Q&A with Dr. Carl Elliott, Part 2: Finding fault with his own university after patient's death
Q: Why give this much attention to just one patient?
A: The fact is that the University of Minnesota's human subject protections are putting other patients at risk. That'sthe larger issue. For those of us like you and me who follow pharmaceutical wrongdoing, we see this story being played out again and again. A company comes out with a drug and promotes it like crazy. It looks like a big success. The company makes billions of dollars. Doctors write scrips like crazy. And then problems come out, and there's litigation. As the litigation progresses, it becomes clear that the company knew about the problems but they had been hiding them or burying them or manipulating their trials so that the problems didn't come out.
Ever since Vioxx, we have seen this story again and again. But what people don't think about is that people have been volunteering to take part in these clinical trials because they think they are contributing to scientific knowledge in some way. They think they are going to save people's lives or make other people's lives better. Some of the people in these trials may not be harmed, but a lot of people are harmed and experience side effects, sometimes very severe ones. And sometimes in the case of the antipsychotics, they kill themselves or they kill other people while they are in the trial. If the trial is being done for a legitimate scientific purpose, that is one thing, but what if they are being done solely to market a drug?
That's why this one death is important. It's representative of a much larger issue with clinical trials. By allowing trials to be used purely as marketing tools, you are misleading doctors and you are hoodwinking patients. Sick patients are being fooled into taking part in something that is no more than marketing tool for a drug that is already on the market.
Q: Isn't there a value in knowing which drug works best, even if the drugs area already on the market?
A: There is a value in knowing that, but what we have found from the American Journal of Psychiatry is that in 90% of the head-to-head trials of atypical antipsychotics that are sponsored by a company, the sponsor's drug wins. The meta-analysis looked at 32 trials. If Lilly sponsors the trial, Zyprexa wins. If AstraZeneca sponsors the trial, Seroquel wins.
What is the point of doing those studies when you already know with 90% certainty what the answer is going to be when the trial is done? There's no point. There could be a good way to design the trial, but you can't leave that in the hands of a company with a huge financial interest in designing the trial so that their product comes out ahead. You could have an agency like the NIH design a study that, and we would have reliable results. This was a post-marketing trial. All of these three drugs were on the market. The placebo-controlled trials were already done.
One of the things found out in litigation in one of those trials was that in at least one trial, they tested Seroquel against placebo and found that it was no better than placebo. Then the question is: what do you do with that information? And the answer was: you just don't tell anybody. "We can't let anybody see this trial." That's not the way science works. If you have a trial that shows your drug fails to beat placebo, other people have the right to see that trial as well. It's not something that you should be able to bury or spin in some way.
Q: So in your mind there wasn't a legitimate reason for the CAFÉ study?
A: If you rig the study so that it is extremely likely to produce certain results, no.
Q: What makes you say that it was rigged?
A: The main problem was that the trial was too small. It only had 400 people enrolled, and about 160 finished. And the atypical antipsychotics being compared were similar enough that the size of trial was unlikely to show a difference between the three. By showing no difference, and by failing to show a difference between the three, it allowed AstraZeneca to claim that Seroquel was as good as the other two.
Q: What should AstraZeneca be doing to design a good trial? These drugs can help people, right? People just like Dan Markingson. Don't these drugs have to be tested in clinical trials?
A: I don't think that the clinical trials and the marketing of the drug should have anything to do with one another. You need to do a clinical trial to find out legitimate information, not to prove the competitive advantage of one drug over another drug.
Q: One of the issues you raise is that the study was double-blinded. Health writers often assume that a double-blinded trial is better than one where the investigators know which drugs are being tested or which patients received a placebo. You seem to be saying that, unfortunately, that can be dangerous for patients.
A: It's true that randomized double-blind clinical trials are the gold standard. And it's a tradeoff. Everybody realizes that as far as patient care goes, it would be better for you to know what drug you're taking.To get better results,though, you need the trial to be blinded. But if you're going to make that tradeoff, than patients need to understand when they are being enrolled in a trial that they are making that tradeoff.
The reason that became a problem in the Markingson case is that he was court-ordered to follow the treatment recommendations of his psychiatrist. He wasn't court ordered to go into a clinical trial. He was court-ordered to get treatment. His doctor did enroll him in the trial, and the doctor's defense was that he was treating him by enrolling him in the trial.
But there's a problem here. You can't say that the treatment you would get in the study and out of the study are the same. If you are out of the study, your doctor would know what drug you are getting. Your doctor would be able to change drugs from one to the other if there were side effects. It's not accurate to say that the care you get in the trial is identical to what you get outside of the trial. That was my point about the blinding. Not that I want to criticize that as a scientific method. But you can't say that going into a clinical trial is identical to getting treatment.
Q: One would think that a public university is a great place to investigate a story. There are a lot of public records. Lots of politics that may be at play and prompt people to talk with you. In what ways was it easier to report the story based at a University of Minnesota and in what ways was it harder?
A: It was easier because Mary Weiss and Mike Howard were so diligent in keeping records. Basically (for) anything I needed and any question that I had, they had a file. It didn't require a whole lot of digging at all. It was harder because I made myself a hated figure at the University of Minnesota, especially in the medical school. People duck into doorways when they see me walking down the hall.
Q: Are there parts of your routine that you've altered as a result?
A: No not really. I used to get more invitations to do things over at the medical school than I do now.Talks and that sort of thing. I should say that the upside is that outside of the medical school there's been a tremendous amount of support on campus. On one side of Washington Avenue is the academic health center, where my office is, and across the street is where the humanities and social sciences and other buildings are. Outside the academic health center there are a lot of people who are very upset about this case and a lot of support for those of us in the bioethics department who have been speaking out.
