Stefan Molyneux is an atheist and anarchist philosopher. He is the host of Freedomain Radio, which is one of the most popular podcast series on philosophy on the internet. He is an author of many books such as “Universally Preferable Behavior: A Rational Proof of Secular Ethics”, “Against the Gods” and “How (Not) To Achieve Freedom”. He is also a popular speaker on many libertarian festivals and gatherings such as New Hampshire Liberty Forum, Libertopia and many others. His arguments about religion, problems with many government programs and peaceful parenting are very persuasive and worth taking a look at. However, dark clouds appeared on the horizon back in early August.
I explored some of the problematic claims that Molyneux put forward on the topic psychiatry a while back in a post entitled A Critical Examination of Stefan Molyneux’s Claims about Antidepressants, where I attempted to correct what I thought where flaws in his arguments about medical psychiatry. While I did not consider him to be anti-psychiatry at the time, one of his latest videos on the topic has made me reevaluate that stance. In a video spanning almost 50 minutes called There Is No Such Thing As Mental Illness, he lays out his case against psychiatry, arguing not only that there is no such thing as mental illness, but that medications against these mental conditions (such as antidepressants and anti-psychotics) are not only ineffective, but actually harmful. He finishes off by asserting that psychiatry is a pseudoscience and should not be taken seriously.
Let us get one thing straight at the start, Molyneux is not claiming that the conditions that the scientific establishment labels as mental conditions, such as depression and anxiety, do not exist. On the contrary, he admits that the anguish and suffering is very real. His problem lies in the notion that these are classified as mental illnesses. He rather thinks it is a reaction to a sick and harmful society. He makes a large number of other claims, that I will be examining in this blog post, one by one.
I also just want to emphasize, yet again, that I am not a doctor, psychologist or psychiatrist. I am just a guy on the internet. Because of that, I fully accept that I can be completely mistaken in everything I say. But hopefully I can present persuasive evidence for the arguments and claims I make in this entry. I will also list the timestamps for the specific claims made so that readers can make sure that I did not falsely characterize his arguments and positions. With that said, let’s get started.
0. An examination of sources used for the video
When discussing scientific matters such as the efficacy of antidepressants, their side effects or other issues, it is vital to back up claims being made with sources. This provides independent support for the statements and arguments being presented, it directs readers to further material on the topic, it shifts the responsibility towards the source if it included a mistake, as the person using the source most likely trusted the source he or she used, it provides credit where credit is due and so on. The sources for the video was posted on the Freedomain Radio forums and can be found here.
Now, evaluating the credibility of sources is not directly an argument against the position or arguments put forward by those source. That would be a fallacy known as the genetic fallacy, where the truth of a claim is being judged based on its origin, rather than merit. However, it is important to be able to critically examine the reliability of those sources.
The first thing that strikes me is that the sources are all miscellaneous internet links, rather than references to the primary scientific literature. This seems strange, because what hopes can one have in evaluating the science behind something if one does not consult the relevant scientific literature? This, of course, does not necessarily mean that the claims made by those sources are wrong. Many internet websites do have reliable content (and I will be using some myself), but there is usually no independent peer-review or fact-checking. After all, anyone can post almost whatever claims they like on the internet.
Some of the links are to popular newspaper articles (in such newspapers as Washington Post, New York Times and Time Magazine) and these, I think, are the more reliable ones out of the links posted as sources. However, popular media often misrepresent science in profound ways, so even here we must be critical.
Other sources are more problematic. One of them is the website Addiction by Prescription Drugs, which just lists 40 “facts” about psychiatric drugs without any sources to the primary research literature at all. Another source used is an article at NaturalNews, perhaps the most pseudoscientific of all so called health sites online, which regularly promote alternative medicine, anti-vaccination and scaremongering about genetically modified foods. The most suspicious source is Antipsychiatry.org, and it is difficult to see how such a source could be unbiased.
A final issue is that a lot of studies are being described in the video, but no source are provided for these. Not even the lead researchers are mentioned. It is very hard to evaluate studies if you cannot find them. This is the case for most of the claims being made in the video, so I cannot confront them all.
