The Placebo Effect

Hi web visitors,

Recently I have referred to the fact that many of the current schools of therapy have only a placebo effect in the treatment of depression and anxiety, and this has drawn some skeptical, and arguably angry, responses from a couple visitors to the website who did not take kindly to the idea that I may be challenging the validity of their favorite school or method of therapy. In addition, I have mentioned recent researchers such as Dr. Irving Kirsch from Harvard, who have concluded that the chemicals called antidepressants may have few, if any, clinical significant effects above and beyond their placebo effects. This has also annoyed some visitors.

In fact, one reader has called me a quack and asked to be removed from the mailing list for my posts! In all fairness, this has just been a couple folks, and I get many heartwarming and wonderfully positive comments from you every day. I really appreciate your support!

But I thought it might be useful to publish an excerpt from my psychotherapy eBook, Tools, Not Schools, of Therapy in order to present some of my thinking in little more depth, because the placebo effect can be confusing to people. The material I am publishing below is a portion of the chapter entitled “The Clinician’s Illusion.” The chapter highlights ways that practicing mental health professionals sometimes developed inflated ideas about their clinical effectiveness.

Although my book is primarily geared for clinicians who want to improve their therapy skills and effectiveness, the material I am publishing here could be of potential of interest to lay individuals as well who are keenly interested in mental health issues. As an aside, if you are interested in ordering a copy of my eBook, CLICK HERE.

I want to emphasize that the placebo effect is not necessarily a bad thing. A placebo effect is a real effect, and might involve, for example, a complete elimination of the symptoms of depression or anxiety, or even a physical ailment of some type. A placebo effect simply means that the recovery the patient has experienced does not result from the specific mechanism you thought, but rather from your belief that the treatment would help.

It is important to understand the placebo effect, both from the perspective of research and clinical work as well, because it creates profound confusion in both arenas.. My goal in publishing this excerpt is to provide an overview of an important phenomenon which has always been an important part of medical and psychiatric healing.

I am an optimist, and feel that the future of psychotherapy is bright. But I also believe the future will belong to clinicians who are practicing empirically based, data-driven therapy, and that is why I have developed TEAM-CBT. In TEAM-CBT we measure the patient’s symptoms with brief, sensitive scales at the start and end of every session, with no exceptions. That way, clinicians can assess their effectiveness, or lack of effectiveness, at every single therapy session. This is like having a psychiatric X-ray machine for the first time, giving clinicians precise and accurate information that can guide the therapy for the first time.  This treatment breakthroughs has allowed us to develop superior treatment techniques that can lead to high-speed recovery from depression and anxiety, effects that are far more powerful than a simple placebo effect. But the placebo effect will always be a welcome and important component of medical and psychiatric treatment..

Please remember that I am posting my thoughts and experiences as part of my outreach work for therapists and for the general pubic. When I write, I try to express myself in crystal clear language, but am sometimes disheartened when I see that some people may not grasp what I am saying, and may even to jump to conclusions that are the opposite of what I am suggesting. It is not my goal to upset visitors to, and it breaks my heart to see some people becoming upset by my posts. If you find my writing disturbing, then this might not be the website for you. But if you hang in there and share your ideas and criticisms with me, I’ll usually do my best to respond.

Please remember that I can only express my best thinking, for better or worse, and that I am often wrong or off-base in my conclusions. My beliefs are subject to error as well as bias, so please feel free to disagree with me and even to reject my thinking.

YMy opinions may not be valid, but I don’t think I can qualify as a quack who’s just talking off the top of his head, saying things for effect. I began my career doing research on brain chemistry at the University of Pennsylvania School of Medicine, working specifically on the chemical imbalance theory of depression and anxiety for several years. During that time, I won one of the world’s top awards, the A.E.Bennett award, for basic research on brain serotonin metabolism.Some early investigators speculated that a defiicient of brain serotonin is the cause of depression–that IS the chemical imbalance theory. But I could find no convincing evince for this theory, in my published research or in my review of the world literature.

I say that to let you know that I am not an outsider throwing stones, but an psychopharmacology insider. I have personally prescribed antidepressants on more than 13,000 occasions, although I treated most of my patients without them. So I do have considerable experience behind my words. In addition, in this blog I will provide you with references to specific studies, so you can do your own research and come to your own conclusions. One book I would recommend for the serious reader is Irving Kirsch’s book, “The Emperor’s New Drugs: Exploding the Antidepressant Myth,” available as a paperback on Amazon and a good and easy read!

