
Louis Menand on Psychiatry 05/31/2010
What this covers
This is a conversation between Russ Roberts and Louis Menand, built around Menand's New Yorker essay reviewing books on psychiatry. It starts with history: the first blockbuster anxiety drugs, Miltown and Equanil, arrived in the 1950s, when Freud's cultural standing was at its height. Nobody saw them as contradicting Freudian theory until the late 1960s. By DSM-III in 1980 the profession had dropped Sigmund Freud almost completely, in favor of a biological model. From there the talk turns to the evidence. Many studies find that drugs, talk therapies and placebo effect conditions do about equally well. The conversation covers Irving Kirsch's meta-analyses, and the repeated finding that sympathetic non-clinicians help outpatients about as much as licensed therapists. It also notes that the serotonin story is not what researchers actually believe.
The second half of the discussion widens out. Menand describes how the DSM) fits into a loop of drug company profit, insurance reimbursement and legal caution. This is the system that Gary Greenberg finds so distressing, and it is why the diagnostic bar keeps dropping. Greenberg's further argument is that we have good reasons to be depressed about the world, and that industry has persuaded us to blame our brain chemistry instead. Concrete cases include the reported dropping of the bereavement exclusion in DSM-5, ECT as an effective but stigmatized treatment, and David Healy's claims that side effects of Paxil have been suppressed. Menand also questions whether medication would dull creativity, and whether dependence on a psychiatric drug is like addiction or like taking a statin.
Roberts keeps drawing parallels to economics, and this is where the exchange turns most contentious. He compares the field to Ed Leamer's critique of econometrics, in which one death-penalty data set yields opposite conclusions depending on specification. He argues that psychiatry, like economics, borrows the look of rigor and is shaped by fads and money. Menand adds the pragmatist view, drawn from Charles Peirce, that inquiry always begins with interests and is corrected only in aggregate. He also holds that questions about grief and authenticity are cultural and philosophical, so science cannot settle them.
Louis Menand argues that psychiatry is a real but deeply limited science. Its treatments work only modestly and about equally well, its mechanisms remain black boxes, and its diagnostic categories are being widened by incentives from drug companies, insurers and the law. He also argues that critics like Greenberg wrongly demand more than any medical science can deliver, and that questions of authenticity and grief belong to philosophy and culture, not to science.
- Studies since the 1930s find nearly all psychiatric treatments roughly equally effective (the dodo bird effect), with effectiveness rates of about 50-60%. Neither the Freudian model nor the serotonin model explains how treatments work.
- Diagnosis is being 'defined down' through a mutually reinforcing system: pharma profits, insurance reimbursement tied to DSM codes, and legal defensiveness.
- Science cannot settle whether to take a 'grief pill'; authenticity is a cultural question. Roberts parallels this with the limits of econometrics (Leamer) and of expert knowledge.
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
The ironic delayed reaction: in the 1950s, when Freud had peak cultural cachet, the first blockbuster mood drugs appeared. These were Miltown and Equanil (anti-anxiety), followed in the 1960s by Librium and Valium, which relieved anxiety by changing brain chemistry. Their acceptance was not seen as contradicting Freudian theory, even though Freud located anxiety in repressed psychic conflict from personal history. Only in the late 1960s did people in the profession point out the enormous disjunction between psychoanalytic and psychopharmacological premises.
“the acceptance of these of these drugs like Milltown was not considered to contradict the claims of Freudian theory”
Peirce held that no single individual can have an objective view of any subject; even observers of a star's position with the same instruments perceive it slightly differently. In aggregate, averaging many biased, partial inquirers yields roughly the way things are, and Menand thinks this is how science should work.
“no single individual therefore can possibly have anything like an objective view of any subject that he studies including things like you know knowing the position of a star”
Electroconvulsive therapy (ECT) is one of the most effective treatments for severe/major depression, though nobody knows exactly why it works. It is rarely used largely because of bad press, such as One Flew Over the Cuckoo's Nest, which gave it a 'Frankenstein's monster' image.
