Articles

Antidepressants and Effect Size

Antidepressant drugs have been getting a bad rap in the media. I’ll just give 3 examples:

  • On the Today show, prominent medical expert :-) Tom Cruise told us Brooke Shields shouldn’t have taken these drugs for her postpartum depression.
  • In Natural News, “Health Ranger” Mike Adams accused pharmaceutical companies and the FDA of covering up negative information about antidepressants, saying it would be considered criminal activity in any other industry.
  • And an article in Newsweek said  “Studies suggest that the popular drugs are no more effective than a placebo. In fact, they may be worse.”

Yet psychiatrists are convinced that antidepressants work and are still routinely prescribing them for their patients. Is it all a Big Pharma plot? Who ya gonna believe? Inquiring minds want to know:

  • Are antidepressants more effective than placebo?
  • Has the efficacy of antidepressants been exaggerated?
  • Is psychotherapy a better treatment choice?

The science-based answers to the first two questions are clearly “Yes.” The best answer to the third question is “It depends.”

In 2008, Erick Turner and four colleagues published an article in The New England Journal of Medicine (NEJM) entitled “Selective Publication of Antidepressant Trials and Its Influence on Apparent Efficacy.” The FDA is able to make sure that drug companies don’t pick and choose which trials, and which outcomes within those trials, get seen. Using clinical trial data from the FDA as a gold standard, Turner, et. al. examined how these same trials were reported in published journal articles, They found that:

…according to the published literature, the results of nearly all of the trials of antidepressants were positive. In contrast, FDA analysis of the trial data showed that roughly half of the trials had positive results.

And some of the negative trials were published with a “spin” that made them appear positive. The data did show that each drug was superior to placebo, but the true magnitude of that superiority was less than a diligent literature review would indicate. They warned that

By altering the apparent risk–benefit ratio of drugs, selective publication can lead doctors to make inappropriate prescribing decisions that may not be in the best interest of their patients and, thus, the public health.

Irving Kirsch has been outspoken about antidepressants’ alleged lack of efficacy. In a controversial meta-analysis published in 1998, he found that placebos provided approximately 75% of the improvement provided by active drug. He suggested that the other 25% is debatable and could be due to an enhanced placebo response when patients experience side effects that convince them they are getting an active drug. In a further study in 2002, he “questioned the clinical significance of antidepressants.”

Kirsch recently looked at the FDA data for 4 of the 12 drugs that Turner examined. In spite of the smaller sample, where Turner found an effect size of 0.31, Kirsch got 0.32. So they got almost exactly the same result. But it was their interpretations of that result that were very different. Kirsch concluded that antidepressants are ineffective, while Turner found that the drugs were indeed superior to placebo. As the figure below shows, each drug’s effect size was positive. Also, none of the confidence intervals overlapped zero. This means that, while there is some probability that the true effect size is zero, meaning that antidepressants and placebo are equal in efficacy, that probability is negligibly small.

 

The discrepancy between Turner’s and Kirsch’s interpretations hinges on what these effect size numbers mean in terms of clinical significance,. Values of 0.2, 0.5, and 0.8 were once proposed as small, medium, and large effect sizes, respectively. The psychologist who proposed these landmarks admitted that he had picked them arbitrarily and that they had “no more reliable a basis than my own intuition.” Later, without providing any justification, the UK’s National Institute for Health and Clinical Excellence (NICE) decided to turn the 0.5 landmark (why not the 0.2 or the 0.8 value?) into a one-size-fits-all cut-off for clinical significance. In an editorial published in the British Medical Journal (BMJ), Turner explains with an elegant metaphor: journal articles had sold us a glass of juice advertised to contain 0.41 liters (0.41 being the effect size Turner, et al. derived from the journal articles); but the truth was that the “glass” of efficacy contained only 0.31 liters. Because these amounts were lower than the (arbitrary) 0.5 liter cut-off, NICE standards (and Kirsch) consider the glass to be empty. Turner correctly concludes that the glass is far from full, but it is also far from empty. He also points out that patients’ responses are not all-or-none and that partial responses can be meaningful.

Incidentally, NICE is no longer using the 0.5 effect size cutoff.

If we followed Kirsch’s interpretation and rejected antidepressants, how would we treat depression? Psychotherapy avoids the side effects of drugs, but it has its own drawbacks: it is expensive, time-consuming, and variable in quality. How effective is psychotherapy? Psychotherapy trials also suffer from publication bias, just like antidepressant drugs. And when one weeds out low quality studies, psychotherapy has an effect size of only 0.22, lower than the value for antidepressants reported by Kirsch himself, So if we reject any treatment below the (arbitrary) 0.5 cutoff, when a mental health care provider is faced with a patient in need of help, is he or she to do nothing at all?

I don’t doubt that antidepressants have sometimes been over-prescribed and used inappropriately for lesser levels of depression where they are less effective or even ineffective, but this is probably true for psychotherapy, as well. On the other hand, it has been estimated that only about half of depressed patients are getting any kind of treatment. Severe depression is a life-threatening disease. A recent study showed that antidepressants reduced the risk of suicide by 20% in the long term. The risk/benefit ratios are still not clear cut for either form of treatment.

Once more, science fails to give us the black-and-white answers we crave. And once again we are reminded that we can’t rely on the media for accurate, nuanced information about medical science.

For his assistance in preparing this article and for providing the figure, I want to thank Erick Turner, M.D., Department of Psychiatry, Oregon Health and Science University; Staff Psychiatrist, Portland Veterans Affairs Medical Center; Former reviewer, FDA.

 

Posted in: Neuroscience/Mental Health, Pharmaceuticals

Leave a Comment (49) ↓

49 thoughts on “Antidepressants and Effect Size

  1. wbremers says:

    ” Some products that have been on the market for decades, like Prozac, are faltering in more recent follow-up tests.” http://www.wired.com/medtech/drugs/magazine/17-09/ff_placebo_effect?currentPage=all
    Placebo is becoming stronger in the last decades. Sounds strange but the article in Wired proofs it.

  2. Carl Graham says:

    As a clinician I’m not sure why the two treatments are so frequently proposed as competing rather than complimentary.

    It seems to me that initiating an antidepressant medication and then getting the patient to improve their problem solving and coping skills in an evidence based psychotherapy would deliver additive benefits.

  3. onekind says:

    Just a quick question – what’s the reference point for measurements of effect size? Does 0.30 effect size mean 30% of patients in remission?

    Also, does 0 mean ‘absolute zero effect’, or is it a baseline, relative to the efficacy of placebo in the group?

  4. draftsentdiscard says:

    As someone who has low-level (occasionally worse) consistent depression (lasted for about four years), the debate on anti-depressants has factored into my decision making process. The non-dire nature of my situation when combined with other factors like possible side-effects, money, the inexact nature of psychotherapy, etc. have all dissuaded me from seeking treatment. When all the time and resources that are involved in the process of arriving at an effective treatment regimen are combined with the unintended consequences of intervention, the means start to seem worse than the ends. At the same time, feeling happy more often may be worth it.

    Thanks for another analysis of the debate.

  5. Rick says:

    Thank You for address this Dr. Hall. I saw an interesting Article in Scientific American regarding this called “Are Antidepressants Just Placebos with Side Effects?” By John Horgan, which looks at two book reviews articles by Marcia Angell, former editor of The New England Journal of Medicine. One of the books who author you mentioned is Iriving Kirsch.

    I was hopeful you would address the issue that serotonin levels for example aren’t typically checked before prescribing an SSRI for example. If I take ACE/ARB’s long term I am going to get a BMP or CMP done, other medications taken long term would require a LFT. Doesn’t seem like the same type of monioring goes on with medications that effect the brain.

    Overall like with any medication, the prescriber must weight the risk and benefit of any prescription. However, with primary care practitioners taking a larger role in dx and treating mental illnesses, I worry how knowledgeable and comfortable they are with this.

