- Home
- About Us
- News
- Advocacy
- Mental Health Information
- NMHA Store
- Calendar
- Affiliate Network
- Support NMHA
- Search
|
|
||
Still Not Happy with CATIE |
||
|
By Mark Ragins MD
Since my reaction to the first CATIE article attracted so much interest and response and even ended up published, I felt obligated to carefully read the two new articles in the American journal this month and share my thoughts. I have to say I’m still disappointed by CATIE. I had such high hopes that this “real world” study would focus on people instead of illnesses. I have to remind myself that even the most ethical, well intended study is intended to collect data, not to treat people and that research, even when done by clinicians, creates a different human interaction than treatment. I just didn’t realize how much of the positive effects of medications would be lost when the human healing factors were changed. It seems that almost the only people who did well in CATIE were in the Clozaril group. This group was really very different than the other groups and not just because they received Clozaril; the human factors were also different. As I understand it they didn’t respond to, but tolerated, their first medication and then chose to be randomized into a Clozaril group. Since most of the poor responders chose not to have even a chance of getting Clozaril, I assume these people actually wanted Clozaril. They, and their doctors, knew they were getting Clozaril, reputed to be the most powerful, but most difficult to tolerate, even potentially lethal, medication we have. And they were seen far more often than the other patients because of the blood draws. As the authors note, any or all of those factors could’ve had a dramatic biasing effect causing them to stay in the study longer and benefit more, so it’s hard to know how good Clozaril really is. (I’m a big believer in the powers of positive expectations, hope, mutual perseverance, and human contact. That may well have been enough to explain their success. Too bad everyone didn’t get them. I wonder how well Seroquel would’ve worked if everyone would’ve thought it was Clozaril.) Beyond the Clozaril group, the outcomes were variations of dismal. Again, few people stayed on their medications or showed substantial improvement and many of them were hospitalized. The worst news is buried in the “Other adverse reactions” section of the second article where they mentioned in passing that two of the Geodon patients and one of the Risperidone patients committed suicide. I was surprised they didn’t comment on that tragedy. I seem to remember that even though we now have a black box warning about suicidality and antidepressants in adolescents that none of the thousands of study patients actually killed themselves. I began to wonder if anyone who didn’t get placebo or taken off medications had ever committed suicide while enrolled in a careful medication study before. I assume that if I had three suicides in my program in a relatively short period I’d have investigators breathing down my neck, but it’s never happened to me. One of the innovations in this study was to organize the data around treatment discontinuation instead of treatment response. The authors were right in their response to my first letter that I ‘m confused about what they mean by “discontinued treatment.” They seem to have elevated this concept far above what I’m accustomed to, using it as their primary outcome measure, dividing it into different subtypes, and even using it to triage people into further phases of the study. I would call learning how medications affect people both positively and negatively and both the patient and I making changes based on what we’ve learned “treatment,” not a series of “discontinued treatments.” Blinding, while useful to avoid biases in head to head trials, cripples our ability to learn from medication responses and may at least partially explain why so few people were treated effectively in this study. One message of CATIE may well be that to get good results for more than a few weeks doctors and patients have to experiment and learn together what medications help people the most with the least side effects. That collaborative process may be at the core of successful psychiatric practice. If so, it’s of great concern that collaborative learning has also been eliminated from clinics all over the country that provide only once every two month doctor visits. An accompanying editorial noted that it’s comforting that this study, financed by the government and conducted by highly experienced, well regarded researchers, more or less confirms the findings of more suspect pharmaceutical company sponsored studies. That may be true qualitatively – yes, Zyprexa does make more people gain weight than anything else does and Risperidone raises prolactin levels - but it’s far from true quantitatively. What happened to the nice 70% response rates and low 10% drop out rates I’m used to with the pretty graphs of how people get better? Why am I looking at graphs of people discontinuing treatment and tables of figures showing them not getting any better? If I didn’t know from my own experience that these medications work well for me I’d be far from comforted. If we assume that these medications do work well – and I do, though
I realize not everyone does – what’s going on here? We have
to remember that the CATIE study was a highly unusual, even courageous study.
It was a serious attempt to study real world patients in real world conditions.
To be fair, it was probably even more difficult in some ways than the real
world because of the elaborate sometimes blinded study protocol and the
need for so many measurements and complicated consents. The two possibilities
I’m left with for what’s going on are: 1) People in the real
world do much worse than I thought and we really do need to overhaul our
system, or 2) This study removed a great deal of the positive healing effects
of real world treatment without realizing it. Either way, the message is
clear to me: We need to get more serious about studying and using the human
healing that often accompanies medications. If CATIE is to be believed the
medications just don’t work very well alone. For additional information or to speak with an expert or individual
with mental illness, please call (703) 797-2588 or email mediainfo@nmha.org.
|
||
|
National Mental Health Association
|
