St. Paul Pioneer Press (Minnesota)
By Paul Tosto and Jeremy Olson
Markingson center and Tamar Bekmedzjian, right, when Dan’s mother Mary
Weiss, left, visited him in Los Angeles in August 2001. (Photo courtesy of Mary Weiss)
When people enter drug studies at the University of Minnesota, they’re supposed to be protected by a safety net keeping watch that the vulnerable are not coerced, that standards of conduct are met and When people enter drug studies at the University of Minnesota, they’re supposed to be protected by a safety net keeping watch that the vulnerable are not coerced, that standards of conduct are met and that researchers aren’t tangled in conflicts that might influence their decision-making.
That system was supposed to protect Dan Markingson.
A schizophrenic, Markingson killed himself in 2004 while enrolled in a study at the U comparing anti-psychotic drugs. Documents surfacing the past year in a lawsuit over his death have raised questions about whether the U psychiatrist running the study followed university ethical guidelines. They also raise questions about why the Institutional Review Board, the internal group charged with protecting people in university studies, didn’t intervene.
University officials say their nationally accredited review board — a volunteer panel of 57 experts in medicine and other disciplines — works well and rigorously reviews studies. They would not talk specifically about the Markingson case to the Pioneer Press. A judge ruled in February that as a state agency, the university and its IRB are immune from the lawsuit.
The legal ruling didn’t allow questions to be explored about who’s ultimately responsible for the safety of research subjects and whether the university did everything reasonable to protect Markingson from harm.
According to the U’s human subjects protection guide, the IRB’s first charge is “to protect human subjects involved in research at the university from inappropriate risk.”
In reality, the IRB operates largely on trust. Trust that researchers will follow the rules. Trust that people will speak up when a safety plan is violated, even if they have professional or financial pressures to stay quiet.
“It’s the people who implement the plan who are responsible for protecting the subjects,” said Moira Keane, the U’s director of research subjects protection programs.
The IRB approves all clinical research — modifying safety rules when necessary — and samples study records every year or so to make sure its conditions are met. It also has the power to shut down projects that aren’t complying with safety requirements or have caused “unexpected serious harm” to subjects.
Keane recalled four studies out of thousands at the U over the past two decades that the IRB stopped.
The lawsuit by Markingson’s mother, Mary Weiss, alleged that the IRB’s trust was misplaced in the so-called CAFE study, led by Dr. Stephen Olson, a U psychiatrist.
A central allegation was whether Olson had too much power over Markingson, and too many conflicts that obscured his clinical judgment. Olson recruited Markingson into the study at the same time he served as Markingson’s treating doctor and advised a Dakota County judge on whether Markingson should be committed to a psychiatric hospital.
Had the IRB followed its own guidelines, it would have discouraged Olson from recruiting his own patient. The IRB Web site states that “doctor-patient relationships between the investigator and participants should be avoided, when possible, to eliminate any power-based coercion.”
It’s impossible to know whether Markingson would have killed himself if he hadn’t enrolled in the research study. He was in a sensitive early stage of his schizophrenia diagnosis, during which the suicide risk is greatest. Even so, the study’s rigid guidelines meant that Markingson received only one anti-psychotic drug to help control his delusions.
Experts hired by Weiss’ attorneys said in court depositions that the IRB missed opportunities to make the study safer.
Dr. Harrison Pope from Harvard Medical School called the IRB’s role an “essential link in the chain of causation that improperly admitted Mr. Markingson into the CAFE study, improperly held Mr. Markingson within the CAFE study, prohibited effective treatment of Mr. Markingson, and thus became a substantial, proximate cause of Mr. Markingson’s death.”
The IRB could insist researchers turn over all complaints about their studies, which might have raised concerns in this case. Weiss had complained in letters to Olson and Dr. Charles Schulz, head of the U’s Department of Psychiatry, that her son wasn’t getting better and was at risk for harm. She had requested that the doctors try other treatments, even if he had to be withdrawn from the study.
The U hired its own national IRB expert to refute Pope’s claims. The IRB had no legal obligation to require someone other than Olson to evaluate Markingson’s competency or his ability to consent to research, said Ernest Prentice, associate vice chancellor at the University of Nebraska Medical Center.
Nor is there a requirement that complaints such as Weiss’ letters be forwarded to the IRB unless there is some unanticipated risk. Had the IRB received complaints, it could have investigated, he said.
Weiss said she’d never heard of an IRB.
The CAFE study was fairly prominent, involving 26 academic institutions and 400 schizophrenic patients. Financed by the pharmaceutical company AstraZeneca, it was worth up to $327,000 to the U, with some of those funds going to Olson’s salary and other study personnel.
U officials said the IRB acted ethically and within its obligations and federal regulations to protect human subjects in this study.
After the suicide, the IRB sought information from Olson on how Markingson consented to the study. But IRB officials said in depositions for the lawsuit that the review board never formally investigated Markingson’s death.
The IRB investigates when there is evidence of misconduct. There was no evidence of that in the Markingson case, said Dr. Richard Bianco, a U physician who oversaw the U’s research subjects program at the time Markingson participated in the study.
Bianco declined a Pioneer Press interview request. But in a court deposition, he acknowledged that the U has some 8,000 studies involving humans — research he estimated was worth about $15 million — but that the IRB doesn’t track the number of people enrolled in U research, only the number of projects approved.
Bianco agreed with Keane that the IRB system operates largely on self-disclosure by researchers.
The U’s top research official says researchers and IRB reviewers “are aware and understand their ethical and moral obligations to do the right thing.
“If people write with concerns and issues, they will be reviewed,” said Tim Mulcahy, the U’s vice president of research. “If the IRB were to become aware of a suggestion of coercion or heavy handedness,” he added, “we have an obligation to act promptly and very directly.”
Olson declined to talk to the Pioneer Press about Markingson’s care.
He said it would be difficult for any researcher to get away with violating research rules because they are observed by so many medical students, residents, nurses and others. However, a 2006 internal audit of the U’s psychiatry department challenges the notion that those workers would speak up.
Nearly 40 percent of the psychiatry department staff responding to the auditor survey said they did not believe they would be protected from retaliation for blowing the whistle on a suspected violation in the department.
Some experts believe the nation’s system of review boards is dysfunctional and in need of reform.
“We have a very haphazard way of overseeing (IRBs) and collecting data on adverse events,” said Dr. Ezekiel Emanuel, bioethics chair at the Clinical Center of the National Institutes of Health and a national expert on institutional review boards.
“There’s no one in America who can tell you how many people are enrolled in clinical research,” he said. “No one can tell you how many people died in (ways) attributable to clinical research. No one can tell you how many people got injured, and no one can tell you over time whether the system is getting less safe.”