More young Australians suicide/self-harm and use antidepressants while experts dismiss FDA warning
Updated: May 19, 2020
In 2004, the US FDA warned that using antidepressants was associated with an increased (approximately doubled) risk of suicidal thoughts and behaviours in people aged under-18 years of age. In 2007 the FDA extended the warning to all people aged under-25. In response, prominent Australian mental health organisations, including Suicide Prevention Australia, Orygen and headspace, along with psychiatric thought leaders, Professor Patrick McGorry and Professor Ian Hickie, minimised the significance of the warning, arguing that, on balance, antidepressants reduce the risk of youth suicide. Over the last decade, Australian child and adolescent antidepressant prescribing and suicide/self-harm rates have jumped alarmingly. It is time to ask some confronting questions: Have Australia’s leading suicide prevention organisations and experts got it horribly wrong? Despite their good intentions, could their advice have contributed to the avoidable deaths of young Australians?
(with responses from Prof Jon Jureidini and Dr Anthony Dillon)
Enquiries to firstname.lastname@example.org
The International Controversy
In 2004, the US Food and Drug Administration (FDA) issued a black box warning (the highest level of warning) that using antidepressants is associated with an increased risk of suicidal thinking and behaviour in children (under-18). In 2007 the warning was expanded to all people under-25 years of age with depression and other psychiatric disorders. The warning was a result of a 2004 FDA analysis of short-term trials of antidepressants, in children and adolescents that showed “a relative risk for suicidal behaviour or ideation of 1.95 (95% confidence interval 1.28 to 2.98) for those treated with antidepressants compared with those given placebo”.
The FDA warning was criticised by multiple authors who suggested that restricting antidepressant prescribing might result in an increase in youth suicide in the USA. However, two American studies, one that examined the experience of 1,600 paediatric patients before and after the warning, and another that tracked youth antidepressant prescribing and suicide attempt rates from 2004 to 2016, both found a strong correlation between young people’s antidepressant use and suicidal thoughts and behaviours. Despite this evidence supporting the FDA’s warning, 15 years later the debate is far from over and, in the USA and in many other developed countries, child and adolescent antidepressant prescribing rates have risen.
The Australian Experts' Response
In Australia, no antidepressant is approved for to treat depression in a patient under-18 years of age. The first Australian Atlas of Healthcare Variation, published in 2015, suggested that, for children, antidepressants are “primarily prescribed for anxiety, rather than depression”. Irrespective of the intended purpose of the prescribing the FDA warning applies to young people with any psychiatric disorder.
In 2005, in response to the FDA’s black box warning, the Australian Therapeutic Goods Administration (TGA) did not issue the equivalent Boxed Warning, but required the rewording of Product Information and Consumer Information leaflets made available to doctors and consumers. This 'softly, softly' approach was (and arguably still is) typical of the TGA’s approach to protecting consumer safety.
Subsequent to the FDA’s black box warning and the TGA’s lower-level warnings, prominent Australian mental health advocacy organisations including Orygen, headspace and Suicide Prevention Australia, and influential psychiatrists, most notably Professors Patrick McGorry and Ian Hickie, have endorsed the use of antidepressants for depression by children, adolescents and young adults. Most have acknowledged the FDA warning but argued that the benefits outweigh the risks.
Orygen/headspace and Professor Patrick McGorry - In 2009, mental health organisations Orygen and headspace, headed by Professor McGorry (who was a co-author), produced an Evidence Summary that noted that no antidepressant was approved for under-18 use, and discussed the FDA warning, but concluded that there were “even greater risks of not treating depression with any type of intervention”.
Orygen’s comparison of “not treating depression” with using antidepressants is a false dichotomy. As the authors of the Evidence Summary acknowledged, there are multiple options for treating depression.
The 2009 Evidence Summary (and the updated 2012 version) recommended that SSRI medication may be used to treat moderate to severe depression “within the context of comprehensive management of the patient, which includes regular careful monitoring for the emergence of suicidal ideation or behaviour”. However, in contrast to Orygen’s recommendation of “comprehensive management" and “careful monitoring”, an earlier (2007) audit of prescribing practices highlighted that antidepressants were being prescribed at Orygen’s own clinic to a majority of depressed 15 to 25 year-olds “before they had received adequate counselling” and that “75 per cent of those diagnosed with depression were given the drugs too early”.
