Remember when the New Yorker didn’t run letters to the editor? I’m guessing it was because Mr Shawn thought it would be vulgar, but whatever the reason, the practice frustrated enough people that there were, as I recall, a number of other places where letters were published, arch little columns. There was probably one in Spy, but who knows? This was before the Internet, I’m afraid, so I can’t be more specific or provide links. And given my multifarious character flaws, I may just be making this up.
Well, times have changed. The magazine has a letters page, and the website…well, the website would make Mr Shawn blanch, what with its open invitation to the hoi polloi to express itself ungrammatically, unconcisely, and with numerals. And my post on the elements blog has elicited quite a few comments (in addition to the surprising amount of email I’ve gotten, much of it vitriolic, much of that from psychiatrists, like the one reproduced here).
Now I suppose I could just reply on the New Yorker’s website, but that would be vulgar. It would convey that I am thin-skinned, that I care what the hoi polloi think, that I want everyone to love me and get my feelings hurt when they don’t, and I just don’t want my editors there to get the wrong idea. But I am all of those things, of course, so I must respond to at least a couple–as in the old days, in a different venue.
First, there are a number of comments about placebos and antidepressants, of which the best is surely this one
The canard that SSRI’s are equivalent to placebo, at least in moderate to severe depression, has been repeatedly and convincingly rebutted (1,2). Paul Krugman calls these sorts of arguments “Zombie lies” –
(1,2) are references to a couple of papers that reanalyzed the FDA data used by Irv Kirsch to show that half of the clinical trials run by drug companies for antidepressants failed to show a therapeutic effect of the drug, and in the successful ones the drug effect was just barely significant, statistically speaking. These papers argued that in moderate to severe depression, the drugs were effective, and that the problem was really that too many mildly depressed people were getting into the clinical trials.
Whatever the virtues of this argument (and it may have some, but it is hardly a conclusive rebuttal, let alone a killer argument), what’s important here is that while some people do conclude from the FDA data that SSRIs are “equivalent to placebo,” I’m not one of them, and in the New YOrker piece all I write is that they have “proven to be no more effective than placebos in clinical trials.” That is not a canard, let alone a zombie lie. It’s just what the FDA data show, no more, no less. I actually believe, as I have written elsewhere, that the SSRIs are powerful psychotropic drugs whose effects don’t show up well on trials with depressed people because their main effects aren’t on mood (which is the industry’s fault; they want the drugs to be antidepressants because that’s a good way to market them).
So why are defenders of the psychopharmacological regime so eager to say that I’m attributing an equivalency? Because they cling to those reanalyses like Titanic passengers to the lifeboats, and unless they can make it sound like I’m saying what I am not saying and don’t believe, and what is just plain stupid to say, then they are just left to flail about in the frigid waters.
Nice article, but it’s explicitly false that there’s “nothing new on the horizon.” Brain network analysis is finally putting a meaningful (although still incomplete) neurobiological face on mental disorders. For example, with depression, remarkable results are being achieved with ketamine (e.g., http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0044799), a broad glutamate antagonist that reduces cortical over-activity. Functional connectivity imaging has now strongly implicated over-active cortical control, rumination, and decision-making networks in depression (http://www.pnas.org/content/107/24/11020.short), and this can be addressed chemically or through other more direct means. Similar stories are emerging w/ other mental disorders.
I like this formulation: “meaningful neurobiological face on mental disorders.” It’s just spooky enough, no doubt inadvertently so, to function as its own critique of the neurobiological approach to mental suffering. Myself, I think the human face is meaningful enough, would much rather gaze into someone’s eyes than into their gray matter. But it’s the ketamine thing I want to respond to. Speaking of lifeboats, this ketamine research has been the great white matter hope for neurobiologists of depression. The research shows that a single dose of ketamine relieves depression. (THis, by the way, is another accidental finding–anesthesiologists have observed that depressed patients on whom they have used ketamine often emerge from the surgery feeling better.)
But two things: first, just because ketamine is active at glutamate receptors and modulates “cortical over-activity,” that, to paraphrase tje New York Times’s Richard Friedman, does not mean that glutamate or cortical overactivity are the cause of depression. Nor can a target as amorphous as cortical overactivity, whatever that is (and how do we know it’s “over-activity”? how much cortical activity is normal? will cortical activity become the new chemical imbalance?), be particularly useful to drug makers–I’m sure that Thorazine reduces cortical activity too. So it’s only the Prozac problem, shifted from serotonin to ketamine. The second thing is that no one in the ketamine world seems to want to talk about the psychological effects of ketamine, which has been known for decades as a psychedelic drug. Euphoria and other effects are sometimes noted as adverse reactions in the studies, but the scientists are so enamored of the notion that biochemistry must be the signal and psychology merely the noise that they ignore the experience of the patient completely. There’s a whole world of meaning in that elision; mayb there’s more going on here than brain chemistry. (I’d also point out that the other drug that seems to relieve depression in a single dose is psilocybin, and at least in those studies they don’t sidestep the question of the actual effects on the subject’s psyche.)
And the third thing: when this commenter writes, “this can be addressed chemically or through other more direct means,” am I the only one who gets a little worried?
Psychiatry is hardly unique when it comes to serendipitous discoveries. The same can be said of virtually every other therapeutic area. What makes the “chemical imbalance” theory compelling is not its scientific foundation but its rhetorical appeal. The same can said of the LDL-HDL (bad cholesterol, good cholesterol) theory. What Greenberg and other sophisticated skeptics often lose sight of, however, is that for many — myself included — these drugs have made an extraordinary difference. For patients like me, the question is not so much “how these drugs may work” but rather “do they work for me?”
It is, of course, true that every medical field has its accidental discoveries. Only in psychiatry, however, have none of those accidents eventually been explained. More significant is the point about how the drugs work. It is of course not fair to say that I have lost sight of this fact, although it is true that I don’t directly address it in the New Yorker piece. (And interesting that in the absence of an explicit statement to the contrary, the assumption is that I think psychiatric drugs don’t make a difference.) But what the commenter points out is one of the deeper scandals of the psychopharmacological era. The researchers have been so busy trying to fit the empirical evidence into the magic bullet model that they have failed to give us a solid account of what the drugs do for people. Of course, and apologizing for two Titanic references in one post, that would steer them dangerously close to the recreational drug iceberg, for they would quickly find themselves talking with people about how the drugs changed their sense of who they are.