The table below describes why clinicians, including those who start new schools of therapy, frequently conclude that their treatment methods are substantially more effective than they really are. I call this problem “The Clinician’s Illusion.” I discuss all of these problems in Chapter 6 of my eBook, but will only publish the material on the Placebo Effect here.

How Clinician’s Fool Themselves

1.   The Placebo Effect. You attribute your success to your theory and methods when the improvement was actually a non-specific placebo effect. 5.   Therapeutic Solipsism. You assume you know how your patients are thinking and feeling, and you’re not aware that your perceptions may be off base.
2.   Self-Fulfilling Prophecies. You feel so certain that something will happen that you subconsciously make it happen. Then you conclude you were right all along. 6.   Therapeutic Arrogance. You assume that your theories and methods are valid a priori, and that you don’t really need to test them empirically.
3.   Sample Selection Bias. You generalize about the effectiveness of your treatment based on your experiences with a highly-selected group of patients who seek you out, as well as those who continue to work with you, forgetting those who dropped out. 7.   Coupling. You couple a new technique, such as EMDR (eye movement desensitization and reprocessing), with an established method, such as exposure therapy and cognitive therapy, and attribute the improvement you observe to the new technique.
4.   Selective Recall. You selectively remember the patients who respond well to your interventions, and selectively forget or discount those who don’t. 8.   The Confirmation Paradox. You assume that your theories are valid because they’re consistent with your clinical observations.

The Placebo Effect

from Tools, Not Schools, of Therapy
by David Burns, MD *
Copyright © 2017 by David D. Burns, MD

Imagine this scenario: a man who’s been feeling worthless and discouraged for several months seeks treatment from his family physician, or from a local psychiatrist. The doctor explains that he’s suffering from an episode of depression that results from a chemical imbalance in his brain. The doctor reassures him and gives him a prescription for an antidepressant that will correct the imbalance.

Four weeks later, the man reports that he’s feeling much better. He’s sleeping better, his mood has lifted, and he’s productively involved in life again. He and his psychiatrist attribute his dramatic improvement to the antidepressant medication. Is this sound reasoning? Can we reasonably conclude that:

  • His depression resulted from a chemical imbalance in his brain?
  • The antidepressant corrected this chemical imbalance?
  • He responded to the antidepressant?

Although many health professionals and patients do draw these kinds of conclusions, his rapid improvement does not provide convincing evidence for any of them. All we can say for sure is that he was depressed and now he’s feeling better. That’s great, but we don’t know what caused his depression or what triggered his improvement. His improvement could have resulted from the passage of time, unexpected events that lifted his mood, the fact that he became more active, the medication, or other factors, including the placebo effect.

What is the placebo effect, and why is it so important? We know that our expectations can have powerful influences on the way we think, feel, and behave. If you’re convinced that something will help, there’s a good chance that it will help, even if it has no real effects at all.

Let’s say that you and I work as marketing executives for a pharmaceutical company. One day, at a press conference, we announce the synthesis of a wonderful new antidepressant called “Placebin.” We emphasize the superior antidepressant effects of Placebin and explain that it has few or no side effects, and virtually no toxic effects. In fact, we’re so excited by this new breakthrough that we’re going to give Placebin to a million depressed people absolutely free of charge in a huge, nationwide clinical trial. There’s tremendous enthusiasm for Placebin and our stock goes up by more than a billion dollars overnight.

Of course, we don’t tell anyone that our new drug is just a placebo, with no active chemical ingredient. How many of the million depressed patients who take Placebin will recover?

Numerous research studies have shown that if you give an inert placebo to people who are suffering from depression, 30% – 50% of them will recover. The precise percentage will depend on how severe or refractory the patients are. This means that within a few weeks, 300,000 – 500,000 of the patients in our clinical trial will have recovered. They’ll swear by the drug and tell all their friends about it. Some may even appear on national television and give glowing testimonials about how Placebin totally cured them and changed their lives. Tens of thousands of people will rush to their doctors to get prescriptions for this remarkable new medication, and controversial books will appear, asking whether it’s ethical to prescribe “happiness pills.”

But of course, Placebin didn’t really do anything for anyone. It was the patients’ expectations, and not the pills, that got them better. The patients actually healed themselves, but didn’t realize it.