  6. Tom S says:

    If NICE can abandon its arbitrary 0.5 as a cutoff for “effectiveness,” why can’t we all get away from the arbitrary p-value of .05 for statistical significance?

    And yes, some of these things are hard to think about, but it is important for both prescribers and PATIENTS to take some responsibility for doing so.

  7. Thanks for this thoughtful explanation Dr. Hall.

    I was diagnosed with mild depression and social anxiety many years ago. I think it’s important to understand that even with a diagnoses of “mild” depression I still contemplated suicide at least once a day at that time, but I was functional, able to work, etc.

    I did take an anti-depressant for a year or two at that time as well as participating in CBT. My understanding from the therapist and psychiatrist was that most people have the more success when using both treatments togehter. As the patient builds more coping skills they may then be able to wean off the the anti-depressants (depending upon the severity/cause of depression). This program genuinely helped me in the long term, even past the time that I stopped taking the anti-depressants.

    One thing that I found interesting about the medication I took (Paxil) was the relief that I received from my typical negative rumination thought process. All of my life, I had had an inner critic that would nag at me almost constantly from the time I woke up in the morning til going to bed and often wake me up and keep me up in the middle of the night. I made various attempt to stop these negative thoughts, but without any luck.

    After a couple month taking Paxil, I noticed that number of these ruminations had greatly diminished. Just knowing that a medication could diminish these thoughts so dramatically gave me a genuine insight into my cognitive/emotional process.

    I will also note that I have taken both St. Johns Wort, Omega-3 as well as an anti-depression supplement cocktail given to me by a friend, none of those offered the relief from ruminations that the Paxil. Perhaps that can be considered my placebo experiment.

    I will say that for me the side effects of the anti-depressants where significant (not-life threatening, but not fun either), which is why I eventually decreased, then ceased medication. But I do think that, for me, it was well worth the side-effects that I experienced, because the insights it offered me and it enabled me to participate in my therapy more thoroughly. That has had a long term benefit. It is possible that it saved my life.

  8. Angora Rabbit says:

    @Rick:
    Good question about serotonin levels. They aren’t checked because the best and most accurate way to do this involves either a headset that does brain microdialysis or parting with part of the brain. Great for animal studies but not many human takers. :) One could measure it in the cerebrospinal fluid, but this is unpleasant, associated with risks, and doesn’t inform what’s happening in the synapse. And thanks to the blood/brain barrier, the body serotonin compartment is distinct and separate from the brain compartment. Thus blood or serum levels don’t inform what’s happening in the brain.

    It may be possible to measure serotonin using sophisticated imaging technology; I have not looked deeply at this literature and defer to someone like Steve Novella here. But even if feasible, it would have to be clinically approved and quite likely might be too expensive for routine clinical use. I defer to the clinicians on this one.

  9. Despite the commercials – treating depression pharmacologically is not treating something as simple as a low level of a single neurotransmitter. Measuring serotonin levels, even if practical, would likely be of no clinical value.

    Depression is a result of poor emotional regulation among various brain regions. Drugs are a blunt tool by which we can nudge brain function in a direction which, for some people, can change this regulation and reduce depression. It’s not really about the levels.

  10. trrll says:

    One thing that crops up in most trials is that the placebo effect of antidepressants is greater than the therapeutic effect. Kirsch argues that because of their side effects, effective blinding of antidepressants is difficult, and that the difference between the placebo effect and the “therapeutic” effect is actually due to incomplete blinding, and that a proper trial would need to be against an “active placebo” with perceptible effects.

    I tend to think that there is a real, if modest, therapeutic effect of antidepressants, based on the face validity of the animal studies. But that’s a pretty weak argument. Let’s suppose that Kirsch is right, and the true therapeutic effect is clinically insignificant, if not nonexistant. In that case, Dr. Hall’s argument, “If we followed Kirsch’s interpretation and rejected antidepressants, how would we treat depression? ” sounds uncomfortably reminiscent of the arguments made by alternative medicine practitioners in defense of such treatments as acupuncture. Psychotherapy, even though in some sense it could be considered to be entirely a placebo effect, provides much of the benefit without the hazards of drugs or the ethical quandary of lying to the patient (when giving a placebo). It would seem that a rational, patient-centric strategy would begin with psychotherapy and add drugs only if the response is inadequate.

  11. Rick says:

    @trrll

    In trials using atropine as the placebo, Krirsch found there was no difference between the antidepressant and the active placebo. Everyone had side effects of one type or another, and everyone reported the same level of improvement.

    In his earlier study and in work by others, he observed that even treatments that were not considered to be antidepressants—such as synthetic thyroid hormone, opiates, sedatives, stimulants, and some herbal remedies—were as effective as antidepressants in alleviating the symptoms of depression. Kirsch writes, “When administered as antidepressants, drugs that increase, decrease or have no effect on serotonin all relieve depression to about the same degree.” What all these “effective” drugs had in common was that they produced side effects, which participating patients had been told they might experience.

    “If we followed Kirsch’s interpretation and rejected antidepressants, how would we treat depression? ” sounds uncomfortably reminiscent of the arguments made by alternative medicine practitioners in defense of such treatments as acupuncture…I picked up on that too.

  12. wlondon says:

    Steven Novella wrote (and explained at TAM9): “Depression is a result of poor emotional regulation among various brain regions.” Makes sense.

    He continued: “Drugs are a blunt tool by which we can nudge brain function in a direction which, for some people, can change this regulation and reduce depression.” How do we know that any anti-depressant drugs do anything to correct “poor emotional regulation among various brain regions”?

  13. delaneypa says:

    Can’t help mentioning the item Google Reader sent me (just before this excellent post) from JournalWatch:

    Commonly Used Antidepressants No More Effective Than Placebo in Patients with Alzheimer Disease

    Sertraline and mirtazapine, commonly used in depressed patients with Alzheimer disease, are no more effective than placebo, according to a Lancet study.

    In a double-blind trial, researchers randomly assigned patients with Alzheimer disease and coexisting depression to one of three treatments: placebo, sertraline (target dose: 150 mg per day), or mirtazapine (45 mg per day). Each group comprised roughly 100 patients.

    At 13 weeks, depression-in-dementia scores did not differ significantly among the groups — all showed improvement in scores. Adverse events, however, were more common among those on the antidepressants.

    The authors conclude that “the drugs from the two classes of antidepressants most likely to be prescribed for depression in Alzheimer’s disease seem to be no more effective than placebo.” They suggest that the drugs might be better reserved for patients whose depression has not resolved 3 months after referral to specialist services.

    http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(11)60830-1/abstract

  14. I just wanted to throw a thought out. It seems that we’ve seperated out depression from other conditions that it is often associated with. Anxiety disorders, Bipolar Disorders, PD, etc. While the criteria used to diagnose these conditions may be distinct in the DSM-? (whatever number we are on) I don’t know that we have any clear cut indication that depression is a unique and separate disease process than other mental disorders (correct me if I’m wrong, please).

    With that in mind, is testing a medication’s effectiveness for depression alone really giving us a real life view of it’s usefulness, or is it giving us an artificial view?

  15. delaneypa says:

    @Rick – “I was hopeful you would address the issue that serotonin levels for example aren’t typically checked before prescribing an SSRI for example.”

    Is there any reason to think that serotonin levels in plasma correlate with those in the CNS, behind the blood-brain barrier? I doubt it, after all we don’t routinely check dopamine levels in Parkinson disease.

  16. Rick says:

    @micheleinmichigan

    Dr. Marcia Angell reviews three books in the her New York Review of Books titled “The Epidemic of Mental Illness: Why?” Two of the three books include other mental illnesses while the Krirsch is solely on antidepressants.

    From the Review:

    …none of the three authors subscribes to the popular theory that mental illness is caused by a chemical imbalance in the brain. As Whitaker tells the story, that theory had its genesis shortly after psychoactive drugs were introduced in the 1950s. The first was Thorazine (chlorpromazine), which was launched in 1954 as a “major tranquilizer” and quickly found widespread use in mental hospitals to calm psychotic patients, mainly those with schizophrenia. Thorazine was followed the next year by Miltown (meprobamate), sold as a “minor tranquilizer” to treat anxiety in outpatients. And in 1957, Marsilid (iproniazid) came on the market as a “psychic energizer” to treat depression.

    When it was found that psychoactive drugs affect neurotransmitter levels in the brain, as evidenced mainly by the levels of their breakdown products in the spinal fluid, the theory arose that the cause of mental illness is an abnormality in the brain’s concentration of these chemicals that is specifically countered by the appropriate drug. For example, because Thorazine was found to lower dopamine levels in the brain, it was postulated that psychoses like schizophrenia are caused by too much dopamine. Or later, because certain antidepressants increase levels of the neurotransmitter serotonin in the brain, it was postulated that depression is caused by too little serotonin. (These antidepressants, like Prozac or Celexa, are called selective serotonin reuptake inhibitors (SSRIs) because they prevent the reabsorption of serotonin by the neurons that release it, so that more remains in the synapses to activate other neurons.) Thus, instead of developing a drug to treat an abnormality, an abnormality was postulated to fit a drug.

    That was a great leap in logic, as all three authors point out. It was entirely possible that drugs that affected neurotransmitter levels could relieve symptoms even if neurotransmitters had nothing to do with the illness in the first place (and even possible that they relieved symptoms through some other mode of action entirely). As Carlat puts it, “By this same logic one could argue that the cause of all pain conditions is a deficiency of opiates, since narcotic pain medications activate opiate receptors in the brain.” Or similarly, one could argue that fevers are caused by too little aspirin.

    Moreover, Whitaker contends, the natural history of mental illness has changed. Whereas conditions such as schizophrenia and depression were once mainly self-limited or episodic, with each episode usually lasting no more than six months and interspersed with long periods of normalcy, the conditions are now chronic and lifelong. Whitaker believes that this might be because drugs, even those that relieve symptoms in the short term, cause long-term mental harms that continue after the underlying illness would have naturally resolved.

    The evidence he marshals for this theory varies in quality. He doesn’t sufficiently acknowledge the difficulty of studying the natural history of any illness over a fifty-some-year time span during which many circumstances have changed, in addition to drug use. It is even more difficult to compare long-term outcomes in treated versus untreated patients, since treatment may be more likely in those with more severe disease at the outset. Nevertheless, Whitaker’s evidence is suggestive, if not conclusive.

    If psychoactive drugs do cause harm, as Whitaker contends, what is the mechanism? The answer, he believes, lies in their effects on neurotransmitters. It is well understood that psychoactive drugs disturb neurotransmitter function, even if that was not the cause of the illness in the first place. Whitaker describes a chain of effects. When, for example, an SSRI antidepressant like Celexa increases serotonin levels in synapses, it stimulates compensatory changes through a process called negative feedback. In response to the high levels of serotonin, the neurons that secrete it (presynaptic neurons) release less of it, and the postsynaptic neurons become desensitized to it. In effect, the brain is trying to nullify the drug’s effects. The same is true for drugs that block neurotransmitters, except in reverse. For example, most antipsychotic drugs block dopamine, but the presynaptic neurons compensate by releasing more of it, and the postsynaptic neurons take it up more avidly. (This explanation is necessarily oversimplified, since many psychoactive drugs affect more than one of the many neurotransmitters.)

    Books reviewed are:
    The Emperor’s New Drugs: Exploding the Antidepressant Myth
    by Irving Kirsch

    Anatomy of an Epidemic: Magic Bullets, Psychiatric Drugs, and the Astonishing Rise of Mental Illness in America
    by Robert Whitaker

    Unhinged: The Trouble With Psychiatry—A Doctor’s Revelations About a Profession in Crisis
    by Daniel Carlat

    Part 1 “The Epidemic of Mental Illness: Why?”
    http://www.nybooks.com/articles/archives/2011/jun/23/epidemic-mental-illness-why/?page=1

    Part 2 “The Illusions of Psychiatry”
    http://www.nybooks.com/articles/archives/2011/jul/14/illusions-of-psychiatry/

  17. Xplodyncow says:

    Measuring serotonin levels would be useless in my case, even if it were helpful in other patient types. SSRIs and SNRIs worsened my symptoms, but abrupt withdrawal improved my mood dramatically for roughly 4 weeks each time. (Note: Listen to your physician. Never quit SSRIs or SNRIs cold turkey.)

    I was prescribed bupropion SR as first-line treatment for MDD (atypical subtype) 4 years ago; it was the only medication that ever worked for me (until it lost its efficacy after several months). Recently, I was prescribed pramipexole off label and have begun to notice a significant improvement.

    As a patient with no medical or scientific training, it is infinitely frustrating to get accurate, reliable, and understandable information on coping with and treating depression—especially on the link between the dopaminergic system and symptoms of depression.

    Thank you for the concise and helpful post, Dr. Hall.

  18. Rick it’s hard to imagine how Whitaker could find reliable enough sources on the progression of schizophrenia before 1950 to make the claim that the natural history of mental illness has changed.

  19. onekind says:

    Dr Hall?

  20. Harriet Hall says:

    @onekind,

    “what’s the reference point for measurements of effect size?
    Does 0.30 effect size mean 30% of patients in remission?
    Also, does 0 mean ‘absolute zero effect’, or is it a baseline, relative to the efficacy of placebo in the group?”

    I’m not a statistician, but there are several explanations of “effect size” on the web. This looks like a good one: http://www.medscape.org/viewarticle/569729

  21. RyanJLind says:

    When reading articles like this I am often compelled to give my anecdote. I recognize the non-value of anecdotal evidence, but I offer my story nevertheless just as an attempt to share my perspective, and not really to try and convince anyone of one thing or another.

    –Warning, long, meandering post to come–

    I have struggled with clinical depression and anxiety for over 10 years, since I was in middle school. It is something that I think got worse and worse as I grew older but never really understood and never sought treatment for. As time went on I found myself getting more and more isolated from society, shutting myself away from what I feared to the point where I was 20, had very few friends, had literally zero relationship experience, and virtually no social life to speak of.

    When I was a teenager I would lie awake for 3-4 hours every night (every night,) fantasizing about suicide and trying to figure out how I could get away from my life without hurting my family. I had at one point convinced myself that I would end my life on my birthday had I not found something to make things better. I was in high school and after college and terrified of meeting new people or having any experiences.

    Common stuff, I guess.

    I flat-out did not want to take drugs. You have to understand that for my generation that word “drug” has a pretty negative connotation. From elementary school the notion of “drugs are bad” was heavily beaten into us and I was a good boy who had never touched a joint or even really taken an aspirin. I did not want to be on drugs and I was afraid to tell my parents that I was going to be taking drugs of any kind, especially since as a student I was on my father’s insurance plan and had to go through him to get reimbursed for any prescription.

    I digress, but the point is that I was a bit predisposed to be anti-drug, And having the anxiety problems I had it was tremendously difficult to reach out for help of any kind. Still, with only a couple days before “the birthday” I finally got in touch with a therapist of some kind. This therapist was not particularly effective in my opinion, but told me something that I believe is true and think about when reading these “therapy vs drugs” style of debates. She told me that the drugs can help you get to a certain base level where psychotherapy can begin to be effective.

