Eight Unanswered Questions about Psychiatric Research in Minnesota

The wait has been exhausting, but it is possible that a flicker of light may finally shine on the dark recent history of psychiatric research at the University of Minnesota.  A controversial external panel commissioned by the University and the Association for the Accreditation of Human Research programs was scheduled to deliver its review to the University by February 15, and in mid-March the Office of the Legislative Auditor for the state of Minnesota is scheduled to deliver its review to the state legislature.

Given these upcoming reviews, it seems like a good time to look back at some of the unanswered questions about psychiatric research at the University that have emerged over the past several years.  What follows is not an exhaustive list, just some of the highlights.

1) How did Seroquel from a University of Minnesota research study get into the oatmeal at a sex offender facility?

Actually, that question is not hard to answer.  It is documented in a Serious Adverse Event report by Dr. Charles Schulz, the Chair of the Department of Psychiatry.

Two inmates at a sex offender facility in Minneapolis were recruited into a clinical trial of Seroquel for Borderline Personality Disorder.  Schulz was the principal investigator.  Apparently these subjects didn’t especially enjoy their daily Seroquel.  One subject – whose name is redacted, so I’ll just call him “McMurphy” — took his study drug back to the sex offender facility, where he worked in the kitchen.  Here is the relevant passage:

“Some residents noticed pink particles in the oatmeal.  After eating breakfast the residents and staff reported feeling sedated and some were ‘knocked out’ for the remainder of the day.  Staff asked McMurphy if he had put the study medication into the oatmeal and he denied it.  After failing a polygraph test McMurphy was re-imprisoned.”

But that’s only the start.  Put aside the fact that federal guidelines have special protections for research on prisoners.  Also put aside the financial conflicts of interest for Schulz, which the IRB noted but did nothing about.  As I pointed out in my letter to the Institutional Review Board last month, the IRB overseeing the study appears to have done no investigation whatsoever into this alarming incident at the sex offender facility. In the continuing review documents I was provided, there is no indication that the IRB even asked how the sex offenders wound up in the study.

“I was incompetent and didn’t know what I was doing,” Robert told the KMSP reporter.  “Then, they say you have a giant medical bill and if you do the research, you won’t have this giant medical bill.”

Robert says he was told that bifeprunox was safe.  But the FDA soon rejected the drug, asking the sponsor to look into the death of a research subject who had died of liver failure shortly after taking it. Several months later, the sponsor halted all bifeprunox studies.

The side-effects of bifeprunox were so severe that Robert considered suicide. He went to the Fairview emergency room three times, once by ambulance, yet Olson apparently dismissed his symptoms as “psychosomatic.”  Olson told the sponsor that Robert’s symptoms were probably not related to bifeprunox.

The University of Minnesota claims that it has reviewed the bifeprunox study, in response to complaints I submitted last year.  Yet it has refused to release the investigation report, and has contracted with a business consulting firm called FTI Consulting for an additional review.

3) Why did the FDA visit Dr. Stephen Olson in November? 

During an FDA panel session at the PRIM&R conference last December, there were hints that the FDA might be re-opening an investigation into the Markingson case.   Several weeks ago, Vice-President for Research Brian Herman conceded that the FDA has visited Dr. Stephen Olson, the Principal Investigator for the study in which Markingson died, but would not say any more.  Several people (myself included) have filed FOIA requests to the FDA, but after five weeks, the FDA has still not responded.

4) Just how many subjects were enrolled in the CAFÉ study?

Dan Markingson committed suicide in an AstraZeneca-funded antipsychotic trial called the CAFÉ study.  When study coordinator Jean Kenney was deposed, she estimated the number of subjects enrolled in the CAFÉ study to be “in the twenties.”

Yet when the FDA conducted an inspection in 2005, Stephen Olson, the principal investigator, reported only 17 subjects. That figure is also the one that Olson used in 2006, when he gave a presentation on the CAFÉ study at an International Congress of Schizophrenia satellite session.

However, the statements of Olson and Kenney both appear to be contradicted by a 2005 document from Quintiles that lists 18 subjects, and payments to the university for 15.

