Rigorous Study Finds Antidepressants Worsen Long-Term Outcomes


A new study conducted by Jeffrey Vittengl at Truman University has found that taking antidepressant medications resulted in more severe depression symptoms after nine years.

The study, published in Psychotherapy and Psychosomatics, examined outcomes over a nine-year period and included initial depression severity as well as other factors. Vittengl divided treatment into categories and compared them to those who received no treatment:

  • inadequate treatment without medication (fewer than eight sessions of therapy)
  • inadequate treatment including medication (fewer than four appointments with prescriber)
  • adequate treatment without medication (at least eight sessions of therapy)
  • adequate treatment with medication (at least four appointments with prescriber)

Of participants with depression, 38.1% received no treatment, 25.2% received inadequate treatment with medication, while 13.5% received adequate treatment with medication. 19.2% received inadequate treatment without medication, and only 4.1% received adequate treatment without medication.

The results were startling. Even after controlling for depression severity, participants who took medication had significantly more severe symptoms at the nine-year follow-up than participants who did not. In fact, even people who received no treatment at all did better than those who received medication. “Adequacy” of treatment did not appear to make much of a difference.

Photo Credit: Philippa Willitts, “A is for antidepressants,” Flickr

These results add to a body of research that indicates that antidepressants worsen long-term outcomes. In an article published in 1994, the psychiatrist Giovanni Fava wrote that “Psychotropic drugs actually worsen, at least in some cases, the progression of the illness which they are supposed to treat.” In a 2003 article, he wrote: “A statistical trend suggested that the longer the drug treatment, the higher the likelihood of relapse.”

Previous research has also found that antidepressants are no more effective than placebo for mild-to-moderate depression, and other studies have questioned whether such medications are effective even for severe depression. Concerns have also been raised about the health risks of taking antidepressants—such as a recent study which found that taking antidepressants increases one’s risk of death by 33% (see MIA report).

In fact, studies have demonstrated that as many as 85% of people recover spontaneously from depression. In a recent example, researchers found that only 35% of people who experienced depression had a second episode within 15 years. That means that 65% of people who have a bout of depression are likely never to experience it again.

Critics of previous findings have argued that it is not fair to compare those receiving antidepressants with those who do not. They argue that initial depression severity confounds the results—those with more severe symptoms may be more likely to be treated with antidepressants. Thus, according to some researchers, even if antidepressants worked as well as psychotherapy or receiving no treatment, those treated with antidepressants would still show worse outcomes—because they had more severe symptoms in the first place.

That is why, in the current study, Vittengl included initial and follow-up depression severity in his analysis, as well as other variables that might provide alternate explanations for the results. This provides a direct counterargument to those who argue that initial severity confounds the results.

To this end, he used data from the Midlife Development in the United States Survey, which tracked depression severity as well as types of treatment utilized over the course of nine years. The data were collected in three waves (1995-1996, 2004-2006, and 2013-2014), and 3,294 participants remained in the study by the third wave.

The survey collected data on depression, generalized anxiety disorder, panic disorder, as well as other medical conditions, family history of mental health conditions, and childhood trauma. Additional data included personality factors, social support, daily functioning, and alcohol use. Because all of this information was included in the survey, Vittengl was able to add it in his analysis.

He found that although these factors impacted depressive symptoms, they did so equally between the groups. That is, initial depression severity does predict lack of improvement—but it does so whether the person is taking medication or not. Therefore, it does not explain how outcomes could be worse with medication.

Perhaps the most notable limitation of Vittengl’s study is his distinction between “adequate” or “inadequate” treatment based solely on the number of sessions (because that was tracked in the survey). This may not be the best indicator of whether participants were receiving sufficient care. However, this does not impact his general findings comparing treatment with medication to treatment without medication and to the group that received no treatment.

Although Vittengl writes that antidepressants may have an immediate, short-term benefit, he argues that long-term use appears to be detrimental. His results suggest that in general, people actually fare better over the long-term if they seek no treatment at all than if they take antidepressant medications. Psychotherapy, on the other hand, appeared to have no detrimental effects. However, even doing nothing was more successful at reducing symptoms after nine years than medication use.



Vittengl, J. R. (2017). Poorer long-term outcomes among persons with major depressive disorder treated with medication. Psychotherapy and Psychosomatics, 86, 302-304. doi: 10.1159/000479162 (Link)

Previous articleDualism and the Mind-Body ‘Problem’
Next articleZombies Must Be Dualists
Peter Simons
Peter Simons was an academic researcher in psychology. Now, as a science writer, he tries to provide the layperson with a view into the sometimes inscrutable world of psychiatric research. As an editor for blogs and personal stories at Mad in America, he prizes the accounts of those with lived experience of the psychiatric system and shares alternatives to the biomedical model.


  1. “Among persons with MDD [Major Depressive Disorder], symptoms were higher after inadequate treatment (d = 0.25), adequate treatment (d = 0.40), or treatment including medication ( d = 0.54) compared to no treatment, and symptoms were higher after treatment including medication versus treatment without medication ( d = 0.43), p s < 0.001. However, symptoms after treatment without medication were no longer elevated compared to no treatment, d = 0.11, p = 0.20."

    Hopefully there will be a new publication on this study because the effect size (d) is not a very visual number. It would take a graph: a picture is worth a thousand words.

    You say: "Psychotherapy, on the other hand, appeared to have no detrimental effects." It seems not to have a favorable effect either. It may be hard to admit, but psychotherapy seems to have, at best, a neutral long-term effect on severe depression. This suggests that psychotherapy, on average, does not have a better or worse effect than any other human relationship.

    We should therefore requalify psychotherapists: "expensive friends who take themself a little too seriously".

    Report comment

  2. It has been 25 years since I recovered from an episode of severe depression and repeated panic attacks. I was hospitalized 40 days. I did about a year of intensive therapy, and then I did a monthly check in with a psychiatrist. However since 2005, I have just popped the pills and lived a normal life. I take the Real Deal Prozac. Not the generic. Generics in many cases don’t work. They are not even regulated. Prozac allows the brain to regenerate brain cells. Exercise can do this as well, but not as well as Prozac does. Combined the two work very well. I was on the generic for a few years and I began to deteriorate. I went back on the Real Deal Prozac and everything snapped back into place.

    This blog exists to encourage people to use a psychologist and discourage psychiatry and medication. For some people a psychologist is a good choice. However for people like myself it will not work. I would be DEAD today. Discouraging people from at least trying medication is incredibly dangerous to some people who can only survive with it.

    Report comment

    • Have you tried cocaine? Freud thought he had found the miracle drug before being confronted with side effects. There is always a period of romance at the beginning of addiction. Then we are disillusioned. What happened to cocaine is coming to antidepressants: it is now proven that antidepressants aggravate depression in the long run. If you do not want to believe in scientific research, it’s your right.

      Report comment

        • Freud became interested in cocaine as early as 1883, and prescribed it to his patients to fight against morphine addiction and to face various psychological or social difficulties. It seems Freud became aware of the dangers of cocaine quite late in comparison with his colleagues, in 1895. One of Freud’s friends died because of his addiction to cocaine. I don’t know if Freud took back cocaine later because of his cancer.

          Report comment

    • This site exists to distribute actual DATA from scientific studies, as well as anecdotal stories such as your own. Your story is data of a sort, but all it demonstrates is that sometimes these drugs work well for some people. The study, however, looked at many hundreds or thousands of people, and it showed that ON THE AVERAGE, antidepressants made no difference or in fact made things worse, and that most people are better off without them. This finding does NOT conflict with your story in the least. It says that your experience is not what most people will experience, that’s all.

      My question to you is this: if you are so comfortable saying that this approach worked for you, why does it bother you when an honest research study is published whose results are not what you expected? Do you believe that your personal experience must translate into everyone’s personal experience? Do you think psychiatry should be based on science, or individual stories? Why is it difficult to accept that a study might find that most people are not having the same positive experience you have had?

      Report comment

    • Reads like a pharma – turf war – generic v ‘real deal’ comment. Watch out for more. It tells me things maybe going in the right direction. Have also noticed a change in the Guardian, they recently removed a piece about personality disorders and how to spot them by a psychologist, which really revealed how patients are actually pretty much hated and demonised, and by doctors themselves. I had a field day in the comments before it was removed.

      Report comment

    • Well, I will state that I’m glad that Prozac works for you but will have to also agree with all the other people who responded to your statement that you can’t assume that your experience is that of numbers of other people.

