Developing a tool to predict suicide and self harm in adolescents
Mental health problems in children and adolescents contribute to outcomes with long-lasting effects, such as suicidality, self-harm, violence towards others and risky behaviours. For each individual, predicting the likelihood of these consequences is challenging. Professionals use a range of methods to assess risk, including structured risk assessment tools. However, there is no standardised method for risk assessment and many tools lack a scientific evidence-base and have poor accuracy. The purpose of this project is to develop a risk prediction tool for adolescents attending mental health clinics.
Part of our Mental Health across the Life-course research theme
Self harm is a serious public health issue in young people, and is increasing.
Self harm and attempted suicide are common in young people with mental health disorders. In children aged 11-16 years with a mental health disorder, 25.5% have a history of self harm or attempted suicide, compared to only 3% of children of a similar age without any mental illness. This figure rises to 48.6% of older teenagers aged 16-17 with a mental disorder.
Predicting which adolescents are at higher risk of suicide or attempted suicide is an important part of case assessment and management. Accurate risk assessment tools could help professionals base their assessments on evidence. They may help guide the allocation of treatment in a way that is tailored to individuals.
Better assessment of risk could inform the use of treatments and interventions at key times in order to reduce the risk of poor outcomes for children and adolescents.
The first part of this project is to assess current risk assessment tools for child and adolescent mental health, asking the question: what tools are available, and have these tools have been developed using appropriate methods? This will help to identify areas where there is a need for better risk assessment, and key improvements which can be made.
The second part of the project will aim to build on this knowledge by developing a new risk assessment tool, based on similar work carried out for adults, and for physical health conditions such as high blood pressure.
Our aim is to create a tool which is easy to use, clinically useful, and which could be implemented within NHS services.
How we are involving patients and the public
Our aim is to engage with existing PPI groups within the Department of Psychiatry to explore PPI insights to inform proposed research questions and methodology, including adolescents and/or their parents’ perceptions of suicide risk and risk prediction.
We will discuss acceptability of modelling risk in the target group.
How we are planning to implement the research outputs
To be completed
- To use the CRIS database to develop a new tool to predict the risk of self harm in adolescents in secondary mental health care.
- A review of the literature on tools to predict self harm in adolescents.
- Development of a tool to predict self harm in adolescents.
- If the tool appears sufficiently robust, the next stage will be to engage with local health decision makers about how such a tool can be used to improve patient care.
This research will contribute to improving the health of young people accessing secondary mental health care services. Our work will help to refine the picture of the risk factors for self harm in young people accessing secondary mental health care.
The research will also progress the use of natural language processing of healthcare records.
Those accurately assessed as being of higher risk can benefit from being offered interventions, and those accurately assessed as being of lower risk may avoid unnecessary interventions, thereby avoiding associated harm and waste of resources. Structured tools have the potential to make risk assessment less subject to bias and susceptible to errors of individual clinicians. They can support discussions between patients, families and professionals about suicide risks.