To sum up, the list of sources did not contain any reference to the scientific literature, whether this be reviews or original research, but rather a variety of website links, some to reputable newspapers (although it is unclear if the themselves correctly represented the science) and some sources that are outright suspicious, such as NaturalNews. This, of course, does not mean in and of itself the claims made are wrong, but it casts a long shadow over the credibility of the video.
1. The mainstream theory of mental illness
Molyneux suggests that the general thesis of psychiatry is that mental illnesses are due to “chemical imbalances” in the brain that can be corrected with psychiatric drugs (00:56). This is wrong. In most introductory textbooks, any given mental condition is explained as a complex interaction between many different biological, psychological and environmental factors. This means that since there are many different causes, there can be many different treatments, not just psychiatric drugs, but many different forms of therapy depending on the specific condition.
Let us look at two examples, namely major depression and anxiety disorders. These descriptions come from the Passer et. al. textbook from 2009 called Psychology: The Science of Mind and Behavior (p. 27 for depression and p. 796 for anxiety disorders).
genetic predisposition, chemical factors in the brain affected by antidepressants, perhaps an exaggerated form of adaptive withdrawal shaped by evolution etc.
negative thought patterns/distortions, pessimistic personality style, susceptibility to loss and rejection, perhaps linked to early life experiences etc.
previous life experiences of loss and rejection, current decrease in pleasurable experiences, increased life stress, loss of social support, cultural factors etc.
evolutionary preparedness to fear certain stimuli, genetic predisposition, over-reactive autonomic nervous system, low levels of inhibitory transmitter GABA, other possible neurotransmitter dysfunctions, possible sex-linked biological factors etc.
displacement of neurotic anxiety, “catastrophizing” appraisals of threatening events, exaggerated appraisals of anxiety symptoms, classically conditioned fear response, observationally learned fear response, negatively reinforced avoidance responses etc.
previous exposure to aversive unconditioned stimuli, traumatic experiences, avoidable fear-inducing conditioned stimuli, exposure to fearful models or to other individuals traumatic experience, fear-inducing media exposure, cultural learning experience etc.
It is important to understand that these factors interact in complex ways, and that putting forward, say, genetic predisposition as a factor that influences mental conditions is not an attempt at dismissing psychological or environmental factors.
So by erecting the straw man that the scientific mainstream position is all about “chemical imbalances”, Molyneux has loaded the dices in favor of anti-psychiatry from the very start. Mainstream psychology and psychiatry absolutely does accept other important factors influencing mental conditions and absolutely does accept treatments other than drugs. As we shall see later on, a combination treatment of drugs and therapy is routinely the best available treatment for mental illness.
2. The growth argument against the efficacy of psychiatric medications
Just a few minutes into the presentation (04:20), Molyneux makes the first argument against the efficacy of psychiatric medications and it goes something like this: if psychiatric medications where effective, we would expect the prevalence of mental conditions to decrease over time, but prevalence of mental conditions do not decrease over time (they are increasing). So therefore, psychiatric medications are not effective. It is an argument that is easy to understand, but it is fatally flawed in at least three respects.
The first problem is that it assumes that an increase in the number of diagnosed cases implies an increase in actual prevalence. However, confounding factors preventing this conclusions exists, thereby making comparisons over time like comparing apples and oranges. Molyneux himself makes this admission earlier in the video, where he says that the number of diagnoses in the Diagnostic and Statistical Manual have increased over time. Let us take autism as an example: due to an increase in awareness and broadening of diagnostic criteria (severe cases of autism was previously labeled as mental retardation and less severe cases where not even noticed as such), this causes the statistical artifact that makes it appear as if actual rates of autism has increased (Gerber and Offit, 2009; Laidler, 2005).
