The Anti-Psychiatry Propaganda of Nathan Shachar
Dagens Nyheter (the largest morning newspaper in Sweden) continues to promote pseudoscience. Last month, the newspaper decided to publish a full-page anti-immigration ad with flawed equivocations, ignorance of confounders and the deceptive framing of published statistics. Out of the ashes into the fire, Dagens Nyheter decided to publish a column filled with anti-psychiatry propaganda by Nathan Shachar. He has previously come under fire for his ignorant ADHD denialism, where he fear mongered about the side-effects of ADHD medications, butchered the complex etiology of ADHD and misrepresented scientific reports on ADHD rating scales.
In his latest pseudoscientific screed, Shachar continues to recycle common anti-psychiatry myths. He claims that the FDA and the Swedish counterpart (Medical Products Agency) is bought by pharmaceutical companies, but the FDA charges pharmaceutical companies to ensure fast and high-quality drug review process. He asserts that modern psychiatry holds that all psychiatric conditions are caused by “chemical imbalance”, when the scientific consensus position is that they result from a complex interaction between biological, psychological and social factors. Shachar makes a false comparison between “pure amphetamine” and ADHD medication, completely without insight that it differs in chemical nature, dosage and release rate. He bases his case against modern psychiatry on the writings of Janne Larsson, a member of the Scientology-based organization Citizens Commission on Human Rights (CCHR). Counter to the myths promoted by Shachar, several large-scale meta-analysis has found that antidepressants are more effective than placebo even when you take publication bias into account. He continues by showing that pharmaceutical companies sometimes behave unethical, but that is an issue of corporate ethics, not the science of psychiatry or the efficacy of psychiatric medication. Finally, he dismisses ADHD as merely a “maturity” issue, despite the well-known biological underpinnings of the condition.
FDA charges pharmaceutical companies to ensure fast and high-quality drug review process
The procedure of drug discovery, research, clinical testing and approval is incredibly slow. From the first stages of clinical trials to approval, it takes around 10-15 years for an average anti-cancer medication (Cancer Research UK, 2013). There are few areas that could occur faster if given more resources. Drug discovery is already done using large high-throughput screens and clinical trials cannot be rushed through (they take the time they take). However, if the drug review process (that usually takes close to a year) is given more resources, the FDA can hire more physicians, chemists, toxicologists, biostatisticians and so on as well as upgrading their IT systems for submissions, reviews and archiving. This ensures that a faster process does not sacrifice quality. The user fees are also used to improve post-marketing safety monitoring for 2-3 years to ensure that any potential negative side-effects (FDA, 2011; 2012).
However, anti-psychiatry proponents promote the flawed conspiracy theory that the pharmaceutical companies bribe the FDA so that they will approve their drugs without any serious review. In reality, the FDA demands that the pharmaceutical companies to pay user fees to the FDA so that the FDA can perform high-quality and critical reviews of drug applications and extend post-marketing surveillance to protect consumer health. There is no shadowy conspiracy and this occurs in accordance with the Prescription Drug User Fee Act (PDUFA).
A similar situation occurs with the Swedish equivalent to the FDA called Medical Products Agency (“Läkemedelsverket”). The Medical Products Agency requires applicants to pay a fee for application, supplementary fees and yearly fees (Medical Products Agency, 2012). This occurs in accordance with the enactment of fees for the governmental regulations of pharmaceuticals (SFS 2010:1168).
Clearly, if pharmaceutical companies what the medical authorities to review their product, they should pay a fee for it and not the taxpayer. There is no mystery here.
The chemical imbalance straw man
Modern psychiatry holds that psychiatric conditions are caused by a complex interaction between many different biological, psychological and social factors. An introductory textbook lists the following factors as contributing to the risk of depression (Passer et. al, 2009, p. 24):
Biological factors: genetic predisposition, chemical factors in the brain affected by antidepressants, perhaps an exaggerated form of adaptive withdrawal shaped by evolution etc.
Psychological factors: negative thought patterns/distortions, pessimistic personality style, susceptibility to loss and rejection, perhaps linked to early life experiences etc.
Environmental factors: previous life experiences of loss and rejection, current decrease in pleasurable experiences, increased life stress, loss of social support, cultural factors etc.
