Suicide prevention

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e5779 (Published 31 August 2012) Cite this as: BMJ 2012;345:e5779
  1. Michael Kaess, child and adolescent psychiatry fellow
  1. 1Orygen Youth Health, 35 Poplar Road, Parkville VIC 3052, Australia
  1. michael.kaess{at}unimelb.edu.au

Early public health interventions may be the answer

Suicide is the tenth leading cause of death worldwide, with a rate of about 14.5/100 000 people a year.1 Suicide often affects younger people than other causes of death, and it is the second most common cause of death among European youth.2 Given these figures, suicide prevention efforts are critical and are one of the World Health Organization’s prime health policy goals.3

A linked paper by Morthorst and colleagues (doi:10.1136/bmj.e4972) reports the findings of the Assertive Intervention of Deliberate Self Harm (AID) Trial, which investigated the effectiveness of an assertive outreach intervention that offered patients motivational support and provided assistance with help seeking, attending appointments, treatment compliance, and crisis management after a suicide attempt.4 The rationale behind this randomised controlled trial was clear: episodes of deliberate self harm often precede completed suicide,5 and most suicides occur in the context of mental illness.6 Adequate assessment of mental state and follow-up care after a suicide attempt are therefore important. In reality, however, only a minority of those who attempt suicide receive adequate care, and facilitating optimum follow-up care after deliberate self harm has recently been suggested as a promising strategy for preventing suicide.7

The findings of the AID trial went against expectations. The intervention did not result in significant differences in the frequency of deliberate self harm in the following year compared with standard treatment. These negative findings may seem disappointing and contradictory to previous studies evaluating the same approach.8 However, attention to an important influential factor may cast a more optimistic light on the findings.

Although treatment compliance was not ensured, control participants were offered standard treatment from a centre of excellence in suicide prevention.4 Standard care was both of apparently above average quality and, more importantly, easily accessible. Standard services for most people who have self harmed may be very different. Clients are commonly lost to follow-up after a suicide attempt. They have to be motivated to seek and access treatment, which is a big ask given that most people who have attempted suicide are ambivalent about staying alive in general and seeking help in particular. Clients often go on to waiting lists for treatment. In such a reality, an assertive outreach intervention like AID might still be beneficial, as has been shown previously in a randomised controlled trial.8

Furthermore, this is not the first randomised trial with subsequent deliberate self harm as a primary outcome that has shown negative results. The intervention group in the current study received more frequent medical treatment than controls, which may have patented detection bias for the primary outcome. Whether increased contact with medical services can contribute to the prevention of future completed suicide is difficult to assess since the relative rarity of suicide makes choosing it as a primary outcome measure difficult.9

A meta-review of suicide prevention strategies found that some methods, such as pharmacotherapy or specific psychotherapy, showed promising effectiveness compared with minimal treatment or “treatment as usual.” However, these effects often disappear when another active standard treatment is the comparator.7 Does this mean that good standard treatment is already as good as it can get at preventing further suicide?

Morthorst and colleagues report further suicide attempts in 10-20% of study participants, depending on group and outcome measure, which should temper excessive optimism. A previous review estimated the one year incidence of a repeat suicide attempt to be 16% and of a fatal repeat attempt to be 2%.10 It seems that even good standard treatment after an episode of self harm does not reduce rates of subsequent deliberate self harm remarkably.

The AID trial raises important considerations for both clinicians and policy makers. Since the findings do not suggest that there were general engagement difficulties in the treatment of this population, those designing services may consider that instead of additional interventions, it may be sufficient to facilitate easier access to support and treatment for young people who have self harmed.11 Aspects of active interventions such as AID might still contribute to closing gaps in treatment after deliberate self harm in the “real world,” but may become less important when adequate standard follow-up treatment becomes more widely available.

It has been suggested that a multilevel approach to suicide prevention (increasing public and clinical awareness, training general practitioners to recognise and refer patients with suicidal ideation and behaviours, facilitating access to treatment, and restricting access to means of suicide) would be most beneficial.7 Suicide prevention strategies may also need to be targeted at younger children or adolescents, before the age when the risk of suicidal behaviour starts to increase and suicide events begin to occur. Prevention and early intervention are crucial since high quality treatment fails to reduce repetition of deliberate self harm in well performed studies. School based awareness programmes, training of gatekeepers (such as teachers), teaching problem solving skills to children and adolescents, and screening interventions to detect those at risk may have potential as early suicide prevention strategies. However, evidence of the effectiveness of such approaches is required and some are currently being investigated in a large European multicentre study.12


Cite this as: BMJ 2012;345:e5779


  • Research, doi:10.1136/bmj.e4972
  • Competing interests: The author has completed the ICMJE unified disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares no support from any organisation for the submitted work; no financial relationships with any organisation that might have an interest in the submitted work in the previous three years; and no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.