Overview

For the first time, healthcare planners and commissioners can have access to valid, reliable information about the expected burden of new cases of psychotic disorder in England & Wales.

Version 1.1 is now released! This version updates PsyMaptic predictions using population data from the 2011 Census of Great Britain. Predictions are available at Local Authority, County and National Levels. Predictions use the same methodology as before, but models now take into account local variation in both population density and deprivation. See our Release Notes for full details of the changes and enhancements made to Version 1+.

Until now, healthcare policy makers, commissioners and planners have not had access to accurate information regarding the level of new cases of psychotic disorder expected in different regions of the England & Wales or different sociodemographic groups. This problem has been particularly noticeable within Early Intervention in Psychosis Services [EIS], launched in the UK in 2002 as the main service point for young people (aged less than 35 years old) experiencing a first episode of psychosis. These services were commissioned on the basis of an anticipated caseload of around 51 new cases per 100,000 people per year (Department of Health, 2002). Since then, some anecdotal reports from some services suggested lower than expected caseloads (Tiffin & Glover, 2007), while others reported caseloads considerably higher than predicted by this rate (Mahmmood et al 2006). Recent epidemiological studies conducted in EIS in rural areas (Kirkbride et al 2012a, Cheng et al 2011) suggest that in more rural communities rates are somewhat lower than predicted by the Department of Health. A uniform rate for planning EIS in England is not optimal given that the incidence of psychotic disorder is known to vary considerably by age, sex, ethnic group and level of urbanisation (Kirkbride et al 2012b). We developed PsyMaptic to take these variables into account in public mental health planning.

PsyMaptic is an epidemiological prediction tool which gives healthcare commissioners and other stakeholders free access to key prediction data on the expected incidence of first episode psychosis in England & Wales every year, based on the unique sociodemographic characteristics of the underlying population. Version 1.0 takes into account age, sex, ethnic group, Local Authority (LA) population density,  the extent of deprivation in each LA and a quadratic term for this measure. Predictions are based on risk estimates using robust empirical data from two large epidemiological studies of first episode psychosis in England (see Kirkbride et al 2006, Coid et al 2008) to predict the expected rate of disorder in other regions of England & Wales.

What’s available?

Our PsyMaptic models are presently limited to the prediction of incidence cases of first episode psychosis in people aged 16-64 years old. Predictions can currently be broken down by broad age group, sex and selected ethnic groups* for all Local Authority Districts and Counties in England & Wales. Other predictions (i.e. by ethnic group or for other geographical aggregations) may be requested as custom data.  

Future versions of PsyMaptic will allow for predictions to be made for specific psychotic disorders, such as schizophrenia, or other non-psychotic psychiatric disorders such as depression or anxiety. These models are currently under development to ensure the accuracy of the data. We will also continue to develop and refine the models which underpin the PsyMaptic prediction tool to improve the precision with which we can make predictions. Key prediction data is currently disseminated free-of-charge to the end-user through our online PsyMaptic maps or as downloadable tables.

*Only prediction data which passed validity checks is released. See Version 1.0 release notes for full details.

About the development of PsyMaptic

The development of PsyMaptic was led by by Dr James B. Kirkbride, while he was part of the EpiCentre group, Department of Psychiatry, University of Cambridge, as part of his Sir Henry Wellcome Fellowship from the Wellcome Trust. The methodology used to develop PsyMaptic has undergone academic peer-review and   is fully described in our accompanying published paper at BMJ Open. Version 1.0 of the tool was developed by Dr Kirkbride in the Division of Psychiatry, UCL where he is currently based. 

Several people and services have contributed to the genesis of this project, which has been funded by the Wellcome Trust and supported by the NIHR through the Collaborations for Leadership in Applied Health Research and Care (Cambridgeshire & Peterborough) [CLAHRC-CP]. Full acknowledgements are detailed here.

Disclaimer

While we have made attempts to provide robust, accurate prediction models of psychosis, our methodology involves a number of assumptions, which you are recommended to read. Further, our prediction models provide estimates of the expected count of new cases of psychosis in a given region and, like all estimates, include a degree of error. This error needs to be taken into account by end users of the data, and we include estimates of this error in the data files you may download from this website. This allows you to interpret the precision or confidence we assign to each prediction. Generally speaking, as predictions become more fine grained (i.e. for more specific sociodemographic groups), the error associated with our predictions becomes larger. Before downloading any prediction data from this site, you will be asked to agree to our Terms of Use. We also strongly advise you to read our disclaimer.