In other words, claiming that modern science considers depression to be due to nothing other than “chemical imbalance” is dishonest, deceptive and unwarranted. Rather, it is merely an anti-psychiatry straw man used in an effort to spread misinformation.
Janne Larsson and the Scientology connection
In his column, Shachar favorable refer to a writer by the name of Janne Larsson. Who is this person and what is his qualifications on psychiatry? A quick Pub Med search reveals that he has been discussed at great length in a letter published in the Swedish Läkartidningen written by author and science journalist Vanna Beckman. Here is my translation of the relevant passage of Beckman (2007):
Janne Larsson, who calls himself a writer or a special education teachers, writes for the Citizens Commission on Human Rights (CCHR). It is an organization that was founded by the Church of Scientology with the mission to concentrate the struggle against psychiatry, which it considers its main enemy — the cause of all evil, from the Third Reich to al-Qaida. For instance, it is apparent from their “Industry of Death” dvd. The scientologists accuse psychiatry of wanting to, in an alliance with the pharmaceutical industry, drug children with narcotics.
Usually, Janne Larsson works in the dark. He requests documents from authorities who have contact with the ADHD-issue, such as The National Board of Health and Welfare, Swedish Prison and Probation Service and the Medical Products Agency — the latter having registered 171 requests relating to Larsson since January 2004. Sometimes, according to employees at agency, several days have been spent for each request in order to copy paper, remove identifications and check with lawyers.
Individual researches such as Peder Rasmussen, Björn Kadesjö, Lars-Olof Janols and Bruno Hägglöf has spent a lot of time and energy on fulfilling his demands that they should give him email conversations regarding, for instance, a certain drug study, to Janne Larsson and his friends.
Then he systematically selects those facts that fit with the CCHR conspiracy theory, and ignores that which argues against it.
The projects have to apply for additional funding just so they can have time and resources to comply with the requests made by Larsson. There is also the problem with credibility and integrity of the research, as Larsson puts all of the material on his website:
Although scientists will ensure that all information about names, addresses, crimes and punishment will not be disseminated, there might be outsiders who put two and two together and could then deduce the identity of the individuals or other circumstances. The entire project can be jeopardized, she [principal investigator Ylva Ginsberg] says, and feels that it is repulsive to see Janne Larsson handle their records the way he does
In other words, the main source used by Shachar is a scientologist who actively tries to undermine psychiatric research project by spamming Freedom of Information requests. Not exactly a credible scientific source.
Antidepressants are more effective than placebo, even when controlled for publication bias
Turner et. al (2008) published a meta-analysis on the efficacy of antidepressants from both published studies and unpublished FDA reviews. They found publication bias, but even when taking that into account, the effect size of antidepressants was 0.31 (0.27-0.35 95% CI), which is enough to conclude that antidepressants are more effective than placebo. This was true on the level of individual antidepressants as well, as all 12 antidepressants studied were better than placebo. An effect size of 0 means no difference when compared with placebo, an 0.3 represents a small to moderate effect. Considering how prevalent depression is, even a small to moderate effect is strong enough to be of considerable clinical significance.
That same year, another meta-analysis was done by Kirsch et al. (2008). Instead of looking at 12 different antidepressants, Kirsch and colleagues only looked at four. They reached an almost identical effect size (0.32). Kirsch should have concluded that antidepressants was better than placebo and of clinically significant effect. Instead, he appealed to a deprecated and arbitrary cut-off for clinical significance (0.5). In that way, Kirsch could make it appear as if antidepressants were not better than placebo, when in fact, they were. In layman’s terms, Kirsch made the claim that since the glass is less than half full, it must be empty (Turner and Rosenthal, 2008). Despite the faulty interpretation, the results of the Kirsch study show that antidepressants are more effective than placebo.
So here we have two large-scale meta-analyses that show that antidepressants are more effective than placebo in a clinically significant way. Yet Shachar repeats the common anti-psychiatry myth than they are not. By spreading this kind of dishonest smokescreens, he distorts the scientific reality of the efficacy of antidepressants.
Pharmaceutical companies lawsuits relates to corporate ethics, not psychiatry
Shachar correctly points out that pharmaceutical companies have lost a lot of lawsuits. However, these issues often related to false advertisements, off-label usages and patents and not outright scientific fraud. In other words, it is an issue of corporate ethics and government regulations, not the credibility of psychiatry as a scientific research field.
ADHD medication: low doses and high-tech capsule technology to ensure slow release
ADHD medications are given in extremely low doses compared with what a drug addict would use. It uses high-tech capsule technology to ensure slow release. Thus, there is no high and no pathological memory learning that precedes addiction. Although ADHD medication such as methylphenidate (Concerta, Ritalin etc.) belong in the same category as amphetamine, it has a fundamentally different mechanism of action (e. g. does not stimulate release of dopamine like amphetamine). In Sweden, amphetamine is only prescribed for ADHD if methylphenidate does not work or the patient cannot use it for whatever reason (Medical Products Agency, 2013). Again, remember that it involves a much lower dose and a much slower release than for drug addicts, so the effects differ.
In other words, when Shachar claims that children will soon be getting “pure amphetamine”, he fails to understand that it is not the first line of treatment and that doses and release time differs substantially.
The biological underpinnings of ADHD
A review paper by Singh (2008) published in the journal Nature Reviews Neuroscience summarized the state of knowledge regarding the etiology of ADHD. Some potential genetic risk factors have been identified, the role of dopamine and deficiencies in executive function is well-established and there seems to be systematic differences in structure and function of some brain areas, such as the dorsolateral prefrontal cortex, the caudate nucleus, the dorsal anterior cingulate cortex and the lateral prefrontal cortex. Although difficulties relating to the heterogeneity of ADHD, progress is being made in understanding the biological underpinnings of ADHD.
Despite this, Shachar callously dismiss ADHD as “a maturity issue”. He then goes on to advance the fact that younger children in a class are more likely to be diagnosed with ADHD than older. This, however, is not an argument against psychiatry as a research field or against the validity of the ADHD diagnosis. Rather, it probably relates to the way teachers interpret child behavior. When diagnosing ADHD, teachers are generally interviewed and they may be holding up children born later in the year to the same behavioral standards as those born earlier (because they are in the same class), despite the fact that two children may differ by as much as 11 months (which means a lot in terms of maturity at a young age). In other words, this is at best an argument for improving the way teachers take into account differences in age and how it affects maturity.
The intellectual responsibility of the Editor-in-Chief
The Editor-in-Chief of Dagens Nyheter, Peter Wolodarski, defends his decision to publish this pseudoscientific piece of anti-psychiatry. Essentially, his response is that people in Sweden should learn to deal with the fact that different opinions are brought to public attention and stop feeling offended or crying for censorship. He also adds that we must always allow dissent.
However, this issue is not about feeling offended or crying for censorship. It is about the fact that one of largest newspapers in Sweden publishes anti-psychiatry propaganda and thereby contributes to the harm done by this pseudoscientific movement. It is great to be open-minded, but do not let it be so open minded that the brain falls out, as the classic adage goes.
As an Editor-in-Chief, Wolodarski has an intellectual responsibility to avoid spreading scientific falsehoods to millions of people. It does not matter that they occur in a column. This is not a free-speech issue. It is an issue of scientific integrity. Based on recent events, it would appear that this is something that Dagens Nyheter is currently hemorrhaging.
References
Beckman, Vanna (2007). Offentlighetsprincipen exploateras. 104(39): 2769-2770.
Cancer Research UK. (2013). How long does it take for a new drug to go through clinical trials?. Accessed: 2014-01-02.
FDA. (2011). Frequently Asked Questions about the FDA Drug Approval Process. Accessed: 2014-01-02.
FDA. (2012). FDA’s Drug Review Process: Continued: The Role of User Fees. Accessed: 2014-01-02.
Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ, et al. (2008). Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the Food and Drug Administration. PLoS Med 5(2): e45.
Medical Products Agency. (2012). Ansökningsavgifter och årsavgifter för läkemedel. Accessed: 2014-01-02.
Medical Products Agency. (2013). Läkemedelsverkets följer upp behandlingen med centralstimulantia vid ADHD. Accessed: 2014-01-02.
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.
Singh, Ilina. (2008). Beyond polemics: science and ethics of ADHD. Nat Rev Neurosci, 9(12), 957-964.
Turner, Erick H., Matthews, Annette M., Linardatos, Eftihia, Tell, Robert A., & Rosenthal, Robert. (2008). Selective Publication of Antidepressant Trials and Its Influence on Apparent Efficacy. New England Journal of Medicine, 358(3), 252-260.
Turner EH, Rosenthal R. (2008). Efficacy of antidepressants: Is not an absolute measure, and it depends on how clinical significance is defined. BMJ 336:516-7.
Pingback:Osakligt om ADHD i DN | Attention Lund
Nice overview.
Concerning the meta-analyses, perhaps you can say something about what the summary statistic represents, and why 0.3 is a non-arbitrary benchmark while 0.5 is not? I hope it is something that relates to actual clinical significance more concretely than, say, the conventional cutoffs for Cohen’s d do (http://www.polyu.edu.hk/mm/effectsizefaqs/thresholds_for_interpreting_effect_sizes2.html).
After re-reading the relevant passage in the above post, I realize that my phrasing was obscure and perhaps even a bit misleading. When I wrote…
…I did not mean to suggest that Turner et. al (2008) used 0.3 as an (arbitrary) cut-off for clinical significance. I just wanted to give readers a general idea about what different standardized effect size values are typically taken to mean. I do not want to imply that I think that arbitrary cut-offs are good or that I reject the relevance of the biological context (see e. g. here).
You and Ellis are right, of course, that researchers need to interpret standardized effect sizes in the relevant scientific context and that a given standardized effect size may have different interpretations (in fact, Turner and Rosenthal (2008) makes precisely this point). As Ellis (2010, pp. 23-24) points out, “the benefits of aspirin consumption expressed in correlational form are tiny, just r = .034”, but “taking aspirin lowers the risk of heart attack by more than 3%” (quotes on different pages). Clearly, even though the correlation is almost non-existent, the fact that so many people have heart attacks per year will mean that even a 3% reduction could save literally thousands of people.
Going back to Turner et. al (2008), they concluded that antidepressants were superior to placebo, but since they are clearly aware of the problems with statistical significance tests and the benefit of effect sizes, this lends to the interpretation that they were not simply saying that antidepressants show a statistically significant benefit over placebo.
On what can the conclusion that antidepressants are clinically significantly better than placebo be based on? Given the fact that around 11% of the population aged 12 and up are taking antidepressants and about 10% of the U.S. adult population meet the criteria for depression, an effect size of 0.31 could mean relief for millions of people (analogous to the aspirin and heart attack argument above). Sure, we do not have perfect information about the benefits/risk ratios for antidepressants and they probably have more side-effects than aspirin (although aspirin can cause internal bleeding), so the antidepressant effect size has to, so to speak, “make up for it”. Patients can switch antidepressant if the benefits do not outweigh side-effects (antidepressants in the same general class have similar mechanisms of action, but different side-effect profiles).
I did not mention in the post that Turner et. al (2008) used Hedge’s g and not Cohen’s d (which Kirsch used in his 2008 meta-analysis), but I want to mention this for completeness. As I am sure you know already, they differ by either using pooled standard deviation (d) or a sample size-weighted pooled standard deviation (g), the latter converging to d when sample size increases.
This is a very intriguing discussion and it deserves to be had, but we should not forget the bigger picture: Kirsch’s entire argument against antidepressants rests on the appeal to a large, arbitrary and deprecated cut-off for clinical significance and this is highly suspect.
It is also worth pointing out that this is not the first time Kirsch has dropped the ball on this issue. The earliest instance of this I can find is the 1998 exchange between Kirsch and Sapirstein (1998) and Klein (1998) were Kirsch and Sapirstein (1) did not cover the issue of differential attrition and the risk of it compromising randomization (essentially differential drop-out between the groups, many of the included studies had a higher placebo drop-out), (2) calculated Cohen’s d incorrectly by having the difference between means start and mean final scores per group in the numerator instead of the difference between the means for the two groups and (3) extremely tendentious selection of studies to include.
The issue of publication bias and unethical behavior by pharmaceutical companies are very real, but it is not useful or credible to base such a position on papers by Kirsch.