Why are placebos so powerful? A few possibilities jump to mind:

  • Hope. Hopelessness is one of the cardinal symptoms of depression. The belief that things can’t change acts as a self-fulfilling prophecy, because the patient gives up. Then nothing changes, so the patient concludes that she or he really is hopeless.Hope works in the opposite way. If you expect to get better, you’ll become more productive and your outlook will change, so you’ll start to feel better. That’s why patients who believe in any new “treatment” for depression or anxiety have such a good chance of recovering, even if the treatment only has placebo effects. In many cases, the belief that you’ll recover, and not the treatment itself, causes the improvement. Hope is the most potent antidepressant in existence.
  • Warmth and empathy. Positive correlations between therapeutic empathy and recovery have been reported in more than 100 psychotherapy outcome studies (Orlinsky, Grawe, & Parks, 1995). Patients who perceive their therapists as warm and understanding recover more rapidly than patients who do not feel accepted or understood by their therapists. Of course, a correlation does not necessarily imply causality. However, my own research, using Structural Equation Modeling techniques, suggests that the correlation results from a direct causal effect of therapist empathy on recovery from depression (Burns & Nolen-Hoeksema, 1992).

Hope and empathy are important keys to recovery, but these variables create confusion about how and why therapy works. You can test any kind of kooky therapeutic intervention you want, and if you can convince people that your treatment will be effective, it probably will be reasonably effective, even if it involves blatant quackery. Whatever we do will be somewhat effective for some of our patients, and amazingly effective for a number of them. As a result, we may conclude that our theories are valid and that our treatments have specific antidepressant or anti-anxiety effects. We’ll attribute the positive outcomes we observe to our theories and to the methods we’re using, and not to the placebo effect. We’ll tell ourselves, “Bob recovered because of the antidepressant I prescribed,” or “the EMDR I used,” or “the family of origin work we did,” or whatever method we tried.

This is not a trivial concern, because new treatments for depression and anxiety emerge all the time. Many of the new schools of therapy attract large numbers of followers, especially if they’re skillfully marketed.

I don’t mean to imply that the placebo effect is a bad thing, or that it’s unimportant. It’s been one of the physician’s strongest medicines for thousands of years, and it deserves more research. If we can learn more about how the placebo effect works, we can develop more potent placebos. But if you had a ruptured appendix, wouldn’t you want the help of a skillful surgeon? If the surgeon had a warm, reassuring bedside manner, so much the better, but you’d need more than just the placebo effect to save your life.

Now, you may be thinking, “Well, this is all rather academic, because we do have empirically validated treatments for depression and anxiety. Drugs and psychotherapy have been proven to be effective.”

In fact, the situation isn’t so clear-cut. Recent studies suggest that antidepressants may have very few real therapeutic effects, if any, above and beyond their placebo effects. For example, in a well-controlled, multi-university study funded by the National Institute of Mental Health, 320 patients with major depression were randomly assigned to treatment with St. John’s wort, sertraline (Zoloft), or placebo. As you can see in the chart below, 32% of the patients who received placebo recovered, as compared with only 25% of the patients who received sertraline and 24% of the patients who received St. John’s wort (Hypericum depression trial study group, 2002).


This study clearly showed that St. John’s wort has no real antidepressant effects above and beyond its placebo effects, a result that was widely publicized by the pharmaceutical industry. However, they didn’t publicize the fact that the antidepressant didn’t fare much better than St. John’s wort! The authors of the study concluded that “the overall response to sertraline on the primary measures was not superior to that of placebo, an outcome which is not uncommon in trials of approved antidepressants” (NIH, 2002).
These results are not consistent with the widely held notion that the chemicals called “antidepressants” have strong, specific antidepressant effects. In fact, based on their review of the world literature, as well as the data that has been submitted to the FDA by drug companies over the past several decades, Kirsch and his colleagues have concluded that at least 75% – 80% of the effects we attribute to antidepressant medications clearly result from their placebo effects (Kirsch & Sapirstein, 1998; Kirsch, Moore, Scoboria, & Nicholls, 2002).

