Abstract
Early intervention (EI) in psychosis refers to the timely provision of professional help to an individual with a first episode of psychosis. The rationale behind EI is that much of the deterioration seen in patients with schizophrenia and other psychotic disorders is not due to an inevitable biological ‘process’, but to psychosocial factors that operate in the first few years following onset of psychosis (National Early Psychosis Clinical Guidelines Working Party, 2010). These psychosocial factors include the disruption of education and vocational trajectory, family stress or breakdown, loss of friendships, depression, demoralization, stigma, disruption of personality development, fear of relapse, substance use and homelessness, and post-traumatic stress disorder. With EI, these factors may be preventable, or at least minimized. EI in first-episode psychosis (FEP) aims to reduce the duration of untreated psychosis (DUP), manage FEP people in specialized services, and focus on recovery.
Evidence for the effectiveness of EI
Reduction of DUP
The longer a person remains psychotic in the community without treatment, the greater the likelihood he or she will damage relationships and drop out of work or education. This commonsense notion is backed by evidence of the association between long DUP and poor outcome in terms of symptoms and functioning (Marshall et al., 2005). Some argue that this association is due to a more severe underlying illness being associated with both insidious onset (and therefore long DUP) and poor outcome. That is, that DUP is merely a marker of poor outcome. This is undoubtedly true in some cases. But DUP is also an independent predictor of outcome (Marshall et al., 2005) and, importantly, a malleable one. Thus reduction of DUP is a target that makes sense, and there is now evidence that not only can DUP be reduced with sustained and vigorous attention to it, but that decreasing DUP is associated with improved outcomes up to 10 years post-service entry (McGlashan et al., 2011).
Management of the first episode in specialized services
Specialized services for FEP patients enable an intensive biopsychosocial approach to treatment. These services have been shown to be superior to generic services at 12 and 18 months, 2 years, and up to 5 years post-entry (Craig et al., 2004, Harvey et al., 2007; Bertelsen et al., 2008; Norman et al., 2011). Two of these randomized controlled trials (Craig et al., 2004; Bertelsen et al., 2008) found that some gains were lost after patients were discharged from specialist services; however, outcomes of rates of living independently and days in hospital were still superior in one study (Bertelsen et al., 2008). Furthermore, if specialist treatment can be maintained, even at a lower level of intensity, over 5 years, there is evidence that good outcomes can not only be sustained but can continue to improve, as demonstrated by a recent Canadian study (Norman et al., 2011).
Recovery phase
An intensive biopsychosocial approach to treatment and a focus on maximizing functional recovery is another component of the early intervention model. Even critics of early intervention acknowledge that such comprehensive treatment would be of benefit (Pelosi, 2010). For example, Bosanac et al. (2010) state ‘Specialized treatment plans for early psychosis patients are arguably no different to those that would be considered best practice by multidisciplinary psychiatry teams, and should be provided to all patients’. But the question is: can they be consistently provided in generic services? Castle argues that they can (Castle et al., 2010). As evidence, he cites a paper that purports to show that a generic service can implement recommendations from the Australian Clinical Guidelines for Early Psychosis (ACGEP) (National Early Psychosis Clinical Guidelines Working Party, 2010). In fact, this paper (Petrakis et al., 2011a) shows, for example, that in only two-thirds of cases was contact with family or carers made, and that in four cases this took over a month. Furthermore, a biopsychosocial assessment of patients was completed in only 65.5% of cases, and that in a substantial minority (29.1%) this process took a mean of over 2 and a half months. This is not a criticism of individual clinicians involved in Castle’s mental health service, but rather a illustration that unless resources are specifically directed at management of FEP patients it is difficult for clinicians to prioritise these tasks as they also juggle the needs of other patients, many with chronic illnesses and complex needs. Additionally, without a focus on early detection of FEP patients, many arrive at generic services with long DUPs and often entrenched symptoms. For example at Castle’s service they report a mean DUP of 24 months, which was actually higher than the DUP of 15 months before their implementation of the ACGEP (Petrakis et al., 2011b).
