Selective prevention

Rather than employing a universal strategy, we think there is greater utility and better use of resources by adopting a selective preventative approach. This public health model would aim to stop the development of new FEP cases from subpopulations at increased social risk. Individuals within these subgroups may be asymptomatic or display nonspecific symptoms of mental distress associated to the social risk factors they have been exposed to. We also believe this work should be children and young people specific, as the onset of psychosis is most common in youth.^{Reference Solmi, Radua, Olivola, Croce, Soardo and Salazar de Pablo35}

Risk prediction–detection modelling

To identify at-risk individuals from within the general population, a new prediction–detection tool will be needed. Through an innovative, data-science-based approach, this tool could be mathematically modelled on existing FEP patients’ sociodemographic information and social determinant data. By using real-world metrics, the tool should be able to: (a) identify neighbourhoods and communities at high risk, in terms of their probabilistic likelihood of containing future psychosis cases; and (b) predict the demographic level characteristics of at-risk individuals within those neighbourhoods. The tool would therefore enable a place-based focus to risk prediction and detection, which would facilitate localised prevention planning. There are existing examples of data-driven tools that utilise either patient^{Reference Baio, Coid, Ding, Dliwayo, French and Jones36} or social determinant^{Reference Oliver, Arribas, Perry, Whiting, Blackman and Krakowski37} data to predict and forecast psychosis cases in clinical and population contexts. Although these digital technologies are not specifically designed to aid selective prevention programmes for FEP, they do provide support for what is achievable in this space through their combination.

Collaborative case identification

An effective preventive strategy will need to consider the mechanisms by which FEP prediction–detection technologies are used to find at-risk cases in the real world. In addressing some of the accessibility issues of ARMS services, selective prevention will need to go beyond the psychiatric care system and reach into the wider social institutions that children, young people and families interact. We therefore feel a localised and coordinated network of institutions across the health and social sector will be best positioned to identify at-risk individuals in the community. Religious; voluntary, community and social enterprise, education and social care services are some of the likely candidates for this network. We also believe there is a role for the NHS, particularly Child and Adolescent Mental Health Services^{Reference Kelleher38} and general practices, because of their specialised or localised focus on child and family health.

In a practical sense, the detection of at-risk cases would involve a whole range of integrated measures across the network of providers. For example, in the health and social care system, a nationally coordinated selective screening programme^{Reference Bobrowska, Murton, Seedat, Visintin, Mackie and Steele39} could be used to proactively invite at-risk individuals for routine mental health screening assessments. Outside of the NHS, voluntary, community and social enterprise organisations and schools in areas of high risk could be trained to spot early cases, leading to supported referral or screening processes.

Multi-layered youth-focused preventative interventions

Existing evidence should be used to decide which preventative interventions are adopted within the selective prevention programme.^{Reference Frieden34} Prevention will also need to be multi-layered, able to intervene on a range of direct and distal psychosocial developmental levels in childhood,^{Reference Bronfenbrenner40} and able to influence key social determinants.^{Reference Kirkbride, Anglin, Colman, Dykxhoorn, Jones and Patalay6,Reference Jester, Thomas, Sturm, Harvey, Keshavan and Davis41}

First, interventions should aim to address the impact of childhood adversity.^{Reference Varchmin, Montag, Treusch, Kaminski and Heinz42,Reference Davies, Segre, Estradé, Radua, De Micheli and Provenzani43} Psychological interventions should be considered because of their effectiveness in targeting the effects of childhood abuse, neglect and victimization. For example, eye movement desensitization and reprocessing has been shown to reduce the symptoms of childhood trauma by adapting negative memory pathways and lessening one's reactivity to traumatic stimuli.^{Reference Moreno-Alcázar, Treen, Valiente-Gómez, Sio-Eroles, Pérez and Amann44,Reference Lewey, Smith, Burcham, Saunders, Elfallal and O'Toole45} Family-focused therapy should also be included, because of its effectiveness in addressing various adolescent mental health difficulties. Furthermore, eye movement desensitisation and reprocessing and family therapy have both been shown to lessen psychotic experiences in clinical and non-clinical populations.^{Reference O'Brien, Miklowitz and Cannon46,Reference Hardy, Keen, van den Berg, Varese, Longden and Ward47}

Second, a preventative strategy should also aim to address the effects of social disconnectedness, such as social fragmentation, social marginalisation and racial discrimination.^{Reference Jester, Thomas, Sturm, Harvey, Keshavan and Davis41,Reference Jongsma, Gayer-Anderson, Tarricone, Velthorst, van der Ven and Quattrone48} Interventions that improve civic engagement should also be considered, including youth-focused social prescribing and educational/vocational participation schemes.^{Reference Kirkbride, Anglin, Colman, Dykxhoorn, Jones and Patalay6} At the neighbourhood level, improving community resources and infrastructure will also be pivotal. Cultural centres, community organisations, outdoor recreational areas and religious organisations are likely to act as protective factors,^{Reference Jester, Thomas, Sturm, Harvey, Keshavan and Davis41} by providing greater community cohesion. Family interventions might lessen youth alienation, by improving family cohesion and connectedness.^{Reference Holt-Lunstad, Robles and Sbarra49}

Finally, strategy should aim to lessen the impact of social economic disadvantage. Some examples might be improving the economic state of families in high-poverty neighbourhoods through direct payment schemes, which have been shown to reduce distress and anxiety in parents and children.^{Reference Holt-Lunstad, Robles and Sbarra49} Neighbourhood regeneration schemes that improve the physical quality of the built environment by planting trees, removing litter and landscaping vacant land should also be included,^{Reference Holt-Lunstad, Robles and Sbarra49} as these initiatives have been shown to lower depressive symptoms and improve self-worth amongst residents.

A placed-based health partnership^{Reference Naylor and Charles50} will be most effective in delivering these interventions. For example, local authorities and public health departments could be responsible for delivering the community and neighbourhood-level components of the preventive strategy, whereas schools and social care organisations could be tasked to facilitate the individual and family level. This collaborative approach to prevention ensures that the most effective interventions are delivered at the right time and by the right provider.

In conclusion, the accessibility of existing preventative strategies for psychosis^{Reference Griffiths, Brown and Kirkbride51} requires us to explore greater diversity in our approach.^{Reference Ajnakina and David11,Reference Kelleher38} What is lacking is the how – the specific strategies that ensure that all communities have equal access to preventative care. We believe that a public health approach employing a selective preventative strategy offers a novel and equitable way to achieve this, by focusing on communities at increased risk in the general population and developing collaboration between the healthcare system and different social organisations. Interventions within such a strategy should be youth-focused and aim to target multiple levels within the life course of the young. Future pilot research is however needed to establish which preventive interventions have the greatest impact in reducing incident rates of psychosis in a population. From this, recommendations for health policy and political commitment can be generated, so that effective interventions can be expanded to the national stage.