You can stop feeling like it's your fault that you haven't found an antidepressant that works. Not that you should have ever felt that way anyway, but most depressed people blame themselves for everything that goes wrong in their lives.

But this one is definitely not your fault, and probably not your doctor's either, for that matter. Last week a paper published in The New England Journal of Medicine showed that publication bias put antidepressants in a more positive light than was warranted by the evidence. To put that in plain English, when drug makers ask the FDA to approve a drug, they have to submit all studies that have been done for that drug, even the ones that show the drug to be less than successful. However, they are under no obligation to publish all the studies in medical journals. They can, and apparently did in this case, cherry-pick the studies and only submit the positive ones for publication. Not only that, some of the negative studies that were published were slanted to make them appear more positive.

Here's the bottom line: if you looked at only the reports that were published, you would see that in 94 percent of the studies, antidepressants performed better than a placebo. But if you added in the reports that the FDA saw, but the public didn't see, that drops down to about 51 percent.

I can promise you that every medical professional who has been prescribing antidepressants is having one of two reactions to this report (or perhaps both). They are incredibly frustrated that they did not have all the information available with which to prescribe antidepressants. These studies are some of the most useful tools they have at their disposal when it comes to making prescribing recommendations to their patients, and they've been relying on them heavily.

At the same time, they are relieved, because they didn't understand why antidepressants were failing with some patients, when, according to the studies, they should have been working. After all, most of the studies that were published showed positive results.

To me, there are several very troubling issues that this revelation brings to mind. First of all, if the FDA saw the studies that demonstrated that the antidepressants were ineffective, why were the drugs approved? My faith in the FDA's impartiality has sunk even lower than it was before.

Secondly, I seriously doubt that this selective publication happens only with antidepressants. I hope that someone does the same research on other classes of drugs. The results would in all likelihood be sobering, but at least we'd have the entire picture.

Now let me make one thing clear - antidepressants can, and do, successfully treat depression in many cases, including mine. I've been on three different antidepressants in the last eighteen years, and two out of the three worked very well for me. I'm thankful every day that they exist. I know that for me they are definitely not sugar pills. I've had my depression return when I took less-effective generics. However, I know a lot of people who have been frustrated because they and their doctors couldn't find an antidepressant that worked for them.