Below, you can see a figurethat represents the kinds of results you’ll find in the most favorable drug company studies that have been conducted. This figure represents almost 20,000 patients who were tested in drug company studies of the SSRI antidepressants, like Prozac, and includes every known SSRI at every known dose. As you can see, in drug company studies, they typically select patients with scores averaging 25 on the Hamilton Rating Scale for Depression (HRSD), indicating moderate or severe depression.  Then these patients are randomly assigned to treatment with placebo versus an antidepressant.[*] Of course, they tell the patients that they won’t know whether they are getting the antidepressant or the placebo.

A 25-point reduction in HRSD scores would be needed for full recovery. As you can see, the patients who received the antidepressants experienced a 10 point reduction in HRSD scores, while the placebo group experienced an 8.5 point drop.

slide1There are two striking things about these results. First, the difference between the drug and placebo groups was only 1.5 points. This is the most improvement you could attribute to the drug itself, and it’s small, especially when you consider the fact that a 25 point reduction is needed for full recovery.

A number of recent researchers have suggested that such a tiny effect may not justify prescribing antidepressants, given the significant side effects, toxic effects, and hazards associated with these agents (Antonuccio, Danton, & DeNelsky, 1995; Antonuccio, Danton, DeNelsky, Greenberg, & Gordon, 1999; Kirsch, Moore, Scoboria, & Nicholls, 2002). I would strongly agree with that. In fact, I have developed more than 75 psychotherapy techniques to help patients who are depressed or anxious. And using the new TEAM-CBT, I often see a complete or near-complete elimination of symptoms in a single, two hour therapy session.

if there was a psychotherapy technique that only caused a 5% or 6% in 12 to 16 weeks of treatment, it would not make my top 1000 list, much less my top 75 list, and I would tell my students not to bother with it.

The second potentially important point is that roughly 8.5 of the 10 points of improvement in HRSD scores in the drug group, or 85%  of their improvement, resulted from the placebo effect, and not from the drug itself. In other words, if the people in the drug group had been in the placebo group instead, they still would have improved by 8.5 points.

Keep in mind that the figure we just reviewed represents the best studies that drug companies published. However, drug companies suppress the results of many studies that don’t come out in the “right” way. If you look at the next figure, you’ll see the results of many unpublished studies in which there were no differences  between the antidepressant and the placebo. Drug insiders will tell you that studies like this are common, but the results are never published. As a result, there’s a highly misleading pro-drug bias in the world literature. This leads to false perceptions about the efficacy of these agents.


Recent researchers have argued that the tiny differences between antidepressants and placebos in even the most “favorable” outcome studies may result from flaws in the way drug companies conduct these studies (Antonuccio, Burns, & Danton, 2002; Antonuccio, Danton, & DeNelsky, 1995; Antonuccio, Danton, DeNelsky, Greenberg, & Gordon, 1999; Kirsch, Moore, Scoboria, & Nicholls, 2002). One problem is called the “placebo washout” period. Prior to randomizing patients to the drug or placebo group, drug companies put all the patients on placebos for a one to two-week period. During this phase of the study, they carefully monitor the patients’ depression scores. Any patients who improve are selectively removed from the study. At the end of the “placebo washout” period, the researchers randomly assign the remaining patients to the placebo or drug groups. In this way, they stack the deck in favor of their new drug.

In case this did not jump out at you, think about it this way: the have removed the placebo responders from the study! In cards, this is called “stacking the deck.”

Can you imagine what would happen if they set the study up in the opposite way? Suppose they put everyone on their new antidepressant for one or two weeks, and then removed all those who began to improve before randomly assigning the remaining patients to the placebo versus drug conditions. The drug companies would cry foul, because you’d be biasing the study against their drug. But that’s exactly what they do when they selectively remove the placebo responders from their studies.

Why do drug companies do this? In a cynical moment, one might argue that their motives could be financial financial rather than scientific. If they can come up with two studies that demonstrate a statistically significant difference between their new drug and a placebo, they’ll receive FDA approval to market the new drug. As a result, their stock will probably increase in value by a massive amount overnight. And they executives of the drug company may get bonuses based on increases in stock value.

They use other tricks they may use to try to stack the deck in their favor as well. For example, patients who enlist in a drug company study of a new antidepressant are informed that they’ll either receive the new drug or a placebo. They’re also informed that the placebo will be chemically inactive and will have no side effects. However, if they receive the new antidepressant, there will be certain side effects they should expect. For example, if they receive Prozac, they may experience upset stomach, diarrhea, nervousness, a loss of sexual drive, and so forth.

