Why Antidepressants Barely Work

When antidepressants are compared to placebos, they do only slightly better. This is not a problem  for psychiatrists. People get better, they can charge money for access — that’s what matters. The rest of us, who would benefit from a better understanding of depression, do not feel bad because we have no idea what we are missing. But the puzzle of weak effectiveness remains. If the theory used to justify the antidepressants is correct, shouldn’t they work better? If the theory is totally wrong, why do they work at all?

John Horgan, a science writer, commented about this recently:

I first took a close look at treatments for mental illness 15 years ago while researching an article for Scientific American. At the time, sales of a new class of antidepressants, selective serotonin reuptake inhibitors, or SSRI’s, were booming. The first SSRI, Prozac, had quickly become the most widely prescribed drug in the world. Many psychiatrists, notably Peter D. Kramer, author of the best seller Listening to Prozac, touted SSRI’s as a revolutionary advance in the treatment of mental illness. Prozac, Kramer said in a phrase that I hope now haunts him, could make patients “better than well.”

Clinical trials told a different story. SSRI’s are no more effective than two older classes of antidepressants, tricyclics and monoamine oxidase inhibitors. What was even more surprising to me—given the rave reviews Prozac had received from Kramer and others—was that antidepressants as a whole were not more effective than so-called talking cures, whether cognitive behavioral therapy or even old-fashioned Freudian psychoanalysis. . . . Psychiatry has made disturbingly little progress since the heyday of Freudian theory.

To psychiatrists, psychiatry has made great progress since Freud. First, it is much easier to prescribe a pill than listen to a patient talk for 50 minutes. Second, the new pseudoscience of serotonin deficiency is far more respectable (more “scientific”) than the old pseudoscience of psychoanalysis (ego, id, super-ego, repression, etc.). It is harder for other doctors to make fun of psychiatrists.

But Horgan was not thinking like a psychiatrist. He was thinking like the rest of us. From that point of view, he should not have been “disturbed” by “little progress”. Antidepressants will never work well. Poor effectiveness is inherent in the situation. Antidepressants must do two things: (a) people must get better and (b) psychiatrists must make a living. Those are different goals (“misaligned incentives”) and they conflict.

Suppose a repairman comes to fix your dryer. One part is broken. The repairman orders a replacement and installs it. Your dryer now works fine. Because you could not diagnose the problem nor fix it, the repairman continues to be necessary. Suppose, on the other hand, the repairman can not replace the broken part. He must do something else. Maybe use duct tape. In this situation, the repair cannot possibly work well. Whatever he does can be better than nothing, but it cannot be a good repair

That is the situation of psychiatrists. I’m sure depression is due to the wrong environment. My work suggests we need to see faces in the morning for our mood-controlling system to work properly. Jon Cousins’ work suggests we need to believe others care about us. Those are two possibilities. Psychiatrists cannot fix the environment. The pieces of the environment we need to be healthy must have been abundant during the Stone Age. This means they must be cheap. Psychiatrists cannot supply things that are cheap and abundant. If that’s what they did, they couldn’t make a living. This means they can only supply something that is not what is missing. Like a repairman who cannot replace a broken part, they are stuck with second-rate solutions. This is the fundamental reason that all mainstream treatments for depression, whether talk or drug, have roughly the same effectiveness — and none of them work very well.

Thanks to James Lucoff.

26 Replies to “Why Antidepressants Barely Work”

  1. Interesting post, but I think that antidepressants are worse than placebos in one crucial way. Antidepressants are known to cause long-term changes in synaptic function. Evidence exists that those changes may be harmful and may actually cause depression (or at least exacerbate it).

    The main thesis of Robert Whitaker’s recent book, Anatomy of an Epidemic, is that the widespread use of antidepressants (and other psychiatric drugs) is causing an epidemic of mental illness. For a shorter treatment, see: Now Antidepressant-Induced Chronic Depression Has a Name: Tardive Dysphoria.

  2. Reminds me of a problem the Xacto corporation was having a few decades back. They were successful company, but profits were modest. They would sell a customer a handheld knife and a pack of refill blades, but that was it.

    Xacto called in a consultant to advise them on what to do to grow. The consultant told them to make the blades out of softer steel.

    Profits exploded, since the blades now needed constant replacement. (Previously, they lasted until one happened to break, which might take months.)


  3. Dearieme,

    Why not try for facing your dog for an hour every monring and record the results?

    All of us here will be highly interested in it.

    Theraphies that work a tiny bit abound. (every change in whatever parameters of body,mind,environment should have. *some* effect). Basically the only plus of antidepresants is that they have huge clinicl trials and have the aura of science. Otherwise, I bet there are very many things that have a tiny effect on almost any complex illness

  4. My normal but unhappy brother in law was put on a cocktail of anti depressants and has now developed bipolar disorder, so they very well may make things worse. I think that talking therapies could be a route to helping an individual change their environment, as this can often involve hard choices.

  5. Talkig therapies work. But most of the effect is in the very act of coming to talk and the talking.

    Methods of talk therapy do not matter much.

    When compared to “placebo” (i cannot remember how they managed to fix a placebo to. Tlk theraoy) there aRe clAims that it gives similar results.

    I am fully supporting using the placebo effect (somehow medicine ignores ts strong effect, maybe not much money to make from. Or the delusion at we need “something real”). Anyway, placebo is superior, as it does not contain harmful chemicals. One only needs to made the psychological staging as strong as possible. (the more professional looking, aura etc. The stronger the placebo effect)

  6. I am an avid reader of this blog and totally respect and admire Mr. Roberts’ sometimes unconventional way of looking at things. But I have to say I was disappointed in the one-sidedness of this article (and the comments on it so far). These opinions do not seem to come from people with personal experience with depression and/or anti-depressants. Why be so judgemental about something you know nothing about?

    Try to believe me when I say that when you are in the midst of a depression (which is NOT the same as ‘being in a bad mood’) and you suddenly feel the clouds lifting you will KNOW it is because of the pills you have been taking, not the result of the placebo effect. Especially if you, like many others, have tried different drugs before you found one that works.

  7. David — a couple of things:

    First, I have some experience with antidepressants. From 1995 to 1998, I worked on the team that managed the antidepressant Zoloft (generic name: sertraline; known as Lustral in the UK and in some other countries).

    Second, I don’t think that introspection is a reliable method to distinguish placebo effects from true drug effects. See, for example, this article, which starts with the story of a person who participated in a clinical trial for an antidepressant. The woman’s response to the treatment was so profound that she was astonished to find out that she had actually been enrolled in the placebo arm of the study. I’ve included the first few paragraphs below.

    Janis Schonfeld recalls the events that started her on her recovery from 30 years of depression with snapshot clarity: the newspaper ad she saw in 1997 seeking subjects for an antidepressant study; the chair she was sitting in when she called UCLA’s Neuropsychiatric Institute; the window she was looking out of when she first spoke with Michelle Abrams, the research nurse who shepherded her through the trial. She remembers being both nervous and hopeful when she arrived at the institute, and a little uncomfortable when a technician put gel on her head, attached a nylon cap shot through with electrodes, and recorded her brain activity for 45 minutes. But most of all she remembers getting the bottle of her new pills in a brown paper bag from the hospital pharmacy. “I was so excited,” she told me. “I couldn’t wait to get started on them.”

    Within a couple of weeks, Schonfeld, then a 46-year-old interior designer, got quickly and dramatically better, able once again to care for herself and her husband and daughter, no longer so convinced of her own worthlessness that she’d consider killing herself. For the next two months, she came back weekly for more interviews and tests and EEGs. And by the end of the study, Schonfeld seemed to be yet another person who owed a nearly miraculous recovery to the new generation of antidepressants — in this case, venlafaxine, better known as Effexor.

    But during her final visit to the institute, one of the doctors directing the research sat her down to deliver some disturbing news. “He told me I hadn’t been taking a medicine at all. I’d been on a placebo. I was totally shocked.”


  8. Definitely good analysis there… And I totally respect people who have used SSRI’s and had great personal success, but you have to look at the problem from a scientific perspective. And if they don’t make sense to all parties when you look at them scientifically, there needs to be a change somewhere. I’m not sure whether that means trying to change public perception, medical practice, or research practices, but something needs to change.

  9. David Perch,

    I accept the possibility that a specific drug that is compatible to a specific person can work well for depression.

    The studies of antidepresants use large grouos of people, so the personal compatibility is probably nonexistent.

    In terms of scientific knowledge, however, we have no information. There are no strong effects in the studies, and we cannot take personal experience as proof.

    Practiclly speaking, i believe that science aside, what works for a person works. And it is foolish to avoid using a drug that proved highky effective for a person for whatever soohisticated thinking possible. (i would say that seth wuld accept this as kind of personal science)

  10. Seth, I am generally in sympathy with the tenor of your article, but I do object to one argument you appear to be making towards the end of your post. You conclude that psychiatry will never succeed in curing depression because (a) depression is due to environmental factors and (b) psychiatry cannot change the environment nor supply the “cheap and abundant” things that were present in the ancestral environment but missing in contemporary society. This is a non-sequitur. Suppose your own hypothesis is correct and depression is caused by a failure to see faces in the morning. The brain would still be implicated in this chain of causality: absence of morning faces will cause certain changes in the brain which would themselves cause people to feel and act depressed. If psychiatrists uncovered the relevant neural mechanisms involved in this process, they could in principle develop agents that mimicked the effect of morning faces in the brain. So it is not true that, because depression is ultimately caused by the environment, it falls outside the scope of psychiatry.

    1. Alex, good quote, thanks.

      Pablo, you write, “If psychiatrists uncovered the relevant neural mechanisms involved in this process, they could in principle develop agents that mimicked the effect of morning faces in the brain.” Think of it like this. Scurvy is due to lack of Vitamin C. But no one is going to make a living by “developing agents that mimic the effect of [Vitamin C]” because Vitamin C is easy to get.

  11. Jazi, thanks for your comment. You hit on something I have been wondering about. I have never really understood the absolute reliance on clinical trials in these matters. They do not seem to take into account that people can react very differently to the exact same thing. If our individual response to recreational drugs like cannabis and alcohol vary so dramatically why can this not be true for non-recreational drugs? Millions of people use paracetamol/acetaminophen successfully but it has never done a thing for me, for instance. Cannabis makes me very relaxed and in touch with my creative side, it makes plenty of other people tired and paranoid. Why would it be different with something like Prozac? It makes me wonder whether we might be missing out on some wonderful drugs just because they did not work on a large enough percentage of people.

    I thought one of the major points that Mr. Roberts is making in his work is that absolute scientific proof is less important than finding something that works. I am not sure if anyone will ever prove beyond a shadow of a doubt that standing on one leg to improve sleep really works but in the meantime it seems like a good idea to try it when you are having sleep problems.

    To clarify: I am not from the US so as a result perhaps not so suspicious of our healthcare system as you guys are of yours. One of the perks of living in what a lot of Americans seem to fear so much: a ‘socialist’ society.

  12. David, we’re not suspicious of our health-care system; we’re suspicious of our government. And the more that the government gets involved with our health care, the more suspicious we will become.

    One man’s perk is another man’s poison.

  13. we should device a clinical trial for morning faces therapy.

    idealy, we need people who handle something (audio) on a computer screen anyway.

    we assign to two groups one with random visuals etc. the other with faces, and measure depression scales ( in a masked way, to avoid them knowing the idea. like asking various inventories)

    I am thinking about where we can find a large enough group of people who can get easily subjected to such an experiment.

  14. Seth, Vitamin C is easy to get now that we can synthesize it cheaply, just like fish/flax seed oil and Vitamin D pills..but human attention in forms that will alleviate depression? Unfortunately that is a commodity too costly in US society for many people, hence the drive to replace therapy with pills. I’m reminded of Emily in Our Town, back from the dead for one day: “Oh Mama, just look at me for one minute as if you really saw….” Did Book TV really work for you?

    Joe, I’m not sure why you don’t trust the government when you trust the highly paid health care administrators who deny people coverage and suck up 40% of our health care bill. The hospital is the only place where a bandage can cost you $50 dollars, and you can’t bring your own. After being without insurance for almost two years, I’m looking forward to the day when people don’t have to “choose” between affording a pain pill and an antibiotic for lack of coverage like the young man in Cincinnati who died. And I work and I pay taxes, thank you. Health care or endless war? I’d like health care, please.

  15. Seth, your suggestion that needing to see faces of others (and to converse with them) in the a.m. is intriguing. It makes sense if we think about the fact that humans are essentially social animals, seek friendships, and crave regular interaction with others.