“it's one of the most effective treatments for people with major depression and part of it it's called ECT and part of the reason that it's not used problem anymore is because it got very bad press”
Depression and especially bipolar disorder have often been associated with creativity, but it is not clear that medicated people are less creative. People may value the manic phase as part of who they are, yet they can't actually do their work while manic. So it's unclear we need extreme mood states to accomplish things. Equally, it's unclear that mass antidepressant use wouldn't turn us all into sheep, cut off from painful but fulfilling aspects of life.
“it's not altogether clear that people who've been medicated to deal with those disorders are less creative under medication”
Averaging biased views doesn't always work. As UCLA econometrician Ed Leamer showed, with the same death-penalty data set a proponent can reasonably show each execution deters 30 murders, while an opponent, using a different specification and slightly different variables, can show each execution causes 12 murders. Leamer's conclusion is not to take the average but that the data don't let us answer the question, and Roberts thinks the same may be true of depression treatments.
“if you favor that argument you can show reasonably that for every execution you deter 30 murders... you can show that actually causes 12 murders every execution”
Psychiatric treatment has a self-fulfilling prophecy aspect: patients who expect to get better tend to get better, so much depends on the patient buying into the treatment's claims. Placebo effectiveness in drug trials shows this, since people who think they're taking an antidepressant often report improved symptoms on a sugar pill.
“there's also a self-fulfilling prophecy aspect to psychiatric treatment which is that if people feel that they're going to get better they tend to get better”
Dependence on a maintenance drug differs from addiction. Menand takes a generic statin daily and doesn't feel dependent: stopping would risk the consequences of high cholesterol but would cause no withdrawal and he wouldn't feel bad. What makes people leery of psychiatric drugs is that stopping them makes you feel sick.
“I take whatever the generic lipitor is statin for cholesterol right I take it every day I don't think that I'm dependent on the drug”
We readily use chemistry to alter moods for life problems, from caffeine to SSRIs, but treat grief as out of bounds. Grief is the only category excluded from depression diagnosis in the DSM: grieving people with depressive symptoms are not supposed to be diagnosed as clinically depressed. The forthcoming DSM-5 reportedly drops this bereavement exclusion, allowing grieving people to be diagnosed and prescribed.
“it is the only category and the DSM that's excluded from diagnosis of depression”
The diagnostic bar has dropped because medications are easily available, and the key player is the insurance industry. The DSM exists to provide clinical categories: if a doctor prescribes based on a DSM diagnostic number, insurers will likely pay. If insurers backed out of that business, there would be far less diagnosis.
“the key player and all this is the insurance industry because the reason we have the DSM is because it provides clinical categories for diagnosis”
The pragmatist view associated with Peirce, William James and John Dewey holds that humans pursue knowledge for interested reasons, materialistic or idealistic, with assumptions, goals and hoped-for outcomes in mind even as empirical inquiry begins. This is as true of empirical studies as of literary criticism.
“the pragmatic view that person's associated with and also William James John Dewey does does assume that human beings pursue knowledge for interested reasons”
The first two DSM editions (1952, 1968) were heavily Freud-influenced and held anxiety and psychic conflict to be at the root of most mood disorders. By 1980 the profession had weaned itself from Freud, marked by Donald Klein's article 'Anxiety Reconceptualized'. DSM-III (1980) struck Freud out completely, and today the DSM shows almost no Freudian vocabulary. Between the 1960s and 1980s psychiatry revolutionized itself, adopting both a biological model and different ideas of effective talk therapy.
“the first two editions of the Diagnostic and Statistical Manual 52 and 68 were very influenced by Freud”
The objectivity of econometric analysis is oversold. For the sake of reputation, ideological bias or profit, researchers distort empirical work or its presentation, or convince themselves they've been objective when they haven't. Similarly, of the two psychiatry authors reviewed, one thinks CBT is the greatest thing since sliced bread and the other thinks it's a total waste of time.
