Monday, August 19, 2013

The Empire Strikes Back

Some of you older folks may remember the case in which geneticist Therese Markow (then of Arizona State, now of UC-San Diego) bled a Native American tribe on the promise of studying diabetes, and then piggybacked some research on schizophrenia and population structure and history onto that promise; except that she didn't tell them about it and they didn't consent to it.  This had been "situation normal" in the field for decades, but in the wake of the Native American Graves Protection and Repatriation Act (NAGPRA, born 1990) and the Human Genome Diversity Project (HGDP, died 1996), it is no longer acceptable practice.

The Havasupai case helped to reframe the relationship between scientists and Native peoples, which had been tested by the HGDP, and which had relied for decades on the assumption that there was a gentleman's agreement between the geneticist and the tribe, and that the geneticist could say anything to get the genetic samples from the tribe, and after it was out of their bodies, it was the property of the scientist, who could then do pretty much anything with it, including research that the tribe had not agreed to, and trading samples to other labs for other research.  But no more.  

In April 2010, the lawsuit brought by the Havasupai against Markow's institution was settled out of court, and the tribe, the university, and the bioethics community were all satisfied with the results.  I wrote it up for Anthropology Today shortly afterwards.

Now that we have blown up the Death Star, however, the Empire is striking back.  Ricki Lewis, the author of a major textbook on human genetics, and thus with the potential to miseducate thousands of students in human genetics, now says that Teri Markow did not actually study schizophrenia, the investigation and report commissioned on behalf of ASU and the Havasupai was a smear job, and the charges against her were entirely false.

So we are compelled to revisit the case once more.  Think of Teri Markow as Alex Rodriguez, and yourself as Bud Selig.

Ricki Lewis carried out a literature search and discovered that Teri Markow never published on the genetics of Havasupai schizophrenia.  From this, she concludes that Markow never actually studied it.

Like the weapons of mass destruction in Iraq, the narrative that the geneticist sought schizophrenia genes in Havasupai DNA became established fact with the repeating. Soon the accounts of the case began quoting and citing each other, as if the original documents that held the truth didn’t even exist.

The trail of original documents is actually rather interesting.  The most significant document is the one produced by Arizona State's investigation at the end of 2003, known as the Hart Report.  Ricki Lewis dismisses it, because she has additional secret information:   

I’ve got more background than went into the blog, and based on it, I wouldn’t trust the Hart report.

Be that as it may, the Hart report was based on extensive interviews and paper trails, and was as comprehensive as could be expected, to the satisfaction of both the Havasupai and the university.  It discusses lots of other original documents, such as consent forms.  In fact, it tells us that in an interview with Markow in 2003, "Markow had indicated, during the course of the interview, that she had lost or misplaced the file containing the informed consents from 1991 on ...".  The dog ate my consent forms.

Another original document is a story that appeared in Phoenix Magazine in 2008 by Jana Bommersbach, which says,

Markow maintains to this day that she had permission to test for things other than diabetes and that her “proof” is the consent forms signed by some of the Havasupai who donated blood. She insists the project had two focuses: diabetes and schizophrenia.

So in 2008, Markow was saying that schizophrenia had been a research focus of hers, and in 2013 she is saying that it wasn't?  Or more precisely, in 2013 Markow is enlisting a shill in the genetics blogosphere to deny that schizophrenia in the Havasupai had ever been a research focus of hers?

Further, the Hart Report says:

Dr. Markow was funded by NlMH and NARSAD to study the role of genetic factors in schizophrenia, a psychiatric disorder "which occurs at a significantly greater rate among the Havasupai (7%) than in any other population (1 %). Interestingly, all cases of schizophrenia occur in lineages tracing back to a single man (a shaman or medicine man) who lived in the 1880's."

NIMH is the National Institutes of Mental Health, not a cartoon.   NARSAD is the National Alliance for the Research of Schizophrenia and Depression, now called the Brain and Behavior Research Foundation.  In other words, among the original documents are funded grant proposals for research on schizophrenia.

The NARSAD grant for 1990-1991 was a big one, but unfortunately, the dog ate that, too.  According to the Hart report,

Information was hard to obtain on this grant. It is our understanding that it was funded in the amount of $92,880.00 and provided funds for the blood collection, processing and analysis that led to the genetics studies undertaken at Arizona State University (and elsewhere)....

Dr. Markow submitted the letter of intent for this grant, which requested funds to "initiate a major research program on the etiology of schizophrenia in a unique patient population in Arizona," the Havasupai. ... She proposed conducting the research in two stages. The first stage was to involve the collection of data, and it was for this stage that she was asking NARSAD, the National Alliance for Research on Schizophrenia and Depression, for funding. She indicated that support for the second, more extensive stage would be sought from NIMH, the National Institute of Mental Health. We have uncovered no evidence to establish that this letter of intent, or other documentation associated with this grant proposal, was presented to the Havasupai.