    And here’s another thought: Why hasn’t the issue of widespread vitamin D deficiency been explored more throughly? There has been a good deal of writing (and some scientific examination) into the fact that vitamin D deficiency is rampant in modern society (especially in climes where sunlight is weaker). And there is apparently evidence that using light therapy, or raising the individual’s blood levels of 25(OH)D through supplementation with vitamin D3, can be effective in treating SAD or other mood disorders. So why hasn’t there been more focused scientific inquiry along these lines? Could it be that advising people to get more exposure to natural sunlight, or otherwise recommending they raise their serum level of vitamin D with supplementation, is just too simple, direct and cheap compared to drugs and psychiatric treatment, which are expensive?

  16. @WCB: Here’s my experience with vitamin D deficiency, for whatever it’s worth. About four months ago, I had routine bloodwork done. The results showed very low vitamin D levels. Since that time, I started taking D3 supplements, and the latest lab results showed normal levels. I have not noticed any change in my mood (though I was not depressed to begin with).

  17. Lots of commenters are dismissing cognitive behavioral therapy. But it has been relatively successful.

    The CBT method appears to be effective even outside one on one therapy.

  18. @Alex: Thanks for your comment about your experience with D3. I”m guessing that over a longer time frame you will likely realize many other benefits, apart from the subject matter in this string. In any event, it’s certainly clear that more research on the topic of finding effective, inexpensive treatments for those who experience depression is needed.

    But more broadly re vitamin D supplementation, based on the reading and research I’ve been doing for quite awhile, and my own experience with adding D3 over some four years and continuing, I’ll just mention that eliminating the deficiency can have a number of other benefits, including prevention (or avoidance) of things we would all like to avoid. But rather than me going OT reciting the litany of potential benefits and cluttering up Seth’s excellent blog, you might be interested in checking out a blog such as that of Dr. William Davis (Heart Scan blog) and the website of the Vitamin D Council. They both provide a lot of useful info, including, inter alia, advice as to the desireable level of serum 25(OH)D for most individuals to realize the full benefits (based, e.g., on Dr. Davis’ own clinical experience) which is significantly higher than the conventional view of something like 25-30 ng/ml.


  19. As Jazi yechezkel zilber said above
    “… placebo is superior, as it does not contain harmful chemicals.”

    I am strongly opposed to using “harmful chemicals” (drugs) of any kind for myself. Despite the severe scoldings I get from doctors, I currently use no prescription medicines, no over-the-counter meds (not even aspirin or vitamins), and certainly no alcohol, tobacco or illegal drugs.

    My resolve may be severely tested this weekend. This Friday, September 30, I am scheduled for some surgery. (If I had my way, it would be done without anesthesia, but the surgeon will not do it without anesthesia.)

    The surgeon has told me that I will need medication for pain after the surgery, and she plans to prescribe Lortab (hydrocodone).

    I told her that I would not take hydrocodone, or even fill the prescription. As I told her:

    “I have only had one hydrocodone pill in my life, and it made me feel so good that I immediately threw the other 17 pills away. As far as I’m concerned, the only thing worse that a drug that makes me feel bad is a drug that makes me feel good, because the feel-good drug is the one that could get me addicted.”

    That screaming you will hear this weekend will be me suffering without pain meds.

    1. Jim Purdy, I pretty much agree, except I have no problem with alcohol. I tried to get a dental operation done without anesthesia but I lasted about 0.5 seconds. I was able to avoid pain killers after the operation, however.


  20. Try to believe me when I say that when you are in the midst of a depression (which is NOT the same as ‘being in a bad mood’) and you suddenly feel the clouds lifting you will KNOW it is because of the pills you have been taking, not the result of the placebo effect. Especially if you, like many others, have tried different drugs before you found one that works.

    Sorry, but that methodology doesn’t work. Assume for the sake of argument that your depression is either environmental, seasonal, or cyclical – something that gets triggered at time T and then sometime much later at time U goes away. What does that look like to somebody who takes pills? You go to the doctor, who prescribes pill A; it doesn’t seem to work, so the doctor prescribes pill B which also doesn’t seem to work, and so on until time U is achieved. Since time U happens a couple weeks after you’ve started drug D, you are now utterly convinced drug D is the “one that works”.

    Regression toward the mean explains much of the apparent effect of antidepressants.

Comments are closed.