Is it time to rethink treatment of early psychosis?

Catts and O’Toole (2016) have thrown down the gauntlet, challenging our profession to alter our approach to people with first episode schizophrenia. They have made several controversial recommendations, often contrasting with the recently published Royal Australian and New Zealand College of Psychiatrists (RANZCP) clinical practice guidelines (CPG) for the management of schizophrenia and related disorders (Galletly et al., 2016).

Their recommendations come in the context of a widening mortality gap among people with schizophrenia. Data from Western Australia show that mortality gap between men with schizophrenia and the general population has widened from 14.5 years in 1985 to 16.4 years in 2005 (Lawrence et al., 2013).

It has been over 10 years since Gavin Andrew’s Tolkien II challenged us to shift to evidence-based care, reminding us that evidence-based care would cost no more than treatment as usual but would increase health gain by two-thirds (Andrews et al., 2003). However, Catts and O’Toole’s article suggests that resistance remains to what actually is evidence-based best-practice for early psychosis.

Catts and O’Toole pose several provocative questions. First, they ask whether schizophrenia is a progressive disease. Despite increasing evidence that people can and do recover from schizophrenia, there are a subset of the consumers we serve who have a deteriorating course of illness. Catts and O’Toole argue that relapse is a mediating factor, citing multiple cohort studies showing a temporal relationship between successive relapses and step-wise deterioration in function, and suggests that this can increase the risk of developing treatment-resistant schizophrenia. They note, however, that it is unclear whether this deterioration would have occurred irrespective of relapse given the dearth of long-term studies of first episode schizophrenia with high treatment adherence rates.

Second, they challenge the current paradigm of delaying early diagnosis of schizophrenia. Their concern is that delayed diagnosis may lead to a less assertive approach to medication adherence and relapse prevention. In contrast to their argument and given (a) the diagnostic instability in people with early psychosis and (b) the increasing availability of intensive Early Psychosis services that can provide monitoring and education, this concern may be unfounded in high resource settings. In areas without access to specialised early psychosis services, a more aggressive treatment approach may be warranted, with the caveat that diagnoses need to be reviewed to prevent unnecessary lifelong anti-psychotic treatment.

Third, Catts and O’Toole argue that cortical grey matter loss is related to schizophrenia itself and that anti-psychotic medications may protect against, rather than hasten this process. Catts and O’Toole argue that cessation of anti-psychotics can increase the risk of relapse, with associated increased risk of suicide and decline in psychosocial function. This contrasts with the RANZCP CPG suggestion of consideration of cessation of medication for people with at least 12 months of full recovery. Catts and O’Toole insufficiently highlight the cardio-metabolic adverse drug reactions associated with anti-psychotic medications and the poor management of cardio-metabolic illness among people with schizophrenia in clinical practice. In the 2010 Australian Survey of High Impact Psychosis, over half of the participants had metabolic syndrome (Galletly et al., 2012). Among the one in five with diabetes, less than half were taking appropriate medications. Given these risks, the RANZCP guidelines around cessation are appropriate, assuming they are done so in an environment of intensive monitoring to recommence treatment if there are early signs of relapse.

Fourth, they promote the role of long-acting injectable anti-psychotics (LAIs). They argue that RCTs of LAIs versus oral medications are conducted in an artificial environment, and that population studies give a more accurate reflection of the benefit of LAIs. They recommend commencing an LAI during the first hospital admission and then continuing it indefinitely. Although this line of reasoning may be ‘logical’, it neglects the importance of consumer choice, self-determination and shared decision-making (Galletly et al., 2016). Extinguishing symptoms is not the only, and not necessarily the most important, facet of a consumer’s recovery journey.

Finally, they propose a new algorithm for treating first episode psychosis. They argue that this should start with an exhaustive screen to exclude other medical causes of psychosis, early diagnosis and commencement of treatment, individual and family psycho-education, commencement of LAIs within 3 months or commencement of clozapine if treatment refractory, with long-term follow-up in a Chronic Care Model. Many elements of this model mesh with the RANZCP CPG (Galletly et al., 2016), notably physical assessment and screening, and individual and family psycho-education. The key area of divergence is their recommendation for early commencement and long-term use of LAIs.

Challenges to the dominant zeitgeist of early psychosis management, such as this one presented by Catts and O’Toole, are an essential part of the process of honing services forthe consumers we serve.

See Key Review by Catts and O’Toole, 50: 1128–1138.