(Note: We have been unable to find evidence of subsequent prescribing audits, and Orygen declined to respond to direct questions, so it is unclear whether Orygen has improved its practices since 2007.)
It is also notable that in 2009, the same year that Orygen produced the original Evidence Summary, it re-published a document. titled Medications for Depression that made no reference to suicidality risks, and contradicted the Evidence Summary by overstating the evidence for the effectiveness of antidepressants. Medications for Depression states: “Antidepressants also work well for less severe types of depression” , whereas the Orygen/headspace Evidence Summary states that, of all SSRI antidepressants, only “fluoxetine is modestly effective for reducing symptoms of depression in young people”. Despite this inconsistency, Medications for Depression is still available on Orygen’s website (as at 13 May 2020).
(Note: Orygen was invited to comment but declined. A copy of the questions Orygen was asked is included below in Appendix 2)
Suicide Prevention Australia - Orygen's/headspace’s 2009 Evidence Summary was cited by Suicide Prevention Australia (SPA) in a 2010 position statement titled Youth Suicide Prevention, which concluded that “balanced against the risk of not treating youth depression, SSRIs offer some potential to reduce youth suicide”. SPA’s position statement included little evidence to support its positive risk assessment. SPA cited a single review (Gould et al. 2003), a literature review of the period 1992 to 2002, as having “shown [SSRIs] to be an effective treatment for youth depression and suicidality”. SPA got this wrong. Gould et al. reported that there was no evidence from “psychopharmacological studies that have specifically targeted suicidal adolescents”, but said it was “plausible” that increased antidepressant prescribing might have decreased youth suicide rates in the US.
SPA’s position statement also stated that “the decreased use of SSRIs in Australia has recently been linked to increased youth suicides”, but did not identify evidence of an Australian decrease in use of SSRIs or who made the link and what youth suicide evidence they considered. Furthermore, the following sentence cited a Cochrane review (Hetrick et al. 2007) of SSRI antidepressants use by children and adolescents as supporting fluoxetine as the “most effective SSRI”. However SPA's paper did not identify that, Hetrick et al. found “an increased risk of suicidal ideation and behaviour (RR 1.80, 95% CI 1.19 to 2.72)” and higher rates of adverse events among children and adolescents prescribed SSRIs.
It is disturbing that, in relation to such an important issue, Suicide Prevention Australia made unreferenced claims, misrepresented the findings of Gould et al.’s paper, ignored key findings of Hetrick et al.'s paper, and minimised the importance of the FDA and TGA suicidality warnings.
(Note: The current board and management of SPA is completely different to the team in 2010 and therefore the current CEO and Board are not personally responsible for the errors and omissions in the in the SPA's 2010 Youth Suicide Prevention position statement. The position statement was available on the SPA website until March 2019 when the current CEO and Board, to their credit, decided to remove it as part of reviewing all SPA policy positions. The position statement is still available on the Northern Territory Parliamentary website.)
Professor Ian Hickie - In 2003, before the FDA issued its warning, prominent Professor of Psychiatry and depression expert Ian Hickie was a co-author of a frequently cited paper (Hall et al., 2003) published in the BMJ, supporting the use of SSRI antidepressants. Hall et al.'s abstract concludes: “Changes in suicide rates and exposure to antidepressants in Australia for 1991-2000 are significantly associated. This effect is most apparent in older age groups, in which rates of suicide decreased substantially in association with exposure to antidepressants. The increase in antidepressant prescribing may be a proxy marker for improved overall management of depression. If so, increased prescribing of selective serotonin reuptake inhibitors in general practice may have produced a quantifiable benefit in population mental health”.
Hall et al. (2003) has been cited by both Professor Hickie, and the TGA, as evidence that increasing antidepressant prescribing rates are likely to be associated with better treatment and fewer suicides. However, the data contained in the results section of Hall et al. do not support this positive assessment.
