Swaran Singh has recently argued for a care pathway for psychosis or schizophrenia. ^{Reference Singh1} We have recently argued for a staging approach to schizophrenia. ^{Reference Agius, Goh, Ulhaq and McGorry2} Such an approach argues that there are different stages in the development of schizophrenia, and that therefore different stages of the illness will require different interventions to optimise treatment, be it pharmaceutical, social or psychological. Furthermore, logically, the different stages will require different goals of treatment and different expected outcome measures. Thus, for example, the aim of treatment in the first or ‘at risk mental state’ stage of psychosis is to prevent psychosis developing, while the aim of the second stage, or the first-episode stage, is to end the psychotic episode and return the patient to work and education.

Staging in schizophrenia also extends to the phase of chronic illness, and here the goal will be, depending on the severity of the illness, to limit the positive and negative symptoms of the illness, to prevent relapse, and to optimise social inclusion, promoting a return to work if possible. Such a staging approach to schizophrenia is underpinned by the neuroimaging evidence, since the loss of grey matter linked with schizophrenia does start in the prodromal ‘at risk’ phase, becomes more prominent in the first episode, and then becomes incrementally more severe in the later stages of the disease. ^{Reference Meisenzahl, Koutsouleris, Gaser, Bottlender, Schmitt and McGuire3–Reference Pantelis, Yücel, Wood, Velakoulis, Sun and Berger5} Furthermore, different stages of the illness appear to be mirrored in different patterns of change in such structures as the hippocampus and the amygdala, ^{Reference Velakoulis, Wood, Wong, McGorry, Yung and Phillips6} as well as changes in pituitary volume. ^{Reference Pariante, Vassilopoulou, Velakoulis, Phillips, Soulsby and Wood7,Reference Garner, Pariante, Wood, Velakoulis, Phillips and Soulsby8} Thus, a ‘staging approach’ to schizophrenia does provide a logical framework for the development of a care pathway for schizophrenia, with different stages or phases requiring the development of specialised teams with different expected outcomes, but who will always, in each phase of the illness, strive to optimise treatment in order to achieve the best results. Hence, such a pathway may include an ‘at risk mental health’ team, which will attempt to reduce the rate of transition to full psychosis in patients who are developing ‘prodromal’ symptoms. This would be followed in the pathway by a first-episode service which will work assertively with patients so as to deal with the first episode and return patients to work and education, and at the other end of the spectrum, assertive outreach teams will work with patients who are difficult to treat who have demonstrated the most serious deterioration in functioning.

What, however, is missing in this care pathway is the treatment of those patients who are returned to community mental health teams (CMHTs) after 3 years in an early intervention service and who are not deemed ill enough to require referral to the assertive outreach teams. These constitute the majority of patients with long-term schizophrenia. Unfortunately, since CMHTs have other priorities, and indeed are oriented to dealing with patients with relatively less severe forms of mental illness, many of these patients may receive suboptimal care, sometimes consisting of the simple delivery of medication within a depot or clozapine clinic, and without the systematic delivery of psychosocial interventions. As a result, in many cases, social inclusion is not optimised as a direct result of the loss of the assertive approach to care. It is therefore small wonder that both the Lambeth Early Onset (LEO) ^{Reference Gafoor, Nitsch, McCrone, Craig, Garety and Power9} and the OPUS ^{Reference Bertelsen, Jeppesen, Petersen, Thorup, Øhlenschlæger and le Quach10} services report a loss of improvement in outcomes within 5 years of first treatment, after patients have been transferred from early intervention teams to the care of CMHTs.

It is of interest that a study in Russia, ^{Reference Zaytseva11} where patients were followed up assertively for 5 years, has shown no such loss of improvement in outcomes. It is urgent that the development of ongoing assertive, specialised teams for psychosis, as suggested by Singh, should proceed in order to complete the schizophrenia care pathway. The CMHT cannot provide such an assertive service, since it is focused on other things. Seen in this perspective, recent suggestions that early intervention and assertive outreach teams should be amalgamated into CMHTs and provide elements of specialised care within the CMHTs must further confuse the focus of the CMHTs and constitute a serious misreading of the evidence.