Daniel Carlat had a strange response to the review that Marcia Angell wrote in the New York Review of Books of his book. Despite what he wrote in his book, that psychoactive drugs are hardly better than placebos, in his response to Marcia Angell, he now says the majority of psychoative drugs are robustly better than placebos, which flies in the face of the evidence, as Angell points out below. He even seems to be critical of Angell, who, on the whole, wrote a rather kind review of his book. What I think is going on with Carlat is rather simple. It’s one thing to fearlessly say what you want in your own book, it’s another thing to see it in print in someone else’s review, particularly when Carlat said many things in his book that are deeply unflattering to his own profession. I think Carlat may be beginning to panic about his standing with his peers. He espouses that psychiatry, as many practice it, is money grubbing in the extreme and there is dubious science behind the drugs that are prescribed. This is probably controversial only with his peers.
Rather than reprint Carlat’s response, I have reprinted Angell’s rebuttal below. (Marcia Angell is the former editor in chief of The New England Journal of Medicine.) I apologize that this post is so long, but the message is important.
Marcia Angell responds to Carlat et al.
All three of these letters simply assume that psychoactive drugs are highly beneficial, but none of them provides references that would substantiate that belief. Our differences stem from the fact that I make no such assumption. Any treatment should be regarded with skepticism until its benefits, both short-term and long-term, have been proven in well-designed clinical trials, and those benefits have been shown to outweigh its harms. I question whether that is so for many psychoactive drugs now in widespread use. I have spent most of my professional life evaluating the quality of clinical research, and I believe it is especially poor in psychiatry.
The industry-sponsored studies usually cited to support psychoactive drugs—and they are the ones that are selectively published—tend to be short-term, designed to favor the drug, and show benefits so small that they are unlikely to outweigh the long-term harms. The problem with relapse studies, like that of John Geddes, which is cited by Friedman and Nierenberg, is that they don’t distinguish between a true relapse and withdrawal symptoms that result from the abrupt cessation of drugs.
Both the pharmaceutical industry and the psychiatry profession have strong financial interests in convincing the public that drug treatment is safe and the most effective treatment for mental illnesses, and they also have an interest in expanding the definitions of mental illness. Even Dr. Carlat, whose excellent book I reviewed, admitted that he and other psychiatrists make nearly twice as much money prescribing drugs as providing talk therapy. In his letter, which seems somewhat inconsistent, he states that the “unequivocal, if perplexing truth about psychiatric drugs” is that “they work” (his italics), and that all the major psychoactive drugs “are robustly more effective than placebos in double-blind controlled trials.” (In fact, the trials yield varying outcomes, many of which fall far short of robustness.) But elsewhere in the letter, he says, “There is no question that among the medical professions, psychiatry is the most scientifically primitive,” and in his book, although he claims anti- depressants work, he comes close to Kirsch in concluding that “much of this response is undoubtedly due to the placebo effect.”
Carlat mischaracterizes Kirsch’s work by suggesting that he contradicted himself. Kirsch did indeed find that the six antidepressants he studied were more effective than placebos, but the difference was very small (similar to the difference found by Turner and his colleagues, in the study cited by Carlat). Kirsch then speculated that even this small effect might not be real, because patients who received the antidepressant instead of an inert placebo would experience side effects that might enable them to guess that they were receiving an active drug, and therefore might make them more likely to report an improvement in their depression. In support of this hypothesis, Kirsch pointed to a few trials employing placebos that themselves had side effects, where no differences were found between drug and placebo. But despite the persuasiveness of his theory, Kirsch acknowledged that it remains to be proven.
The UK’s National Institute for Health and Clinical Excellence (NICE) develops treatment guidelines for the National Health Service on the basis of benefits and costs. It concluded that because improvements in the 51-point Hamilton Depression Score (HAM-D) of less than three points are not clinically discernible, antidepressants that on average fail to provide at least that level of improvement could not be recommended. While that cut-off is indeed arbitrary, as Carlat says, so are many other conventions in medicine, e.g., the number of symptoms required for a diagnosis of a major depressive episode or the accepted standard (P less than 0.5) for statistical significance. The NICE cut-off strikes me as eminently reasonable. Friedman and Nierenberg point out that a reanalysis found a 2.68 point difference instead of a 1.8 difference, but that is still below NICE’s threshold for clinical significance.