Hall et al (Table 2) reported the antidepressant use rate by five-year age groupings (no data were available for children aged 14 or younger), in three time bands, 1990-1991, 1995 and 1998-2001. There were massive increases in antidepressant prescribing to all Australians, particularly younger Australians, from 1990-1991 through to 1998-2001.
The suicide rate data were examined in three different time bands; 1986-1990, 1991-1995 and 1996-2000. Over the period 1986-90 to 1996-2000, the per-capita suicide rate (calculated from Hall et al. Table 1) rose by approximately 16% for Australians aged 15 to 45. There was a 15% fall in the per-capita suicide rate for Australians aged 45 or older. When combined, the per-capita suicide rate for all Australians (aged 15 or older) rose by 3%. Obviously, because of the growth in suicides among Australians aged 15 to 44 (who made up 59% of the population of Australia aged 15 or older), the life years lost to suicide will have risen by significantly more than 3% from 1986-90 to 1996-2000. (Note: Similar patterns in suicide rates changes are evident when comparing the data for 1991-1995 to 1996-2000.)
Given the results detailed in the body of Hall et al. (2003), it is difficult to understand why the authors, including Professor Hickie, highlighted such positive conclusions in their abstract about the relationship between antidepressant use and suicide.
Professor Hickie compounded the problem when he cited Hall et al. as evidence of a causal effect. He wrote that “increased treatment of depression reduces suicides” [emphasis added] in a debate piece that was published in the BMJ in 2007. Professor Hickie's debate piece also stated:"Although there has been much hype and regulatory concern about increased prescribing of the new drugs [SSRIs], there is little hard evidence of harm to a significant number of people. The real harm, as evidenced by the suicide statistics, comes from not receiving a diagnosis or treatment when you have a life threatening condition like depression.”
In a similar vein, in a 2010 televised debate, 'Is Depression Being Over-Diagnosed?', Professor Hickie mocked as "absolute total nonsense" the assertion that, on occasions, normal sadness was being pathologised, and depression was being over-diagnosed and over-treated with medications. Professor Hickie, in a passionate but rambling answer, went on to praise growing rates of medication use and psychological treatments. He concluded without any reservation, "when depressions been treated, suicide goes down, in this country, and many other countries - well demonstrated!" Professor Hickie, was not alone in putting an overly positive spin on the Australian evidence of the relationship between antidepressant use and suicidality.
The TGA also cited Hall et al. (2003) in Suicidality with SSRIs: adults and children which was part of a 2005 Australian Adverse Drug Reactions Bulletin as evidence that “increased prescribing of antidepressants in Australia during 1991-2000 was associated with decreasing suicide rates”. As outlined above, this was wrong. Even Hall et al. (in the body of the article) stated “The total suicide rate for Australian men and women did not change between 1991 and 2000 because marked decreases in older men and women were offset by increases in younger adults, especially younger men”. In Suicidality with SSRIs: adults and children the TGA identified that there was a decrease in the suicide rate among older Australians, but did not acknowledge the more than offsetting increase in suicide by younger Australians. While the TGA recognised that Hall et al. did "not demonstrate a causal relationship”, it reinforced Hall et al.’s suggestion that increasing SSRI prescribing rates may be one indicator of “improved overall management of depression”.
All of this invites an obvious question: Did the TGA just accept the misleading conclusion in the abstract of Hall et al. (2003), without checking the rest of the paper?
Professors Hickie and McGorry have held and continue to hold prestigious and influential positions. In 2006 Professor Hickie was named by the Australian Financial Review in its list of the Australia’s top 10 cultural influencers for his leadership in mental health and depression in particular. In 2010 Patrick McGorry rose to national prominence when he was appointed Australian of the Year. Few would dispute that Professor’s McGorry and Hickie have been the two most influential mental health key opinion leaders in Australia in the 21st century and have made mental health a prominent issue on the national political agenda.