This controversy erupted two years ago, when I asked Frances Lawrenz, the Research Integrity Officer for the university, to investigate a puzzle surrounding the “evaluation to sign consent forms” used to determine whether psychotic subjects in the CAFÉ study were mentally capable of consenting to research.

Here was the puzzle. Two slightly different “evaluation to sign consent” forms had been entered as evidence in the litigation over the death of Dan Markingson.  In most ways, the forms were identical: same handwritten answers, same date, same signatures, even the same minor mistakes.  But there was one obvious difference: the blacked-out marks in the space where the initials of the research subject were supposed to be entered.  The blacked-out marks were differently shaped, and one had Dan’s initials written above the mark.  Why?

When I posted these two forms online, a number of other family members of research subjects contacted me privately, saying that they had found strikingly similar “evaluation to sign consent” forms in their records.  One  family even sent me the form they had found, which I posted online.

Parts of it were identical to Dan’s form, which raised a larger, more disturbing question.  Were researchers simply using the same “evaluation to sign consent” form for different subjects, with the answers already filled in?

You can compare the form this family sent to me with the “evaluation to sign consent” forms for Dan Markingson, which can be found here and here.  The answers on each form are exactly the same; the handwriting is exactly the same; the signatures are exactly the same.  Even the blacked-out mark at the top is the same as the mark on one of the forms for Markingson.  How could such near-identical forms be legitimately used for different research subjects?

Predictably, Lawrenz refused to investigate the concerns I raised.  Instead, she turned the matter over to the Office of General Counsel, which issued a press statement denying any wrongdoing.  Mark Rotenberg claimed that the form I posted was “merely another photocopy of the Markingson form, on which someone has clumsily whited out identifying features.”

Rotenberg said that the University had “reviewed the files of all CAFE trial participants who actively participated in the study” and that there was no evidence that any of the “evaluation to sign consent” forms had been duplicated. But the University denied my open records request for the forms and has refused to make any of the forms public.

Rotenberg’s response inadvertently created an additional puzzle.  His press statement claimed that there were only sixteen subjects in the CAFÉ study. Why would Rotenberg make that claim, when Olson reported seventeen subjects and Quintiles paid for eighteen?

7) Did Michael Swanson go untreated before his killing spree because he was going to be enrolled in a research study? 

Before 17 year-old Michael Swanson stole his parents’ car in 2010 and killed two convenience store clerks in Iowa, he was seen by University of Minnesota psychiatrist Dr. Jonathan Jensen.  According to a Fox 9 KMSP News report, Jensen diagnosed Swanson with bipolar affective disorder and recommended the antipsychotic drug Abilify.  “There’s a real concern about his use of force,” Jensen wrote. “It would seem to me that likelihood of his repeating the theft of guns, robbery of people is probable and that without anti-psychotic medication, he may carry out these behaviors.”

Instead of treating Swanson, however, according to Fox 9, Jensen recommended that he be enrolled in a research clinic at the University of Minnesota. Swanson’s family did not see Jensen’s report until 4 months later when attorneys got a copy for the murder trial.  Then they learned that Jensen was concerned their son was a “risk to society if un-medicated.”  Swanson’s mother, Kathy, is convinced that Michael went untreated so that he could be enrolled in a research study.  But KMSP News could uncover no hard evidence to support her belief.

8) Why did the University of Minnesota destroy the records for Study 41?

Study 41 was a controversial clinical trial of Seroquel XR that emerged in litigation against AstraZeneca.  The University of Minnesota was a site for Study 41; Charles Schulz, the Chair of the Department of Psychiatry, was a site investigator.

Unfortunately for AstraZeneca, the study showed Seroquel XR to be no better than placebo for patients with acute schizophrenia. In emails, AstraZeneca officials referred to Study 41 as a “failed study,” or “code red.” Advisors and employees were instructed not to discuss the study.

AstraZeneca decided to bury Study 41 and re-do it in the developing world.  That trial, which was called Study 132, turned out positively. So in 2007 AstraZeneca published study 132 in the Journal of Clinical Psychiatry.  Schulz was not an investigator for that study, but he was listed as second author anyway.  Schulz also presented the data at the annual meeting of the American Psychiatric Association and served as the academic point man in company press releases.