Video: The flawed “study” that says serotonin doesn’t cause depression

I have a problem with anxiety and depression. I think that’s pretty normal for anyone who spends a lot of time thinking about climate change, but a few years back it got to the point where I was worried about my heart. I ended up going on an SSRI(Strategic Serotonin Reuptake Inhibitor). I’m not sure which, but I think it was Wellbutrin. The first couple weeks I was on it were great. For whatever reason, my brain suddenly worked! I had been expecting it to take the edge off my anxiety (which it did), but suddenly my problem with procrastination was just gone. It was like there had been a loose wire in my brain for my entire life, and someone had suddenly fixed the connection. That went away pretty quickly, and I was back to having to fight my brain to do just about anything, but my anxiety was under control, and that made a huge difference in my quality of life. Managing that didn’t fix my depression, but it made it a lot easier to handle.

Since then I’ve been taking them off and on. In the U.S. it was because of money. When I got to the UK, it just took my a while to get around to getting a prescription, because as I’ve mentioned, that part of my brain doesn’t seem to work quite right. I’ve been on fluoxetine (generic Prozac) since March of 2020, and while I did get a productivity kick when I went back on it, that faded as before. My own experience has been that SSRIs work.  That’s why it was odd to see headlines saying otherwise. I didn’t bother digging into it much because I’m aware of how inaccurate headlines can be, and I’ve had other things on my mind. I also knew that Rebecca Watson had looked at the issue in the past, and would likely do so again. I was not disappointed.

But before we even get to that, the title should also give away the fact that this was NOT a review to determine whether or not antidepressants work (which, again, they do). This was a review to determine HOW antidepressants work. Because in all fields of science, researchers accept that sometimes we know that something works but we’re not yet sure why, like when James Lind performed a randomized controlled trial that found that citrus fruit cured scurvy in 1753. Vitamin C wasn’t discovered for another 150 years.

The brain is a complicated lump of meat, so it makes sense that how it works remains a big question. In the late 1980s, doctors found that depressed patients could be helped by giving them fluoxetine, more commonly known as Prozac, which is a selective serotonin reuptake inhibitor, or SSRI. SSRIs, to put it very simply, change the way your brain processes serotonin, causing serotonin to stay for a longer period of time in your synapses.

This led some researchers to think, quite understandably, that maybe people are depressed because they need more serotonin. That hypothesis got popular in the years that followed the introduction of Prozac and it certainly became popular amongst the general public, but serious researchers knew that it was always going to be more complicated than that. They already knew that depression has many causes: yes, there’s an imbalance of chemicals in the brain which might include serotonin, but there’s also depression that comes from bad life events, or from physical ailments. Just because it helps a person when you adjust their serotonin levels, doesn’t mean the problem was the serotonin levels. I’ve heard several doctors recently explain it like this: taking Advil may help your headache but it doesn’t mean that your headache was caused by a lack of Advil in the brain.

I will probably die angry about “reporting” like this. While I think the sigma against people with mental health problems has decreased in my lifetime, it is far from gone, and that means there are a lot of people who still have to fight constantly just to get the people in their lives to treat them with respect. Stuff like this makes that a lot harder. It’s irresponsible reporting, and it can do real harm to people.

Watch the video or go to Skepchick for the transcript and sources, and if you see people saying that SSRIs don’t work, please push back, even if it’s just linking them to Watson’s breakdown.



  1. invivoMark says

    My grad school pharmacology course was downright fascinating when it came to psycho/neuro drugs.

    We had specialist lecturers for most major categories of drugs we have, and they would go through the molecular mechanisms of action – exactly how, on a chemical level, the drugs change cellular function and how that ultimately changes the body’s function. And the details were pretty in-depth for almost all of them – statins, ACE inhibitors, beta blockers, beta agonists, anti-cancer drugs, anti-inflammatory drugs, antihistamines, proton pump inhibitors, etc. The keen-eyed reader might notice that many of these drug types are named after the mechanism of action.

