Wednesday, March 4, 2009

Conflict of Interest - Considerable Room for Improvement


"Conflict of Interest," read the first part of a commentary headline appearing in this month's American Journal of Psychiatry, published by the American Psychiatric Association. "An Issue for Every Psychiatrist," ran the second.

More than 25 prominent psychiatrists (and one prominent PhD researcher) were listed as authors. These included Robert Freedman MD, Nancy Andreasen MD, PhD, John Rush MD, Myrna Weissman PhD, and others.

Clearly, they had something important to say.

"To many psychiatrists’ dismay," the commentary opened, "unresolved conflicts of interest between parts of our profession and the pharmaceutical industry continue to be a focus of concern. ... Our standards should address not only the conduct of high-profile opinion leaders, but also our responsibility as individual physicians to deliver to our patients the highest-quality evidence-based medicine."

So far, so good.

The commentary acknowledged reality, namely that if journals didn't publish industry-supported clinical trials, there would be little in the way of new meds research to report. As a bulwark against the industry pushing its own PR disguised as research, the commentary pointed to the AJP's rigorous editorial review standards.

Hold on a second ...

In Feb 2006, the AJP published an Eli Lilly-supported study that found:

"Compared to placebo, olanzapine delays relapse into subsequent mood episodes in bipolar I disorder patients ..."

In other words, according to the study, patients on Zyprexa stayed well much longer than patients taking a sugar pill.

The study also revealed that only 21 percent of the Zyprexa patients actually completed the long-term (48-week) phase of the study. These were patients who had responded well to the med during the initial phase of the study.

To phrase this another way, eight in ten patients stopped taking their Zyprexa. Transposing this result to life in the real world, a doctor might think twice before prescribing Zyprexa over the long term. After all, even the best med in the world is useless if patients refuse to take it. Or, based on this information, a doctor might spend extra time informing the patient of the value of staying on the med.

Surely, such an important finding would be highlighted toward the beginning of the article. Surely, it would appear in the article abstract (which is the part most doctors actually read).

Well, no. The finding was buried deep in the study.

Are these the high editorial standards this month's AJP commentary was referring to? If so, heaven help us all.

2 comments:

cretin said...

Poor Zyprexa gets such a bashing, deservedly or not. The problem as with all antipsychotics is the litany of side effects. The latest analysis from CATIE shows that there is a high risk with the atypicals for akathesia parkinsonism and dystonia as is seen with the older "typical" antipsychotics. I would not be surprised if that 80% that dropped out experienced one of these. If one is lucky to be free of these side effects or is able to tolerate them, Zyprexa can really clear the mind of delusions and depression. It seems to be one of those drugs of last resort. For me, nothing else worked (and I spent 15 years trying everything besides the atypicals). For me, the side effects of Zyprexa are far easier to handle than the symptoms of bipolar disorder. Sometimes you have to take the bad with the good.

John McManamy said...

Hi, Cretin. From what I have heard, Zyprexa and Clozaril are the most effective of the atypicals, but also the ones with the greatest side effects. If pdocs would spend more time with patients, my guess is the compliance rates would be a lot higher. For instance, if pdocs told their patients in advance that the med is a huge weight gain risk, then patients could - with physician help - could implement low-cal diets in advance instead of having to contend with huge weight gains further down the line.

I'm very glad the Zyprexa worked for you. Perhaps you can share your experiences, as this would help many other readers.