Advancing Psychiatric Mapping Translated into Innovations for Care: PsyMaptic-A (360G-Wellcome-101272_Z_13_A)

My current and previous Wellcome Trust Fellowships (Henry Wellcome, 2009-13; Henry Dale 2014- present) focus on important aetiological questions about psychotic disorders, using epidemiological data. Psychotic disorders are a debilitating set of mental health disorders, characterised by hallucinations, delusions and cognitive deficits. My research demonstrates that these disorders have a robust, replicable social aetiology, with higher incidence rates observed in young people,1–3 men,1–3 ethnic minorities2–7 and people exposed to greater social disadvantage.8–11 In my previous fellowship, I established the largest epidemiological study of first episode psychosis [FEP] in England since 1999, to demonstrate that these substantial mental health inequalities also exist in more rural populations (East Anglia)3,12; rates are over twice as high as expected,3,13 with deprived rural communities experiencing the highest psychosis incidence. This study has generated new Page 5 of 18 aetiological clues, for example by showing that people at "ultra-high risk" of psychosis are exposed to similar social and spatial markers of social disadvantage as FEP patients,14 implicating an aetiological role for social adversities prior to onset. I have also demonstrated that migrants face greatest FEP risk when immigrating in childhood,15 an important period of sociocognitive development. I am attempting to replicate this in my current Fellowship, in a larger longitudinal cohort using Swedish national register data. Using this data, I have already shown that refugees are at elevated psychosis risk compared with other migrants from the same region of origin,7 providing further insights into the possible social determinants of psychosis. Epidemiological data can also inform mental health service planning. In England, Early Intervention in Psychosis [EIP] services assess and treat people with suspected FEP, offering evidence-based multidisciplinary care to improve downstream clinical and social outcomes, shown to be highly costeffective.16 Unfortunately, original policy implementation guidance17 made no provision for the heterogeneity in incidence described above, with services commissioned on a uniform expectation of 15 new cases per 100,000 people-per-year. This was at least half the true incidence,1,3 and over three times lower than the overall referral rate for all suspected FEP, including "false positive" (nonFEP) referrals,3 who still require appropriate psychiatric triage and signposting, and consume additional EIP resources not factored into original guidance. In response, I demonstrated that epidemiological estimates of psychosis risk could be used to better predict the expected FEP incidence in the population at-risk in England,13 nationally and regionally. The tool, known as PsyMaptic, has had substantial impact on policy and commissioning since it was freely-released in 2012 (www.psymaptic.org).16,18–22 Most recently, it has been used to inform national EIP workforce calculations23 following the introduction of Access and Waiting time standards,19 as part of the Department of Health’s commitment to achieving parity of esteem between mental and physical health by 2020.24 Whilst I have demonstrated, via PsyMaptic, that it is possible to translate epidemiological data into effective public mental health,25 some vital methodological limitations require empirical attention. I therefore seek Wellcome Trust enhancement funding to answer four empirical questions to develop and apply novel statistical prediction methodologies to generate sustainable, dynamic populationbased models of future mental health need.