Cochrane review of antidepressants vs placebo

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

BabyPsychDoc

Full Member
10+ Year Member
15+ Year Member
Joined
Apr 22, 2007
Messages
622
Reaction score
1
came out recently. Funny how there is no press coverage of this one, especially if you remember how much fuss there was about a (rubbish) study claiming "antidepressants are no more effective than placebo" last year...

http://www.mrw.interscience.wiley.com/cochrane/clsysrev/articles/CD007954/frame.html

Members don't see this ad.
 
The conclusion is that SSRIs are "effective in the treatment of depression" but the NNT is 7, the duration of all the studies was only 6-8 weeks, and the measured outcome a reduction of symptoms on an invented and questionably validated scale.

I still have to wonder HOW effective are they: do the apparent reductions in symptoms as measured in the studies make a difference to the patient's subjective experience, and if so are any of these results durable?
 
My old program had a grand rounds speaker who brought in data from several other medications that had just the same problems with SSRIs such as the statins, yet those meds didn't get the media craze.

It also turned out that the media blitz to present SSRIs as ineffective was to some degree mushroomed because the Churh of you know who helped to push that message along. Several news outlets also had psychiatrists talk about it who were clinicians, and not researchers, and didn't have much data to counter back because they are not actively engaged in research. I'm sure those docs could've mentioned to you about their patient's sexual side effects, but had little else to say. I bet several of these clinicians were just so happy to be on the national news, they jumped at the oppurtunity without thinking of their responsibility to educate the public.

And an important question that was not addressed by the article that brought up this SSRI issue was that several studies that were not considered by the FDA may have been pulled out because they were not properly conducted. The paper simply pulled in all the data in other studies that the FDA did not consider, but they did not investigate the reasons why.

I completely admit that several pharm companies do a few studies, and only one of them passes the FDA test while others did not, and that med gets the approval, however like I said some of the studies that wouldn't have passed weren't done properly. Also with 95% statistical standard, you are going to get plenty of studies that didn't pass that mark, but would have passed the 90, 85, or 80% mark. Does that mean the medication doesn't work when you get other studies that say there was significance at 90%?

IMHO a better way to look at SSRIs is the STAR*D which didn't use the FDA measures, but took a very large body of subjects and followed them over time. It clearly indicates that SSRIs are not going to help everybody, nor get most people back to where they were depressed, but it does show that in several people, antidepressants do work, and should be considered as a treatment option.
 
The conclusion is that SSRIs are "effective in the treatment of depression" but the NNT is 7, the duration of all the studies was only 6-8 weeks, and the measured outcome a reduction of symptoms on an invented and questionably validated scale.

I still have to wonder HOW effective are they: do the apparent reductions in symptoms as measured in the studies make a difference to the patient's subjective experience, and if so are any of these results durable?

I completely agree with whopper's post above -yes, NNT is 7, but we do not expect the psychotropic medication to help EVERYBODY, there is just so much about the brain chemistry that we do not know yet. They do help many people, though.

The duration of studies was 6-8 weeks - but this is the time when we usually expect to see some effect in most SSRIs/TCAs, so I do not see this as being problematic.

The scales you mention do measure some of the patients' subjective experiences.
 
Top