Q: What about the backlash from patients who have been helped by the drug study? Or by Seroquel? Often when I write about a doctor or a treatment in a critical way, I hear from patients saying they were saved by that doctor or couldn't live without that drug.
A: I haven't heard from a single patient who says they were helped. Just the opposite.I have received letter and email and phone call one after the other from people who have had horrible experiences, and they want me to write about them.
Q: What do you do with those?
A: I'm still thinking about what to do. Some are so far outside my area of work or expertise there's nothing I can do very much to help. Sometimes people call or email, and I just don't think they are right. I don't think that the kind if injustice has been done that they think has been done.
Q: Do you respond to all of them?
A: I do respond to all of them, but sometimes I have to put them in touch with someone else. There are a couple of them that I am trying to work on myself.
There's a crazy story, actually, an old one, that I'm hoping to bring to the attention of people in the bioethics world. Dr. Tom Harbin wrote a book called Waking up Blind about something that happened at Emory. I met him recently. It's a horrible story of a guy who was the chair of ophthalmology at Emory in the 1980s and was extremely talented and ambitious but just drove himself to the point where he was seeing way too many patients and doing a terrible job. He was seeing two to three times as many patients as anyone else in the department.
Soon it became clear that he was pushing too hard. He was submitting false bills to Medicare and doing surgery that was unnecessary. There was eye tissue coming back from the pathology lab showing he was removing healthy tissue. Then he operated on the wrong eye of a patient who only had one eye. He then lied about it. There were a couple of vocal colleagues of his in the department who decided that something needed to be done and started making complaints.
The short version of the long story is that they were both punished and disciplined and lost their jobs while the chair of the department flourished and now has an endowed chair at the University of Texas Southwestern. Tom Harbin is trying to get some redemption for the whistle-blowers. He's an ophthalmologist himself, and he thinks these whistle-blowers were unfairly punished. One was formally reprimanded by Emory, and they were forced out of their jobs and had their careers damaged. I heard about that case after the Markingson case, and I am trying to help him with that.
Q: You bring up a good point about exoneration, which is something that came up in the Markingson case and comes up a lot in investigative reporting. A hospital or a university is in trouble and they say that they have been cleared by the FDA or the DEA or the FBI, etc. What made you doubtful in the Markingson case and prompted you to really dig into the FDA's inspection report?
A: Because the facts are outrageous. You have an acutely psychotic patient being coerced into a clinical trial by virtue of a court order over the objections of his mother. She repeatedly tries to get him out of the trial, telling his doctor that she's afraid he's going to kill himself, but she is ignored. Then he eventually kills himself, and the FDA comes in and says, "We think everything is just fine here."
Nobody could possibly look at this case – nobody in my world and in bioethics, at least – and say nothing went wrong. On the one hand, either the FDA has access to something that I'm not seeing that makes them think everything was done exactly right, or they just screwed up. So I got a copy of the inspection report to see what they knew that everyone else didn't. I found out that they just chose to ignore the facts.
Q: Did you FOIA it?
A: It was in the litigation from Mary Weiss. They did the report in 2005. The litigation wasn't until 2008, and so that's when it finally came out.
Q: Did you get any records from the FDA?
A: No. They still won't talk to me.
Q: What about AstraZeneca?
A: They have no comment.
Q: Have you had luck in the past getting pharmaceutical companies to discuss difficult cases like this?
A: I don't think I've tried. Most of the stuff I have written about pharma has been on a theoretical level. Bioethics scholarly articles, not actual reporting like this.
Q: You say the FDA ignored the facts. For example, you say that the inspector failed to note that "during the period leading up to his enrollment, Markingson had been repeatedly judged incapable of consenting to neuroleptic drugs. On November 14, 2003 Dr. Olson signed a commitment document stating that Markingson "lacks the ability to make decisions regarding such treatment." This may sound like a dumb question. But how do we know that the FDA had access to all these records. Aren't agencies themselves often in the dark?
A: We don't know exactly what they knew. But it doesn't seem as if these documents would have been that hard to get. The inspector did know that Markingson was under a commitment order because that was mentioned. But she just sort of breezed over it. She knew that it was a clinical trial for acutely psychotic patients. You would ordinarily assume, unless it's shown otherwise, that those patients might have a problem consenting because they are delusional and are hearing voices. These are not people you would ordinarily presume would be capable of consenting. If I were the one looking into this case for the FDA, the question of the competence of the subject would be one of the first things I would be interested in looking into carefully.
Q: But how does a researcher go about getting informed consent form someone who is severely mentally ill?
A: Some people can be mentally ill but competent. So when you are evaluating them, you need to find out some basic things. Do they understand what the clinical trial is?Are they capable of communicating about their choices? Do they understand the risks and the benefits of a trial? Do they truly understand what they are getting into?
The problem with Dan Markingson is not just that he had this vast delusional system where he thought he was being controlled by Satanic forces that were ordering him to kill people, it's that he didn't believe he was mentally ill. If he doesn't believe he is mentally ill how can he consent to a trial comparing treatments for his mental illness?
At the very minimum for someone to give fully informed consent for a mental illness, you have to acknowledge that you are mentally ill. But in a much broader sense, someone can be incompetent to consent and still be enrolled in a trial if they have a competent surrogate. This is the case of children below a certain age who are not capable of understanding the complexities of a clinical trial. So in those cases you have a surrogate decision-maker, typically the parents.
The question here is: why wasn't a surrogate decision-maker appointed for Dan Markingson? The surrogate decision-maker would have ordinarily been his mother, who was his closet relative and the person who had the most knowledge of him and who loved him most. She was telling them that she didn't want him in the trial, but then they said he was capable of consenting himself. The consequences were severe. And if trials continue to be conducted this way, there will be more patients who die as a result.