    I think that this is true, at least for some circumstances, and I will explain why:

    With me, it is the anxiety that is probably the biggest issue. If I’m invited to a party, and I couldn’t find a way to get out of it, I would begin to plan things as much as possible. Map out all of the people I know who would be there, think about the possible conversation topics that may arise, plan out possible responses to those topics, think about what responses I would get to each response I give, think about what response I would then give to those responses, depending on who said them and so on and so on. I know and knew this to be a worthless exercise. None of the plans ever come close to fruition and within minutes of being at the party I would always find myself unable to use any of my planned things I was going to say. Having to come up with things on the spot and not having the time to think about how others may respond led to tremendous anxiety, which is where the physical problems kick in. Feeling nervous, I would start to sweat, I would notice my heart beating faster, I would sometimes start shaking, and so forth. Worse, conscious of my sweating, I become hypersensitive to the people around me. Do the people around me notice that I’m sweating? Can the person next to me hear my heart beat? This amplifies the nerves and makes me sweat more. And so forth.

    This is a very brief window into the anxiety issues I’ve had for most of my life, very familiar I am sure to psychologists and to people who have similar issues. There are many more aspects to it but I am going to stop there firstly because I am sure you are bored to tears if you’re still reading this, and secondly because I’ve gotten to the point of this anecdote where I want to explain what I meant several paragraphs ago.

    Because the first thing that I noticed after being on anti-depressants (Paxil) for awhile was that those physical symptoms stopped. And that, for me, was tremendously important. I might still feel nervous, but if I’m not sweating profusely then I’m not self-conscious about my sweating, which means I don’t sweat more and that vicious cycle never begins.

    After I finished college, but before I was on Paxil (I started on Cipralex which I found unhelpful,) I was working at a local software company as (what else?) a computer programmer. Like at most companies, at least software companies, we had weekly meetings where we talked about our progress, discussed roadblocks, and so forth. At the end of these weekly meetings we did something called a “round-table.” During the round-table each member of the team was to share a personal item about himself that wasn’t related to work. This is of course a fairly transparent attempt at “team-building” that was to “bond” us by having us talk about what we did on the weekend, what grade our kid is starting, and all that other nonsense.

    I am not sure how adequately I can put into words how terrified I was of the round-table. I would spend days in advance of the meeting trying to think of something to say that would be interesting to the people of the team and not give a negative impression of myself. Everything I could think of I would carefully analyze and think about what each person in the room’s reaction would be to my piece of personal trivia. What would be my co-workers impression of me afterwards? Does revealing this information adequately reflect my personality? Does it reveal too much personal information that would allow others to infer other personal information that I don’t want them to infer? Things like my age, my relationship situation, and so forth? And on and on and on. Usually failing to come up with something interesting I would first attempt to get out of the meeting or at least leave before the end. Maybe I can have someone call me so I have to duck out early. Before the -haha-lighthearted roundtable that is supposed to be the easiest, most “fun” part of the meeting.

    All of this is to say that if I had to be there for round-table, the physical symptoms were ridiculous. I would be almost literally gripped with fear as I sat in my chair and watched it move around to me, such that I wouldn’t even be able to hear what my co-workers are saying because I’m focused on oh my god my heart is beating so loud and my palms are soaked and what am i going to say and can Dave tell how nervous i am oh my god this is so embarrassing and. But it’s pretty much just fear. The same kind of fear I would presume one with claustrophobia would have if locked in a closet, or that character at the end of 1984. I am not afraid of spiders, heights, closed spaces, open spaces, or anything like that. It was people that terrified me, people and their judgment.

    I am rambling again but it was only after I had been on anti-depressants for a few months that for the first time in my life I could sit in a situation as above and not have those physical symptoms of anxiety. Sure, I might still be nervous. But I wasn’t PANICKED. I didn’t feel dizzy; I wasn’t sweating like I was in an oven, and the lack of physical symptoms allowed my mind to focus more on rational thinking than worrying about who can notice how worried I am.

    This is what I think the therapist meant by “base-level.” With the drugs taking care of some of the physical symptoms, I am more able to focus on the Cognitive Behavioural Therapy type stuff that I’ve read about online and in books. I haven’t gone back to another psychologist*, but the difference between my life today and my life even two or three years ago is night and day. I can take girls out for dinner now. I am still nervous, as most people are on first dates, but at least my mind feels clear and focused ENOUGH that I can attempt to assuage the anxiety with the techniques I have learned from reading about psychotherapy (replacing negative thoughts with positive ones, and so on.)

    I am not going to sit here and say that my life is perfect now. I don’t jump out of bed in the morning or strut around smiling at people all day. I still have issues with self-confidence and self-worth. I still have anxiety problems. I still have suicidal fantasies, particularly when trying to sleep. I am still prone to having complete emotional meltdowns when something bad happens like getting rejected or what have you. But I take risks now that I never would even consider before. I might feel like killing myself after a bad date, but the mere fact that I went on that date to begin with is a huge step that I wouldn’t even consider in the days where the mere thought of spending anything like intimate time with someone would cause me to shake and feel dizzy.

    In short, I believe that the drugs that I have been prescribed have been effective at minimizing physical symptoms of anxiety that have allowed my mind to be more focused to work on the other aspects of the anxiety that I need to work on to be a functional member of society. I am not “cured,” and probably never will be, but I think to dismiss a small effect as being the same as “no effect” is extremely ill-advised, in my opinion.

    I am sorry that this comment is longer than Harriet’s article, but I’m just trying to say that my personal experience essentially agrees with what conclusions one should probably draw from it. I think the drugs are helpful, perhaps even life-saving, but they are hardly going to magically turn a depressed person into a confident one. The small effects they may have can help maybe get them to a base-level which allows the therapy to penetrate and take effect over long periods of time.

    *not true, I had three sessions with a registered psychologist who tried to heal me with meditation and helping the chi flow through my chakras, but that’s another story for another day…

  22. evilrobotxoxo says:

    I’m an MD/PhD neuroscientist in my last year of psychiatry residency, and my take on it is this:

    The main problem with using this kind of research to guide clinical decision-making is that placebo-controlled antidepressant trials to obtain FDA approval are designed to determine whether a drug is effective beyond placebo in an idealized population of research subjects, not to measure the effect size in a realistic patient population. For example, subjects are often chosen so that their depression does not have comorbid anxiety, substance abuse, medical problems, etc., many of which are the norm in clinical practice. Also, it is commonly stated that these studies show that only patients with the most severe depression benefit from antidepressants. In reality, patients who actually have “severe” depression by any reasonable clinical definition (e.g. active suicidality, poor self-care, depression with psychosis, etc) are excluded from these studies. The patients who have moderate depression are labeled as having “severe” depression, those with mild depression are labeled moderate, etc. Meanwhile, the only large clinical trial of antidepressants on a realistic patient population, the STAR*D study, is so badly designed as to lack placebo controls or even full randomization. So the bottom line is, the data clearly shows that antidepressants are not a placebo, but no one really knows what the effect size is. Antidepressants clearly have some effect, but they’re obviously not a miracle cure either.

    A second issue is the question of which indication antidepressants are being used for. In particular, there is data supporting the use of antidepressants for anxiety disorders such as panic disorder, OCD, generalized anxiety, etc, and it is clear that they are effective for these conditions. The “antidepressants are placebos” argument looks narrowly at using antidepressants to break a major depressive episode, while ignoring their other uses. Of note, the effect sizes for antidepressants in preventing relapse of depression are much higher than in actually breaking acute depression. People often try to write off those studies by saying that this is due to discontinuation syndrome, but that argument is pretty unconvincing because the effect is seen even when people are slowly tapered off their meds, and also because in the placebo group the relapse frequently occurs months after the active medication is stopped, long after any discontinuation symptoms are gone.

    A final issue, and perhaps the most important one, is that depression is not a monolithic disorder, not even a single syndrome, but a broad collection of related syndromes. Looking at things like effect sizes in terms of scores on rating scales like the HAM-D tacitly assumes that depression is a single entity that should respond the same way to a given treatment. However, any psychiatrist will tell you that some patients improve on traditional antidepressants while some get worse, some patients do better with mood stabilizers, some need thyroid hormone supplementation, some need psychotherapy, some need more exercise, some need their sleep adjusted, some need a light box, some need ECT, etc. Usually there’s a lot of trial and error involved, but most psychiatrists think that they can predict which interventions will work at better than chance levels, and I actually think that’s true to some extent; humans are excellent at pattern recognition. However, the fact remains that only a fraction of this clinical decision-making is actually supported by any real data. Not because we’re in the pocket of big pharma, and not because we don’t think evidence is necessary, but because it’s hard to formalize clinical impressions, and many practical issues including the high variability in the data and large number of possible interventions make it difficult to get good statistics in a clinical study. Therefore, we’re stuck doing the best we can with relatively low-quality data to guide us.