The second problem is that it assumes that psychiatric medications are claimed to cure individuals permanently, rather than suppressing the physical, emotional and behavioral symptoms (and improve the efficacy of talk therapies). No serious scientists have put forward psychiatric medications as a cure-all miracle substance. Compare this with antiretrovial (ARVs) medication against HIV. These do not usually cure the patient, but merely suppresses the virus in different ways thereby delaying the onset of AIDS leading to a longer and better life (Cameron et. al. 1998; Mocroft et. al. 2000). Does this mean that ARVs are completely worthless against HIV/AIDS? Hardly.
The third problem is that some psychiatric medications, such as antidepressants, decreased suicide rates among individuals with clinical depression. If this sort of thing is the gold standard for measuring efficacy, this shows that antidepressants are effective (compared with placebo). As I wrote in a previous blog post about this, the Mayo Clinic (Mayo Clinic, 2011), most antidepressants are generally safe although, the FDA requires that all antidepressant medications carry black box warnings, which is the strongest warnings that FDA can issue for prescription medications. The warnings note that in some cases (children, adolescents and young adults 18-24 years old) may have an increase in suicidal thoughts and behaviors when taking antidepressants. However, does this state that there is a mere possibility, rather than a documented fact, or is it an uncommon side effect? However, the increase that media reported was just from 2 to 4% and this may have been due to increase in reports (Hall, 2009). Also, after the prescription rate fell by 18-20%, suicides increased by 18% (Hall, 2009). Of course, we have to keep in mind that just because B follows A does not mean that A causes B, but is an important fact to keep in mind. The Olfson et al. (2006) study that appeared to show an increase in suicidal thoughts and behaviors was problematic, since it made the incorrect assumption that the two groups had the same risk for suicide, whereas it was likely that the group treated with antidepressants had more severely depressed patients and thus a higher risk for suicide. Furthermore, Bridge et. al. (2005) showed that suicidal behavior and thoughts in antidepressant tests where similar to psychotherapy trails and Simon et. al. (2006) showed that suicide rates before starting antidepressant treatment where higher and that this declined progressively after starting medication. Cuffe (2007) describes this situation in additional detail.
If even one of these objections are reasonable, it would be lethal for the growth argument.
3. Confusing estimates of heritability with genetic mapping
At 08:50, Molyneux makes the assertion that “psychiatrists have yet to conclusively prove that a single mental illness has a biological or physical cause, or a genetic origin”. To support this claim, Molyneux quotes a number of scientific authorities on the problems and uncertainties with genetic mapping of genes that contribute to mental disorders.
The first problem is that you cannot conclusively prove anything in science like you can in mathematics, where proven conclusions flow deductively from stated premises. In science, the best you can hope for is evidence from many independent sources converging on the same general conclusion. This does not mean that scientific conclusions are uncertain, just that they are well-supported, but can never be “conclusively proven”. By setting the bar so arbitrarily high, Molyneux thereby excludes most scientific conclusions as “not conclusively proven”.
Molyneux, however, is correct on one thing: the genetic mapping of genes that contribute the mental disorders is fraught with insecurities and anomalies. One very common method used is called genome-wide association studies (GWAS), which attempts to find correlations between single nucleotide polymorphisms (SNP, variations in single nucleotides across a population) and the conditions. The problem with this is that results will vary depending on what population you will study. The reason for this is that GWAS studies generally do not look at gene interactions (would make the models terribly complicated for comfort) or copy number (missed in microarray assays), thereby missing out on factors of vital importance.
This is a problem with finding which genetic variations work as risk factors for mental conditions. It has absolutely nothing to do with finding if there are genetic risk factors for mental conditions. The previous question is addressed by molecular geneticists using techniques such as GWAS. The second is addressed by psychologists and biologists using twin and adoption studies for estimating the heritability of mental conditions. The heritability is defined as the proportion of phenotypic variation that is due to genetic variation. This can be calculated by comparing how often the second identical twin has the condition if the first twin has it with what the correlation would be if the trait was completely heritable or by comparing the concordance rates in identical and fraternal twins. To take clinical depression as an example, it is usually the case that identical twins (sharing all of their genes and having the same shared environment) have a concordance rate of about 67%, whereas fraternal twins (sharing only half of their genes and having the same shared environment) only have a 15% concordance rate (Gershon et. al. 1989). This demonstrates that while depression is not completely determined by genetics by any stretch of the imagination, there are genetic risk factors for depression (Donaldson, 1998; McGuffin et. al. 2005; Wender et. al. 1986; Barondes, 1999; Davidson, 1998 etc.).