Once the study begins, patients who experience the anticipated side effects usually conclude they got the new drug. In contrast, patients who don’t experience any side effects usually conclude that they’re in the placebo group. Studies indicate that if you ask patients which group they’re in, they’ll be correct as often as 80% to 90% of the time.

So the studies aren’t really “double blind” at all, because the patients know very well whether or not they’re receiving the new “antidepressant.” This flaw tends to bias the results because the patients who think they’re getting the new antidepressant become more hopeful and optimistic, thinking they’re getting a wonderful new medication for depression, so they improve because of feeling greater hope and optimism. In contrast, the patients who think they’re only getting the placebo may feel disappointed, so their depression scores can get worse. This artifact can create statistically significant differences between the drug and placebo groups, even when no differences exist in reality.

Another problem is the use of badly flawed assessment instruments like the HRSD. This scale is one of the worst psychometric instruments ever developed and it boggles my mind that anyone would take it seriously, much less use it in research! There are many fatal flaws with this instrument, including the fact that it focuses almost exclusively on non-specific somatic symptoms, which are poor indicators of depression.

For example, three of the 16 items on the HRSD ask about insomnia. As noted earlier in the book, many factors other than depression can cause insomnia. Furthermore, any medication with sedative properties will cause an apparent “improvement” on the HRSD, even if the medication has no antidepressant properties at all! This is not a trivial problem, because many antidepressants have significant sedative side effects.

These flaws are particularly egregious because all these problems are easily solvable from a research perspective. For example, you could simply ask patients what group they think they’re in soon after the study begins, and control for this variable when you analyze the data. Or, you could use active placebos rather than inactive placebos. For example, if the drug being tested causes sedation, you could use an antihistamine like Benadryl for the placebo, since this medication causes sedation. Or, if the drug causes stimulation and diarrhea, like Prozac, you could use caffeine for the placebo.

But drug companies refuse to use more refined assessment instruments or implement any corrective strategies because they know they’ve got a good thing going and don’t want to rock the boat. In fact, the only studies in the world literature that have used active placebos have failed to show any differences whatsoever between placebos and antidepressants (Kirsch & Sapirstein, 1998). Clearly, drug companies don’t want to wander into this territory!

Recent reports have dealt even stronger blows to the psychopharmaceutical industry. Investigators have shown that all the new antidepressants, including the SSRIs, appear to cause substantial increases in the rates of successful suicide in children (Garland, 2004; Jureidini et al., 2004; Whittington et al., 2004) and in adults (Healy, 2003).Below, you can see the rates of successful suicide in depressed adults who were randomly assigned to SSRIs (such as Prozac) or placebos in drug company studies. If these were true antidepressants, why would they cause an increase in suicide rates?[2]


Many people find these studies hard to swallow, and simply cannot believe them at first. We all know someone who’s said, “Prozac worked for me. It saved my life. The effect has to be real!” But remember that 30% – 50% of the people who receive a placebo will say exactly the same thing.

My reading of the literature, as well as my clinical experience over the years, has led me to conclude that true antidepressant medications may not yet exist. We have chemicals with side effects that are called “antidepressants,” such as Prozac and Paxil, but the antidepressant effects of these kinds of drugs appear to be underwhelming at best.

I don’t want to throw too many stones at the drug companies, because a critical reading of the psychotherapy outcome literature reveals similar problems. Many forms of psychotherapy do not seem to have any strong or specific antidepressant effects above and beyond their placebo effects, either in the short-term or the long-term. This suggests that most of the effects we attribute to each “brand” of therapy may be non-specific placebo effects as well.

Where do we end up? One potentially important conclusion is that the needs of science clash with the needs of marketing. When someone is selling a product, their research needs to be considered with considerable scrutiny, because all human beings are subject to corruption, especially when huge amounts of money are involved. But this is equally true in psychotherapy as in psychopharmacology. If someone has staked his or her career on promoting this or that new brand of psychotherapy, it may be just as difficult for that person to acknowledge or publish research showing that the new treatment has few or specific antidepressant effects above and beyond the placebo effect.

So they first question is–are the chemicals know as antidepressants significantly more effective than placebos? The conclusion I have come to, through clinical experience and my reading of the research literature, is–probably not.

And the second question is–are most psychotherapists, and most schools of psychotherapy, more effective than placebos in the treatment of depression?  The conclusion I have come to is the same–probably not.

