health & science

Media council ruling on anti-depressant story


[1] Ronan Whyte has complained about a story in in the FutureLearning section of Newsroom headlined ‘A case against antidepressants’. That section contains articles from academia and researchers, writing on their areas of expertise.

[2] The article was co-authored by Dr Bruce Cohen, Maria Bradshaw, Professor Roger Mulder, Professor Peter Gotzsche, Robert Whitaker and David Carmichael and was published on February 14, 2018. The complaint was received the next day.

[3] Beyond an email acknowledging receipt of the complaint, Newsroom did not reply to the complaint until April 3 due “over-stretched” resources, “inadequate” processes and the need to seek responses from the whole group of academics.


[4] Whyte does not specify which principle he is complaining under, but his central concern is about the accuracy of the article. The Media Council is considering the complaint under Principle 4 Comment and Fact. While not mere opinion pieces – and not labelled as such – the articles presented in FutureLearning are often arguments or interpretations of research and events being made by experts and so they do not fit under Principle 1 which applies principally to news reports. Principle 4 allows considerable latitude to the expression of opinion, but requires that where opinion is based on facts, those facts must be accurate.

[5] Whyte is primarily concerned that the article is “misleading” and “could result in harm to both the general public and vulnerable persons with depression”.

[6] Some of Whyte’s complaints and the ensuing debate between him and the researchers require a level of understanding in the field that goes beyond the expertise and mandate of the Council. We are not in a position to delve into the footnotes and in-depth medical research required to do that debate justice. What’s more, during the debate with the authors, the complaint expands. We can however engage with some of Whyte’s complaints as they pertain to accurate reporting of the scientific research that is the subject of the article.

[7] At first, Whyte says his concerns “focus largely” on one paragraph:

"A 2016 analysis of the data produced in drug company antidepressant trials undertaken by Professor Peter Gøtzsche, director of the prestigious Nordic Cochrane Centre, found that antidepressant use doubles the risk of suicide and violence in young people."

[8] Whyte says that does not accurately reflect the findings in the research. While the article says “antidepressant use doubles the risk of suicide and violence in young people”, the research talks about “suicidality”, which includes suicidal thoughts and attempted suicide. The complainant argues “suicide” and “suicidality” are not the same, yet the authors have used them interchangeably.

[9] In that research – published by one of the co-authors of this article, Prof. Peter Gotzsche – “suicidality” and “violence” are included in a list of “harms” that include anxiety, tremors and bad dreams.

[10] Whyte says “Professor Gotzsche’s definition of 'violence' is significantly different to the general public’s understanding. In fact there are no events that would meet a reasonable definition of violence (assault, homicide, aggression etc) reported in the trials he covers.”

[11] Second, Whyte complains that the article describes Gotzsche as a director of the “prestigious Nordic Cochrane Centre” without acknowledging that the Centre has publicly distanced itself from the professor’s views on this precise topic. A British Medical Journal article from 2015 reports the Centre “has stated unequivocally that it does not share the views of Peter Gotzsche, the director of the Nordic Cochrane Centre, on the benefits and harms of psychiatric drugs”.

[12] It continues that “Gotzsche was an experienced researcher and was free to interpret the evidence as he saw fit but that he had an obligation to distinguish sufficiently in public between his own research and that of Cochrane… There is a wide range of views within Cochrane on the benefits and harms of psychiatric drugs, of which Professor Gotzsche’s is one.” Whyte says: “Referring to him as a director of the ‘prestigious’ institute falsely suggests an alignment between their views”.

[13] Third, the authors write in the article:

“According to New Zealand drug regulator Medsafe, the doubling of suicide risk is a well-known fact, and cognitive impairment is an acknowledged and often permanent side effect of a range of psychiatric drugs”.

[14] Whyte says this is a misquote. He contacted Medsafe directly and says Medsafe does not think the doubling is a well-known fact and neither do they accept that cognitive impairment is a side effect, permanent or otherwise.