    And then there were the lectures on psych drugs, analgesics, and anesthetics, and suddenly that all went out the window. Our best guess for why general anesthetics work is that they increase the permeability of the cell membrane in neurons, but that could also just be an incidental side effect. And for 70 years we thought that pain was caused by a neurotransmitter called “substance P,” but blocking that chemical turned out to have zero effect on pain.

    We’re lucky to know the molecular mechanism of action of some drugs like SSRIs (literally, they block the reuptake of serotonin in neural synapses – again, it’s in the name), but that’s a far cry from knowing the full story, because there are so many types of cells that use serotonin as a signaling chemical – and therefore, there are so many types of cells that are affected by SSRIs. (That’s why all drugs have side effects – whatever their molecular target is, it’s used by multiple systems in the body).

    And that’s the big reason that there has been so little progress in treating mental illnesses. (There are other contributing reasons as well, such as the difficulty of developing drugs that cross the blood-brain barrier.) We just don’t know what molecular targets we should be aiming for.

    Not knowing a mechanism means that we only know they work based on human patient data. And our only measure of their success is based on subjective patient response.

    With a cancer drug, you can tell it works by whether the cancer physically shrinks compared to placebo, and whether 5-year survival rates improve. You know that all your trial subjects had cancer, and you can objectively measure the outcomes. But mental illnesses aren’t diagnosed based on any molecular or cellular characteristic. So results from studies are confounded by the fact that drugs could be effective on only a subset of patients, and the fact that diagnosis is subjective means that your results will change based on things like culture and environment.

    The fact that we know SSRIs work (at least on a subset of people) is a minor miracle, and a triumph of scientific rigor. But we definitely need a lot more science, both to find improved psych drugs, and to improve our diagnostics so we’re getting the drugs to the people they will actually work for.

  2. says

    Thank you for the comment! One of my favorite parts of doing this is when people add their own experience and expertise. It moves the blog from being the opinions of an inexpert generalist to something where people can learn stuff they wouldn’t encounter in their daily lives. I really appreciate that.

    The narrative I was given on how SSRIs work was that blocking reuptake means there’s more of it just floating around, so the serotonin receptors are always saturated. That signals to the brain that it needs more serotonin receptors, and so it sort of creates a feedback loop that makes your brain more responsive to good things in general.

    I have no idea if that’s true, but it certainly sounds plausible from what little I know. I don’t think I’ll ever be able to match it against my own experiences, because so much about my environment has changed, and I have no interest in recreating the living situation in which I had a panic attack that made me worried about my heart. I’m quirky that way.

    It does feel like my ability to cope with the world, even just on the SSRI, has increased, but there’s no way to tell how much of that is my perspective changing, or me leaving the US, or moving out of a roach-infested apartment complex, or being forced by circumstance do the thing I really want to do in life, or, or, or.

    So yeah, I 100% believe that it’s fiendishly difficult to figure out.

    I had a teacher in high school who started an electronics course by having us try to draw the internal shape of a literal black wooden box, based on the sound a marble in the box made rolling along whatever was in there. Everyone’s box was different, and few of us got it quite right. Treating mental health seems to be like that, but instead of a simple box with a plank added, it’s a complex three-dimensional structure that’s always shifting, and instead of one marble, you’ve got about two thousand ball bearings of differing size.

    So yeah – I’m really glad that SSRIs got figured out, and I honestly wish I’d sought them out earlier in life.

  3. garnetstar says

    This is like saying that lithium doesn’t work for treating bipolar disorder, because bipolar isn’t caused by a lack of lithium.

    And again, whatever lithium does to treat mania and depression, after all these decades, is anybody’s guess.

    BTW, Wellbutrin isn’t an SSRI, selective for serotonin. It inhibits the uptake of norepinephrine and dopamine. If that was what helped you, it’s good to know that your brain is responsive to those two neurotransmitters as well, might help in the future.

    The stigma is such that, when Wellbutrin was found to help is smoking cessation, they called it and prescribed it under its generic name, buproprion, so that smokers and their acquaintances wouldn’t know that they were shamefully taking an antidepressant.

    The evidence is overwhelming that neurotransmitter reuptake inhibitors work. Simplistic, reductive explanations for that are usually not accurate.

