select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to antidepressant use and suicidality

The effect of specific types of antidepressants on mortality, suicide and aggressive behaviours is the subject of a paper published in The BMJ in which the authors report an increased risk of suicidality and aggression in children and adolescents taking those drugs.


Dr Paul Ramchandani, Consultant Child and Adolescent Psychiatrist with CNWL Foundation NHS Trust and Reader in Child and Adolescent Psychiatry, Imperial College London, said:

“I think this paper is really limited in what it can tell us about treatment of depression in the UK.  That’s not to dismiss all the findings, and risks of suicidal thinking and acts of harm and aggression should be discussed with young people so that they and their parents can make an informed decision, but the findings should be viewed with extreme caution.

“There have been concerns for a number of years about the way in which some trials of medicines or drugs for depression were originally reported, particularly when the treatment was for children and adolescents. This new paper has looked again at some of the key reports and found that rates of suicidal thinking and self-harm and aggression are higher in those taking antidepressant drugs in trials than those taking placebos. The absolute increases in what the authors call suicidality (which includes acts of harm and suicidal thinking) are increased by about 2-3 in 100 people taking the antidepressant drug. That means that about two to three people in 100 taking the medication will have a harmful act such as an overdose or thoughts of suicide which they may not have had taking the placebo. Although this study has found more detailed data, the increase in risk is similar to what has been known about for some time, and described in the NICE guidelines going back to 2005.

“For aggression the increase in risk is slightly bigger – about 3 in 100 people taking antidepressants compared to placebo. For comparison the estimated benefit of taking antidepressant treatment in young people is that about 10-20 out of 100 will benefit more taking medication compared to placebo.  So these risks and benefits have to be weighed together.

“There are three important things to consider though. First, there were no suicides in young people in any of these trials, so this is talking about suicidal thinking and behaviour, not actual suicide. Suicidal behaviour and thinking is a cause of serious concern and this small increase in risk is important, but is not the same as an increase in suicide.

“Second the studies included here covered a range of antidepressant drugs.  This included at least one drug (Paroxetine) that is not recommended for use in young people as we already know about the increased risk of suicidal thinking. In fact within this analysis most of the strongest effects were seen with this one drug (4 of the 5 strongest effects for suicidality and 3 of the 5 for aggression) so it is possible, and in fact seems likely that the increased risk described in young people in this paper does not properly relate to the drugs actually used in the UK for depression in young people.

“Third, these increases in risk have to be balanced against the possible benefits of using antidepressant medication. Depression is a serious condition and young people with depression deserve access to all the forms of treatment that may help them. This includes psychological or talking treatments, like Cognitive Behavioural Therapy and Interpersonal Therapy. However for some people antidepressant drugs may be helpful.  For these young people and their families, knowing more about the potential risks of medication, as well as the potential benefits, is important. Being able to discuss this with a qualified and knowledgeable professional is key to helping them reach a balanced decision about what is best for them.”


Dr Michael Bloomfield, Clinical Lecturer in Psychiatry, MRC Clinical Sciences Centre and UCL, said:

“This new study highlights the challenges faced in measuring the intended benefits against the potential harms of psychiatric medicines in young people along with the urgent need to improve outcomes in this potentially vulnerable patient group. However, many of the treatments included in this analysis are currently neither routinely used nor recommended in young people.

“It is existing best practice for any young person commencing treatment with an antidepressant to be seen very regularly by their doctor during the initial phase of their treatment.

“Unfortunately, severe depression can be potentially life-threatening but there is evidence that early treatment with antidepressants together with psychological therapies for the young person and their family are effective. As such, no young person should stop taking an antidepressant without first discussing this with their GP or psychiatrist.”


Dr Mara Parellada, Specialist in Child and Adolescent Psychiatry at Hospital General Universitario Gregorio Maranon and the Centre for Biomedical Research in Mental Health Network (CIBERSAM) at Complutense University of Madrid (UCM) and member of the Scientific Committee of the European College of Neuropsychopharmacology (ECNP), said:

“The results from this study do not allow us to state that ‘Antidepressants double the risk of aggression and suicide in children’. There was no single death by suicide in children and adolescents in the 70 trials reviewed for the article. Indeed, there was no increased risk of mortality for all causes for active drug vs placebo (Table 2).

“Second, stating that ‘Antidepressants double the risk of suicide’ can be misleading. Even if replicated, it is necessary to consider the absolute risk, not just the relative one. Increasing by 100% looks like a large difference, but if it’s one in a thousand being doubled to two in a thousand then it is still a very small amount.

“The editorial, by Joanna Moncrieff, from the Critical Psychiatry Network, that summarizes the article by Sharma and colleagues by stating (as in the discussion of the paper) that it ‘confirm[s] the association between suicidal behaviour and use of modern antidepressants in children and adolescents’, is also misleading and potentially harmful. The conclusions of the referred article state: ‘Because of the shortcomings identified and having only partial access to appendices with no access to case report forms, the harms could not be estimated accurately. In adults there was no significant increase in all four outcomes, but in children and adolescents the risk of suicidality and aggression doubled. To elucidate the harms reliably, access to anonymised individual patient data is needed’. However, suicidality is not the same as suicide (the former includes many non self-harm-seeking behaviours).