[15] In an email to Whyte, Medsafe Group Manager Chris James acknowledges:

“…there may be an increased risk of suicidality in children and adolescents, as previously published in Prescriber Update. Information on risk of suicidality is included in data sheets provided by the companies and published on the Medsafe website. The level of risk is estimated in the Arrow-Fluoxetine data sheet has an increase from 2% to 4% (which is a doubling), but most data sheets do not estimate the level of risk. The concept that ‘doubling of risk is already known’ was in a previous version of the arrow-fluoxetine data sheet, but is no longer included and is different to it being a ‘well-known fact’ as quoted [in the article]. Other symptoms such as anxiety, agitation, panic attacks, irritability, hostility, impulsivity for example are also included in data sheets but cognitive impairment is not a side effect listed in data sheets. Medsafe has not acknowledged cognitive impairment as a side effect permanent or otherwise of antidepressant medicines”.

Editor’s Response

[16] Newsroom co-editor Tim Murphy responded on behalf of the website, apologising that it took nearly seven weeks to reply substantively to Whyte’s complaint. During that period however Newsroom deactivated the article as the process dragged on.

[17] The co-author’s response came from its convenor, Dr Bruce Cohen of the University of Auckland, and it stood by the article. Cohen also raises the concern that the article was taken down while they replied, as it threatened freedom of the press and the group’s academic freedom.

[18] Cohen reasserts that it has been “very well documented that the risk of suicide in children and adolescents is doubled on antidepressants” and points to a book by Gotzsche. It says many suicides and attempts have been omitted by drug companies in their trials, “thus the risk of suicide is likely much worse than what has been reported”. In direct reply to Whyte’s first complaint, Cohen says the distinction between suicide and suicidality (suicide attempts and suicidal behaviour) is a “red herring that the drug industry has invented. A suicide starts with thoughts about suicide, preparations, attempts and suicide”.

[19] Second, Cohen writes that it is “false” to say Gotzsche’s views are not supported by “the Cochrane Collaboration”. Gotzsche does not have “views”, he draws conclusions from the best science. Further, the Cochrane Collaboration does not have views on antidepressants and therefore neither supports nor disagrees with Gotzsche’s conclusions.

[20] In relation to the views attributed to Medsafe, Cohen says the group was drawing on a March 2017 OIA request by Maria Bradshaw, one of the authors. He reinforces that “an increase in risk of suicidality also increases the risk of suicide”, but does not want to misrepresent Medsafe and so writes “we are happy for ‘risk of suicide’ to be replaced with ‘risk of suicidality’.

[21] However Cohen does not accept Medsafe’s stance on cognitive impairment, writing, “It has been well documented that antidepressants impair higher brain function, including cognitive functions”.

[22] On behalf of NewsRoom, Murphy offers Whyte the chance to write an article countering the Cohen group’s conclusions and agrees to amend the piece as the group suggests and reinstate it to the site.

Discussion & Decision

[23] As noted, the Media Council focusses on the parts of the complaint that come within its mandate around the accuracy of the reporting and leaves the more complex scientific debate to the scientists.

[24] On the points under consideration, we can identify what seem to be words, meanings and viewpoints in the article that differ from the words in the supplied research.

[25] The key to this complaint lies in what exactly Gotzsche’s 2016 analysis shows to have “doubled”. The article says “the risk of suicide and violence in young people”. However that seems inadequate to explain the research’s findings.

[26] What the research says doubled was a series of “harms”, which, according to the study was defined as “adverse events that were either suicidality or violence or were considered pre-cursor events to suicidality or violence in the literature… We categorised the harms as suicidality, violence, activation events, psychotic events and mood disturbances”.

[27] As Whyte argues, that range of events – even though they do sometimes lead to suicide – are not accurately captured in the three words “suicide and violence”. The public understanding of those words would not extend to bad dreams and anxiety; suicidal thoughts (or indeed tremors etc) do not always lead to suicide.

[28] Further, the research itself repeatedly uses the word “suicidality” rather than “suicide”, so it is odd that one of the authors has in effect misquoted himself. The Council welcomes the group’s and NewsRoom’s willingness to change that. But problems remain with the use of the word “violence”, given the range of events it is meant to describe. For most readers, it would not be accurate to use “violence” to mean bad dreams, “caffeine feeling” and some of the other harms listed.