      I myself almost lost my life to the use of the good ol’ SSRI’s. I ended up taking them for a long time and then began doing some very strange things and eventually tried to kill myself. I have no doubt at all that this was due to my swilling those damned “antidepressants”. Once I got off the damned things all the suicidal feelings and urges disappeared. This is my personal take on these things and no one else has to accept it.

      Report comment

    • Generics are exactly the same as the `real deal’. That’s why they only appear AFTER the patent has expired. And they ARE regulated. If you found yourself relapsing after years on a SSRI without a drug company label, that was the result of the drug – back to Prozac that you believed in meant your belief was what did it for you. That’s called placebo – there’s nothing wrong with that, in fact several universities are doing major research into the action of placebos, it’s just that a good placebo would be harmless and Prozac is not.
      I don;t know where you got the information that Prozac or exercise can increase the number of brain cells, but I suggest you take another look.

      Report comment

  3. I believe it’s more the acceptance of the biomedical model of “depression” that comes with antidepressant usage, than the antidepressants themselves, that causes people to worsen in the long run. If people have been tricked into forgetting that sadness is a normal response to unavoidable life struggles commonly experienced by everyone, such as disappointment, failure, or loss, then they will be less likely to be driven by their painful feelings to adaptively solve whatever problems upset them. So instead of productively identifying, understanding, and devising plans to tackle/master their challenges, they will be more likely to wrongly see themselves as powerless to help themselves. Wrongly believing their painful feelings to be “symptoms” occurring for no reason, they will likely instead “battle depression or anxiety” itself. This may work in the short-term due to an antidepressant’s placebo effect, but in the long-term their real-life problems will get worse due to being ignored. When the placebo effect fades and they are forced to face up to this, of course they are going to become more depressed than what they were like before they started “treatment”, when their life was still mostly ahead of them. Psychiatrists have set them up for certain failure, making them permanently “ill” in order to produce more permanent customers for their assembly lines.

    Report comment

    • dr kelmenson—-I think you have presented very important ideas here…
      I have another idea to add that is also important…many people are stressed…they have many behaviors that are not healthy…one of those behaviors is not eating healthy…they are not getting certain key nutrients such as tryptophan and omega-3….any life problem that happens will easily take them down and anxious…what do you think about this biologic factor added to your real life thoughts…


      Report comment

  4. I am not sure how you conduct studies like this without making the assumption that depression is a thing to be studied, that people with depression are a homogenous group in respect of those aspects that could contribute to low mood, that “depression severity” as subjectively experienced by them is something all can agree on, that they all made sense of their distress in the same way and that in nine years there were no other variables that impacted on how the feel about themselves and their lives

    Report comment

    • That’s a very good point. Unfortunately, psychiatry has defined the terms of engagement, and if one wants to make an impact within the system, one has to start from their definitions, no matter how irrational or subjective. The good news is, even using their own terms and measurements, their drugs fail miserably! Proving such may be a key part of establishing a new paradigm where recognition of the idiocy that underlies defining “diseases” by behavioral or emotional manifestations becomes the new reality.

      Report comment

  5. I cannot see how this study shows what the author concludes. I can only acces the front page of the article so it may be my mistake, but it seems that he data on treatment was from the past year (being year 8 to 9). In which case the resuts just shows that the sicker you are, the more treatment you recieve and that the treatment is not sufficient to alleviate the worsening. All this is to be expected and changes nothing to current knowledge. What could be interesting is to see what the treatment effect was to those who recieved treatment from year 0 to 1 and then normalize the data vis-a-vis the present treatment to understand how treatment effects may change over time?
    In any case however, everyone should be very clear that you cannot infer casual relationsshios with these types of data. The risk of residual confounding is substantial and alone calls for considerable moderation in the intrepretation no matter what.

    Report comment

    • You can access the study via http://sci-hub.cc/

      Copy-paste DOI and let’s go.

      The article is not very detailed, and it is hoped the study will be subject of a more complete new publication, however it concludes that all conditions being equal, including the severity of the depression, people who do not undergo any psychiatric treatment have the best results in the long run.

      Notably, drug treatment has the worst negative effects on long-term symptoms of depression, and non-drug treatments have neutral effect, equal to no treatment at all.