In other words, even if we do not precisely know how certain genes work as risk factors for depression from genetic mapping studies, we can know that there is a genetic influence for depression from looking at twin and adoption studies. This is no different from the fact that we can know that common descent is true, even though we do not know, and perhaps will never know, the exact way this happened for every single species in molecular detail.
4. Classification of psychiatric conditions
It is true that there is no blood test for mental disorders. But the same goes for migraine. No one would make an argument stating that just because there are no blood test for migraines, then migraine is a dubious diagnosis. Mental disorders are diagnosed by criteria in texts such as DSM-IV-TR or ICD-10. If criteria are too loose, a lot of individuals that do not actually have depression will be diagnosed with depression. If criteria are too strict, then individuals that genuinely have depression will not be diagnosed with depression. This means that it probably can be improved, but of course we should not through out the baby with the bathwater.
There are also many counterexamples one can use such as Huntington’s disease, that is a form of neurodegenerative disease that leads to cognitive decline and dementia. It is caused by an autosomal dominant mutation in the Huntingtin gene that causes an expansion of trinucleotide repeats. When a critical threshold is reached, the protein encoded by the gene because causes the pathological changes. So here we have a disease that has severe psychological symptoms, yet have a clear and demonstrated physical, biological and genetic basis. I have also written a rebuttal to the “well, that is just a brain disease” objection.
Molyneux points out that homosexuality used to be classified as a mental disorder, yet it was removed many decades ago. This, he thinks, suggests that classifications are arbitrary. However, the reason it was removed was because it was no longer rational to hold such a position. Whether something is a mental disorder or not is decided based up on the three Ds: distress, dysfunction and deviance. Old and religiously-based assertions were debunked, and so homosexuality was no longer considered a mental illness. This is evidence that psychiatry has progressed scientifically, not that the label of mental illness is completely arbitrary.
5. Efficacy of antidepressants
I have discussed this a lot of this blog, so I will settle for re-posting an earlier discussion on the topic.
Fournier et. al. (2010) only looked at six studies out of several hundreds. Only three of these looked at an SSRI drug (Paxil). The other three looked at a tricyclic antidepressant (imipramine), which has not been the standard for over a decade. This is important because different SSRIs have different efficacy and side effects, so the results from Paxil cannot naively be extrapolated to most or all SSRIs. The study also used several arbitrary inclusion criteria such as available of patient level data. As most RCTs do not include this, the arbitrary criterion excludes most studies. The initial analysis identified 23 studies, but as the researchers could only gain access to the data in 6 studies, so they ignored the other 17, which may have biased the results (Tuteur, 2010).
Kirsch et. al. (2008) closely reproduced the findings of earlier studies such as Turner et. al. (2008). The effect size of all drugs tested where, compared with placebo, positive. None of the calculated confidence intervals overlapped zero, meaning that it is very unlikely that antidepressants tested and placebo are no different in efficacy. However, Kirsch made a radical new interpretation of those findings. Whereas Turner et. al. drew the conclusions that antidepressants where more effective than placebo, Kirsch drew the exact opposite, namely that antidepressants where not better than placebos, using an arbitrary cut-off standard for clinical significance of 0.5 devised by National Institute for Clinical Excellence, a standard which they no longer use. While it is true that a glass that is 1/3 full is not 1/2 full, a 1/3 glass is not empty. If Kirsch’s interpretation was reasonable, we would have to reject psychotherapy as a treatment as well antidepressants, because psychotherapy alone has an even lower effect size than antidepressants alone. This is why a lot of treatments for depression uses both antidepressants and psychotherapy. They work better together than any of them work alone (Hall 2010a, Hall 2010b).