“I believe that the objectivity of of econometrics analysis in the case of economics is oversold”
Freudianism is embarrassingly 'touchy-feely', since no surgeon has found the id or superego. Brain-chemistry biology has a natural appeal in the competition among scientists because it looks more scientific. This parallels economics' push to become more mathematical, as if using a rigorous science's tools made it rigorous, even though psychiatry hasn't actually opened the black box.
“it reminds me in economics of the attempt to make economics more mathematical as if that makes it more rigorous”
Psychiatric diagnosis and medication form a self-reinforcing system: the drug industry, insurance industry and medical profession all support each other. It is very hard to break out of because everybody has a stake in its success, and this is what particularly distresses Greenberg.
“every every entity from the drug industry to the insurance industry to the medical profession basically supporting each other and this in this system and it's so it's very hard to break out”
People romanticize science, social science and medicine, assuming changes in treatment result from careful study. In both economics and psychiatry, fads, the appearance of scientific rigor (mathematics, pharmacology), and money and self-interest shape what is believed about causal relationships.
“people have a romance about science and sometimes social science and certainly about medicine”
A repeatable finding, going back to the 1940s, is that depressed, anxious or mood-disordered outpatients do equally well talking to sympathetic professors with no psychiatric training as talking to licensed psychotherapists. Simply talking to someone who seems to care and empathize helps, and the listener's training doesn't seem to matter.
“people do equally well these are just talking to sympathetic professors who don't have any training in psychiatry we're talking to licensed psychotherapist”
It is impossible to distinguish empirically between sadness caused by a life event (death, divorce, job loss) and clinical depression with no exogenous cause. Brain imaging of people sad over a layoff and of people with the same symptoms but no identifiable event shows the same kind of brains.
“it's impossible to distinguish empirically sadness over some life event death or divorce or losing your job and what you're calling clinical depression”
Psychoanalysis is the classic case of a bogus-theory placebo. Very few people now accept Freudian premises, and the profession has completely discarded its theory of what goes on in the mind. Yet for years people felt better from it, though some spent years and lots of money for no benefit.
“psychoanalysis is a classic case I don't think anybody very few people of what should say not anybody very few people except the premises of Freudian psychoanalysis but it was used for years”
David Healy, a psychiatrist and prolific historian of psychiatry, believes important bad side effects of antidepressants, particularly Paxil, are being suppressed by the pharmaceutical industry, including suicidality in younger patients. This claim has been debated since Prozac's licensing around 1987-88. Peter Kramer (Listening to Prozac) argued the side-effect claims were exaggerated or baseless.
“there are important bad side effects of antidepressants in particular paxil that are being suppressed and that they can cause suicidality particularly in younger patients”
Diagnosis creep is also driven by the legal environment. Fear of misdiagnosis lawsuits makes the DSM an attractive way for doctors and insurers to insulate themselves by 'following best practices', even if those best practices are nonsense.
“the DSM is a very attractive way for doctors and insurance companies to insulate themselves because they're quote following best practices even if those best practices or nonsense”
Greenberg argues we have plenty of good reason to feel depressed about the state of the world: the environment, corruption, politics. Industry and capitalism have persuaded us that the problem is our brain chemistry rather than the world, so we take pills, put on smiley faces and ignore real problems.
“his claim is that we have lots of good reason to feel depressed about the state of the world and that industry and capitalism has tried to persuade us that it's not really the world that's the problem”
An enormous number of studies compare psychiatric treatments: talk therapy vs. medication, CBT vs. medication, psychoanalysis vs. other talk therapies, and drugs vs. placebo. Most conclude that all treatments are about equally effective or equally ineffective. Rosenzweig noted this in the 1930s and called it the 'dodo bird effect', after Alice in Wonderland's 'everybody has won and all must have prizes'.
“most of these seem to conclude that that they're all about equally effective or equally ineffective this was actually something that was noted back in the 1930s”
The popular biological model, that depression is caused by a lack of serotonin which SSRIs remedy, is not what medical scientists believe. Giving people serotonin doesn't make them feel better. Drugs like Prozac (fluoxetine) seem instead to trigger a poorly understood cascade of other brain events. So the medical model is also full of black boxes and not much better than Freud's, except that brain scans can show things like shrinking hippocampi associated with depression.
“no medical scientist actually thinks that that's what that's what causes depression or that's what fixes depression you can give you can actually give people serotonin and it doesn't make them feel any better”
Irving Kirsch (The Emperor's New Drugs) has done meta-analyses of antidepressant trials for about 12-13 years. He concludes that placebo usually does as well as the antidepressant. Where the drug statistically outperforms placebo, often by a small margin, he attributes this to an enhanced placebo effect: patients on the real drug experience side effects, infer they are on the real medication, and so feel better.
“most of the time the placebo will do as well as the antidepressant”
The deeper problem is not that science can't do everything but that data and understanding of what works may be distorted by the profit motive, by third-party payment ('other people's money'), and by the legal environment. You can be a friend or enemy of capitalism and still see this as unhealthy.
“whether the data and our understanding of what works and doesn't work is being distorted in some way either by profit motive”
A recent JAMA meta-analysis found that placebos were as good as antidepressants for mild or ordinary outpatient depression, but antidepressants did significantly better for severe/major depression. Menand calls the study odd for three reasons. It tested only one drug, Paxil, out of about 25 antidepressants on the market. It pooled only about 700 patients, against millions of users. And its severity split is hard to square mechanistically: if altering brain chemistry helps major depression, it obviously does something.
“for patients with mild depression or normal outpatient depression placebos were just as good as antidepressants although for patients with severe depression and major depression eddie's presidents actually had a significantly better effect”
Genuine (apparently) endogenous depression clearly exists and warrants intervention by talk therapy or medication. The real problem is 'defining the category down': diagnosis comes to include conditions only marginally connected to clinical depression, or ones with identifiable other causes such as job loss.
“the question really has to do is whether we're defining that category down to include all kinds of other things that are either only marginally connected”
Different antidepressants work better or worse depending on the individual, and some people must try three or four before one works. There is no evidence that talk therapy is better or worse than Prozac, or that CBT is better or worse than either.
“some people have to try three or four to find one that works but there's actually no evidence that talk therapy is better or worse than that taking Prozac”
The pharmaceutical industry runs its own drug tests, and even the ones the FDA is supposed to do are paid for by industry. Companies need not report negative trial outcomes and need only two positive trials for licensing, which raises the question of what the unreported data show.
“the drug industry does not have to report negative outcomes in their trials for new medication they only have to come up with two positive trials for the drug to be licensed”
Fear of malpractice suits reinforces DSM use: following its categories lets doctors and insurers insulate themselves by appearing to follow 'best practices', even if those practices are nonsense.
“they're quote following best practices even if those best practices or nonsense”
Few genuinely new psychiatric drugs are in the pipeline. Instead companies tweak an existing molecule, give it a new brand name, earn enormous profits during the period of exclusivity, then tweak again when it goes generic. Despite industry claims about R&D spending, there seems to have been little real development recently.
“there's not a lot of new drugs in the pipeline so what tends to happen is they tweak the molecule give it a new brand name”
Nobody has a bigger romance about science than the critics of medical science. They think that if it's called science it must have all the answers, for example that it isn't science unless reported symptoms can be traced to an underlying organic cause. We can't do that for most symptoms, but that doesn't mean we shouldn't treat them or that treatment is unscientific; it is a best guess. Greenberg in particular expects too much of science, especially medical science dealing with minds.
