Issue 33 / 8 September 2014

SUICIDAL behaviour, including attempted suicide, is more common than schizophrenia, a major reason for emergency department admission and hospitalisation, and is relatively poorly researched.

It causes more deaths than motor vehicle accidents in young men and adults. Little wonder that Suicide Prevention Australia has set a goal to reduce the suicide rate by 50% in the next decade.

Studies using psychological autopsies in Australia indicate that as many as 77% of people who suicided were in contact with their GP in the 3 months leading up to their suicide. Since almost 80% of the Australian public visits a GP at least once a year, GPs have the opportunity to detect those at risk of suicide, assess the level of risk, manage the situation and make appropriate referrals.

Educating GPs to assess and identify people at risk for suicide is an effective strategy and, if more widely implemented in general practice could have the potential to lower suicide rates by up to 10% in the short term.

Doctors might be hesitant to assess for suicide risk because individual prediction of those who do suicide essentially appears impossible, given the low base rate and the high number of people presenting with depression. However, the task required is to identify suicide ideation and make decisions about referral and management.

European studies indicate that after GP training to detect depression and suicide risk is implemented, the rates of suicide and suicide attempts decrease, although some reviews show results are not always consistent while others report overall effectiveness.

A caveat is that these studies do not detect the effect of GP education in isolation, but in conjunction with broader interventions such as community awareness.

There is also the issue of the cost-effectiveness of GP education. A recent UK study found that GP education in suicide prevention was highly cost-effective. The authors said: “The costs of training UK general practitioners in suicide prevention, and the ensuing costs of psychological and pharmaceutical therapy they may prescribe to at-risk people, are outweighed by the savings to the public purse owing to the roughly 600 deaths averted”.

About 6000 people die from suicide in the UK each year, so the training program has been associated with a 10% reduction in the suicide rate. If translated to Australia, a 10% reduction in suicide rates would save 250 lives a year, based on our current 2500 suicide deaths each year.

It’s difficult to know if identification and assessment for suicide is common in Australian general practice. Research based on BEACH (Bettering the Evaluation and Care of Health) data indicate that of 1 479 300 general practice encounters between 1998 and 2013, only 406 suicide encounters (suicide ideation or attempt) were recorded. However, suicide screening and identification may be occurring within depression assessments.

A clear barrier to detection and referral in general practice is time. However, GPs are willing to be trained. An Australian study found that many medical students and GPs rated themselves as least competent on skills-based suicide prevention capabilities.

A recent unpublished survey of GPs by the Black Dog Institute found that they identified suicide detection as a gap in their skills base.

For time-poor GPs, an alternative intervention might be to adopt self-screening via tablets or smart phones in the GP setting, prompting self-identification by patients at the time of an appointment. This means GPs would provide tablet computers for patients to screen themselves through a patient health questionnaire while waiting to see the doctor, and then discuss the results with the GP.

GP education on suicide prevention works, is cost-effective and may reduce suicide by 10%. The GP setting provides opportunities to detect risk. A core competency for all doctors is suicide prevention education and training. The provision of new training materials, developed by GPs for GPs is urgently needed.

It’s time this education and training was implemented to significantly accelerate Suicide Prevention Australia’s goal to reduce suicide rates.

Professor Helen Christensen is the director and chief scientist of the Black Dog Institute, director of the NHMRC Centre for Research in Suicide Prevention at the Black Dog Institute and professor of mental health at the University of NSW.


Should more emphasis be given to detecting suicide risk in medical education and training?
  • Yes – specific training (72%, 34 Votes)
  • Yes – with depression (23%, 11 Votes)
  • No – current emphasis is adequate (4%, 2 Votes)

Total Voters: 47

Loading ... Loading ...

2 thoughts on “Helen Christensen: Preventing suicide

  1. Philip LP Morris says:

    One way of improving detection rates for potential suicide is to ask questions about suicide in a graduated non-threatening manner.  First ask if the person has times when, if things continue with the way they are feeling, they would be better off not being here.  Then if that provokes a positive response a more detailed question about having thoughts of self-harm could be asked.  A positive response to that question would then be followed-up with more explicit enquiries about suicidal intent, plans and timing.  This information then informs a physician response plan.  It is the first question that is the key and this should be on the history-taking agenda for all patients attending with emotional distress (of whatever type, and especially in the setting of formal mental illness, substance abuse, onset of serious physical illness, and social situations that generate guilt or shame).   

  2. Ian Watts says:

    When I raised my acutely suicida thoughts with my GP, in my view he needed 3 things in addition to specific mental health skills and knowledge. 1. Good micro-skills in communication (like ‘breaking bad news’); 2.effective funding (I had enough money and would  NEVER allowed him to claim a mental health item); 3, a good prior relationship to understand why I wouldn’t be admitted and how safe I would be when I wasn’t. Are any of these possibly confounding variables studied?

Leave a Reply

Your email address will not be published.