    A final point is that it is frequently stated that psychiatrists are strong believers in the efficacy of antidepressants, but in reality it is primary care practitioners who prescribe the bulk of antidepressants, and most psychiatrists almost exclusively see patients who were referred after a trial of antidepressants *didn’t* work. So trust me – psychiatrists are well aware of the limitations of these medications. But to say that they’re placebos is absurd and clearly contrary to the existing evidence.

  23. evilrobotxoxo says:

    @Rick:

    The content of the Angell book review is partly true and partly false. It is true that antipsychotics (e.g. chlorpromazine) were discovered by chance, and it was later found that their mechanism of action is blockade of the D2 receptor. And it was later confirmed by PET imaging that schizophrenics who are actively psychotic actually are in a hyperdopaminergic state. However, no one believes that schizophrenia is *caused* by excess dopamine, only that it is associated with a hyperdopaminergic state and that blocking the D2 receptor improves psychotic symptoms, and this is true for psychotic symptoms arising from other disorders as well. It’s worth noting that most schizophrenics still don’t do that well in the long term, but that’s because the cognitive and negative symptom clusters aren’t helped by D2 antagonists, and those are ultimately more debilitating than the psychotic symptoms. Also, it’s flat out absurd to claim that schizophrenia used to be a transient condition that people spontaneously recovered from. Before the development of thorazine in the early 1950s, a huge fraction of every state’s budget went to supporting massive state hospital complexes housing huge numbers of chronic schizophrenic patients. The desperation arising from the inability to treat this large pool of difficult-to-manage patients is part of what led to the acceptance of the lobotomy in that era.

  24. evilrobotxoxo – what thorough and thoughtful comments. Thanks so much for sharing your perspective.

  25. Rob Tarzwell says:

    I’ve given a lot of thought and attention to the question, “Do antidepressants work?” Unfortunately, you can’t get there from here is the shortest answer.

    First, we need to define depression. Even if we accept the DSM-IV-TR version, where you need 5 of 9 symptoms, one of which must be symptoms 1 or 2, basic combinatorix tells us that is 70 ways to be depressed (2*(9 choose 5)). Which of these 70 types is more common? Are these commonly clustering kinds, and if so, what are their manifestations at the level of neural substrate? If there are naturally occurring depressive subtypes, it seems reasonable to suppose that these require different treatments, perhaps SSRI’s for some, SNRI’s for others, stimulants for still others, and so on.

    Next, when we turn to the literature, we find that the most commonly used tool for measuring changes in depression, the Hamilton Rating Scale for depression (HAM-D), we find it has 21 items, not 9, many of which simply don’t appear in the DSM. Also, rather than simply being present or absent, the items have severity scales. So, what syndrome(s) is it we are investigating? Should we assign hypochondriacal depressives with prominent GI symptoms into their own study, for instance, and this opens the door to legitimate questions about how well we carve nature at its joints in these one-size-fits-all medication trials.

    Then there is the ongoing debate about the medications themselves. Do they work? They appear to, modestly. Yes, when compared with active placebos, this unfortunately appears to vanish, though the published active placebo trials are generally pretty small and have significant methodological problems.

    Finally, as has been noted above, the patient in the consulting room is a lot different from the person who answers a Google ad recruiting volunteers for the medication study. The people in the clinic have rotten depression, rotten childhoods, rotten lives, and heaps of other problems. Did the depression get better because he took Paxil, or was it because he found a job and a girlfriend after a layoff that cost him his home and his marriage? I don’t know, but I know for sure he’d be shown the door in a hurry if he tried to enrol in a drug trial.

    To summarize: I’ll be able to tell you whether antidepressants work as soon as you tell me what the heck depression is and what the heck we’re measuring in drug trials, and whether or not that relates at all to the folks in my waiting room.

  26. Rob Tarzwell says:

    Sorry, that formula should be 2*(7 choose 5), but the answer still comes out to 70.

  27. Tell it like it is says:

    “Psychotherapy avoids the side effects of drugs, but it has its own drawbacks: it is expensive, time-consuming, and variable in quality.”

    ‘Expensive’ is relative and a price is based upon what the market will bear. You pays yer money and you takes yer choice. If any commodity or service is considered too ’expensive’ people ‘vote’ with their feet (well in Britain they do anyway) and the punter does not return unless or until the price drops.

    ‘Time consuming’ is relative and irrelevant to the argument. Heart surgery is ‘time consuming’ – how relevant is ‘avoiding’ needed heart surgery because it is ‘time consuming’ to carry out.

    ‘Variable in quality’ – a misnomer – do your mean variable in ‘effectiveness’ (it works) and ‘efficacy’ (it fixes it for good)?’

    Hypothesis testing is the art of testing whether a variation between two sample distributions can be explained by chance or not. For clarity on the ‘0.5’ measure (i.e. the half full half-empty postulation) cited, is evaluating a thing called the ‘null hypothesis’.

    Ignoring the argument as to whether statistical analysis has any worth, from the ‘Bayesian’ point of view, a type I error is one that looks at information that should not substantially change one’s prior estimate of probability, but does. A type II error is that one looks at information which should change one’s estimate, but does not.

    For clarity:

    A ‘type 1’ error (known as ‘the error of excessive credulity’) would show a woman had cancer when she did not (false positive)

    A ‘type two’ error (known as ‘the error of excessive scepticism’) would show a woman did NOT have cancer – when in fact she DID and requires urgent treatment (false negative).

    Which is the worse error?

    In statistics, all effort is made to minimise the occurrence of a Type I statistical error!

    “How effective is psychotherapy?” About as effective as a Tarot reading – but a Tarot reading is more fun – and less costly – and so gives better value.

    In the behavioural sciences it is well established that temperament and disposition determine character. All ‘traits’ – be it colour of hair, shape of lip, or choleric disposition are determined to be genetic. This was first postulated in a seminal work by an Austrian monk called Gregor Mendel – which completely shot down Charlie Darwin’s pangenesis poppycock (humans evolved from apes – yeah right – everyone knows its crocodiles).

    It is now shown that the laws of psychology and sociology are derived from ‘biological’ laws. The logic is ‘Exclusive OR’ or XOR – and the rule is ‘different = TRUE, same = FALSE. *I will demonstrate this and its relationship to Mendel as a footnote below.

    Sadly, two guys – Sigmund Freud and Ivan Pavlov – almost succeeded in throwing psychology and psychiatry back into the dark ages.

    To illustrate, ‘beriberi’, also known as Wernicke-Korsakoff syndrome, is a brain disorder caused by a lack of vitamin B1 (thiamine) in the body that results in a number of neurological symptoms that lead to psychosis, confusion, hallucinations, and severe depression. Inherited beriberi creates a dysfunction in a person’s ability to take up and use thiamine.

    The most common cause of beriberi in the United States is alcoholism. Heavy consumption of alcohol inhibits the body’s ability to absorb thiamine, leading to the dreaded symptoms I have just cited. However, in the United States beriberi can also be found in formula-fed nursing infants whose formula is inadequate in thiamine.

    Freud would have us believe the condition was motivated by lust, and Pavlov would have us believe it was motivated by hunger (AKA longing). How prejudice is THAT? Did both men have a grudge against the other six deadly sins and the seven virtues?

    Moving on: “I don’t doubt that antidepressants have sometimes been over-prescribed and used inappropriately for lesser levels of depression where they are less effective or even ineffective.”