In the lead-up to the 2010 Federal election, Professor McGorry partnered GetUp and addressed candle-light vigils organised to highlight concerns about youth suicide. That same year, he and Professor Hickie co-authored an opinion piece on youth depression in the Medical Journal of Australia that acknowledged the FDA warning of "suicidal ideation" but implied that antidepressants use reduced suicide risk. Arguably, Professors McGorry and Hickie have helped normalise the use of antidepressants by children, adolescents and young Australian adults.
There have been isolated critical voices of their promotion of pharmacological interventions, including Australian psychiatrists Professor Gordon Parker and Professor Jon Jureidini and prominent American psychiatrist Professor Allen Frances. These criticisms have extended beyond the promotion of the use of antidepressants by young Australians. They include concerns, outlined in a 2006 article in Time magazine, about Professor McGorry's experiments with the use of antipsychotics, by young people who had never been psychotic, but were perceived to be at elevated risk of developing psychosis.
Political support for Professors McGorry and Hickie has been almost universal, with the exception of the first author (Martin Whitely) when I was a Member of the Western Australian Parliament. In 2011, I expressed concern at the influence of Professors McGorry and Hickie on the direction of mental health policy, and cautioned that: "We may see an expansion of services that prescribe more SSRIs, which in turn leads to more youth suicides" and more candles at the next vigil. In 2012, just before retiring from state politics, I was critical of Professor McGorry, the media, and my fellow politicians (of all parties), telling the WA Parliament that "personalities, rhetoric and charisma are driving the direction of mental health rather than science and evidence". (see 'Whitely slams Patrick McGorry for disease mongering' video above)
A decade on: What does Australian ‘real world’ evidence tell us?
There is now nearly a decade of real-world Australian data since Orygen published its Evidence Summary. Multiple sources (see Appendix 1), including Orygen, have identified rapidly rising child, adolescent and young adult antidepressant prescribing rates and/or increasing rates of suicide and self-harm by young people over the last 10 years to 15 years.
Suicide rates - According to the Australian Bureau of Statistics, 279 Australians aged under 25, died by suicide in 2009. Between 2009 and 2017 (the latest year for which data was available at the time of writing this blog) the per-capita suicide rate for under-25s rose by 40.4% (428 Australians aged under 25, died by suicide in 2017). As demonstrated in the following graph:
If the per-capita rate of under-25-year-old suicide had remained at 2009 levels, 699 fewer young Australians, would have killed themselves between 2010 and 2017.
Antidepressant use - Table 1 below was provided by the Australian Commonwealth Department of Human Services to PsychWatch Australia on 2 April 2019. Because of changes in methodology, care needs to be taken interpreting the data. The Department of Human Services began including the numbers of unsubsidised patients in April 2012. Therefore, data for the period 2003 to 2011 were collected on a different basis from the data for the period 2013 to 2018. The data from 2012 are a hybrid of both methods.
Despite the limitation outlined above, there are clear trends:
From 2003-04 until 2007-08 there was a 37% fall in child (0-17), and a 28% fall in young adult (18-27) per-capita antidepressant use rates.
From the low point in 2007-08 until 2010-11, there was there was a 23% increase in child (0-17), and a 16% increase in young adult (18-27) per-capita antidepressant use rates.
From 2012-13 until 2017-18, there was there was a 36% increase in child (0-17), and a 23% increase in young adult (18-27) per-capita antidepressant use rates.
Because of the change in method in 2011-12, it is not possible to calculate the exact per-capita growth in antidepressant use by young Australians, since the upswing began in 2008-9. Our best estimate* is that, beginning in 2008-09 through to 2017-18, the per-capita antidepressant use rate by Australians aged 0 to 27 grew by 61%. (*We projected growth rates between 2011 and 2013 based on the average growth rates for the two years before and after this period.)
So beginning in 2008-09, an increase of approximately 60% in per-capita antidepressant use rates by young Australians (aged 0 to 27) has been associated with a 40% increase in per-capita suicide rates by young Australians (aged 0 to 24).
Prior to these rises, from 2002-03 to 2008-09, per-capita antidepressant use rates for young Australians (aged 0 to 27), fell by 27%. During this period under-25-year-olds' per-capita suicide rates were volatile, but trended downwards, falling by 17%.