“nobody has a bigger romance than the critics of medical science because they think if we call it science it has to have all the answers”
We want to fix some limitations and contingencies so people don't suffer, while preserving others for ourselves even when they are painful. What constitutes authenticity or autonomy is a matter for philosophy, literature and religion, not science. It is a cultural question that varies as cultures vary.
“what constitutes authenticity or autonomy it's a matter for philosophy and literature and religion it's not really a matter for science to tell us it's a cultural question and it varies as cultures vary”
The current state of knowledge does not permit us to infer a right answer to the questions being asked about depression treatments. Moreover, science will never answer the fundamental questions about autonomy, authenticity and life choices, and we shouldn't even want medical science to answer them.
“science is never going to answer the fundamental questions about our autonomy and our authenticity and the choices that we make in life”
Respect for authenticity is felt by most people, not just the religious; the tricky part is locating the line. God (or nature) gave Menand very bad eyes, yet he feels no compunction about wearing glasses, so there is some point at which one should accept a limitation rather than fix it.
“I don't think that that the respect for authenticity is just something that religious people feeling”
Like most medical science, psychiatry deals with black boxes: we know some things work better than others without understanding why. It is trickier still because it treats conscious life, and the patient is consciously involved in repairing her own conscious life. Yet the fact that treatments are only more or less effective doesn't mean they aren't worth pursuing.
“in the end it's like most medical science its kind of deals with a lot of black boxes”
The grief pill parallels Robert Nozick's 'experience machine', which lets you live any fantasy life that feels real. Some refuse because it's not natural, and there is also a religious element: avoiding grief or plugging in betrays God's path and your purpose. Even for non-religious people the goal of authenticity is very real around big areas like death and life's purpose, despite our cheating in many other domains.
“Robert Nozick brought up called the experience machine where you would hook yourself up to this machine and while you were hooked up to it you could fulfill any fantasy you wanted”
Michael Sandel, drawing on his work on bioethics during the Bush administration, argues that a meaningful life involves accepting one's limitations. Menand holds that where authenticity and autonomy lie is a cultural question for philosophy, literature and religion, not science, and that it varies across cultures.
“it's a matter for philosophy and literature and religion it's not really a matter for science to tell us it's a cultural question and it varies as cultures vary”
The 'grief pill' thought experiment: a pill would eliminate the physical symptoms of bereavement (insomnia, depression, the hollow in the stomach) without lessening love for or honor of the dead person. Most people Menand has asked say they would not take it. Their reasons: we are creatures who grieve, so short-circuiting a natural process seems wrong even biologically; and bereavement is itself a period of connection with the lost person, so taking the pill would be like losing them again.
“most people would say they wouldn't take the pill”
Beyond the expansion of diagnostic categories, there is a multiplication of subsets within them (ADD, ADHD, numerous syndromes and variations). Desperate parents of struggling children want to believe the doctor has evidence and that prescription equals science. As a defender of markets, Roberts is struck by how willing people are to surrender their own and their children's autonomy to experts who don't know very much.
“it fascinates me how willing people are to give up their autonomy in their autonomy of their kids two experts who in this as we've been talking about for the last half-hour don't know very much”
The harm of addiction to mood drugs is that most such drugs have limited effectiveness. With continued use they stop working, yet stopping causes withdrawal, which is what happened with Valium and Librium. People dislike being 'hostage' to a medicine to manage a mood disorder and would refuse an antidepressant they could never stop without withdrawal.
“these drugs most drugs like this you know have a limited effectiveness you keep taking them they stop working on you but if these things not taking them and have withdrawal pains then you really are addicted”
Psychiatry resembles economics. Both study a complex system (the brain, the economy) where you can't hold everything constant, so it is inherently difficult to assess whether an intervention, such as fiscal or monetary policy or a therapy, has an impact and how big it is.
“you can't hold everything constant that you'd like to to assess whether fiscal policy or monetary policy has an impact”
In the 1950s Freudianism dominated medical-school psychiatric training, run by psychoanalysts. Practicing psychoanalysts mostly had private practices, while big mental hospitals were not psychoanalytically inclined. This dominance faded in the 1960s and 70s, as documented in Joel Paris's book Fall of an Icon.