    First: ‘Less effective’ than what?

    Second: If antidepressants are ‘ineffective’ then they do not work – period.

    “On the other hand, it has been estimated that only about half of depressed patients are getting any kind of treatment.”

    Apart from begging the question how this allegation is ascertained, what is your point?

    “Severe depression is a life-threatening disease. A recent study showed that antidepressants reduced the risk of suicide by 20% in the long term.”

    Now: in the context of the sentence, ‘recent study’ and ‘long term’ is incongruent. If you were to say a ‘long term study recently published’ it ‘might’ have some credence.

    However! How can you or anyone else possibly claim a ‘reduction’ of suicide? I and my fellow contributors have not committed suicide – does this bring the incidence rate of suicide down? I think not.

    *George Boole devised several logical ‘gates’ to define logic and gave the world ‘Boolean Algebra’. His mechanical ‘differential engine’ – a wonder to behold – can be seen working in the British Science Museum.

    Each logic gate validates a ‘truth’ by comparing two ‘assertions’.

    For the descriptive purposes, I will use only TWO assertions: Assertion ‘A’, and Assertion ‘B’, and the ‘validation’ I will call ‘C’.

    The Boolean gates are: AND; OR; XOR; and NOT

    Each gate has an ‘inverse’ gate: NAND; NOR; XNOR; and IS, which are ‘mirror images’ of the gates and used in ‘inverse logic’ calculations – a bit like a negative quantity in algebra.

    Each gate has a ‘Law’, a ‘truth table’, and a ‘mathematical operator’ to represent its application, and a ‘pictogram’ to facilitate the drawing of schematics when constructing electronic/bio circuits that can ‘think’ – such as a digital microchip or an organic culture.

    The AND gate (the fastest and most potent) evaluates CONTRADICTION of the STATEMENTS under evaluation.

    The law: Any FALSE = FALSE. This equates to ANY LO = LO

    The truth table:

    A is FALSE AND B is FALSE = ASSERTION is FALSE (all of what is being stated is false)
    A is FALSE AND B is TRUE = ASSERTION is FALSE (one party contradicts the other)
    A is TRUE AND B is FALSE = ASSERTION is FALSE (the same contradiction rule)
    A is TRUE AND B is TRUE = ASSERTION is TRUE (everything being stated is true)

    The mathematical operand is MULTIPLY (where 0 X 0 = 0; 0 X 1 = 0; 1 X 0 = 0; and 1 X 1 = 1)

    The XOR gate tests CONGRUENCY of the ASSERTION (are BOTH assertions DIFFERENT to each other?) and does NOT evaluate the validity of the STATEMENT being asserted (yes it is weird – it is the ‘Sophist’ gate – not interested in validating the subject under discussion – merely interested in destroying an opponent).

    The law: Both the SAME = FALSE. Both DIFFERENT = TRUE.

    The truth table:

    A is FALSE AND B is FALSE = ASSERTION is FALSE (both parties AGREE and so to say there is a ‘difference of opinion’ is FALSE)
    A is FALSE AND B is TRUE = ASSERTION is TRUE (one of the assertions is saying something DIFFERENT to the other – it is therefore TRUE that there IS a difference of opinion)
    A is TRUE AND B is FALSE = ASSERTION is TRUE (again – one of the assertions is saying something DIFFERENT to the other)
    A is TRUE AND B is TRUE = ASSERTION is FALSE (both parties AGREE so to say there is a ‘difference of opinion’ is FALSE)

    The mathematical operand is ADD (where 0 + 0 = 0; 0 + 1 = 1; 1 + 0 = 1; 1 + 1 = 2

    Confused? In BINARY arithmetic, the ‘2’ is ‘carried over’ to the next position so the assertion at THIS position = the output is ‘not’ TWO so the test yields FALSE. It is the VALIDITY OF SAMENESS that is being examined – hence the ‘law’ clarifies the intent.

    Depressed? Back to the Mendelian XOR gate.

    Mendel discovered that when crossing white flower and purple flower plants, the result is not a ‘blend’ as Charlie Darwin postulated. Rather than being a mix of the two colours, for EVERY trait, each offspring has a ‘dominant’ characteristic and a ‘recessive’ characteristic. The dominant gene, such as the purple in Mendel’s plants, will hide the recessive gene – the white – but BOTH are still present in the succeeding offspring.

    Pick two traits? Say – bad eyesight and blonde hair. In each individual, we have a dominant characteristic and a recessive characteristic. The ‘odds’ of each trait turning up in siblings follows the XOR gate logic.

    Given the four states, from examination of the truth table we see

    There is ONE occurrence (25%) when a sibling will NOT have bad eyesight and will NOT have blonde hair.
    There is ONE occurrence (25%) when a sibling will HAVE bad eyesight but will NOT have blonde hair.
    There is ONE occurrence (25%) when a sibling will NOT have bad eyesight and WILL have blonde hair.
    There is ONE occurrence (25%) when a sibling will have bad eyesight and ALSO have blonde hair.
    There are THREE occurrences (75%) that determine whether a sibling has bad eyesight.
    There are THREE occurrences (75%) that determine whether a sibling has blonde hair.

    This law contributes to the genetic variability of progeny.

    Does this mean that if a mother gives birth to four siblings all the above ‘possibilities’ will turn out?

    Answer NO! The ‘odds’ are the same for EVERY child – a 25% chance of NOT having bad eyesight and a 75% chance that the child will have blonde hair.

    Mendel summarized his findings in two laws; the ‘Law of Segregation’ (chromosome crossover) and the ‘Law of Independent Assortment’ (AKA as ‘The Inheritance Law’).

    Going a little deeper: A mutation in a single gene can cause a disease to be inherited. Such diseases include beriberi, cystic fibrosis, and sickle-cell anaemia.

    The push is to determine the relationship from biology to chemistry, and the associated psychosomatic states. Once we understand the root cause, we can devise appropriate treatments.

    Right now we remain with pills, potions, and placebos until someone makes a stand and devotes their life’s energies to resolving this paradox. Is it YOU!

    Regards to one and all

    TILLIS

  28. daedalus2u says:

    I have been traveling and just got back from TAM. I made a detour north and got samples from a couple of hot springs, looking for thermophilic ammonia oxidizing bacteria, which is why I did not comment on this thread sooner.

    I disagree that anecdotes are of no value, people and their brains are complicated enough that each person is a unique individual and in terms of mental health there are only anecdotes unless you simplify things to cartoon levels. My story is not that different than Ryan’s. The details are different, but not that different.

    If you don’t have experience with depression, then you don’t understand it and the Dunning Kruger effect looms gigantically large. It is extremely difficult for a lay person to understand depression unless they have personally experienced it. Clinicians can develop an understanding without being depressed, but it takes significant study, experience and empathy from working with and understanding people who are depressed.

    I think the hypothesis Dr Novella mentioned, that “Depression is a result of poor emotional regulation among various brain regions.” is not correct. I appreciate that the statement isn’t really meant as a “hypothesis”, just more of a description of what is going on and isn’t meant to really be a falsifiable hypothesis. However, I think it is unfortunate in that it frames “depression” as something “wrong” with the person who is depressed. I missed most of the early part of his talk at TAM (I was out getting supplies for my expedition) but I did catch the part about physiology and traits existing on a distribution, and two sigma from the mean is considered to be “abnormal”. I don’t think there is anything “abnormal” about depression, where “abnormal” is defined to mean that the brain is behaving or exhibiting behaviors or effects that it did not evolve to produce. In other words, I think that depression is a “feature” that has been positively selected for by evolution.

    I see depression as the necessary aversive state between the “normal” at rest state and the euphoric state of near death metabolic stress, the kind of stress one gets when running from a bear. That extreme metabolic stress state has to be euphoric if organisms are to be able to escape and survive. If organisms could enter a euphoric state easily they would and uselessly risk death and injury. There has to be an aversive state between the at rest state and the euphoric state. I think that aversive state is depression and is brought about by physiology. Evolution has minimized the sum of deaths from being caught by the bear, by running yourself to death because it feels good, and by suicide.