There are many studies looking at the efficacy of antidepressants that show that they are more effective than placebo. I mentioned this in an earlier blog post, but one such example was published in NEJM (Walkup et. al. 2008), that compared the efficacy of sertraline (an SSRI) alone, cognitive behavioral therapy (CBT) alone, placebo, and sertraline and CBT together and the results where: improvement with CBT alone (59.7%), sertraline alone (54.9%) where both better than placebo (23.7%) and a combination of CBT and sertraine (80.7%) was the best option. Side effects where roughly equal in the group recieving the SSRIS treatment and the group receiving placebo. This study is not perfect, but it independently converges with other such studies showing that SSRIs are by and large effective compared with placebo.
In other words, large scale meta-analyses performed by Turner and Kirsch actually show that antidepressants are better than placebo.
6. The cross-cultural argument
At 21:22 Molyneux explains how certain WHO studies have shown that improvement in industrialized countries is worse than improvement in developing countries. He then argues that since psychiatric medicine is more common in industrialized nations than developing countries, this must mean that psychiatric medications are ineffective.
There are two problem with this argument. First, it is a correlational fallacy. Just because improvement correlates with less reliance on drugs, does not mean that drugs caused less improvement or no improvement. This is no different from the fact that just because ice cream sales and drowning accidents correlate does not mean that drowning accidents cause ice cream sales. There is a confounding third variable in this toy example, namely the seasons. When it is hot, ice cream sales go up and so does bathing (and therefore drowning accidents). So correlation does not imply causation.
The second problem is that a very important confounder has not been controlled for in this argument, namely level of social support. The problem is that industrialized nations are more individualistic whereas developing countries are more collectivist with a stronger social support and a stronger social support increases the chance of recovery (Passer et. al. 2009).
So how does Molyneux know that this correlation implies causation and how does he known that the confounder of social support is not creating the results as a statistical artifact? He does not say.
7. A host of correlation fallacies
Similar correlational fallacies appears in most subsequent arguments about schizophrenia, psychology and race, school shootings etc.
- Ethnic minorities presumably has more social and environmental risk factors for mental conditions, which needs to be controlled for to see if psychiatry is racist.
- Many famous people have been diagnosed with a mental disorder and taken psychiatric drugs and taken their lives. This however, does not mean that the drugs caused the suicide.
- Even if many school shootings where performed by individuals on SSRIs, this does not mean that the SSRIs caused the school shootings. It is ludicrous to deny that the persons mental disorder could have played a part in it. You have to look at how many people on SSRIs do not commit school shootings or similar crimes.
8. Risperdal against PTSD
This is perhaps one of the strangest arguments that Molyneux presents. At 29:45, he says that soldiers with PTSD (an anxiety disorder), who where given Risperdal (an antipsychotic drug used to treat individuals with schizophrenia) did not improve compared with placebo and that this shows that psychiatric drugs are ineffective. Just let this argument detonate in your brain: because antipsychotic drugs used to treat schizophrenia are ineffective against anxiety disorder such as PTSD, this means that psychiatric drugs are ineffective. That is like saying that treating a patient with a viral pneumonia with antibiotics (which does not work) shows that antibiotics are completely useless. Patently silly.
9. A blatant contradiction about addiction
At 32:20, Molyneux makes a massive contradiction without seemingly noticing it himself. Earlier, he has said that it was wrong to suggest that mental conditions are caused by “chemical imbalances” in the brain. Yet now he says that anti-psychotics block dopamine receptors and that this creates, wait for it, a chemical imbalance in the dopamine system, leading to symptoms of withdrawal if suddenly stopped. This particular slide ends with the sentence “If you’re mucking up the dopamine system, you’re increasing the risk of psychosis”. He makes similar claims with respect to antidepressants. The irony is almost unbearable. So, which is it? Can chemical imbalance cause or not cause psychological problems? He cannot have it both ways.