[29] The Council also notes that while “doubling” is accurate, it would be useful to inform readers that doubling is from two to four percent. Scale matters.

[30] Professor Gotzsche’s standing with the Cochrane Centre is also pertinent to readers, given that the authors introduce him as director of the “prestigious” group. Clearly they are expecting that prestige to be associated with him. To omit then the fact the Centre does not share his views is misleading, especially when the Centre has publicly called on Gotzsche to distinguish between his “views” and the Centre’s. (NB despite Cohen’s argument that Gotzsche does not have “views” but rather “draws conclusions”, we note the Cochrane Centre describes Gotzsche’s conclusions as views and we would agree, as Cohen argues, that there is “no Supreme Court” on science. Rather, it is perfectly reasonable to debate how one might view the research, as Whyte does).

[31] The article also misrepresents Medsafe’s position, misquoting the OIA response. Medsafe does not call “the doubling of suicide risk” a “well-known fact”, as the article claims. In the OIA Medsafe says “the doubling in risk of suicidality in children and adolescents with major depressive disorder who are taking antidepressants is already known”. As Medsafe says in its reply to Whyte, “well-known” and “already known” have different meanings. Further, Medsafe’s stance has qualifications around the doubling that the article omits.

[32] The Council does not have a view on whether antidepressants cause cognitive impairment, but the authors are clearly wrong about Medsafe’s opinion. Medsafe has not acknowledged cognitive impairment as a side effect, permanent or otherwise, whether it has been “well documented” or not.

[33] The Media Council shares the authors’ concerns about press and academic freedom. We defend their right to express their interpretations of the research with vigour. However freedom comes with responsibilities and our standards require that opinions must be based on accurate facts. In this case the authors have not met that standard so the complaint is upheld under Principle 4.

[34] The complainant has asked for certain words to be changed, but the Council rarely does that. NewsRoom and the authors have agreed to change the word “suicide” to “suicidality”, which is a welcome start. As the usual sanction for an upheld complaint the Council will require the summary of this ruling and a link to the full ruling be placed on its FutureLearning page and, should NewsRoom choose to reinstate it, on the story page as well.

[35] NewsRoom is in a difficult position. As Murphy says, “NewsRoom published the article in good faith relying, as per our ongoing arrangement to publish academic research and opinion articles, on the University of Auckland’s academic research and editing process”.

[36] The publication of articles by academics for a more general audience is something to be encouraged and the Council accepts it is nigh impossible for a website to be able to fact-check numerous articles submitted by experts and which rely on extensive research. Yet that can be no excuse for inaccurate reporting. While we cannot pass judgement on the accuracy of the research, we can rule on whether the research is accurately reported in any article. We expect any publication or site to ensure the accuracy of its content and, when errors are made, to correct those errors promptly.

[37] To that end, the Council would encourage NewsRoom to stress the importance of accuracy to submitters given the ferocity of some academic debates. It would also be useful to ensure those contributors are aware of the Council’s principles.

[38] The website also needs to establish processes so that complaints can be dealt with in a timely fashion, even if the topics are complex and a response involves contacting a range of contributors.

[39] NewsRoom should also ensure it adheres to Principle 4’s requirement that articles that are “essentially comment or opinion should be clearly presented as such”. The Council urges the website to prominently label the content as commentary or something other than a balanced article and consider a disclaimer that acknowledges that the FutureLearning section hosts stories based on contestable research. While that does not excuse NewsRoom fromMedia Council standards, it would tell readers this material should be read critically and that it exists as part of ongoing academic debate, sometimes in contentious fields.

Media Council members considering the complaint were Chris Darlow, Liz Brown, Tiumalu Peter Fa’afiu, Jenny Farrell, Hank Schouten Marie Shroff, Christina Tay, Tim Watkin, Tracy Watkins.

Craig Cooper stood down to maintain a public member majority.

Newsroom is powered by the generosity of readers like you, who support our mission to produce fearless, independent and provocative journalism.


Newsroom does not allow comments directly on this website. We invite all readers who wish to discuss a story or leave a comment to visit us on Twitter or Facebook. We also welcome your news tips and feedback via email: Thank you.