    Neurodiversity is (I think) the most important and characteristic human trait of humans. My hypothesis is that human neurodiversity is built-in, in utero, and mostly during the first trimester when the most fundamental structures of the brain are forming. Because the maternal pelvis is limited in size, the infant brain at birth is limited in size. Because what ever behaviors a brain can do requires neuroanatomy to instantiate those behaviors and (in general) neuroanatomy used for one thing can’t easily be used for something else. Because the infant brain is limited in size, the infant brain must be optimized for the cognitive tasks the individual must do over its lifetime. A large brain is so valuable from an evolutionary standpoint, that humans evolved under conditions (pre-medical C-section) where a few percent of women died per pregnancy due to cephalopelvic disproportion. What evolution did was minimize the sum of death and non-reproduction from having too small a brain to compete with your peer cohort and from dying (and killing your mother) from having a brain that is too big.

    The reason that big brains are valuable, is because big brains can do things (make stone tools, weave rope, weave baskets, ferment grain, use fire, use language, hunt, etc.) that are valuable. Doing something for the first time is always more difficult than doing it a second, third and nth time. A tribe doesn’t need multiple people that can do multiple things for the first time, the tribe needs one person who can do it for the first time, and then for the rest to learn how to do it from that first person.

    A tribe with individuals that each are expert in one of many different things and non-expert in many minus one things, will do better than a tribe with all individuals having equal expertise in all things. Specialization of labor allows the tribe to draw on the best that each individual can produce. Specialization allows individuals to dedicate larger volumes of brain for their specific expertise (and become much more expert), while relying on the rest of the tribe for what they are non-expert in.

    Because members of a tribe are related and so share many genes, the mechanism for the differences in cognitive abilities can’t be genetic (because the tribe shares the genes that make the tribe successful). Evolution must have configured physiology to use non-genetic mechanism(s) to generate the dispersion in cognitive abilities. I think that mechanism is stochastic resonance in utero, coupling neurodevelopment to the environment at the level of noise.

    I think it is unfortunate that researchers are using MZ and DZ twin differences as a measure of “genetically” inherited cognitive attributes. MZ twins share an in utero environment and sometimes even a placenta. Development is a non-linear product of the combination of the genome and the environment at the level of noise. Brownian motion of molecules will have effects on development that are not predictable or reproducible.

    I think that brain development is so sensitive to environmental effects in utero because a tribe needed neurodiversity. Tribes that had experts in many different things did better than tribes that had only one type of expert.

  29. badrescher says:

    “Just a quick question – what’s the reference point for measurements of effect size? Does 0.30 effect size mean 30% of patients in remission?

    Also, does 0 mean ‘absolute zero effect’, or is it a baseline, relative to the efficacy of placebo in the group?”

    I noticed that Dr. Hall’s link hits a pay wall, so I thought I’d answer this. Excuse me if I missed the answer in other comments.

    What an effect size of .3 means differs with the measure, but in general it means that 30% of the variability in depressive symptoms can be attributed to the treatment condition (placebo verses drug).

    IMO, effect sizes are not as meaningful as many researchers suggest. There is no standard way to interpret effect sizes because they depend on context (e.g., experimental design and controls). Even comparing them (e.g., drug vs. psychotherapy) is somewhat meaningless. Also, mediating and moderating variables differ among treatments and most remain unidentified.

    The expected value of any effect size is zero, but the variance is unknown and different in each case.

  30. trrll says:

    The standard animal assay for antidepressant efficacy is to put a rat or mouse in a beaker of water, and measure how long it struggles to escape before it “gives up” and floats passively. As a model of depression, this has certain face validity. However, it is worth noting that in this model, the animal cannot escape, so struggling is futile, while floating and conserving energy is arguably the more adaptive behavior. So it may be that depression evolved as an adaptive mechanism, but that like the immune response, it can have adverse effects in extreme or abnormal situations (which on a Darwinian time scale, could characterize much of the way humans live today).

    Of course that is if you believe that antidepressants really have some genuine efficacy against depression.

  31. Tell it like it is says:

    @ daedalus2u What you have written is pure poetry. So beautifully put.

    We see from mitochondrial DNA studies passed on only through the female line that mitochondrial DNA mutates about 20 times faster than normal nuclear DNA, which substantiates your ‘big-brain little-womb’ statement. You only need to visit the ‘Axe factory’ in the ‘Rift Valley’ to see your ‘give a man a fish and he will eat for a week – ‘teach’ a man to fish and the ‘tribe’ will eat forever’ allegory.

    There is a fascinating dialogue in ‘Tristram Shandy’ by Laurence Sterne, on breach-birth and caesarean birth – the postulation is that the caesarean/breached children are inherently brighter because these infants’ do do not suffer compression of the head when they are born.

    That aside, to me there are many values in life and I believe that we are not just aiming at a ‘rational end’, and that ‘hitting bottom’ is necessary for the full development of the personality; and that despair and other negative emotions are part of the ‘human condition’ that must be satisfied to make emotional and spiritual progress.

  32. tills “That aside, to me there are many values in life and I believe that we are not just aiming at a ‘rational end’, and that ‘hitting bottom’ is necessary for the full development of the personality; and that despair and other negative emotions are part of the ‘human condition’ that must be satisfied to make emotional and spiritual progress.”

    I agree that negative emotions are a useful thing and contribute to growth. But I also think that positive emotions are needed to make life worth living. The problem with depression or an anxiety disorder is that over time the negative emotions and thoughts consume such a large portion of ones life that there is very little or no pleasure.
    This is not a good motivation for learning, growth or continued life.

  33. Pops, I mean TILIS, sorry for misspelling your acronym.

  34. oops, not pops, damn Mrs. iPad.

  35. Dr. Rob Tarzwell’s comment above should be required reading on this topic.

  36. wlondon asked Dr. Novella, “How do we know that any anti-depressant drugs do anything to correct ‘poor emotional regulation among various brain regions’”? I will presume (ha!) to answer: we don’t know, of course. It’s tinkering.

    Humans have a long history of finding ways to mess with our own heads. There are times in life when almost any change in mental state feels like an upgrade.

  37. evilrobotxoxo says:

    @Paul:

    It absolutely is tinkering, and we don’t know HOW it works, but we do know THAT it works, at least for some people. We know from clinical trials like the ones discussed by Dr. Hall, and we know from functional imaging studies before and after treatment. It’s interesting that knowledge of mechanism is not always strongly correlated with actual effectiveness. There are drugs like lithium that are more effective than SSRIs, but their mechanism is even less well understood.

  38. @eveilrobotxoxo, yep, agree. Although I also think that there is an interesting and profound disconnect between knowledge of mechanisms and their therapeutic implications, it’s sure not surprising in this case. With SSRIs, we know “a” mechanism of “an” action — they block the reuptake of serotonin, whoopty-do — but we know nothing about how this may (or may not) lead to improving a patient’s unpleasant mental state. No one takes SSRIs because they are feeling like they could really use a little more serotonin knocking around their synapses, because depression is very clearly not a disease of low serotonin— hyposerotoninism?

  39. evilrobotxoxo says:

    @Paul:

    If depression were caused by low serotonin, then SSRIs would have a rapid and dramatic clinical effect. Obviously, that’s not the case. However, there are medications that do have a rapid and dramatic clinical effect, like benzos for anxiety, but even the fact that benzos work by boosting GABA transmission doesn’t mean that anxiety is caused by a deficiency of GABA. It’s pretty clear that’s not true. I think mechanistic understanding is important, but ultimately overrated, in the context of diseases involving intrinsic dysfunction of human physiology because it often does not provide a useful therapeutic target. For example, there are numerous underlying causes underlying routine hypertension encountered in the outpatient setting, and the drugs we use to treat it (usually successfully) aren’t actually based on correcting the underlying disorder. All that matters is that we bring the blood pressure down; it doesn’t matter how we do it. Similarly, I think the future of psychiatric treatment will not need to focus so much on figuring out the various mechanistic underpinnings of individual cases of depression, for example, but focus instead on finding useful therapeutic targets in the general circuitry of mood regulation so that we can fix it without ever understanding exactly how it was broken.

