Subtotal: $0.00

No products in the cart.

Select Page

It was out of the blue

Trying to determine if there are predictable indications of risk before the suicide of children and young people

Article in Review: Rodway, C., Tham S-G., Turnbull, P., Kapur, N. and Appleby, L. (2020). Suicide in children and young people: Can it happen without warning? Journal of Affective Disorders, DOI: 10.1016/

Summary & Relevance: Friends and family members of children and young people who die by suicide often say that there was little or no warning, and the phrase “out of the blue” is frequently used to describe the suddenness with which the death has occurred. Risk indicators can include self-harm, the communication of suicidal ideas or contact with relevant mental and social health care providers. Could it be that those families have misunderstood the warning signs or missed them completely? A 2020 study has sought to determine if there are indeed some young people who die by suicide with minimal or no warning.

In the United Kingdom, there were 595 suicides by young people (aged 10-19 years) between 2014 and 2016. For this study, documentation and recordings from the coroner, police and other investigative bodies were obtained for 544 of these deaths. The recordings included statements from families, friends and professionals about the adversaries and stresses the young people were experiencing prior to their death.



In their assessment of these files and recordings, the authors found a substantial minority of young people had given no indication of suicide risk indicators. A majority of 112 (70%) had no known contact with any relevant service. A total of 161 young people, 30% of the sample, were not known to have expressed suicidal ideation nor had previously self-harmed. This group was also less likely to have been involved with support services. Of this minimal warning group, 83% were male. A total of 107 young people who were considered to have minimal risks also had lower rates of other indication risks that were counted and considered. These additional criteria included:

  • Diagnosis of mental illness
  • Excessive alcohol and illicit drug use
  • Diagnosis of mental illness
  • Social isolation
  • Suicide-related internet use
  • Recent life events (relationship breakups or family arguments)

Explanations for findings

Explanations for such a considerable amount of deaths in the minimal warning group could have been examples of where emergent risks were missed, or emotional distress was concealed. However, another explanation is that the onset of suicidal feelings escalated quickly amongst the minimal warning group. This may reflect immaturity in the adolescent brain whereby adverse events cause a rapid onset of despair.

Limitations of the study

The authors recognised that methodology may have influenced findings. Information analysed had been contributed by friends and family members of those that had died. Because of the sensitive nature of their accounts, it is possible that information regarding the young people was misreported or downplayed. It is generally accepted that men are less likely to disclose or seek help for emotional problems, and as most of the minimal warning group were male, it is possible that had not disclosed more long-term stresses and concerns. Families may be distressed by the thought that they could have missed signs to suggest they could have intervened.


The article asks us to consider the question “how can we prevent suicide when there are few warning signs or when suicide risk escalates rapidly?” and closes with three recommendations.

  1. Professionals, schools and families should seek to recognise lesser degrees of distress in children and young people.
  2. Crisis services should be available to support young people where a risk has escalated rapidly.
  3. A multifaceted approach of teaching emotional awareness, along with public health and media can help young people to recognise rapid onset risk of suicide in themselves and their peers.

Request a Prospectus

  • This field is for validation purposes and should be left unchanged.