Early intervention services are effective and must be defended

Three world leading researchers in the field have written a comprehensive review1 of the current evidence for improving outcomes in first‐episode psychosis. As they point out, based on the Global Burden of Disease study, there are currently 23 million people worldwide living with schizophrenia. Half of these people are untreated, and the majority of the other half are likely to receive suboptimal treatment. Schizophrenia is ranked as number 12 among conditions leading to years lost to disability. This summary of the current evidence was highly needed, and the obvious next step could be a set of recommendations launched by World Health Organization and endorsed by health ministers all over the world. The main focus of our commentary is to identify which findings call for immediate implementation.

Interventions to shorten duration of untreated psychosis should certainly be implemented. Partly due to lack of agreement on operationalized criteria, this duration varies widely among studies, but it has consistently been found to be a predictor of short‐ and long‐term outcome after a psychotic disorder. The effect of duration of untreated psychosis on outcome could both be biological (permanent changes in brain function) and of a psychosocial nature (as the disintegration of the patient's social network prior to treatment could have a long‐term effect).

A short duration of untreated psychosis means that treatment is provided within the early course of illness. Patients suffer from severe social and clinical consequences of absent or insufficient treatment in early phases of psychosis, and therefore can be especially receptive to interventions in these early years. If the now established specialized early intervention teams want to exert their maximal effects, widespread interventions to reduce the duration of untreated psychosis should be implemented. The Treatment and Intervention in Psychosis (TIPS) study showed how society level interventions could reduce that duration and affect the long‐term outcome2, and we are awaiting the results of a German study testing the effect of society level awareness campaigns combined with specialized early intervention teams3.

Fusar‐Poli et al1 conclude that there is sufficient evidence to recommend that specialized early intervention programs be implemented. It is beyond doubt that the current huge variation in implementation worldwide cannot be justified by lack of evidence. Even in high‐income countries with large health budgets there are large variations, spanning from almost complete nationwide coverage in Denmark and England to almost no services available in many other European countries. These differences are likely due to local traditions more than scientific evidence.

Health economic analyses of specialized early intervention find it to be cost‐effective on the long term4, but this intervention requires more resources “up front” and this leads to its achievements always having to be defended from cuts in funding by politicians and health administrators. This is why it is imperative to have high‐quality research proving its efficacy.

Further, we need to protect the specialized early intervention teams from “drifting back to ordinariness”. This drift could be due to political pressure regarding measurable productivity goals, but also the sentiment of the team members. It is therefore important to engage the clinical staff in an ongoing debate both with researchers and among themselves. To ensure that the treatment provided still lives up to the standards originally tested, it is central to develop fidelity measures. Fidelity scales work as a safety mechanism and should also be viewed as a tool to empower the clinical staff in defending the treatment from cuts in funding and inauspicious reorganizations.

The duration of specialized early intervention programs is an important issue. The Early Assessment Service for Young People with Early Psychosis (EASY) trial, comparing 3 vs. 2 years of specialized early intervention, has recently published 5‐year follow‐up data and, while there was initial evidence that the prolonged intervention had effect on negative and depressive symptoms, these gains were, as in prior trials, lost when the intervention was terminated5.

We have recently published data from our trial (OPUS II) comparing 2 vs. 5 years of specialized early intervention6, and there are further ongoing trials testing similar prolonged interventions7, 8. In the first OPUS trial (OPUS I), we found that participants randomized to the intervention group relapsed when transferred to standard treatment after 2‐year OPUS intervention. We therefore anticipated that this relapse could be prevented by prolonging the intervention in the second trial (OPUS II). Surprisingly, we found no sign of relapse neither in the control nor in the intervention group, and the overall pattern in both groups was that participants improved over time on most functional, psychopathological and cognitive outcomes. The prolonged intervention did not improve further on this already positive trajectory, except for better user satisfaction and working alliance. The most likely explanation for this finding is improvements in the treatment‐as‐usual arm. Participants randomized to this arm were in most cases referred to community health centres after termination of their 2‐year OPUS treatment, and 20% of them received assertive community treatment. We therefore concluded that the early gains seen in treatment of first‐episode psychosis by specialized early intervention teams can be upheld either by prolonging the treatment or by providing high‐resource standard care with the possibility of assertive treatment for the most debilitated patients.

Regarding the duration of antipsychotic medication for remitted patients, Fusar‐Poli et al1 mention that treatment reduction may be an option for first‐episode psychosis patients who have achieved clinical remission and are not at high risk of relapse. Recent studies based on long‐term results from the OPUS trial document that such patients exist, and that a proportion of remitted patients with schizophrenia and schizophrenia‐like psychosis will discontinue antipsychotic medication with or without doctors being involved in their decision9, 10. In order to protect patients from side effects, further studies are necessary focusing on the identification of the subgroup of patients who can stay in remission without antipsychotic medication.

Finally, we want to point out that the early intervention services now provided in many countries have been part of a very significant change in the view of mental disorders and mental health care. The evidence supporting the introduction of these services is strong and should lead to even more widespread implementation. What is needed is a long‐term, high‐resource commitment by policy makers to develop and uphold the positive gains obtained. It is time to rally behind the banners, protect the ground already covered, and push forward.

Merete Nordentoft, Nikolai Albert
Mental Health Centre Copenhagen, Copenhagen University Hospital, Copenhagen, Denmark