  40. Rob Tarzwell says:

    Hey all, I notice part of the discussion focusing on the monoamine hypothesis (depression = low serotonin). That actually hasn’t really been part of any serious research in the field in at least 5 years. Current models of depression are gruesomely complex (surprise, surprise) but ultimately more powerful.

    The monoamine hypothesis made a lot of sense in the 60′s when Arvid Carlsson first demonstrated monoamines in the brain and in particular when he demonstrated the link between dopamine and the severity of parkinsonian symptoms, ultimately leading to his Nobel Prize.

    It was a legitimate agenda to pursue, and so it was pursued. However, it’s essentially fallen out of favour because it simply didn’t withstand scrutiny (e.g. Reserpine, which is an antihypertensive with antidepressant effects actually *depletes* serotonin in the synapse).

    Any psychiatrist who has kept up with the research really only uses “chemical imbalance” as an explanatory metaphor, because most patients’ eyes would glaze over if we started talking about modification of serotonin and noradrenaline receptor levels via secondary messenger systems to promote amplification of receptor protein translation and transcription. My eyes glazed over just writing that basic sketch.

  41. evilrobotxoxo says:

    @Rob Tarzwell:

    You’re correct that no one who actually thinks seriously about mechanisms of depression considers it to be a “chemical imbalance,” deficiency of serotonin, etc. However, it’s also not a molecular-level condition involving intracellular signaling pathways as you suggest. I think the best way to think about depression, or psychiatric disease in general, is by analogy to cardiac dysrhythmias. This is because neuropsychiatric symptoms ALWAYS correspond to some abnormal pattern of action potential firing in the brain. In the most crude sense, if you have an obvious lesion, there’s an abnormal activity pattern because the cells are dead. However, psychiatric conditions, almost by definition, are syndromes where you get abnormal activity patterns in the absence of an identifiable lesion. I wouldn’t say that psychiatric conditions are brain dysrhythmias per se because activity patterns in the brain are more complicated than in the heart, and the “correct” activity patterns are not always rhythmic, even though activity in much of the brain is rhythmically organized.

    In any case, people typically don’t think about the pathophysiology of cardiac dysrhythmias as being related to the agents used to treat them. The agents are merely used to tip things in one direction such that the heart’s dynamics, or at least the dynamics of the ventricles, go back into an activity pattern that yields an acceptable range of cardiac output. This is essentially what we try to do in psychiatry: use medications and psychotherapy to adjust the brain’s dynamics back into a range where there are fewer symptoms, better functioning, etc. The heart is a lot simpler than the brain is, and it’s possible to make direct measurements of the electrical dynamics as well as the hemodynamic functioning of the organ; in psychiatry, all of this still has to be done solely by clinical observation.

    As a result, most MDs have a fairly clear understanding of what the relationship is between abnormal activity patterns in the heart and the medications used to treat them. Likewise for antiepileptics used in neurology. The relationship in psychiatry is exactly the same, but many people, including most MDs, somehow think it’s different.

  42. Rob Tarzwell says:

    Your explanatory simile of cardiac dysrhythmia is compact, efficient, and lovely. Thanks for offering it.

    I’m not sure what most MD’s think the relationship between psychiatric medication and psychological difficulties is, but certainly some of them are disquieted by it, and others scream about it. I’ve often wondered if the Deep Dark Fear is that, once you admit that human personality, the very Cartesian self, is just as tractable to pharmacologic approaches as the heart, you’ve revealed the awful truth that *gasp*: men are machines!

    It’s purely speculation on my part, but people get very, very anxious to even contemplate that their actions, their very thoughts, are comprehensible, even expected, given their developmental arc. I remember once explaining Bowlby’s concept of secure and insecure attachment, including its predictive power over life success, satisfaction, and psychopathology to a friend, an exceptionally intelligent woman. Her reaction was, “Wow! I always thought I had free will.” The conversation topic quickly got changed.

    Now, while I agree that psychotherapy does change neuronal activity patterns, I’m always mindful of taking the concrete step into some kind of neuro-phrenology. Just because psychotherapy changes neuronal firing patterns does *not* mean psychotherapy is a neurological intervention. Learning French or taking a wine appreciation course also changes neuronal firing patterns, but nobody would seriously defend the notion that this is therefore a species of neurological intervention.

  43. evilrobotxoxo says:

    @Rob:

    Thanks for your kind words. I think you hit the nail on the head, that a big part of it comes down to people refusing to accept the fact that at the end of the day, we’re all meat robots. I think your issue with psychotherapy can’t avoid getting into the semantics of the phrase “neurological intervention.” I would argue that psychotherapy is an intervention if it is undertaken with therapeutic intent. How “neurological” it is is the question. I think that once you accept the fact that information storage is inevitably physical, including in the brain, you realize that any sort of learning or experience involves permanent physical change to your brain, whether it’s learning French or going to a wine bar (assuming you can remember it later). I don’t know if that makes it “neurological” or not, but I’m also not sure that it would matter if it did. It’s funny because I actually feel like a private language tutor sometimes in therapy sessions, except instead of teaching people to talk, I’m trying to teach them to manage their emotions, control their symptoms, and minimize self-destructive behaviors. In a way, teaching someone French is training their brain to do something it couldn’t do before, and I think that psychotherapy is actually a lot like that.

  44. Rob Tarzwell says:

    Yeah, well said. It boils down to our conceptual analysis of both “neurological” and “intervention.” Surely an SSRI is a neurological intervention (with a psychiatric purpose). Surely a wine appreciation course has a neurological impact (without being either “neurological” or “interventional” by intent).

    Once we decide we are monists from skin to marrow, and this is fundamentally a philosophical question, not yet a scientific one, then indeed, all interactions are neurological. Grade 1 affects the brain, as do SSRI’s, as does that horrible bout of pneumonia (treated with antibiotics which, as it happens, do *not* directly affect the brain, though the experience of rapid recovery profoundly affects the brain via the teachable psyche).

    The shoreline of wonder indeed outstrips the island of knowledge, but the island is still real, and it grows ever larger.

  45. libby says:

    I solved my depression issues in the following way.

    I first assessed the function of the depression, in my view the flip side of anger, realizing it has a function.

    If you have low self-esteem, it allows you to feel comfortable with this distorted self-view. The muttered phrase, “I don’t give a shit” is mollifying, allowing you to avoid all sorts of issues brewing within the unconscious. So it functions partly as a tool of avoidance.

    But it also functions as a problem solver.

    Once I discovered this, I realized that my unconscious, the seat of such pseudo problem-solving, is very bad at actually solving any issue it thinks needs attention.

    This ‘problem-solving’ can take extreme measures, such as in the Colombine massacre. To the participants, it did solve a number of issues; revenge against the teachers and other students, and altering their life path, permanently. It was quite an effective solution, albeit undesirable.

    I noticed a chemical component to depression, feeling depressed without cause. I could also feel quite content after taking a pharma drug for moderate pain, again without cause.

    So I concluded that instead of handing over to the unconscious any problem-solving duties, I would identify the real problem, and then consciously deal with them. If I noticed any chemical nose-dive, I again identified it, and then ignored it for what it was.

    None of this admittedly has any scientific basis, but it is how I successfully dealt with depression. It’s quite easy once you find out how to re-frame your belief system.

    Perhaps what is needed in mild cases is that people are given, or better, discover for themselves, tools to deal with the issue.

    It would be interesting to discover how effective such an idea would be for more severe conditions.

Comments are closed.