Of course many medications, no matter if they are psychiatric or not, that come with the risk of addiction. That is why you are not suppose to quit cold turkey, but to follow the advice of the doctors and get weened off the medications, avoiding withdrawal symptoms and relapses. This almost goes without saying.
10. Psychiatry as pseudoscience?
Molyneux suggests at psychiatry is a pseudoscience at 34:46. He lists some criteria for something to be called science, such as parsimony, testability, falsifiability, changeable, progressive and tentative. It is ironic that Molyneux has implicitly shown that psychiatry fulfill most of these criteria throughout the video. When he suggested that psychologists changed their minds and remove homosexuality as a mental disorder or when he described how individuals are being evaluated according to criteria, he showed that psychiatry is changeable and tentative. When he discussed studies looking at the efficacy of antidepressants, he showed that it is testable and falsifiable. When he argued that the number of categories for psychiatric disorders are increasing, he showed that it is progressive.
To be sure, psychiatry is less scientific than say, cardiology, but progress is being made towards making the discipline more scientific. Let us not through out the baby with the bathwater.
In the rest of the video, Molyneux attempts to connect what he think he has demonstrated about psychiatry to his particular flavor of market anarchy and anti-statism, which is more towards philosophy and politics, than the natural sciences, so I will not concern myself with those arguments.
11. Summary and conclusions
Molyneux incorrectly describes mainstream explanations of mental disorders as being just about “chemical imbalances”, when even introductory textbooks show that there are biological, psychological and environmental risk factors for most mental disorders. He asserts that the incidence of mental illnesses are increases, but does not attempt to eliminate confounders such as increased awareness and expansion of diagnostic criteria. He also confuses the molecular genetics approach to finding the precise genetic regions that influence mental conditions from twin and adoption studies attempting to find what the heritability of mental conditions are. He denies that there are any evidence for a biological, physical or genetic basis for mental conditions, yet do not seem to know about easy counterexamples, such as Huntington’s disease. He makes correlational fallacies on many occasions without trying to eliminate for confounders such as social support, socio-economic status and the mental condition itself. While claiming that chemical imbalances cannot cause psychological problems, he asserts that psychiatric drugs cause chemical imbalances that in turn cause psychological problems, which is a clear contradiction. Finally, despite himself implicitly explaining how psychiatry fulfills the criteria of science, he think that it is pseudoscience.
It is unfortunate that a otherwise reasonable person has been taken in my anti-psychiatry nonsense and promoting it on this philosophy show that literally has millions of downloads. It is sad to contemplate how many of his listeners who have a mental condition has gone of their medications and perhaps hurt themselves or even committed suicide. There are risks and consequences of spreading pseudoscience, and it comes down to personal responsibility. This will be a burden Molyneux will have to carry for the rest of his life.
12. References and Further Reading
> Barondes, S. H. (1999). Mood genes: Hunting for the origins of mania and depression. New York: University Press.
> Bridge JA, Barbe RP, Birmaher B, Kolko DJ, Brent DA (2005). Emergent suicidality in a clinical psychotherapy trial for adolescent depression. Am J Psychiatry 162: 2173-2175
> Cameron DW, Heath-Chiozzi M, Danner S, Cohen C, Kravcik S, Maurath C, Sun E, Henry D, Rode R, Potthoff A, Leonard J. (1998) Randomised placebo-controlled trial of ritonavir in advanced HIV-1 disease. The Advanced HIV Disease Ritonavir Study Group. Lancet.351(9102):543-9.
> Cuffe, Steven P. (2007). Suicide and SSRI Medications in Children and Adolescents: An Update. DevelopMentor. Acessed 2011-08-10.
> Davidson, R. J., Pizzagalli, D., Nitschke, J. B. & Putnam, K. M. (2002). Depression: Perspectives from affective neuroscience. Annual Review of PSychology, 53, 545-574.
> Donaldson, D. (1998). Psychiatric disorders with a biochemical basis. New York: Parthenon.