Many factors are likely to impact suicide and self-harm rates, and correlation does not prove causation. However, the evidence from falling young Australian antidepressant use and suicide rates (2002-03 to 2008-09), followed by rising young Australian antidepressant prescribing and suicide rates (from 2008-09 onwards) invites some confronting questions:
Have Australia’s leading suicide prevention organisations and experts got it horribly wrong? Despite their good intentions, could their advice have contributed to the avoidable deaths of young Australians?
This is a possibility that many suicide prevention experts seem reluctant to consider. Instead they hypothesise alternative explanations for rising youth suicide rates. For example, in 2016 Orygen, in collaboration with representatives from eleven other organisations (including Beyond Blue, the Black Dog Institute and headspace), produced a report, titled Raising the bar for youth suicide prevention. The report unambiguously identified that youth and child suicide rates were rising.
The report’s authors hypothesised multiple possible causes (e.g. increased use of social media, homophobia and untreated mental illness). In the entire body of the 57-page report, the word ‘medication’ is mentioned once, and antidepressants and SSRI’s are not mentioned at all. The highly plausible hypothesis that, in line with the FDA warning, rising antidepressant prescribing rates are at least in part responsible for rising youth and child suicide rates was not considered.
Self-harm - More recently, in February 2019, research published in BMJ Open (Cairns et al, 2019) found “a concerning increase in child/adolescent self-poisoning in Australia” that mirrored an increase in psychotropic prescribing rates, particularly antidepressants. Cairns et al. also found that there was "substantial overlap between the most dispensed psychotropics and medicines most commonly used in self-poisoning episodes."
Cairns et al. presented evidence of increasing antidepressant prescribing rates from 2009 to 2016 in two time-frames. They quoted prior research led by one of the co-authors (Karanges et al., 2014) that found, from 2009 to 2012, antidepressant use by Australians under the age of 25 increased by 25%, and among this group grew fastest in children 10-14 years of age (35.5%). Cairns et al. also detailed that, from July 2012 to June 2016, the number of individuals dispensed SSRIs “increased 40% and 35% in those aged 5–14 and 15–19, respectively”.
Figure 1 from Cairns et al, 2019 - Trends in intentional poisonings in children and adolescents reported to the New South Wales Poisons Information Centre and the Victorian Poisons Information Centre, 2006–2016.
Cairns et al. also reviewed data from 2006 to 2016 for New South Wales and Victorian poisonings. They found there was an increase in intentional annual poisonings of 98% from 2006 to 2016 with most of the growth occurring after 2011. Over the same period, they found a much lower “overall increase of 15% in self-poisoning in persons aged 20 years and over” .
These results are consistent with the hypothesis that antidepressants increase the risk of suicidality and self-harm in young people. Furthermore, they provide compelling evidence that the antidepressants prescribed to children and adolescents are frequently the means of self-harm.
Guardian Australia asked Professor McGorry for his views on Cairns et al.'s research. He hypothesised a number of possible factors contributing to rising self-harm rates, including: “the impact of smartphones, online bullying, and a lack of meaningful face-to-face relationships [and young people’s concerns about] climate change, the casualisation of the workforce, HECs debt, financial pressures, and social and environmental changes”. In addition to identifying these possible contributing factors, Professor McGorry unequivocally identified that the inability of headspace to meet growing demand and a blow-out in waiting lists is “part of the reason why we are seeing increases in self harm and suicidal behaviour”.
So, among other factors, Professor McGorry attributed increasing self-harm and suicidal behaviour to under-funding of headspace, a service he has been instrumental in establishing and expanding, without acknowledging the possibility that the off-label prescribing of antidepressants to under-18s that he and Orygen/headspace endorsed might be part of the problem.
Psychiatric practice is notoriously variable. As demonstrated in the first (2013/14) Australian Atlas of Healthcare Variation, there are massive geographical differences in antidepressant and other psychotropic drug prescribing rates. It would be unwise and unfair to attribute total responsibility for the rise in Australia’s youth antidepressant prescribing rates to a few individual key opinion leaders and organisations, or even the psychiatric profession alone.
