Improved early psychosis detection system may halve adolescent risk

Doctors have developed new data mining techniques to detect many young people with new mental illnesses. A new method based on advanced data mining to pick up early risk signs from schools, hospitals, and general physicians was presented at the ECNP Virtual Conference and published in peer-reviewed journals. Schizophrenia Bulletin..

Psychosis is a condition in which you lose contact with reality and suffer from hallucinations and delusions. There are many possible causes, including migration, social stress, and trauma. Drug abuse,Such. It represents an important long-term care burden, affecting about 20 million people and costing about € 94 billion annually in Europe (2011 estimate).

In clinical experience, the best way to manage it is to stop developing it. Over the last 25 years, doctors have developed ways to detect young people at risk of developing mental illness and predict which young people are likely to develop the disease, so they have been able to take steps to reduce the risk. It was. However, the way clinicians find young people is not systematic and may have missed many at risk. Now, UK doctors have developed new data mining techniques that can potentially detect most people at risk of developing mental illness. This will then allow them to provide prophylactic psychological interventions that can halve the risk of developing a full-blown mental illness.

“Prevention is the most promising way to improve,” said Professor Paolov Sapoli, a research leader at the Institute of Psychiatry at King’s College London. mental health Of young people. Mental health in this generation is particularly stressed and faces ongoing concerns about COVID-19 and requires urgent intervention. The future of people at risk of mental illness is to intervene before the disorder occurs. “

“We have developed a data mining method using natural language processing, Medical record For those at risk of developing a mental illness. Many medical records are fairly unstructured, and mental health information is hidden in sections that do not allow systematic research. Our data mining system more completely searches the records of people referred to hospitals (secondary care) for keywords such as weight loss, insomnia, cocaine, and guilt. You can search for 14 different terms. Next, assess the risk of mental illness. At that point, the patient may be invited to a one-on-one interview. It has been found that prevention can reduce the risk of developing mental illness by half. “

The system evaluated 92,151 patients over a long follow-up period. They were able to confirm that their method was working well to detect young people at risk, but Husarpoli said, “These results are other before entering the clinical routine. More replication is needed in the country, but it will look very promising. Replication is an international research consortium funded by ECNP, such as the Prevention of Mental Illness and Mental Health Promotion Network. “

Professor Fusar-Poli suggested that the discovery of these young people was the first step towards prevention. Preventive interventions in these people offer several benefits:

“This is a real benefit. Although the initial cost of establishing a professional service to detect young people at risk for mental illness is high, intervention before the onset of mental illness is specific social health. In addition, it leads to reduced treatment and reduced hospital stays, which means saving about £ 1000 per patient diagnosed with NHS. Our detection system saves many other young people at risk of mental illness. These benefits can be extended to. “

Fusapoli will chair a session on the prevention of mental illness (see below) at the ECNP Conference and announce its achievements.

He went on to say, “We are working on the prevention of mental illness and mental health promotion in the ECNP Special Group, and in collaboration with the European Union Brain Research Area (EBRA), we have set up a pre-warning system across Europe. I am. Young people There is a risk of mental illness. Bringing the best expertise to address this issue is essential and anyone can learn from the experience of others. “

Professor Andreas Meyer-Lindenberg (Mannheim), a member of the ECNP Board, commented: For biological data and biomarkers, the results show the benefits that clinicians can benefit from when the wealth of written information they create in their daily work is mined using an innovative approach. “

Professor Mayer Lindenberg is not involved in this study.

For more information:
Use natural language processing in electronic medical records to enhance detection and prediction of psychotic risk, Schizophrenia Bulletin, 2020.