The NIMH grant was for "Genetic Analysis of the Dopamine Receptor Gene Family" and provided funding for Markow from 1992 through 1995.  The Hart report says,

the two primary goals of the project were, first, to determine the amount of genetic variation present in the dopamine receptor gene family in the Havasupai and, second, to statistically analyze the data to determine if any genetic variation was associated with the development of schizophrenia.

Unfortunately, the dog ate that grant proposal as well.  Teri Markow had a very hungry dog.

So what of this bizarre blog post by this textbook author?  The research didn't get very far, and didn't come up with anything publishable, which is why Ricki Lewis did not find any publications on it.  But how on earth can anyone claim that there was no research?  When she asks rhetorically, "Why did articles twist events to seem as if she had intended all along to study schizophrenia?" the obvious answer is that because Markow had indeed been interested in studying schizophrenia, and had received grants to do it.

So I pointed that out to Ricki Lewis, and she responded:  "just because you are funded from a certain source with a disease name in its title does not mean that you are or are intending to do research on that particular disease."   

Read that again, slowly.  A major science textbook author is telling human genetics students that they can apply for cancer funding from, say, the National Cancer Institute, and get it, while not actually having any interest in cancer or performing any research relevant to cancer.  I think she's wrong.  In fact, I think the National Cancer Institute would consider that fraudulent.

Teri Markow herself jumped in, but didn't say, "No, we did not study schizophrenia" - which was presumably the point under contention.  Instead, she said blah blah blah:

First of all, the consent forms speak for themselves with respect to the breadth of studies. However, without sufficient and appropriate genetic variants, association studies of any disease are not possible. The first step therefore, with our funding, was to seek genetic variability that could be useful in studies of disease. While we did not find sufficient variability with the techniques available at that time to perform association studies, we at least were able to provide people with feedback about their health status with respect to diabetes, because the test for this is simple. Had there been sufficient variability for stringent association studies for any disorder, we could have proceeded. 

However, even if there had been the requisite variability, stringent diagnostic testing for schizophrenia would require a major effort, including lengthy interviews, that never could have gone undetected by the participants. This fact, plus the fact that I have never published a paper on schizophrenia in the Havasupai, answer your question quite definitively.

This is from someone who received hundreds of thousands of dollars in grant money to study the genetics of schizophrenia in the Havasupai, but came up with nothing publishable.  But whether they came up with anything publishable is not the issue, this issue was whether research was carried out on the topic, without the knowledge or consent of the Havasupai.  The latter clause has not been challenged; the Havasupai were not aware that their blood was being used for any studies other than diabetes.

Those of us of a certain age can remember the "non-denial denial" from Richard Nixon's office in the old days of Watergate. 

So I tried again: "What I am hearing is that Dr. Markow sought and received funding from an agency that focuses on schizophrenia, without any intention of actually studying schizophrenia, and ultimately performing no science related to that illness."  Neither Ricki Lewis nor Teri Markow responded to that.  I suspect that if Markow publicly agreed with it, she might be incriminating herself.

Teri Markow seems to have had little difficulty in obtaining research grants or faculty positions subsequent to the Havasupai case, and now studies Drosophila.  She is probably very nice, and I've never thought that she did anything worse than what was common practice at the time.  It was the common practice that was being challenged.

One of my favorite examples of that practice is from a puff piece that Time Magazine ran on the HGDP, in which Luca Cavalli-Sforza was trying to show himself off as a swashbuckler.  And as S. J. Perelman once said, no man ever buckled a better swash.

On one occasion, when Cavalli-Sforza was taking blood from schoolchildren in a rural region of the Central African Republic, he was confronted by an angry farmer brandishing an ax.  Recalls the scientist, “I remember him saying, ‘If you take the blood of the children, I’ll take yours.’ He was worried that we might want to do some magic with the blood."  (Time Magazine, 16 January 1995)

When people come after you with an ax, that is usually a clue that you have not received their voluntary informed consent.  That is why there is no HGDP anymore, and why Teri Markow now works on flies.

Bioethics serves to make scientists responsible social actors, especially if they don't want to be responsible social actors.  Time was, that a scientist could do anything they wanted to anyone they wanted, for any reason they wanted.  Now they can't.  The lesson we learned around the middle of the last century is that the progress of science is great, but when it bumps up against human rights, human rights wins, hands down.  That's a good lesson for genetics students to learn. 

And what is the author/blogger Ricki Lewis's stake in all this?  Why bother to revisit a dead issue, and mount such a preposterous defense of Markow's work?  Maybe she's one of those people still fighting the Science Wars, who thinks that bioethics is just a set of obstacles to scientific research.  She does seem to hold some unusual views about science: "Science has nothing to do with belief, it is about data and evidence."  Actually, science has a lot to do with belief; for example, if you believe that you can do science on people without their consent, then you don't get to do the science.

1 comment:

  1. I don't think I ever want to meet that dog.

    On a related topic: