Time to move on? Commentary on the early intervention in psychosis debate

The evidence

A recent Cochrane review found there was ‘some support for specialized early intervention services’ (Marshall and Rathbone, 2011). This is a conservative conclusion reflecting the rigorous selectivity adopted by the Cochrane Library and, as pointed out by McGorry (2012), additional evidence excluded by Cochrane deserves not to be ignored. The value of Cochrane reviews is in determining efficacy, not dictating policy. Where evidence for efficacy falls short of Cochrane standards, non-efficacy is not proved, but instead further studies with improved design based on larger samples are generally needed. In fact, both protagonists in the recently published debate agree that, overall, the evidence points towards better outcomes for specialized EIP services in the short term, but that by 5 years these effects have dissipated unless good-quality intervention is continued. Could there be a consensus here that good-quality intervention for psychosis needs to be early and sustained?

Implementation

There is disagreement about EIP services being delivered by specialist versus generic mental health teams and setting an upper age limit. These are secondary issues that concern ‘how’ rather than ‘whether’. Certainly, there are equity considerations in limiting EIP to young people only. Cohesive specialist teams may arguably be better able to maintain the integrity of their programs but they do run the risk of ‘silo’ creation. However, healthcare services have been moving to subspecialty teams for decades; patient transfers and communications between teams are conducted well or poorly depending on management expertise and resourcing. Under-resourced services develop self-protective barriers that impede the smooth flow of patients and information, and hence continuity of care, so part of the solution here is adequate resource provision. In fact, a key determinant of the extent to which EIP services are implemented is the amount of identifiable funding committed specifically to those services (Catts et al., 2010). If other service components are then, by comparison, under-resourced then barriers will naturally arise. Adequate funding and good clinical governance are thus critical in ensuring service quality and maintaining continuity of care, whether the EIP service is specialized or integrated within a generic mental health team.

Intervention for ‘ultra-high risk’

It is unfortunate that this genie has popped out of the bottle and into the draft DSM-5. The central problem is the low predictive value of the clinical criteria for ‘ultra-high risk’ (UHR) status and what is to be done when UHR is identified. There is also no published comparative information on rates of transition to psychosis among mental health presentations that fail to meet the UHR criteria. An unpublished audit conducted in Newcastle found a rate of transition to psychosis of 16% over 10 years in initially non-psychotic patients who were assessed and found not to meet UHR criteria. The onus is on the advocates of pre-psychotic intervention to improve their predictive capacities, perhaps by supplementing the clinical criteria with other measures (e.g. Belger et al., 2012). Interventions (including, but not solely, antipsychotic medication) intended to prevent transition to psychosis in UHR patients are experimental and ought to be conducted only in the context of ethically approved research studies; they have no place in ordinary clinical services. Patients who meet UHR criteria and have diagnosable mental disorders (commonly affective, anxiety and substance use disorders) aside from their UHR status ought to be treated on their merits in ordinary clinical settings regardless of UHR status, which should be regarded as merely incidental. It is critical in these settings that patients not be prescribed antipsychotic medication with the aim of preventing psychosis; this would be leaping too far ahead of the scientific evidence base and has huge potential for adverse consequences. Such a position is consistent with the Australian Clinical Guidelines for Early Psychosis.

EIP and mental health reform

The EIP movement has certainly acted as a catalyst for mental health policy debate and sparked national service developments. There has been criticism over the extent to which reform may appear to be excessively driven by EIP advocates. Coherent mental health policy does require a broader perspective in relation to the psychoses and the recent Australian national survey of psychotic disorders may help to inform policy discussion that extends beyond EIP to encompass the many other challenges to be addressed in policy and service development for the psychoses.

Two peripheral issues in the debate include duration of untreated psychosis (DUP) and cost effectiveness. DUP has too many conceptual and measurement problems for the debate to dwell on this issue, and there remain enormous difficulties in sorting out the causal relationships between reduction in DUP, EIP and outcome. The published cost-effectiveness studies are not yet of sufficiently high quality, and require greater rigour in design and much larger samples. Claims of cost savings that could be redirected to other services are not only premature but do not take into account structural constraints on redirecting funding and certain fixed costs of services.

For over three decades mental health services gave priority to ‘severe mental illness’. In practice, this has meant florid psychosis, suicidality or danger to others. Without at least one of these, acceptance for care and treatment was resisted. Many families complained bitterly about how setting the bar this high worked against the mentally ill receiving timely and effective care. If EIP does nothing else but rid mental health services of this pernicious practice it will have achieved laudable success.