At-Risk Mental State Services for Psychosis Should Not Be Delivered in Secondary Mental Health Services

Abstract

Traditionally, preventive approaches in mental health have lagged behind somatic medicine (Arango 2019); however, early detection and subsequent treatment for individuals who present with an At-Risk Mental State (ARMS) for psychosis are now firmly established worldwide (Yung et al. 1996; Murray et al. 2021; Fusar-Poli et al. 2021). Our understanding of ARMS (also referred to as prodromal; clinical high-risk; or ultra-high risk) has expanded significantly over the past two decades to capture a pre-psychotic construct (Oliver et al. 2024). ARMS individuals have a substantially greater risk of developing psychosis compared with the general population (Tandon et al. 2008; Dragioti et al. 2023), with research consistently demonstrating that individuals have a 22% probability of developing emerging psychotic disorders (Webb et al. 2015; Fusar-Poli et al. 2016, 2017) over a relatively short period of 2 years (Kotlicka-Antczak et al. 2020).

Successful delivery of clinical services relies on efficient detection (WHO 2004; Oliver et al. 2020), accurate prognosis of outcome (Fusar-Poli et al. 2015; Bonnett et al. 2019; Salazar de Pablo et al. 2022) and effective preventive treatments (Bosnjak-Kuharic et al. 2019; Devoe et al. 2020; Catalan et al. 2021), which have all led to current national and international treatment guidelines for this population (NICE 2014; NHS England 2023; Schmidt et al. 2015) with substantial advancement of ARMS clinical services over the past 10 years.

Unfortunately, research into specific care pathways for ARMS individuals is sparse and neglected (Allan et al. 2021) with ARMS identification being complicated and under-recognised (Strelchuk et al. 2021). Maric et al. (2018) found an unequal and inconsistent development of Early Intervention in Psychosis (EIP) services throughout Europe. Whilst Australia (pioneers of early detection services) has chosen to adopt a broader front-line context of providing ARMS provision within youth stepped-care mental health services (McGorry et al. 2007; Nieman and McGorry 2015).

In England, frontline clinical services have received additional funding through transformation monies (NHS 2019) to help implement ARMS services. Some major cities across England have dedicated ‘stand-alone’ ARMS services, but for most, detection, prognosis, and treatment now fall within the remit of EIP services in secondary care (NICE 2016). This care pathway should involve the following steps: an initial referral to triage services or primary care liaison services (PCLS); a triaged referral to EIP services (if psychosis is indicated) and an assessment by EIP services within two weeks (NICE 2016).

NHS England (2023) published updated guidelines which set out key aims for ARMS provision. They restate NICE (2014), (2016) guidance for individuals experiencing ARMS, including two-week waiting times; specialist assessment by either a consultant psychiatrist or trained specialist with experience in ARMS; individual CBT with or without family intervention; and interventions for co-existing mental health problems. NHS England recommends a period of active treatment lasting up to 2 years (with a further year of monitoring). Monitoring should include a structured and validated assessment tool such as the CAARMS (comprehensive assessment of ARMS), offered to individuals aged 14–35 years and stipulates that individuals should not be treated with antipsychotic medications (NHS England 2023).

Whilst at first glance the NICE (2014) CG178 guidelines appear to be concise, when scrutinised these recommendations contain a degree of ambiguity and controversy. For example, ‘behaviours suggestive of possible psychosis’ is vague, open to interpretation and has no additional assessment guidance for clinicians. Equally, having a first degree relative with psychosis may slightly increase an individual's predisposition or vulnerability towards developing psychosis, but this concept lacks validity (see Boyle and Johnstone's (2020) critique of biomarkers in the context of psychiatric research).

Two options for the delivery of ARMS provision are offered by NHS England (2023). Stand-alone or integrated, with commissioners and service leads jointly working to identify the most suitable model for their local population. The Stand-alone model has a designated team that receives referrals from other mental health services, is multi-disciplinary (i.e., can provide all recommended ARMS care) and has strong links with EIP services. The integrated model either has EIP staff with dedicated ARMS caseloads (with shared EIP management and resources) or EIP staff with mixed caseloads of individuals across EIP and ARMS care pathways (also with shared EIP management and resources).

NHS England (2023) provide some benefits to each model although it is not clear whether this is anecdotal or empirical observations. Furthermore, neither model seems to consider whether to position ARMS services within the primary or secondary mental health settings. Once again, there are vague and ambiguous statements about stand-alone models having the benefit of ‘reducing potential iatrogenic harms’ or ‘mechanisms should be in place to monitor access rates for all people entering the service, as well the quality of the care and support provided’ (NHS England 2023 36), although the guidance falls short of explaining what the potential iatrogenic harm might be, how impactful they are or which specific mechanisms are needed to monitor access rates.

Furthermore, given that at least two-thirds of individuals who meet the at-risk criteria will never transition to full psychosis (Fusar-Poli et al. 2013) there is a legitimate argument to consider the moral and ethical implications of preventing psychosis weighed against the possible over-pathologising of normal human behaviours and odd, atypical presentations. Indeed, concerns continue to be raised that many ARMS individuals might be receiving unnecessary treatment, and that the NHS may be using costly interventions (such as Cognitive Behavioural Therapy, CBT) for people who may not need it (Bonnett et al. 2019; Frances 2013).

Concepts such as ‘prodromal psychosis’ and ‘attenuated psychosis syndrome’ have led to intense debate and some controversy (Hutton and Taylor 2014), and whilst considered for inclusion in the Diagnostic and Statistical Manual 5th Edition (DSM-V APA 2013), ultimately neither were included (Tsuanga et al. 2013). Frances (2013) 222 went further by suggesting that ‘new psychiatric diagnoses are now potentially more dangerous than new psychiatric drugs’.

Malhi et al. (2021) suggest that ARMS is a problematic paradigm in that its symptoms are transdiagnostic and lack a pathophysiological basis, making diagnosis extremely difficult, whilst offering treatments which predominantly focus on symptom reduction rather than addressing the underlying cause. Indeed, traditional medical diagnosis of ARMS could lead to unnecessary and harmful use of powerful antipsychotic medications (Stain et al. 2019; Fusar-Poli and Yung 2012) and require huge service reconfiguration (within secondary care) to create ARMS care pathways set up to offer CBT interventions by trained, accredited and supervised CBT therapists. This will add additional strain on an already stretched NHS (Bonnett et al. 2019) and vastly under-resourced secondary mental health services.

These pressures of service configuration, alongside specialist training requirements for staff (such as the CAARMS or accredited CBT training) and the requirement to offer individuals a minimum of 2 years and a maximum of 3 years involvement with secondary mental health services can feel potentially overwhelming for secondary mental health services to take on and successfully deliver. It even raises uncomfortable questions of whether ARMS provision, in its current form, is clinically and ethically reasonable and whether it is potentially financially unsustainable to the NHS?

Strelchuk et al. (2023) suggest that clinical guidelines for ARMS populations are not being met. They found most ARMS services in England are delivered within secondary mental health settings despite high acceptance thresholds and scarce treatment availability. There is also evidence to suggest that an ARMS diagnosis elicits stigmatising responses among the public as well as self-stigma from individuals themselves (Colizzi et al. 2020); thus, negatively impacting wellbeing, engagement with services and may in fact exacerbate any emerging symptomology (Malhi et al. 2021).

Nieman and McGorry (2015) 831 suggest ARMS treatment be delivered in a stepped-care model, to promote help seeking and tackle stigma, primarily through psychoeducation. They recommend low-stigma settings and a ‘clinical staging approach…supported by substantial health system reform and investment’. Surprisingly, there does not appear to be much debate about the impact of positioning ARMS services with secondary mental health services. The authors propose the need for a discussion that optimises benefits and minimises harm for young people- given the prognosis rates, diagnostic uncertainty, ineffective treatments (CBT) and stigma of bringing an ARMS population into secondary mental health services.

Whilst early detection and intervention represent an opportunity to prevent the onset of psychosis for a minority, the majority (at least two-thirds) of individuals who meet the ARMS criteria will not develop psychosis and never will have. Further research and consideration are required into why and how commissioners or service leads identify the most appropriate model for their local population and whether these decisions are clinically or operationally led.

We do understand and support the need for early detection and support for individuals at risk of psychosis; however, we feel ARMS as a preventative approach is much more aligned to naturally sit in primary care, with third sector and voluntary providers such as Rethink or Turning Point offering alternative service settings that are less stigmatising for young people to access (Baxter and Fancourt 2020). At the very least, it is clear that existing NICE (2014) ARMS guidelines need reviewing and updating, with the voices of people who have received ARMS support within secondary care services taking centre stage.

The authors declare no conflicts of interest.