doing it we demonstrated that it is practical and effective," Foster said of the agreement. "The question now is, can we expand it to other groups?" In their paper, Foster and his colleagues outline a general approach for determining the relevant social hierarchy, and then working with it to obtain a consensus about the research from the group. From this starting point, the researchers feel, a workable consensus can be reached for any culturally identifiable group being studied for medically useful genes.
But not all agree that Foster's approach is generalizable, or even beneficial to the groups in question. In an editorial appearing in the same journal, Eric T. Jeungst, Ph.D., associate professor at the Center for Biomedical Ethics at Case Western University in Cleveland, argues that this method will not actually provide the protections it promises to these groups.
Jeungst argues that by defining politically recognized social groups in genetic terms, we can make it easier to support racism. Human genetic variation doesn't smoothly follow socially identifiable groups, but by equating those social groups with the genetics studied, we make it seem as if these groups are fundamentally defined by their different genetics.
Jeungst also argues that the protections are meaningless, since the human race is sufficiently widespread and overlapping that any group's decision could be bypassed. For instance, American citizens descended from an immigrant population could be studied over the objections of the population of the "old country."
Philip Reilly, M.D., president of the Shriver Center for Mental Retardation in Waltham, Mass., said that even if the approach can be generalized (which he isn't convinced of), he doesn't think there is a significant threat to individuals from this kind of research. Nor does he feel that increasing layers of protection, which makes research more difficult and costly, needs more of a reason than a suspicion of risk. Obtaining some kind of community consent may be laudable, he said, but should not be raised to the level of ethical or legal obligation.
Reilly also worries that making autonomous adults such as the Apache into something equivalent to a vulnerable population requiring extra protection is inherently demeaning. "Every member of a population is an adult," Reilly points out. "We posit that he or she has the capacity to assess risk in entering the study and make a decision."
Commercialization
Foster and Reilly sharply disagree on the merits of offering financial stake in the results of the research to the groups that are studied.
"It's increasingly obvious that you will need to give them some financial interest," Foster said. While some case law exists saying that a donated sample ceases to be your property, Foster said he expects that will eventually be challenged and donors will have more rights to benefit from the donation.
Reilly described himself as "very uncertain" about going down that road. "I think there is a danger of over-valuing the contribution of tissue samples," Reilly said.
Reilly makes the following analogy: If a sculptor is given 5 pounds of clay from his neighbor's backyard and makes a valuable statue, should his neighbor have a share of that statue's value?
The mere act of participating is already beneficial, Reilly argues. Because the study focused on that population, they are the group most likely to benefit from the research. "What about the principle of altruism?" Reilly asked. "Blood donation is premised on altruism. Why shouldn't altruism be the central premise of operating human subjects research?"
-- Laurent Castellucci
Culturally Identifiable Populations Raise Thorny Issues for Researchers
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||