Evidence gaps
Some FEP patients do not respond to EI and develop poor outcomes. Individuals presenting with high levels of negative symptoms are particularly at risk (Simonsen et al., 2010). Early detection of these vulnerable patients is needed and research into optimal care required. The duration of care needed by a specialist team also needs to be determined. Some FEP patients may not need antipsychotics to return to good functioning. Identification of these individuals is also a priority.
Criticisms
Negative long-term trials
The loss of some early gains by EI services once the specialist teams were withdrawn is a major source of criticism of EI (Bosanac et al., 2010; Castle et al., 2010). These negative results are used as evidence that EI cannot change the trajectory of psychotic disorder. Unfortunately in some cases this is likely to be true. But this does not mean that EI is not justified in these people. Attention to minimizing disruption of peer and family networks and improving quality of life is still important. These negative trials showed that no benefit of EI after withdrawal of specialist services in relation to positive and negative symptoms. But symptoms are not the only aspect of outcome that should be considered. Indeed, subjective quality of life has found to be determined not by positive and negative symptoms but by affective symptoms, functioning, and social relations (Melle et al., 2011). Furthermore, even in Alzheimer’s disease, much attention is given to medications that improve quality of life, despite the recognition that the long-term trajectory will not alter. Additionally, it may be that trajectories of a range of outcomes are actually improved by EI, provided good care can be sustained. This has been shown by the 5-year Canadian study (Norman et al., 2011).
Specialist services not needed
As noted above, critics claim that good care can be provided in generic services. However, although this may be so in an ideal world, evidence is lacking at present that this is possible, as described previously. Some criticize the narrow age range of EI services such as EPPIC, which had an upper age limit of 25. However, other specialist services, such as LEO (Craig et al., 2004), include patients up to 40 years. In support of the call to include a wider age range, Petrakis et al. (2011b) note that their service included two FEP patients aged between 55 and 64, but also state how the 55-year-old had an DUP of 40 years, making him 15 at the time of first onset. Early detection may have saved this man several decades of suffering.
The Cochrane review
A Cochrane review of the effectiveness of early intervention for psychosis has recently been published (Marshall and Rathbone, 2011). The notoriously conservative Cochrane review system excluded the DUP study, historical controls trials, and other two other trials (Craig et al., 2004; Grawe et al., 2006) due to some patients (about 20%) being recent onset but not first episode of psychosis. And yet the review concluded that there was some support for specialized early intervention services and for employment programs and family therapy.
Disruption of continuity of care
Another criticism of EI services is that they disrupt continuity of care as patients are discharged once their tenure of treatment (usually 18 months to 2 years) is finished. This is true, and ideally EI services would routinely manage patients for 5 years or longer. Sadly, budgetary restrictions have not enabled this. Unfortunately resourcing problems also mean that many generic services are unable to provide continuity of care, with staff turnover and pressure to keep case load down by discharge to community options once acute symptoms are reduced. This hampers ability to provide psychosocial management of the recovery phase.
Resources being diverted from other programs
Critics assert (Pelosi, 2010) that support for EI services will lead to reduced funding for people with more chronic illness. However, there is evidence that EI services may actually be associated with a reduction in health costs (Mihalopoulos et al., 1999). Money can be saved through EI that can be freed up to better support later stage care.
Sources of confusion
A persistent source of confusion is that EI teams aim to prevent psychotic disorder through treatment of non-psychotic individuals who may be at risk of developing psychotic disorder. This mixes up the ultra high risk (UHR) area (Yung, 2011) with management of established psychotic illness. Allegations that UHR clinics routinely treat these non-psychotic individuals with antipsychotics are untrue. UHR services do exist, and early trials of low-dose antipsychotics were crucial in determining that other treatments such as CBT, supportive therapy, and fish oil were just as effective as antipsychotics. These trials led to the recommendation in the ACGEP against antipsychotic use in this group. For those UHR patients who do develop FEP, DUP is minimal and therapeutic engagement has already occurred.
Conclusion
Debate is needed in the field of EI for psychosis. It ensures that research continues into this important area and that evidence gaps are acknowledged. But critics must also recognize that evidence is emerging for the efficacy and cost effectiveness of EI for psychosis. No such evidence exists for late intervention.