      Report comment

      • Thanks for the link. I stand with my previous comments as it is unclear to me how data on the treatment relates to the results. It seems that the baseline data is already treated patients in which case they demonstrate only that the more depressed people are, the more treatment they recieve. You certainly cannot conclude much with regard to treatment effects. The author states: At best, treatment was insufficient to overcome liabilities among persons with MDD in the current sample. That is true, however a negatively biased conclusion. Second he concludes: Moreover, treat-
        ment including medication may have worsened depression. This could be technically true as “may have” allows for wide intrepretation, but given the huge selection bias it is not fair.
        Unlike the headline the report in itself is not rigorous, but should be better explained, so that it is much more clear what we really can learn from the data – and much less what is the authors idea/opinion on the matter.
        These issues are really complicated, so adhering to what data says is paramount, not adhering to our own opinion!

        Report comment

        • I think you can at least conclude that the treatments offered don’t improve the situation. If you looked at knee pain, people with more severe knee pain might get more pain relievers on the average, but one would assume that the overall TREATMENT for knee pain would REDUCE the knee pain over time. What is the point of “treatment” if more severe sufferers don’t accomplish a bigger reduction in suffering from receiving increased treatment? The “more depressed people get more antidepressants” argument doesn’t hold water.

          Report comment

    • Jeffrey R. Vittengl monitored 20 psychological and social variables, including the severity of depression, and found that those who did not take antidepressants had significantly better results in the long run, all other things being equal.

      The pharmaceutical toxicomania is a problem that belongs to the culture and not to the medicine, even if doctors are involved in the drug traffic. Antidepressant consumption is similar to the recommended cocaine use by doctors in the early 20th century. Scientific denialism and the double standards regarding medical and illegal psychotropic drugs also belong to the culture.

      Report comment

  6. In my view, antidepressants appear to have caused my bipolar disorder. Of course, my treating psychiatrists have all claimed they just “uncovered” my true psychiatric disease. Obviously, correlation doesn’t prove causation but I have reviewed my medical records extensively and truly believe that psychiatric treatment made me bipolar.

    Report comment

    • Yes indeed if you had akathisia from the AP, maybe didn’t even know it was the AP, this can be treated with benzo’s and anti-psychotics or even straight onto lithium, if you come off lithium too fast you can go into mania. It is an appalling thing to do to a single person. A single psychiatrist has done it to very many through out their career. It is very sickening to realise the seriousness of it, especially through your own experience which is usually horrific.

      Report comment

    • I think this is the case for almost anyone who makes the mistake of seeing psychiatrists, who are largely ignorant and dangerous.
      I assume you were then put on cocktails of drugs including anti-psychotics and then when this made you worse, shock was suggested and caused more damage??

      Report comment

  7. I didn’t need a “rigorous study” to figure this out. All I needed was the example of my own life while I was on these devil’s tic tacs. I nearly killed myself because of these damned things, doing things I never would have done without them. These toxic drugs are dangerous and need to be done away with but unfortunately the public is easily convinced that they need to “ask your doctor” about Zoloft if you’re feeling down. And better yet, combine the Zoloft with Abilify so that the Abilify will make the Zoloft work faster and better! When my former roommate fell for that one he spent four days with horrible hallucinations, all induced by the combined effect of these two toxic drugs. What a bunch of bull feces.

    Report comment

  8. I had an overnight acute suicidal reaction to Prozac, after 2 weeks ingestion, the psychiatrist’s response was 66 ECT treatments. a diagnosis of bipolar, with drugs +++, serious brain damage and a ruined life. A second venture into an SSRI produced the same reaction after 2 weeks = another 20 ECT treatments – 2 years of zombie land THEN AWAKENING – ditched psychiatry & drugs – now well and certainly NOT bipolar or depressed but have a lingering Stress disorder. But launching a new career in my old age.
    So many still believe the hype, `you have a chemical imbalance’, the best PR slogan since `God is Love’.

    Report comment

    • The big lie told over and over becomes fact. We don’t need bumper stickers, everyone already believes it. Best Pr slogan since God is Love. And big pharma has ridden to its position as the richest industry in history on its back. And on the shattered lives of millions of people.

      Report comment

    • But then the psychiatrist would have had to actually do some real work, rather than putting you on the devil’s tic tacs. The psychiatrist could just have easily written out directions for the blood tests, but instead out came the toxic “antidepressants”. It truly is criminal behavior and it’s time to begin putting these people in jail where they belong.

      Report comment