New online psychosis risk calculator could help identify candidates for early intervention services
Dr Paolo Fusar-Poli is a researcher at King’s College London’s Institute of Psychiatry, Psychology & Neuroscience. His research focuses on the detection, assessment and treatment of adolescents and young adults at risk of developing psychosis. Dr Fusar-Poli is working to identify the specific factors which make an individual vulnerable to developing psychosis, leading to better, more tailored preventative treatments and improved outcomes for the most severe psychiatric disorders.
I have been working with colleagues at the EPIC (Early Psychosis: Intervention & Clinical-detection) lab at King’s College London and South London and Maudsley NHS Foundation Trust (SLaM), to develop a new psychosis risk calculator which is available online at www.psychosis-risk.net.
Psychosis is a mental disorder which causes people to perceive or interpret things differently from those around them, sometimes involving hallucinations or delusions. If left untreated, it can potentially develop into a severe mental illness, and existing treatments for psychosis aren’t always effective.
As such, preventing psychosis from developing in the first place could improve the lives of many adolescents and young adults. To do this, we need to first identify those who are at risk, and treat them with preventative strategies.
Although there are existing clinical tools available for detecting psychosis risk, these assessments are usually used only when an individual is already accessing specialist health care services such as SLaM’s OASIS (Outreach and Support in South-London) service. Current clinical practice means that young people are not always referred to these specialist services early enough, and their illness may have already progressed to the point that it may be too late to benefit from preventative treatment.
Our new psychosis risk calculator allows us to predict an individual’s risk of developing psychosis based on simple clinical information routinely collected by mental healthcare services, such as age, gender, ethnicity and initial diagnosis.
To select which factors were likely to best predict an individual’s risk of developing psychosis, we consulted the best available meta-analytical evidence from research in this field. We then developed and fine-tuned our model by testing it on anonymous real-life patient data, accessed through the NIHR Maudsley Biomedical Research Centre’s Clinical Record Interactive Search (CRIS) system.
Our evaluation of the risk calculator – published in the peer-reviewed scientific journal JAMA Psychiatry – showed that it is more useful at identifying adolescents and young adults at risk for psychosis than current clinical practice. Our tool has been externally tested and validated to rigorous international standards, and it is cheap and easy to implement in clinical practice.
Our hope is that the calculator can be used by mental health professionals in specialist (secondary) mental healthcare services to detect which of their patients are more likely to develop psychosis, allowing them to be signposted to specific preventative services such as OASIS for in-depth assessment and tailored preventative treatment.
The calculator will be soon piloted in some South London & Maudsley services to test its real world effectiveness. The EPIC lab is currently testing refined versions of the calculator, as well as proceeding with further external validations in other clinical scenarios. If these tests are successful, our risk calculator has the potential to improve the lives of many people, especially adolescents and young adults who are at risk of developing the most severe psychiatric disorders.
Paper reference: Development and Validation of a Clinically Based Risk Calculator for the Transdiagnostic Prediction of Psychosis, Paolo Fusar-Poli, Grazia Rutigliano, Daniel Stahl, Cathy Davies, Ilaria Bonoldi, Thomas Reilly, and Philip McGuire, JAMA Psychiatry, DOI:10.1001/jamapsychiatry.2017.0284
Tags: Clinical and population informatics - Informatics - CRIS - Publications -
By Paolo Fusar-Poli at 13 Apr 2017, 13:18 PM
Back to Blog List