At-Risk Mental State (ARMS) servicesReference Yung, McGorry, McFarlane, Jackson, Patton and Rakkar1 continue to remain the most established approach to preventing psychosis in the Western world.Reference Kotlicka-Antczak, Podgórski, Oliver, Maric, Valmaggia and Fusar-Poli2 According to the definition set out by the World Health Organization3 (Table 1), ARMS services are classified as indicative primary prevention,Reference Fusar-Poli, Correll, Arango, Berk, Patel and Ioannidis7 as they aim to stop the onset of first-episode psychosis (FEP) in those with specific clinical or genetic risk markers, also known as the ultra-high-risk criteria. In England, UK, ARMS services are expanding, with all regional Early Intervention in Psychosis teams expected to deliver psychosis prevention to 14- to 35-year-olds.8,9 Despite such advancements, there has been little debate about the suitability of ARMS services for all FEP patients.Reference Murray, David and Ajnakina10,Reference Ajnakina and David11
Table adapted from Kirkbride et alReference Kirkbride, Anglin, Colman, Dykxhoorn, Jones and Patalay6 and Fusar-Poli et al.Reference Fusar-Poli, Correll, Arango, Berk, Patel and Ioannidis7
The challenges of ARMS prevention
There are two main reasons why ARMS services are criticised for being the sole approach to prevent psychosis. First, there remains a lack of clarity about the proportion of patients who benefit from ARMS clinics. It is estimated that about a third of patients experience no ARMS symptoms before the onset of psychosis, and so would be ineligible to access ARMS care even if they were to seek help during the prodromal phase.Reference Shah, Crawford, Mustafa, Iyer, Joober and Malla12 Furthermore, transition rates from ARMS services to FEP is low (8–17%),Reference Morrison, French, Stewart, Birchwood, Fowler and Gumley13–Reference Malla, de Bonneville, Shah, Jordan, Pruessner and Faridi16 and so it remains unclear how sensitive the ARMS criteria is to those who truly are at risk. The second critique of ARMS services is their constrained appeal. Only a small proportion (4.1%) of patients presenting to psychiatric care with a diagnosis of FEP come via ARMS services. ARMS services are most likely to be accessed by patients who voluntarily seek help from the healthcare system,Reference Ajnakina, Morgan, Gayer-Anderson, Oduola, Bourque and Bramley17 and are disproportionately underused by individuals from Black ethnic backgrounds (African, Caribbean and British)Reference Byrne, Codjoe, Morgan, Stahl, Day and Fearon18 despite this group being at increased psychosis risk.Reference Kirkbride, Errazuriz, Croudace, Morgan, Jackson and Boydell19,Reference Jongsma, Turner, Kirkbride and Jones20 This may be caused by cultural differences in help-seeking preferences;Reference Ajnakina, Morgan, Gayer-Anderson, Oduola, Bourque and Bramley17,Reference Byrne, Codjoe, Morgan, Stahl, Day and Fearon18,Reference Morrison, Stewart, French, Bentall, Birchwood and Byrne21–Reference Kirkbride, Stochl, Zimbrón, Crane, Metastasio and Aguilar23 alternative beliefs about the causes of psychosis;Reference Anderson, Fuhrer and Malla24,Reference Halvorsrud, Nazroo, Otis, Brown Hajdukova and Bhui25 or the result of a more acute form of psychosis onset, resulting in urgent, involuntary and coercive psychiatric treatment.Reference Burnett, Mallett, Bhugra, Hutchinson, Der and Leff26,Reference Singh, Brown, Winsper, Gajwani, Islam and Jasani27 Collectively, these points raise questions about the accessibility and sensitivity of the ARMS preventative model, which is further concerning given the National Health Service's (NHS) commitment to preventative healthcare and reducing health inequalities.28
Public health approaches to psychosis prevention
Although there is greater recognition of the need for complementary approaches to ARMS services,Reference Murray, David and Ajnakina10,Reference Ajnakina and David11,Reference Jongsma and Kirkbride29–Reference Anderson31 rare is there a discussion about how this can be achieved.Reference Gordon RS4 Selective and universal public health preventative approaches (Table 1) have the potential to overcome the limitations of the ARMS model, as preventative care is directly targeted at the general population, in what is referred to as ‘upstream’ working.Reference Williams, Costa, Odunlami and Mohammed32 These approaches are likely to be more accessible and have a wider reach, as they exist outside of the boundaries of the psychiatric care system. They are also more likely to be acceptable and therefore more appealing, as they offer care in less stigmatising, less coercive and more culturally attuned settings.
One of the overarching mechanisms by which selective or universal prevention could act to stop psychosis transition is by addressing the social factors that predispose healthy individuals to psychosis, known as social determinants. These determinants act at the individual, neighbourhood and environmental levels, comprising of factors like socioeconomic disadvantage, childhood adversity and trauma, migration, discrimination, neighbourhood socioeconomic disadvantage, social capital, social fragmentation, ethnic density and cannabis use.Reference Kirkbride, Anglin, Colman, Dykxhoorn, Jones and Patalay6,Reference Murray, David and Ajnakina10,Reference Schäfer and Fisher33 Public health interventions are effective in acting on the social determinants of psychosis.Reference Kirkbride, Anglin, Colman, Dykxhoorn, Jones and Patalay6 Despite this, there continues to remain a lack of evidence demonstrating the direct effect of public health interventions in reducing future psychosis incident rates in the real world, and no clear agreement about a model of service delivery.
Future considerations
According to Frieden'sReference Frieden34 Six Components Model, innovation is central to the effective design and implementation of any public health programme. Building on this premise, we outline our considerations for building a public health preventative strategy for FEP.
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 patientReference Baio, Coid, Ding, Dliwayo, French and Jones36 or social determinantReference 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 ServicesReference 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 programmeReference 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 partnershipReference 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 psychosisReference 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.
About the authors
Luke Brown, BSc, MRes, PhD, ClinPsyD, is an assistant professor at the Centre for Applied Psychology, School of Psychology, University of Birmingham, Birmingham, UK; assistant professor at the Institute of Mental Health, School of Psychology, University of Birmingham, Birmingham, UK; and a principal clinical psychologist with the Early Intervention in Psychosis Service, Black Country Partnership NHS Foundation Trust, Dudley, UK. Siân Lowri Griffiths, BSc, MSc, PhD, is an assistant professor at the Institute of Mental Health, School of Psychology, University of Birmingham, Birmingham, UK.
Data availability
Data availability is not applicable to this article as no new data were created or analysed.
Author contributions
The ideas within this commentary were jointly developed by L.B. and S.L.G. Both authors contributed to the original and final versions of the manuscript.
Funding
This research received no specific grant from any funding agency, commercial or not-for-profit sector.
Declaration of interest
None.
eLetters
No eLetters have been published for this article.