“in the 1950s Freudianism was very well established in medical schools”
Peirce's idea that wisdom lies in the marketplace of ideas rather than in any one brain is very Hayekian, as Roberts's former philosophy professor, Peirce scholar Richard Smith, confirmed. Menand adds that it corresponds well to nineteenth-century laissez-faire thought.
“this idea that wisdom is is in the marketplace of ideas rather than in any one person's brain is very very Hayekian”
Freud himself regarded psychoanalysis as a medical science. He thought he was describing, as a scientist and biologist, what really goes on in the mind, not merely doing talk therapy. Today we think nothing in the mind corresponds to his categories.
“probably thought of it as a science thought of it as a medical science that was his training”
Szasz's critique differs from Greenberg's: it concerns diagnosis used to categorize, marginalize and control personality types or attitudes the state or society finds repugnant. The two are similar in that people's lack of knowledge is exploited by those with self-interest, whether power or profit, distorting culture, norms and incentives.
“that has more to do with the use of diagnosis as a way of categorizing and ultimately marginalized controlling controlling people”
There is a disconnect between how scientists discuss their work, how the media discusses it, and how people perceive it; these are three very different things.
“there's a disconnect between the way the scientists themselves discuss their work the way it gets discussed in the media and the way people perceive it”
Michael Sandel, drawing on his work in Bush-era bioethics on topics like cloning, argues that part of the meaning of life is accepting its limitations. Someone who can't do that would not be respected by most people.
“Michael Sandel is written a good book... he said you know the meaning of life is to accept the limitations of your life”
The biological model's one advantage over Freud is observability: brain scans can show physical correlates such as shrinking hippocampi, which are one symptom of depression.
“you can now do brain scans and you can actually see things like shrinking hippocampi which are one symptom of depression”
An apocryphal Enrico Fermi horseshoe story ('they say it works even if you don't believe in it') illustrates a striking point. A sugar pill plus a convincing but bogus explanation of how it fixes the black box may genuinely make a person better, if they believe the theory.
“they give you this totally bogus but you believe that theory of why it's going to make you better that it will make you better it's a really extraordinary thing”
Historical drug harms include thalidomide, prescribed as a sedative before its side effects were realized, and Valium and Librium, which turned out to be addictive. Roberts provocatively asks what the harm is in being addicted to something that makes you feel good.
“phylidia might was actually prescribed as a sedative and then people realized what the side effects were”
Freud himself considered psychoanalysis a medical science. He thought of himself as a scientist and biologist describing real processes in the mind, but we now believe nothing in the mind corresponds to his categories.
“he thought he was really describing stuff that goes on in the mind and he was describing it as a scientist as a biologist”
Pharmaceutical use on schoolchildren for attention deficit disorder has exploded. Since kids' minds normally wander when material is boring, clinical ADD is a matter of degree. Medicating these children has had an unknown, though certainly large, good and bad effect on them, and an enormous effect on drug-company profits.
“the use of pharmaceuticals on children has exploded in particular schoolchildren who are say having some sort of a D D attention deficit disorder”
Pharma spends a lot on R&D, but what it chooses to research is partly determined by what is subsidized rather than what works best, making the system very messy.
“what they choose to spend the money on is going to be unfortunately partly determined by factors of them what works best it's what subsidized”
Gary Greenberg's Manufacturing Depression is deeply skeptical of psychiatry's claim to be a science. It attributes the proliferation of its claims essentially to a desire to make money: industry figured out that persuading people their moods were diseases treatable by a pill was very profitable.
“people figured out that if you could persuade people that that their boots were diseases and that use some chemical that you had put it into a pill what makes him feel better you could make a lot of money”
Thomas Szasz has criticized psychopharmacology and psychiatric diagnosis as tools the state has used, particularly but not only in communist countries, to control citizens. These uses are justified as science but are really fake science serving self-interest.
