in my opinion, the placebo effect happens because someone cares enough to ask one is feeling, and is genuinely interested in the response.

I've often wondered if a drug's side-effects enhance the placebo effect as well, where patients who note the side effects are more likely to believe the drug is kicking and addressing the depression, and more likely to report improvement. It seems that could easily explain the remaining 14% improvement attributed to the drug.

I've opined many times the need for an active placebo in such studies.

The entire CNS drug development arena is a mess. The study designs are flawed in almost every case due to lack of an active placebo arm, washout issues, and ridiculous efficacy measures. Compared with cardiovascular/metabolic/anti-infective and even oncology, CNS is joke science at best.

I seriously doubt any real progress will be made until psychiatry ceases to be the "study of psychotropic drug effects on the brain". Personally, I don't see this happening. Too much is vested in pharma and govt money, the law, and psychiatry's need for legitimacy (note people aren't a concern here...)

A step in the right direction would be black box warnings on all psychotropics and a ban on off-label use.

doctors can't even tell you what depression is so how can they treat it?

Another way to look at it is this:
21% of people actually need real help, that can't be provided by a placebo. 14% were not helped by a placebo, but were helped by a real anti-depressant. 7% were not helped by an anti-depressant, either. So an anti-depressant actually helped two-thirds of people who needed help, which isn't a terrible result.
The key may be figuring out in advance the 79% of people for whom a placebo is sufficient.

Whoo Hoo now THAT looks like fun.

JT

The size of the placebo effect in this experiment is unclear because there was no untreated control group. Maybe 79% of the subjects would have gotten better without any treatment.

I think Paul is on the right track - black box warnings on all off and on label uses, and a ban except for proving it is the drug of last resort on those that hit a certain lethal ceiling criteria.

We did pharmacogenomics studies looking at placebo effect and there is a population that is more likely to respond to placebo. You can google "placebo respons LL LS" and find a variety of things. It's basically useless imo.

I don't believe you can run a proper study. It might be possible to design a proper study, but you can't find case matched controlled patients, the efficacy measure are utter BS, and the it's just not possible to washout as you might with other drugs. Lastly, the underlying scientific premise is flawed, so penultimately it's useless and farcical.

obvious article is obvious

I'm wondering where you get that "usual 25 to 30% (effect size of antidepressants over placebo) in shorter antidepressant trials." Here's an article by Danny Carlat that points to only a 10% difference between antidepressants and placebo (http://www.nytimes.com/2007/11/25/magazine/25memoir-t.html; also available on ssristories at http://www.ssristories.com/show.php?item=2203). Frankly the difference seems to be all over the map depending on the length of the study, what other confounding drugs are allowed, and all sorts of other factors. And yes, until biopsychiatry took its iron grip on mental health care, most people considered depression, even serious depression, to be self-limiting. With the adverse effects that characterize antidepressant use I'm surprised that placebo doesn't outperform antidepressants all the time, both short term and long term, and I think the fact that antidepressants sometimes do outperform placebo is a testament to the power of suggestion (especially in a clinical trial) and the short term stimulant effect they have.

sara, the 25 to 30 percent effect size is pretty much the consensus among researchers i've talked to on the matter, and it's what the turner paper from earlier this year came up with and i consider that one of the best evaluations of anti-dep performance ever since it was based on many, many thousands of people in clinical trials. the main part of star-d also comes up with an effect size of about 25 to 30 percent. as such i use this figure as for go-to for this issue, especially since it doesn't make me look particularly radical to say anti-deps work in one of four patients in short term trials (the true believers think i am a murderer for not telling people that the effect size is like 60 percent). i know some studies show a lower effect size and i report on them when i see them and i am pondering revising downward the standard effect size i use when writing here.

I am technically supposed to be too busy to be commenting on anyone's blogs but I had to respond to this.

I have known in my gut all along this placebo effect is what was happening in the majority of folks.

The therapeutic effect of antidepressants is being told you are taking a magic pill that will anti your depression. It is suggestion and belief.

The patient trusts that the information is true, has faith that their doctor is trying to help them and is swayed by the knowledge that such meds have helped others with the same problem. Is swayed by the knowledge that the drug is science based medicine proven in studies.

Suggestion>Assumption>Belief>Faith> The formula for placebo healing is laid right out by putting your head care in the hands of those supposed mental health wizards.

I have found myself guilty of falling for it too Philip. Although I was a cynic when it came to psych meds I did subject myself to a bunch of other things.

I actually tried crystal and gem healing believing that they would help increase my energy by raising my vibrations and heal my blockages and all that new age stuff.

For a few weeks I felt great. Buoyed by the knowledge that I was using *ancient shamanic medicine* I knew the medicine bag and crystals around my neck contained major mojo.

Then reality hit home after awhile and depression came back.

I realized that during the two-three week period of time that I thought I was doing something to heal myself that I became energized, open to possibility, I had faith that it would work.

I knew it would work and I knew that I would get better as a result. I lined up all the factors needed for the placebo effect and managed essentially by positive thinking to temporarily lift myself out of depression under my own power by changing my thoughts moment to moment.

That's when I realized the power to heal could and would come from within and not from without.

I have seen this placebo effect occur in all manner of alternative medicine and new age healing stuff too.

For the same reasons that psych patients trust their Pdocs, people trust their spiritual healers and guinea pig all sorts of remedies and cures and they feel that glow in their hearts when they start treatment that they are using magic to heal themselves. Hey myself being one of them!

It turned out I was right in the end at least as far as my own depression since it's been gone now for years.

So much for the healing power of meds and crystals.

A few remarks on the Khan et al study.

The study reports that "Significantly more patients entered the continuation phase from the antidepressant group than the placebo group" (p. 793): 35% response in placebo group versus 51% response in antidepressant group.

The study also reports that "Significantly more patients in the placebo group relapsed in the continuation phase as compared to the antidepressant group" (p. 794): 79% of placebo responders compared to 93% of antidepressant responders did not relapse during the continuation phase.

So, according to this study, antidepressants are still significantly better than placebo. It's uniqueness lies in its use of trials that continue the placebo arm throughout the trial, and they have established that the placebo effect is not transient as previous supposed. However, in terms of efficacy, antidepressants are still better.

We all know that antidepressant efficacy is 50-80% placebo effect, but the statistics indicate that there is also a significant effect of the active medication over and above placebo, especially in severe depression over the long-term.