This is not necessarily bad, because the placebo effect is quite strong, and helpful to 50% of the individuals who seek treatment for depression. But the reason why I’ve created TEAM-CBT is to try to find answers for the other 50%. I’m optimistic we’re doing that, but we’ll have to wait for the result of current research studies to find out for sure!


[*] This table, and some of the others in this chapter were adapted from Preskorn, S. H. (1997). Clinically relevant pharmacology of selective serotonin reuptake inhibitors: an overview with emphasis on pharmacokinetics and effects on oxidative drug metabolism. Clinical Pharmacokinetics, 32 Suppl. 1: 1-21.

[2] The increase in suicide may result from the stimulating effects of these antidepressants. Some patients become more agitated and anxious or even experience akathisia when taking SSRIs, and this may trigger suicidal feelings. Another possibility is that the lack of antidepressant effects may make some patients feel more desperate. They may reason, “Even this powerful new drug isn’t helping me. I must be a hopeless case.”


Antonuccio, D.O., Burns, D., & Danton, W.G. (2002). Antidepressants: A triumph of marketing over science? Prevention and Treatment, 5, Article 25. You can read this article online:

Antonuccio, D.O., Danton, W.G., & DeNelsky, G.Y. (1995). Psychotherapy versus medication for depression: Challenging the conventional wisdom with data. Professional Psychology: Research and Practice, 26, 574-585.

Antonuccio, D.O., Danton, W.G., DeNelsky, G.Y., Greenberg, R., & Gordon, J.S. (1999). Raising questions about antidepressants. Psychotherapy and Psychosomatics, 68, 3-14.

Burns, D.D., & Nolen-Hoeksema, S. (1992). Therapeutic empathy and recovery from depression in cognitive-behavioral therapy: A structural equation model. Journal of Consulting and Clinical Psychology, 60(3), 441-449.

Garland, E. J. (2004). Facing the evidence: antidepressant treatment in children and adolescents. Canadian Medical Association Journal, 170, 489-491.

Healy, D. (2003). Lines of evidence on the risk of suicide with selective serotonin reuptake inhibitors. Psychotherapy and Psychosomatics. 72, 71-79.

Hypericum depression trial study group. (2002). Effect of Hypericum perforatum (St. John’s wort) in major depressive disorder: A randomized, controlled trial. Journal of the American Medical Association, 287, 1807-14. You can read a summary online at:

Jureidini, N., Doecke, C.J., Mansfield, P.R., Haby, M.M., Menkes, D.B., & Tonkin, A.L. (2004) Efficacy and safety of antidepressants in children and adolescents, British Medical Journal, 328, 879-883.

Kirsch, I., Moore, T.J., Scoboria, A., & Nicholls, S.S. (2002). The emperor’s new drugs: An analysis of antidepressant medication data submitted to the U.S. Food and Drug Administration. Prevention and Treatment, 5, Article 23. You can read this article online:

Kirsch, I., & Sapirstein, G. (1998). Listening to Prozac but hearing placebo: A Meta-Analysis of Antidepressant Medication. Prevention and Treatment, 1, Article 0002a. You can read this article online:

Krupnick, J.L., Sotsky, S.M., Simmens, S., Moyer, J., Elkin, I., Watkins, J., & Pilkonis, P. (1993). The role of the therapeutic alliance in psychotherapy and psychotherapy outcome: Findings in the NIMH treatment of depression collaborative research program. Journal of Consulting and Clinical Psychology, 64(3), 636-643.

NIH (April 9, 2002). NIH News Release., D.E., Grawe, K., & Parks, B.K. (1995). Process and outcome in psychotherapy–Noch einmal. Chapter 8 in A. E. Bergin, & S. L. Garfield (Eds.), Handbook of Psychotherapy and Behavioral Change (pp. 270-376). New York: John Wiley, & Sons, Inc.

Preskorn, S.H. (1997). Clinically relevant pharmacology of selective serotonin reuptake inhibitors: an overview with emphasis on pharmacokinetics and effects on oxidative drug metabolism. Clinical Pharmacokinetics, 32, Supplement 1, 1-21.

Whittington, C.J., Kendall, T., Fonagy, P., Cottrell, D, Cotgrove, A, & Boddington, E. (2004). Selective serotonin reuptake inhibitors in childhood depression: systematic review of published versus unpublished data. The Lancet, 363, 1341-1345.


%d bloggers like this: