ADVERTISEMENT
Unanswered Questions: What Psychiatry May Lose in the Aftermath of the New NIMH Rules
In a recent blog post, the National Institute for Mental Health (NIMH) announced that it will no longer fund clinical trials that don’t simultaneously examine the biological mechanisms underlying the disorder the treatment is designed to address. In other words, just showing that something reduces symptoms will no longer cut the mustard and won’t get researchers like me the cash.
Our research group has spent years using clinical trials as probes into the underlying biology of depression, so I’m not displeased by the NIMH proclamation. But I have a strange nostalgic sadness for all the crucial clinical questions that will likely remain forever unanswered now that biology has been mandated to take center stage in the world of mental health research
Let me say it differently. Forget the high science. I just wish we had a better idea of what we are doing when we give a depressed patient an antidepressant.
Let me bring this point home with an example of how, after thousands of trials, we don’t know what might be the single most important thing when it comes to treating depression with medications.
First some quick background. About a year ago, charged with the task of creating a CME talk on the treatment of depression, I decided to take the occasion to really scour the literature to see what we do and don’t know about treating major depressive disorder (MDD). I’m embarrassed to say how shocked I was by my ignorance—embarrassed because I’m supposed to be an expert.
My exhaustive literature review lets me say this: here is something we actually know. The old idea that antidepressants take six to eight weeks to work is dead wrong. Neuroimaging studies show that antidepressants impact brain function almost immediately, and clinical trials show that the biggest difference in response between antidepressants and placebo doesn’t occur at six or eight weeks, but rather at one week and two weeks. In fact, 60% of the overall improvement that occurs with antidepressants happens in the first two weeks of treatment (1).
Here is why this matters so much: improvement in the first couple of weeks of treatment is by far the strongest predictor of who will and will not ultimately go into remission with the antidepressant they are taking. The size of this effect is really staggering. For example, a study of more than six thousand patients found that people who don’t show 20% improvement in symptoms at the end of the first two weeks of treatment have only a 4% chance of going on to achieve a stable remission (2).
These findings—found in study after study—suggest that if someone starts responding to an antidepressant within the first couple of weeks that nothing further need be done except continue to monitor to ensure that a good clinical response is achieved. But what we really want to know is what we should do when someone doesn’t show a good early response.
And this is exactly what we have no idea about, as far as I can tell. We don’t know whether early non-response is a call to action for us to change what we’re doing or a sign that this is someone we’re not going to be able to help with an antidepressant. We don’t know if we should increase the dose of the antidepressant, add something, change the antidepressant, or just fatalistically hope for the best, against the odds.
As far as I know, here is the very straightforward study that has never been done (and if it has and I’ve missed it, please respond to this blog!). Start subjects on an antidepressant and treat for two weeks. Take everyone who has not had a 20% improvement and randomize them to stay on the same dose, increase the dose, switch to another antidepressant, or add an augmenting agent. Follow the subjects for another four weeks and see which strategy wins and whether anything predicts whether any given type of individual will do better with one strategy vs. another.
If one strategy won, then we’d know in a general sense what to do when people don’t get on the “fast responder track”. If nothing was any better than anything else, then we’d know that people who don’t have an early response to an initial antidepressant are people not likely to benefit from pharmacotherapy.
As I type this, I am flying into Washington, DC, home of the NIMH. I can see the Capitol and the Washington Monument out the plane’s window. I understand as well as anyone how desperately we need high science if we are going to move forward in any real game-changing way in psychiatry. But I hope we don’t as a field make the “perfect the enemy of the good” as Obama was wont to say early in his tenure as commander-in-chief. I hope we find a way to continue exploring the type of down-and-dirty clinical questions that we so desperately need to have answered.
And I hope that I am wrong and that I somehow missed the study that would guide us on what to do for people who don’t have the early antidepressant response that is so essential for long-term good outcomes. If anyone knows of such a study please respond and we can make that my next blog topic.
References
Charles L. Raison, MD, is an associate professor in the Department of Psychiatry, College of Medicine and the Barry and Janet Lang Associate Professor of Integrative Mental Health in the Norton School of Family and Consumer Sciences, College of Agriculture and Life Sciences, University of Arizona, Tucson, AZ. He is also the behavioral health expert for CNN.com, and he is a Psych Congress Steering Committee member.
The views expressed on this blog are solely those of the blog post author and do not necessarily reflect the views of Psych Congress Network or other Psych Congress Network authors.
