Children and adolescents who present to hospital following self-harm have an increased future risk of suicide compared to young people in the general population and the risk may persist for several years, according to a study published in The Lancet Child & Adolescent Health journal.

The observational study of more than 9,000 children and adolescents aged 10-18 years who presented to five English hospitals for non-fatal self-harm spanned 16 years and looked at future risk and cause of death in these patients. Of the 9,173 individuals who presented for self-harm, 55 (0·6%) died by suicide.

While suicide was uncommon, the findings suggest that the annual rate of suicide in the patients who self-harmed was more than 30 times that expected in other young people in the general population. The increased risk remained at a similar level over more than 10 years of follow up.

“Although our findings should be interpreted with caution, due to the relatively small number of deaths by suicide that occurred during the study, they emphasise a need to identify strategies to try to prevent self-harm in children and adolescents in order to reduce future risk of suicide. They also underline the importance of provision of good care for children and adolescents who present to hospital following self-harm.” says lead study author Professor Keith Hawton, Centre for Suicide Research, Department of Psychiatry, University of Oxford, UK. 

The risk of suicide was greatest for boys

Self-harm is the most important risk factor for subsequent suicide in young people. In the UK, suicide is the leading cause of death in people aged 15–24 years, and general population rates of suicide have increased among people aged 10–19 years since 2010.

The new study investigated subsequent risk of death after people attended hospital for non-fatal self-harm, looking at the cause of death – suicide, accidental death, or death by other causes. It also estimated the incidence of suicide by time after hospital attendance, adjusting for gender, age, and previous self-harm.

The study included 9,173 young people (6,828 girls and 2,345 boys) aged 10-18 years who attended hospital a total of 13,175 times between 2000-2013. Patients’ follow-up period varied depending on when during the study patients were admitted to hospital, with a maximum follow-up period of 16 years and a minimum follow up of two years.

The study included patients from five hospitals (one in Oxford, three in Manchester and one in Derby). The authors note that since most young people who self-harm do not attend a hospital, these findings only relate to those adolescents who do, particularly since there tends to be large differences in the patterns of self-harm between the two groups.

Within the 16 years follow up, the study found that 124 (1%) out of 9,173 young people had died.  Of the total 124 deaths recorded, 55 (44%) were due to suicide, while 27 (22%) were accidental and 42 (34%) due to other causes.

The authors found that the rate of suicide in the first year after hospital presentation for self-harm was more than 30 times higher than the expected rate in the general population 10–18 year-olds in England and that the increased risk remained at a similar level over the 10 years of follow-up.

Psychosocial assessment following presentation for self-harm is needed

The risk of suicide was greatest for boys, those who were in older adolescence (16-18 years) when they presented to hospital, and those who repeated self-harm. The authors also note that some forms of self-harm were more strongly linked to subsequent suicide but advise against including detail of this kind in media reporting.

Furthermore, the authors also found that these patients also had a long-term risk of death from accidental poisoning, particularly involving drugs of abuse, and especially in boys.

Study author, Dr Galit Geulayov, Centre for Suicide Research, Department of Psychiatry, University of Oxford, UK makes some recommendations for improving care following admission to hospital for self-harm: “Careful psychosocial assessment following presentation for self-harm is needed, including evaluation of the individual’s problems, needs and risk as specified in national guidelines, to plan appropriate follow-up care. This should also incorporate assessment for possible substance use.” 

The authors note a further limitation of their study, namely that since the study looks at patients in only three cities in England, the findings may not necessarily apply to the whole of the country.

Writing in a linked Comment, Dr Rohan Borschmann, University of Melbourne, Australia, emphasises the importance of the risk: “Timely responses to mitigate this risk will require social scaffolding and support for young people and their families, their schools, their workplaces, and their peers. Such responses will require commensurate coordinated policy and practice efforts across the health, education, and social care sectors. Given the well documented social gradient in health, evidence informed efforts to reduce adolescent self-harm will disproportionately benefit vulnerable subgroups of young people, thus directly addressing the Sustainable Development Goal of reducing inequalities.” 

Support can be found at the Samaritans on 116 123, or email: or visit For those outside the UK, Befrienders Worldwide also provide support: