Posted by Racer on January 5, 2008, at 18:27:25
In reply to Re: STAR*D confirmed what patients already knew, posted by linkadge on January 5, 2008, at 15:24:47
> >The methodology could not include placebo. It is >an ecological study. Has your doctor ever >offered you a placebo? Do you consider placebo >treatment to be a common clinical practise?
>
> Well, then change the methodology. Its not so much that doctors can offer placebos, but if they knew that medications were essentially no better than placebos, they might decide against riskier treatments in favor of other treatments. For instance, they might choose lower (safer) doses, or choose drugs with a more benign side effect profile.>
> >The methodology was not designed to show what >works. It was designed to deal with non->responders.
>
> Well, strong response to a placebo might have helped support the finding that pretty much all choices are essentially equivilant.
>
> I do think it really matters that doctors know the true incidence of placebo response.
>
> >These variables may go some way to predicting >responsivity, something that the artificial >construct of a placebo-controlled efficacy trial >can never show.
>
> You could include these same types of analysis in a study that includes a placebo. It might even help to uncover exactly what factors predict placebo response.
>First of all, what you say in that last short paragraph is right: it might be useful to know what factors predict a placebo response, and how strong that response might be. That's a good start for another study some day, once someone decides how to do it.
As for your very first sentence, though -- that the methodology should have been changed in this study -- I disagree strongly. This study was done to explore treatment algorithms for those who fail to respond to a trial of a single first line anti-depressant. That is the purpose of this study, and the methodology used seems appropriate to me for that purpose. This was not a study of the placebo response in depression treatment.
Your suggestion regarding lower doses directly contradicts what the study found: that in many cases, increased doses of tolerable medications increased response. That non-response may be a question of inadequate dosing in some (or even many) cases. As far as I'm concerned, that's something good to know.
As for the basic premise that response to antidepressants is no better than response to placebo, I'm going to argue that.
First, as Dinah said, most of us can point to effects of these medications. Many of those effects do mitigate our symptoms, even if they do not "cure" the disorder or even result in complete remission. Still, if the effects of the medication affect our symptoms, isn't that response to the medication? Or do you still consider that to be a placebo response because it may not have resulted in complete remission?
It is your stated opinion that anti-depressants are no more effective than placebo. Regardless of what studies you can cite to support this opinion, it is a valid opinion, and it is your right to hold this opinion -- regardless of whether it is right or wrong in fact. In the opinions of others, however, antidepressant medications are more effective than placebo. Some of the others who hold that opinion are researchers, doctors, other patients who take those medications, etc. It is the right of those who hold the opinion that antidepressants are more effective than placebo to hold that opinion. It is also a valid opinion. Perhaps their opinion is wrong.
Perhaps your opinion is wrong.
I'd also like to point out that the whole "placebo response rate" question involves statistics. In a large, double blind, placebo controlled study, the placebo effect may be higher than some pharmaceutical companies might like. In the real world, though, most of us don't care if 30% or 70% of the participants in a research study responded to a particular drug: we're looking for a 100% response rate with an n of 1. Do *I* respond to a particular medication? That's quite a different question, don't you agree?
I know that I have experienced placebo responses, good and bad, to medication. When I felt that a doctor was not hearing me, not listening, not treating me with respect -- and I've had some bad experiences with doctors in my life -- I have not responded well to certain medications. I've tried some of those medications again with similar results, although not as thoroughly negative as the first time around. I know that tolerability, for me, is strongly influenced by my relationship with my doctor. That is a placebo response. The outcome is the same, though, either way: either I feel better or I don't. That's what matters to me. And that hasn't been influenced, as far as I can tell, by how well I get along with my doctor.
Frankly, I'd be inclined to cheer that anyone has done a real world style study of the way depression is treated, if only because it's a nice relief from the usual sort of short term study one usually sees. Is this particular study perfect? I don't think so, because I don't think it's possible to create a perfect study when you're dealing with the diversity of the population being studied. For that matter, I'm not sure what sort of study could be done perfectly in anything involving any living creature, just because of the diversity involved.
What say that it would be nice to have a study into the value of the placebo effect in treatment of depression, with especial interest in the factors which predict response to a placebo, and that we all fervently hope that someone designs and carries out such a study at some point?
And then maybe we can look at STAR*D and say, "wow, someone finally did a study into treatment algorithms for depression that more accurately reflected real world treatment." Regardless of one's opinion regarding the efficacy of placebos in the treatment of depression, or one's opinion regarding the uselessness of antidepressant medication for the treatment of depression, isn't it nice that someone did study the outcomes in a longer term design, with emphasis on what to do next if the first choice didn't work?
I think I've had my say for now.
poster:Racer
thread:804126
URL: http://www.dr-bob.org/babble/20080105/msgs/804507.html