The point of this post is to bring your attention to the writings of some fellow bloggers, particularly 1 Boring Old Man (1BOM). For the past 6 months , but particularly in the past month, he has brought attention to a conflict of interest with David Kupfer, the head of the APA’s DSM-5 task force (here, here, here, and here).
This is the outline of the story. From its inception, there was an interest in adding a dimensional component to the DSM. When I first learned of this, I thought they meant that this would replace the categorical diagnoses. What I have learned, however, is that the dimensions are intended to supplement the categories so they would allow us to quantify the degree to which one suffered from, let’s say, Major Depressive Disorder. I will leave aside for now the merits of this approach except to say that if a category is inherently suspect, adding dimensional ratings will not necessarily improve it validity.
In any event, there is a long history of research in this field. One leader is Robert Gibbons who is the director of the Center for Health Statistics at the University of Chicago. He has developed a computer based test that allows for flexible questions. This would replace traditional questionnaires that have a set number of questions that everyone answers. In these modern tests, the answer to each question determines what each ensuing question will be. The researchers argue that these questionnaires are more efficient. They are able to provide a diagnosis in as little as 12 questions (as compared to the 25 or 30 typical of the fixed question models). In order to do this, complicated computer algorithms are employed. Dr. Gibbons has received about $5 million dollars from the NIMH to develop the algorithms, which they call computerized adaptive tests.
Dr. Gibbons argues that the efficiency of these programs will allow busy clinicians to arrive at a diagnosis much faster. A patient could sit with a tablet in the waiting room, answer a few questions, and walk in to the doctor’s office with a diagnosis. Again, I am going to set aside the relative merits of this type of efficiency except to say that I sometimes wonder if the people who develop these ideas ever spend time talking to actual human beings who go to a doctor’s office seeking help for some personal crisis.
I will add that this type of approach will require access to the computer program with the algorithm. That is a product that could be made available in the public domain or could be sold.
So this is where it gets interesting.
Dr. Kupfer is a leading researcher and a big proponent for these types of assessments. The initial goal was to include them in the DSM-5 but in the end they were just included in a section indicating they needed more research.
Dr. Kupfer and some colleagues, including his wife, Ellen Frank, have been part of a research team with Dr. Gibbons. They have written several papers on their tests. One of them was published in Archives of General Psychiatry (now known as JAMA Psychiatry) in 2012. When an author writes a paper for a journal, he is asked to sign a conflict of interest statement. Dr. Carroll learned that Drs. Kupfer, Gibbons, and Frank has all entered into a business, Psychiatric Assessment, Inc. that would sell the computer program. They failed to disclose this. When Dr. Carroll brought this to the attention of the journal, an apology was published. They then brought this to the attention of the APA, who issued their own report. What is particularly concerning is that along the way, Dr. Carroll wrote a letter to the Archives questioning the validity and value of this line of research. The response from the authors was to suggest that Dr. Carroll was raising these questions not because of scientific concerns but because of his own COI, he is the author of a standard fixed assessment. They wrote this during a time where they appear to well along the path to establishing their business.
1Boring Old Man has written his own open letter to the APA. He has also detailed the timeline in his blogs which can be found in the links above.
One of the questions he has asked is why is this not receiving more attention. I held off on writing about this. First of all, it is complex and I wanted to make sure I understood the issues. I have not been shy about posting but I admit to having worries about being naive and getting into something deeper than I appreciate at this time. Secondly, I am not sure I have any significant influence. I have no idea who reads this, who reads 1BOM or any of these other blogs. I also think that the audience here is already convinced that psychiatry is corrupt. I am not sure what impact this will have.
But in the end, I decided that I had an obligation to use whatever platform I have to bring more attention to this issue. For me, one of the biggest problems we have in psychiatry is unacknowledged COI. Given the sensitive nature of the work we do, the power we can wield over people’s lives, the history of our improper relationships with Pharma – even if it turns out all rules were followed – that is not good enough. We should be above reproach.