“With regard to the main objective of the article, harm can only be estimated fully in comparison with benefit. Comparisons can be made between chemotherapy and antidepressants. In the first, adverse effects of chemotherapy can only be estimated in comparison with benefits in terms of decreased mortality; likewise, the harm of antidepressants can only be estimated in comparison with the decreased risk of suicide associated with mental illness.

“On the numbers involved:

36 out of the 70 clinical trials were conducted for an indication different from major depressive disorder (MDD). How many of the events occurred in subjects with a MDD? We do not know.

The studies included 10,917 patients. There were no deaths in children/adolescents. In adults there were deaths: 1/4277 subjects on duloxetine; 0/456 subjects on fluoxetine; 0/1766 subjects on paroxetine; 0/3165 subjects on sertraline; 1/1263 subjects on venlafaxine; 2/6832 on placebo; 4 deaths ‘misreported’).

Where are the data of expected deaths by suicide in depressed patients? In the article there were 49 suicidal events in 6,832 subjects treated with placebo and 86 events in 10,917 subjects on active treatment.”

“On the terms used: Suicidality is described in the Sharma article as follows:  ‘We counted all attempted suicides, including intentional self-harm (for example, slitting of wrists), intentional overdoses, and obvious preparatory events (for example, putting a knife to the wrist or neck, but being stopped before any harm).’  The range of severity of the different events counted as suicidality is huge. How many of the ‘suicidal events’ were threatening by putting a knife to the wrist? How many were jumping out of the window? Can these be counted together?

“Thus, this article really refers to the shortcomings in reporting harm effects in clinical reports. Nothing else. The risk and potential harm of extrapolating this mean-based data to the individual treatment and making general statements, as in the editorial, is huge. For example, the articles mixes up a middle-age MDD highly suicidal male with a vulnerable young male trying to manipulate his girlfriend’s behaviour by threatening to slash his wrist under the influence of alcohol (and with an SSRI prescribed, not necessarily taken).

“Sticking to the data, no conclusion can be taken with regard to the global benefit/harm associated with antidepressants.”


Dr Paul Keedwell, Consultant Psychiatrist and Senior Research Fellow in the Neurobiology of Mood Disorders at Cardiff University, said:

“The findings provide a further lesson in how professionals must carefully scrutinise drug company summaries for data on adverse events.

“Although the headline is potentially alarming, a doubling of the normal population risk for suicide would still be a rare event, and less than the risk of suicide in untreated depression. Nevertheless the increased rate emphasises the need for careful monitoring in the first few weeks of treatment. The risk of an adverse reaction must also be weighed against the devastating effect of untreated depression on emotional, social and cognitive development.

“Overall, timely antidepressant prescribing has been shown to reduce the incidence of suicide across the world. No treatment is entirely risk free, but the risks must be weighed against the potential benefits.”


Prof. Shirley Reynolds, Professor of Evidence Based Psychological Therapies, Director of Charlie Waller Institute, University of Reading, said:

“This study highlighting the risks associated with two types of antidepressant medication (SSRIs and SNRIs) is extremely important. It should inform the treatment of children and young people with mental health problems and will be of great interest to parents and young people themselves. The research was well conducted and the results are likely to be trustworthy.

“The authors integrated detailed clinical results from 70 separate studies that compared 4930 patients who were taking an SSRI or an SNRI with 3531 patients who received a placebo (i.e. a dummy or fake drug). They compared the number of deaths overall, suicides, suicidal attempts, aggressive behaviours and ‘akathisia’ (an extreme form of restlessness. Patients included children, adolescents and adults. Overall (i.e. across all patients) there was no increase in deaths or suicide.

“They made two startling discoveries specifically relating to children and young people. First, that children and young people who were taking SSRIs or SNRIs were more likely to attempt suicide and were more likely to think about suicide. Second, children and young people taking SSRIs or SNRIs were more likely to exhibit aggressive behaviour than those taking a placebo.

“Obviously these results will make doctors, parents and young people themselves think harder about taking anti-depressant medication. But do the results mean that children and young people should never be prescribed anti-depressant medication? No. There are alternative treatments and all young people should be offered an evidence-based psychological treatment immediately. However, anti-depressants should be available when a young person does not respond to psychological treatment or does not want psychological treatment. Combining anti-depressant treatment and psychological treatment is associated with improved outcomes and can lead to more rapid reduction in symptoms. But only a specialist child and adolescent psychiatrist should prescribe anti-depressant medication to children and young people and all children and young people who are prescribed anti-depressants must be carefully and regularly monitored.

“Children and young people who experience mental health problems are a group at particularly high risk, with or without anti-depressant medication. These data highlight that risk and the urgent need to increase resources to provide rapid access to evidence based treatments.”


Suicidality and aggression during antidepressant treatment: systematic review and meta-analyses based on clinical study reports’ by Sharma et al. will be published in The BMJ on Wednesday 27th January. 


Declared interests

Dr Paul Ramchandani: I have no conflicts of interest to declare.

Dr Bloomfield: None received

Dr Parellada: “No conflicts of interest”

Dr Keedwell: None received

Prof. Reynolds: “My post at the University of Reading is partly funded by the Charlie Waller Memorial Trust and Berkshire NHS Foundation Trust.   I am co-director of the Oxford Academic Health Science Network Anxiety and Depression programme.  I have received grant funding for research from NIHR and MRC.”

in this section

filter RoundUps by year

search by tag