Contrary to Dr. Oldham, I did not say that mental disorders were invented in order to create a market for psychotropic drugs. What I did say is that the boundaries of mental illness are being stretched for a variety of reasons—to increase drug company sales, to enhance the income and status of the psychiatry profession, and to get insurance coverage or disability benefits for troubled families. It may be that, as Oldham says, the disorders that these medications treat have been around for all of recorded history, but they weren’t necessarily considered “disorders,” rather, simply emotional states or personality traits. Just as a cigar is sometimes only a cigar, unhappiness might have been considered just that, not a medical condition.
The letter by Drs. Friedman and Nierenberg is filled with inaccuracies and assertions masquerading as fact. They are simply wrong in asserting that psychiatry, in using drugs to treat signs and symptoms of illness without understanding the cause of the illness or how the drugs work, is no different from other medical specialties. First, mental illness is diagnosed on the basis of symptoms (medically defined as subjective manifestations of disease, such as pain) and behaviors, not signs (defined as objective manifestations, such as swelling of a joint). Most diseases in other specialties produce physical signs and abnormal lab tests or radiologic findings, in addition to symptoms.
Moreover, even if the underlying causes of other diseases are unknown, the mechanisms by which they produce illness usually are, and the treatments usually target those mechanisms. For example, we may not know what causes arthritis, but we do understand a great deal about the mechanism, and we know how anti-inflammatory agents work. Even when there are only symptoms, such as nausea or headache, other medical specialists, unlike psychiatrists, would be very reluctant to offer long-term symptomatic treatment without knowing what lies behind the symptoms.
Contrary to Friedman and Nierenberg, I do not “deny that depression has any biological basis at all.” I know very well that all thoughts, emotions, and behaviors have their origin in the brain. But it is a great leap from recognizing the obvious fact that mental states arise in the brain to knowing why and how they arise. Friedman and Nierenberg make much over recent advances in neuroscience research, but so far this research hasn’t produced much improvement in diagnosis and treatment.
In fact, Allen Frances, the chairman of the task force that wrote the current version of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), opposed undertaking the ongoing revision because he thought there had not been sufficient new data on the biological causes of mental illness to justify a new edition. As for the chemical imbalance theory of depression being a straw man, I still hear it invoked frequently. Even Oldham seems to entertain it in his letter, saying “…there is no consensus on whether these imbalances are causes of mental disorders or symptoms of them.”
Friedman and Nierenberg are right that the National Comorbidity Survey showed very little change in the prevalence of three particular types of mental disorders in adults between 1991 and 2003, although the increase in the percentage of people treated was dramatic. But the frequency of some diagnoses, such as bipolar disease and autism, has soared. Moreover, the survey showed a prevalence of mental illness of about 30 percent, which surely represents either a major epidemic or rampant overdiagnosis. One of the most remarkable findings was that 20 percent of randomly selected adults were undergoing treatment for emotional disorders at the time of the later survey, about half of whom did not even meet the DSM-IV criteria for a mental disorder.
Friedman and Nierenberg refer to the death of Rebecca Riley, who was diagnosed with bipolar disorder as well as ADHD when she was just two years old, as a “tragic anecdote.” While that is true, I believe it should also be seen in the context of the extraordinary epidemic of juvenile bipolar disease that was stimulated largely by the teachings of some of Dr. Nierenberg’s colleagues at the Massachusetts General Hospital. Three of them were recently disciplined by the hospital for not having disclosed some of their hefty payments from drug companies.
If readers check the NYR website, they will see that Dr. Nierenberg discloses his external sources of income, which include consulting arrangements with some of the major manufacturers of psychoactive drugs. While I am not in a position to, and will not, comment on Dr. Nierenberg’s consulting work, it seems to me that in general, one of the risks of close collaborations with industry is that even the best of physicians might develop an insufficiently critical attitude toward a company and its products, as well as to pharmacologic treatment generally.
Dr. Friedman seems to agree. In a review of a book by Alison Bass, published in The New England Journal of Medicine (June 26, 2008), he refers to the handsome payments by drug companies to physician researchers who test their drugs, and goes on to say, “Bass’s riveting and well-researched account of these disturbing ties should be widely read by members of the medical profession, many of whom continue to believe, despite all evidence to the contrary, that they are immune to the influence of drug companies.”
Finally, Friedman and Nierenberg accuse me of downplaying the devastating consequences of untreated psychiatric illness. I do no such thing. But it is no favor to desperate and vulnerable patients to treat them with drugs that have serious side effects unless it is clear that the benefits outweigh the harms.