> Fournier JC, DeRubeis RJ, Hollon SD, Dimidjian S, Amsterdam JD, Shelton RC, Fawcett J. Antidepressant drug effects and depression severity: a patient-level meta-analysis. JAMA. 2010 Jan 6;303(1):47-53.
> Gerber, J. S., & Offit, P. A. (2009). Vaccines and Autism: A Tale of Shifting Hypotheses. Clinical Infectious Diseases, 48(4), 456-461.
> Hall, Harriet. (2009). Psychiatry-Bashing. Science-Based Medicine. http://www.sciencebasedmedicine.org/index.php/psychiatry-bashing/. Accessed 2010-06-26.
> Hall, Harriet. (2011a). Antidepressants and Effect Size. Science-Based Medicine. http://www.sciencebasedmedicine.org/index.php/antidepressants-and-effect-size/. Accessed 2011-07-19.
> Hall, Harriet. (2011b). Angell’s Review of Psychiatry. Science-Based Medicine. http://www.sciencebasedmedicine.org/index.php/angells-review-of-psychiatry/. Accessed 2011-08-10
> Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ, Johnson BT. Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med. 2008 Feb;5(2).
> Laidler, J. R. (2005). US Department of Education Data on “Autism” Are Not Reliable for Tracking Autism Prevalence. Pediatrics, 116(1), e120-e124.
> Mayo Clinic. (2011). Depression (major depression): Treatments and Drugs. Mayo Clinic. http://www.mayoclinic.com/health/depression/DS00175/DSECTION=treatments-and-drugs. Accessed 2011-08-10.
> McGuffin, P., Owen, M. J., & Gottesman, I. I. (Eds). (2005). Psychiatric genetics and genomics. New York: Oxford University Press.
> Mocroft A, Katlama C, Johnson AM, Pradier C, Antunes F, Mulcahy F, Chiesi A, Phillips AN, Kirk O, Lundgren JD. (2000) AIDS across Europe, 1994-98: the EuroSIDA study. Lancet. 356(9226):291-6.
> Olfson M, Marcus SC, Shaffer D (2006). Antidepressant drug therapy and suicide in severely depressed children and adults. Arch Gen Psychiatry 63:865-872
> Passer, M., Smith, R., Holt, N., Bremner, A., Sutherland, E., & Vliek, M. (2009). Psychology: The Science of Mind and Behavior. New York: McGraw-Hill Education.
> Simon GE, Savarino J, Operskalski B, Wang PS (2006). Suicide risk during anti-depressant treatment. Am J Psychiatry 163:41-47.
> Turner, E. H. M.D., Annette M. Matthews, M.D., Eftihia Linardatos, B.S., Robert A. Tell, L.C.S.W., and Robert Rosenthal, Ph.D. Selective Publication of Antidepressant Trials and Its Influence on Apparent Efficacy. N Engl J Med 2008; 358:252-260 January 17, 2008.
> Tuteur, Amy. (2010). Study shows antidepressants useless for mild to moderate depression? Not exactly. Science-Based Medicine. http://www.sciencebasedmedicine.org/index.php/study-shows-antidepressants-useless-for-mild-to-moderate-depression-not-exactly/. Accessed 2010-06-26.
> Walkup, J. T. M.D., Anne Marie Albano, Ph.D., John Piacentini, Ph.D., Boris Birmaher, M.D., Scott N. Compton, Ph.D., Joel T. Sherrill, Ph.D., Golda S. Ginsburg, Ph.D., Moira A. Rynn, M.D., James McCracken, M.D., Bruce Waslick, M.D., Satish Iyengar, Ph.D., John S. March, M.D., M.P.H., and Philip C. Kendall, Ph.D. Cognitive Behavioral Therapy, Sertraline, or a Combination in Childhood Anxiety. N Engl J Med 2008; 359:2753-2766 December 25, 2008.
> Wender, P. H., Kety, S. S., Rosenthal, D., Schulsinger, F., Ortmann, J., & Lunde, .I. (1986). Psychiatric disorders in the biological and adoptive families of adopted individuals with affective disorders. Archives of General Psychiatry, 43, 923-929