One of the things that seems to really get psychiatrists’ dander up about my book is my contention that the invention of antidepressants leads to the invention of depression. They argue back, sometimes impolitely, that antidepressants were invented by doctors seeking to treat a common and under-recognized scourge that has afflicted people in all times and places. But in a recent study, researchers inadvertently support my hypothesis by proposing precisely what I suggest: that psychiatric diagnosis be changed to reflect what drugs are doing.
First, a little recap. While it appears to be true that some portion of the population (I would say a much smaller portion than falls under the diagnostic criteria for major depression, dysthymia, adjustment disorder, etc.) suffer from the “inexplicable unhappiness” that Hippocrates observed, it is also true that antidepressants weren’t created to treat those people. Rather, as my book recites in painful detail, they were discovered when people took new drugs and unexpectedly felt better than well. The allure of such drugs is obvious. What is less obvious is how to market them in a society reluctant to let people simply take drugs to feel better—a problem exacerbated by the Marsilid scandal of the 1950s (an early “antidepressant” that killed a few people and came to be denounced as a “pep pill”) and by the Valium/Xanax scandals of the late 1960s and early 1970s, in which public scolds, some of them Congressmen, railed against the widespread use of minor tranquilizers and denounced users as little better than potsmokers (and their doctors as drug pushers). As a result, for many years, antidepressants wandered around in the pharmaceutical wilderness, a treatment in search of a disease.
When various historical forces—the FDA’s insistence that drugs be specified for particular diseases, the American Psychiatric Association’s turn to a descriptive psychiatry, Nixon’s War on Drugs, the invention of SSRIs—converged to anoint depression as that disease, antidepressants, as we all know, hit it huge, and suddenly there was an epidemic of depression. But there remained a problem for the drug companies: as I’ve described here, the connection between the drug and the disease has always been tentative and contingent. The poor performance of antidepressants in clinical trials is probably in part the result of trying measure them with a test (the Hamilton Depression Rating Scale) that is poorly matched to the actual effects of the drug. Sticking with the Hamilton bolsters the case that these drugs are antidepressant medication, but at the high cost of amassing data indicating that the drugs are not effective—data once suppressed but now embarrassingly public.
The predictable backlash—claims that antidepressants are simply tricked out placebos—misses the real point, which is that researchers have been so intent on insisting that the drugs cure depression, and cooking the books accordingly, that we don’t have any systematic understanding of what antidepressants actually do. But that doesn’t mean there isn’t an emerging consensus. Consider the data recently mined by a group of Canadian scientists. They looked at an epidemiological study of over 20 thousand people, of whom 1441 had been prescribed antidepressants in the past year. Only 718 of them, fewer than half, met the criteria for a psychiatric diagnosis within a year of their prescription. And one quarter of them didn’t qualify for a diagnosis at any point in their lives. These results correspond nicely with those of a group of French data miners. They looked at insurance company records and determined that 53 percent of patients who received antidepressant prescriptions had not been diagnosed with any of a long list of disorders for which antidepressants are indicated or used off-label. So it may not matter as much as you think that antidepressants don’t work for depression or their other official or unofficial indications, because most of the time, that’s not what they are being used for.
So what are the actual uses of antidepressants? Why are people getting the prescriptions and taking the drugs? While they don’t meet diagnostic criteria, still, according to the Canadian researchers, “results suggest that these individuals …have other forms of mental health difficulties and needs for mental health treatment that may not be recognized in the DSM-IV.” In other words, people aren’t just randomly showing up at doctors’ offices and emerging with prescriptions for Prozac; they are “experiencing difficulties” with which they and/or their doctors think the drugs will help.
But what kind of difficulties? Another group of researchers suggests an answer. They gave Paxil to 120 depressed people, placebos to sixty, and cognitive-behavioral therapy to another 60. Then, instead of just comparing their Hamilton scores, they also measured changes in the subjects’ personalities. As expected, there wasn’t much difference in the effects of the various treatments on depression. But Paxil changed people’s personalities much more than the therapy or the placebo. Paxil takers scored lower on something called Neuroticism and higher on something called Extraversion than they had at the beginning of the trial, while the other groups’ personalities remained stable. The strongest effect of antidepressants, in other words, is on your personality. The antidepressant effect, the researchers suggest, is secondary, a side effect due to the drug making you happier with who you are.
Of course, that’s the last thing the drug companies want you to think. They’ve spent an awful lot of money reassuring the public that their drugs don’t change our personalities. If that’s the main effect of the drugs, then they’ve got some explaining to do. Or at the very least, they’d better get busy inventing a new disease for their drugs to cure—and it better be a better disease than Prozac-deficit disorder. And here, to get back to the point I started with, is where the Canadian researchers are offering some help. Lest their their study lead to a dangerous conclusion—that “the prescribing of antidepressants is appropriate among individuals lacking psychiatric diagnoses”—they suggest that “the nosological approach should take lower level psychiatric symptoms into consideration so that treatment options can be studied and administered.” In other words, psychiatrists should do exactly what they claim they never do: rejigger their diagnoses to accommodate the drugs they prescribe. Better they should lower the bar to entry than admit what is really going on. Better they should pathologize more of us than to relinquish some of their hard-earned (if ill-deserved) power.
Excellent observation – certainly ties up a lot of loose ends with these drugs. But as an extension, to encapsulate Whitakers work on long term problems for perhaps 15% of people on these drugs, isn’t it possible that for these individuals there is a overall negative effect on their personality traits eg. apathy, emotional blunting,compromised libido etc. such that they end up much worse off? Hence to take these drugs is really to take an unpredictable gamble on how ones personality traits may be rearranged. While individual patients, perhaps a pluarality, may report a benefit, the extent and severity of negative outcomes experienced by the minority represents I believe, as with cigarettes, an unacceptable societal cost.
That’s an interesting way to look at it–that the overall costs of widespread antidepressant use are higher than the benefits, even if only a relative few people pay those costs.
In Manufacturing Depression, I report on a Harvard economist’s study, in which he comes to exactly the opposite conclusion: that the benefits of antidepressants are so great that even if everyone were given them, we’d collectively be better off. Of course, his major parameters were financial costs.
Here, however, the question is different. There are really two kinds of costs (other than monetary) that can be attributed to antidepressants. First, there are the side effects, especially the sexual blunting and the withdrawal effects. Second are the effects on personality.
I assume you’re talking about Robert Whitaker’s Anatomy of an Epidemic. I reviewed that book for The American Scholar, but I don’t remember his discussing personality effects of antidepressants. He seemed much more concerned with more frank neurological problems, and with people suffering from schizophrenia and bipolar disorder. Nonetheless, I do think his analysis is interesting here. But we have to bear in mind that the negative effects of antidepressants, unlike those of antipsychotics, are largely ideological. Or to put it another way, you don’t have to make controversial assumptions about how we are supposed to function to think that tardive dyskinesia or worsening psychosis is a bad problem. But you do have to do this to think that the kind of personality change wrought by SSRIs–blunted affect, detachment, unrealistic resilience, etc.–is a negative one. Given the large numbers of people who put up with side effects to get those results, I think it’s clear that many people think tha=t these are positive changes. The marketplace has spoken.
So it’s up to critics to first analyze exactly what it is that antidepressants are doing, and then to argue why it is a bad thing–a much less straightforward exercise than with cigarettes and lung cancer.