M.A.P.P.E.D. IMproving the Accuracy of Psychosis PrEDiction using primary-care consultation data

Date of Approval: 
2018-08-07 00:00:00
Lay Summary: 
Psychosis is a serious, long-term mental illness in which sufferers experience hallucinations and unusual thoughts. Many sufferers do not recover well. They may always have symptoms, have a poor quality of life. Poor recovery is less likely if they receive treatment quickly when first ill. Many people who start to feel unwell go to their GP. The GP needs to recognise the early stages of a psychotic illness and quickly refer to mental health services. Many GPs find this difficult because they see few people with this problem and many of the first symptoms of psychosis could lead to many other mental health problems. We are developing a tool to help GPs recognise the early signs using new indicators of risk, which may help to detect these signs much earlier. We will develop and test a tool. An effective tool would tell GPs about the risk of developing psychosis for each patient and therefore who to refer and lead to better use of the NHS's resources. A quicker referral will improve the outcome for patients and decrease pressure on the NHS by reducing the number of times that patients visit their GP or other services like A&E.
Technical Summary: 
People with psychosis, including schizophrenia, frequently have poor outcomes. Approximately 25% relapse within 36 months and many suffer residual symptoms. Quality of life is often low. Repeated relapses are associated with increased distress and greater cognitive impairment. The costs of treating psychosis are high and increased by poor outcomes. Many risk factors for a poorer outcome are difficult or impossible to modify. Increasingly research has focussed on duration of untreated psychosis (DUP) because it is negatively associated with outcome and potentially modifiable. Most sufferers enter specialist services via primary-care, therefore it is important for GPs to recognise emerging psychosis. However, many GPs find this difficult because the early symptoms of psychosis are non-specific. Our data shows that this group presents more frequently to primary-care in the prodromal phase, particularly in the final 6 months, representing an opportunity for earlier diagnosis. There is a need for a primary-care predictive tool. We have already defined candidate predictors. The next stage is to develop and validate a prediction model. An effective prediction tool could expedite referral, decrease DUP, improve outcomes and reduce pressure on primary-care and A&E. A tool with greater specificity and sensitivity may lead to more efficient resource allocation resources.
Health Outcomes to be Measured: 
Primary outcome-a coded diagnosis of a psychotic disorder
Application Number: 

Sarah Sullivan - Chief Investigator - University of Bristol
Sarah Sullivan - Corresponding Applicant - University of Bristol
Daphne Kounali - Collaborator - University of Bristol
David Kessler - Collaborator - University of Bristol
Glyn Lewis - Collaborator - University College London ( UCL )
Irwin Nazareth - Collaborator - University College London ( UCL )
James Robinson - Collaborator - Avon and Wiltshire Mental Health Partnership NHS Trust
Richard Morris - Collaborator - University of Bristol
William Hamilton - Collaborator - University of Exeter

HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation