TGA conducts safety investigation into antidepressants use in young people

pills medicine graph

The review was sparked by concerns that increased antidepressant prescribing could be linked to a rise in suicide rates among young people in Australia… but what did they find?

The strength of the current evidence is insufficient to draw a causal relationship between prescribing of antidepressants and increasing rates of youth suicide, the TGA has concluded following a safety investigation.

However increasing use of antidepressants in young people is a concern and there is further research to do in this area, the agency said.

Most antidepressants in Australia are registered for use in adults only. The exception is sertraline and fluvoxamine products, which have an indication for treatment of obsessive-compulsive disorder (OCD) in children aged 6 or 8 years and over respectively. Outside of these approved indications, use of antidepressants in individuals under the age of 18 is considered “off-label” in Australia.

The TGA conducted the safety investigation in response to a review article by Martin Whitely, Melissa Raven, and Jon Jureidini, in July 2020, which postulated that increased antidepressant prescribing may be linked to an increased rate of suicide in young people in Australia.

Whitely, et al., compared changes in PBS antidepressant dispensing from 2003 to 2018 for young Australians (aged <28 years) with youth suicide rates (for people aged <25 years) from the Australian Bureau of Statistics (ABS).

Australian youth suicide rates have trended upwards in the previous two decades and there has been an increase in antidepressant dispensing in youths in the same period, they found.

“Of course, correlation does not prove causation, and many factors impact suicide rates. However, given that the FDA warned that antidepressants were associated with an approximately doubled risk of suicidality relative to placebo, we are not surprised that rising dispensing rates have been accompanied by increasing youth suicide rates,” the authors said at the time.

In light of limitations of the Whitely, et al., paper, which were acknowledged by the authors, the TGA
conducted its own analysis to explore the issue in more detail.

Its analysis of PBS dispensing data confirmed a steady increase in the rate of antidepressant prescribing to young people in Australia, both male and female, between 2013 and 2019.

Meanwhile, it found rates of suicide of young people in Australia have also increased—but with a more complex trajectory, with male and female suicide rates showing differing trends.

“These complexities are important to consider when studying the relationship between antidepressants and suicide deaths,” said the TGA.

“Aggregated trends are useful to generate hypotheses, but are not a sufficient base for making causal statements. Analysis of linked PBS data and suicide data is required to further explore any potential causal relationship between antidepressants and increased rates of youth suicide in the Australian context.”

Clinical worsening of depression and suicidality is a recognised potential risk with the use of antidepressant medicines, based on clinical trial data that showed an increased risk of suicidality in children, adolescents, and young adults (ages 18-24) with major depression and other psychiatric disorders.

As a result, PI and CMI for antidepressants in Australia contain warnings about the risk of worsening depression and emerging suicidality, and advise patients and caregivers to closely monitor for worsening symptoms and suicidal thoughts.

However, the TGA adds that no suicides occurred in these trials and it is unknown whether the suicidality risk in children and adolescents extends to use beyond several months.

“Patients with depression may experience worsening of their depressive symptoms and/or the emergence of suicidality, whether or not they are taking antidepressant medications, and this risk may persist until significant remission occurs,” it said.

The TGA’s safety investigation has led to two recommendations: firstly, that it will liaise with relevant professional colleges, particularly around the need for additional clinical guidelines for general practitioners on the management of depression and prescribing of antidepressants to children, adolescents and young people.

Dr Martin Whitely, a mental health researcher at Curtin University in Perth, told AJP he is pleased with the its second recommendation, which says the TGA will, in early 2021, commence analysis of linked PBS, MBS, hospital and death data to further investigate the clinical journeys of young people prescribed antidepressants in Australia and the relationship between antidepressants and rates of youth suicide, subject to access to data.

It will also consider the established datasets held by the Australian Bureau of Statistics and the Australian Institute of Health and Welfare.

“The correlation is concerning, no question about that, particularly in light of the FDA’s warnings based on RCTs,” said Dr Whitely who has argued that, as antidepressant treatment in Australian children is off-label use, it should be treated with caution.

“But you can’t really establish causality until you have forensic analysis of many individual experiences of those that have used antidepressants, which is exactly what recommendation two will do,” he said.

The TGA’s commitment to doing a forensic analysis in 2021 using the best available linked data is exactly what should happen.

The TGA also sought advice from the Advisory Committee on Medicines (ACM) on the issue.