“sahs has been very critical of psychopharmacology because it's been used by the state particularly communist countries”
Diagnostic categories are both expanded (dumbed down) and multiplied into subsets, as with ADD becoming ADHD and its variations, and with the many mood syndromes.
“not only do you get a dumbing down of expansion of the category you get a multiplication of the number of subsets within the category”
Patients do have a rational stake in the system: when insurance subsidizes a remedy, they might as well try it.
“the patient does have a stake in it to the extent that the remedy is available and the insurance company will cover it you might as well try it”
The DSM has been criticized ever since its creation in 1952, and especially in its fourth edition. The core contemporary debate is whether psychiatric diagnosis and treatment are medicalizing normal life mood changes.
“the DSM which has been criticized really ever since it was created back in 1952 but particularly in its most recent edition the fourth edition”
Even if depression cannot be seen in the brain, we can recognize it in behavior. People who suddenly cannot get out of bed or function, with no tragedy or layoff in their lives, plausibly have something clinically wrong.
“we can't see it in the brain but we know it when we see it in people's behavior”
In Sylvia Nasar's A Beautiful Mind, John Nash or his family at times refused treatment (possibly shock therapy) out of respect for his brain's gift. They felt it would be immoral for him to lead a more placid life if that meant cutting off his creativity, a worry people also have about pharmaceuticals, caffeine or drinking affecting something precious.
“Nash refuses therapy shock therapy if I remember correctly or his family refuses”
We don't feel the same aversion to lifelong heart or cholesterol drugs. If it's affordable or someone else pays, people shrug at taking it forever, suggesting we treat our brains differently from other organs.
“we don't feel that way about a heart drug if I said you know I'm sorry you got a cholesterol problem and for the rest of your life you got to take this drug you shrug”
As a defender of markets, Roberts is disturbed by consumers being manipulated for profit, and looks for incentives in the system that distort natural choices. What fascinates him most is how willing people are to cede autonomy over themselves and their kids to experts who know little.
“it fascinates me how willing people are to give up their autonomy in their autonomy of their kids two experts who ... don't know very much”
When Kirsch began doing meta-analyses 12-13 years ago, his methods were severely criticized within the profession for statistical errors and unreliability. Meta-analyses, then uncommon and dismissed, are now more common, but Menand doesn't know whether they are more reliable.
“his methods were severely criticized by people within the profession for statistical errors and just general unreliability”
Psychiatric treatments have effectiveness rates of roughly 50 or 60 percent. Menand stresses he is estimating off the top of his head: they are relatively effective but not 100% effective.
“their effectiveness rates are 50 or 60 percent I mean I'm starting that out of my head”
It seems a stretch, and rather sweeping, for Kirsch to claim that every case in which a medication outperforms a sugar pill is just a placebo effect. Patients do seem genuinely to get better on some of these medications.
“it does seem to be a bit of a stretch to claim that in every case in which a medication outperforms a sugar pill it's just it's just a placebo effect”
Roberts recounts an apocryphal story about Enrico Fermi keeping a horseshoe because 'they say it works even if you don't believe in it'. He uses it to illustrate how a bogus but believed explanation of how a sugar pill fixes the black box could actually make a patient better.
“they say it works even if you don't believe in it”
Folk wisdom from Roberts's Eastern European great-grandmother, 'if you're not in a good mood go out and talk to a rock', suggests you may not even need empathy, just the act of talking.
“if you're not in a good mood go out and talk to a rock”
If you are a talk therapist, talk therapy is better than Prozac; by the same logic, a professor of English thinks English is better than sociology.
“if you're in this professor English is better than sociology”
Roberts notes the pharmaceutical industry had done fantastically over the previous 15–20 years and asks what its future holds.
“which is doing fantastically for the last 15 or 20 years”