In 2014/15, the vast majority (90.4%) of antidepressant prescribing was done by general practitioners. Psychiatrists were directly responsible for only 6.5%. Whether GPs are properly trained, or have sufficient time to inform young people of the risks of antidepressants and monitor their response, or explore other options, are important questions. Nonetheless, individual clinicians should be accountable for their prescribing practices, and should not exempt from responsibility for harm caused, just because lots of others are just as bad.
The TGA (and therefore successive Australian Governments) should not be exempted from responsibility either. In 2016, the TGA issued a Medicines Safety Update highlighting concerns about SSRIs and suicidality in children and adolescents. It discussed research that found patients and carers were very often not informed about potential risks of antidepressants, including suicidality risks. The TGA’s 'softly, softly' responses to the initial FDA warning and its careless, and frankly wrong, interpretation of the Hall et al. 2003 paper are likely to have contributed to the low level of consumer knowledge. (Note: More blogs highlighting the failure of the TGA to do its job properly are scheduled in coming months.)
Australia-wide, there appears to be a culture of uncritical ‘group think’, led by thought leaders, that extends beyond the mental health sector, the TGA and the medical profession, to sections of the media and most politicians. Frequently the media have echoed the message that, depression is very common and easily treated, so people who are troubled need to ‘seek help’. This sounds entirely reasonable, and many who propagate this message are undoubtedly well-meaning. However, the reality for too many young people is that ‘help’ is nothing more than a short consultation with a GP, a script for an SSRI antidepressant, and (if they are lucky) a few words of caution about possible side-effects.
Regardless of who is responsible for increased antidepressant prescribing rates, the facts are clear. More young Australians are taking antidepressants, and more young Australians are killing themselves and self-harming, often by intentionally overdosing on the very substances that are supposed to be helping them.
Only wilful ignorance will allow current practice to continue, without exploring the very real possibility that, in relation to youth suicide prevention, Australia has followed very bad advice and got it horribly wrong.
Response from Professor Jon Jureidini
Young Australians, suicide, antidepressants and key opinion leaders.
Martin Whitely and Melissa Raven do well to alert us to the blindness in mainstream medical and lay discourse to the possibility that prescribed antidepressants contribute to suicide.
We know from FDA analyses that these drugs increase suicidal thinking and actions in adolescents in randomised controlled trials (RCTs); and we have plausible accounts of increased rates of completed suicide in adults in RCTs. We must therefore be vigilant to the possibility of doing harm, especially as the drugs in question are unlikely to be effective in treating depression.
Cipriani and colleagues in The Lancet conclude that the risk–benefit profile of antidepressants in the acute treatment of depression does “not seem to offer a clear advantage for children and adolescents”. In fact, there were actually a number of factors giving an overly favourable impression of benefit vs harm from antidepressants in Cipriani et al.’s analysis. For example, the authors were not able to factor in the additional problems and consequences of probable data misrepresentation by the companies that did the primary studies; and the suicidal event data available to them were likely to substantially underestimate the events occurring in the drug groups.
It is not uncommon for key opinion leaders to advocate antidepressant prescribing when patients do not have access to evidence-based psychotherapies. In these circumstances, prescribing might help doctors feel as if they are doing something, or help parents feel that something is being done. However, careful prescribing of fluoxetine (let alone badly monitored prescribing, which often occurs in clinical practice) has not been established as more efficacious than a relationship with a benign, supportive clinician, engaging in watchful waiting. Watchful waiting, or active monitoring, is less potentially toxic, and encourages clinicians to engage with a young person, and to explore and, wherever possible, to act on the predicament(s) driving their distress.
Antidepressant Prescribing Rates
2000 - 2015 (from OECD 2017)
Of 33 OECD countries, Australians (all ages) were the second highest (behind Iceland) per-capita consumers of antidepressants in both 2000 and 2015.
Australian per-capita daily defined dose of antidepressants grew 130% over this period.
2009 - 2012 (from Karanges et al. 2014)
From 2009 to 2012, antidepressant use by Australians aged under 25 years increased 25%.
The most rapid percentage increases in antidepressant dispensing occurred in children aged 10-14 years (35.5%).