The ACM acknowledged that clinical worsening of depression and suicidal thoughts is a recognised and well-known risk observed in clinical trials of antidepressant medicines and can occur in a small number of patients during the early phases of treatment.

It further advised that there is a valid and important role for SSRIs and SNRIs in current clinical practice in treating moderate-to-severe depression in children and adolescents, supported by professional guidelines. 

However, in referring to Dr Whitely’s findings, the Committee said that it “did not support the authors’ interpretation of a possible causal relationship between the dispensing of antidepressants on the PBS and rates of youth suicide in Australia”.

Dr Whitely was critical on this point, arguing that the ACM should not rule it out as a possibility since “it’s an unresolved question”.

“We’ve all got to enter into it with an open mind, that there is the possibility that there is causality or not – follow the evidence.

“What we need to do is to engage in the sort of research outlined in the TGA’s recommendation two and then we’ll have more definitive evidence. But the ACM ruling causality out as a possibility… it’s the wrong way to approach science,” he said.

See the full TGA findings here

If you or someone you know needs help the following 24/7 support services are available: 

Lifeline 13 11 14

Suicide Call back Service 1300 659 467

Beyond Blue 1300 224 636 

MensLine Australia 1300 789 978 

Previous Getting in early
Next Looking back and looking forward

NOTICE: It can sometimes take awhile for comment submissions to go through, please be patient.


  1. Peter Bayly

    look carefully at cause and effect. a more likely hypothesis is an increase in rates of depression because of lockdowns and with it more suicides, with or without antidepressant treatment

    • Andrew

      That’s a talkback radio trope that’s been well and truely debunked.

      No increase in suicides, no increase in antidepressant prescribing.

      • Jarrod McMaugh

        agreed – why would such a trend affect only young people?

        this is clearly not a valid theory

      • Anthony Tassone


        Whilst there may not have been an increase in the suicide rate contributing to overall death rate during 2020 from the lockdowns for COVID restrictions, there is evidence of an increased rate of self-harm.

        There have been reports of an increased rate of self harm in Victoria this year of 17% (refer News Corp article below):

        Researchers from the University of Sydney’s Brain and MInd centre authored a report: “Road to recovery: Restoring Australia’s mental wealth”

        Under their own modelling and ‘best case scenario’ they have forecast an increase in the suicide rate in Australia of 13.7% over the next five years factoring in the economic climate and impacts of employment, stability of housing etc.

        This is an evolving and complex situation and too early to draw any firm conclusions (especially ruling anything out).

        Anthony Tassone
        President, Pharmacy Guild of Australia (Victoria Branch)

        • Andrew

          Hi Anthony
          With the increased casualisation of the labour force and insecure work being identified as a major contributor to distress, and likely a significant contributor to the stats you’ve provided;

          Does Guild have any plans to revisit or revise their position on casualisation and pay-rates for staff given the real-world evidence for public health harm of these policies?

    • Sheshtyn Paola

      Hi Peter,

      The researchers were looking at trends in data from from 2003 to 2018, and the TGA was looking at data between 2013 and 2019.

      Lockdowns have only been occurring this year. This particular hypothesis and research has nothing to do with the lockdowns.

      Kind regards,
      Sheshtyn Paola (author)

  2. Brad Sparling

    Probably a long overdue review, especially if you have read any of the books by Peter Breggin, or David Healy’s “Let them eat Prozac”, which were written over a decade ago.
    There have been concerns about suicide and SSRI’s since at least the early 2000’s and I sometimes find it ironic that we make patients sign up to an intense monitoring program when they are prescribed clozapine (which has been incredibly effective at reducing fatalities due to agranulocytosis and myocarditis) but we are happy to hand out SSRIs like lollies with very little follow up – ‘here is six months worth, come back and see me… if you can get in for an appointment’

    There are potentially many factors at play and it would be great to see things like pharmcogenomics taken into account. A study from a few years ago showed that patients that had their antidepressant selection guided by pharmacogenomic considerations had a 70% rate of remission from depression Vs about 28% in those who were given a random antidepressant by their GP.

    Another study by a NSW Psychiatrist showed more than 90% of patients referred for pharmacogenomic testing due to akathisia had at least one diminshing mutation for drug metabolism (this also included more detailed analysis of 10 of these patients that had been involved in either homicide or suicide).

    Either way, it is a good thing that the TGA is taking a closer look and hopefully they can approach it with an open mind, like they claim.

Leave a reply