  4. says

    Garnetstar – thanks for the info! it’s been years since the first thing I took. I know I was on wellbutrin for a bit, but it’s been fluoxetine for the last few years, and the overall effect has been the same.

    I’d love to be able to get treatment for my ADHD, but getting diagnosed is out of my price range.

  5. invivoMark says

    The narrative I was given on how SSRIs work was that blocking reuptake means there’s more of it just floating around, so the serotonin receptors are always saturated. That signals to the brain that it needs more serotonin receptors, and so it sort of creates a feedback loop that makes your brain more responsive to good things in general.

    That’s kinda close, although there are flaws in the way it’s stated: the receptors won’t be “always saturated,” and there are multiple feedback loops working in both directions that get involved (for instance, a neuron producing serotonin can also have serotonin receptors itself; it will activate those receptors with its own serotonin production, which leads the cell to stop making serotonin).

    But the biggest problem with the explanation is that there are 13 known serotonin receptors in the human genome, across 7 different classes, each with different downstream effector molecules, each regulating different sets of genes in different cells, and most of which are expressed in multiple types of cells and tissues throughout the body, not just in the brain. (And when they are expressed in the brain, they may be expressed in different types of brain cells and have different functions on each of them, etc.) The table here should give a hint to the complexity:

    But, hey, at least we’re pretty sure that bumping up the function of at least one of those receptors is probably involved. It’s a start! Now we can try poking at each of the receptors one by one and see what happens. Or at least, we can find their equivalents in mice and rats (if, indeed, there is truly an equivalent) and poke at those.

    And then hope that what we learn from mice and rats is applicable in humans.

    There’s a lot of science getting done, and we’re slowly learning. This review article looks pretty neat:

    If you’re inclined, maybe you could take a look. It goes into stuff that wasn’t known when I was taking classes.

    Then, of course, there’s electroconvulsive therapy – back firmly in the territory of “no clue how THAT works.” Apparently it actually does work for depression, particularly when it’s medication-resistant. It’s got such a stigma around it that I don’t know if it’ll ever catch on (well, again), but I’m a big believer in evidence – if the data say it works, then it works!

  6. Allison says

    What all this leaves out is that social conditions can cause depression all by themselves.

    One of the big criticisms of the medical profession’s way of framing mental “illness” is that it identifies the “source of psychological suffering in the individual” and “overlooks the undeniable social causation” (more or less quoted from Dr. van der Kolk’s The Body Keeps the Score.) To me, the reason is obvious: if you consider the possibility that depression (for example) is caused by a persons circumstances, it would require them to question the power structures (on both small and large scales) that are an integral part of how our society works.

    It seems pretty obvious that if a person is living in circumstances that would depress anybody, changing the brain chemistry is not going to fix it. The best you can hope for is that it might make them a little better able to function. Of course, alcohol and various drugs work, to, more or less, and sometimes as well or better than the psychotropic drugs.

  7. Allison says

    (I left out a step in paragraph #2.)

    After “it would require them …how our society works”, add:

    “Since the medical profession is part of society’s power structures and has traditionally been tasked with maintaining them, even at the cost of individuals’ well-being, they have a strong motivation not to consider society’s role in mental illness.”

  8. says

    Agreed. A lot of the reason I’m doing better right now is that my immigration status excludes me from the conventional labor market, so I’m doing work that’s important to me, and I’m doing it on my own schedule.

    And I’ve gotten FAR more done in this setting, and while my brain still makes work difficult, it’s a lot easier.

    But that doesn’t solve the environmental anxiety problem, and my situation isn’t currently available to most people, so we need meds to function.

  9. garnetstar says

    Allison @6-7, agreed. And depression brought on by societal circumstances may even be more common than intrinsic illness, I don’t know. Doctors should not medicalize all depression.

    But, there is depression and other mental illness that is intrinsic and not situational. Medicines for that are life-savers, and more info on how they function would really help, even in helping not-prescribe them to people whose troubles are societally-based.

  10. says

    I also think it’s a terrible idea to suggest anyone NOT take meds because society is the cause of their problem. Any version of that has to come after society changes.

Leave a Reply

Your email address will not be published. Required fields are marked *