La schizophrénie est une maladie chronique dont le pronostic n’a que peu évolué durant les dernières années. La durée de psychose non traitée (DPNT) fait partie des variables les plus étudiées actuellement dans les premiers épisodes psychotiques, ce dans le but de déterminer l’effet d’une intervention précoce sur le pronostic de la schizophrénie. La DPNT définit la durée entre le début de la phase psychotique et l’initiation du traitement. Celle-ci peut être plus ou moins longue en fonction de plusieurs facteurs. Les méthodes d’évaluation de la DPNT sont multiples et leur fiabilité reste difficile à déterminer. Jusqu’à présent, aucune méthode d’évaluation de la DPNT (entretien clinique compris) n’a prouvé sa supériorité par rapport aux autres méthodes. Quel que soit l’outil d’évaluation employé, dans la schizophrénie, une DPNT plus longue est associée à un plus mauvais pronostic, à court et à long termes et ce dans plusieurs domaines. Le concept de « durée de maladie non traitée » (DMNT), période associant la phase prodromique et la DPNT, a émergé. Certaines interventions auprès de sujets « à ultra-haut-risque » de transition psychotique (c’est-à-dire durant la période « prodromique ») suggèrent des pistes prometteuses pour réduire le risque de transition, diminuer la sévérité de la maladie et plus généralement améliorer le fonctionnement des individus. La poursuite des efforts déployés dans ce domaine est nécessaire afin de comprendre la physiopathologie de la maladie et d’améliorer son pronostic.

Prognosis of schizophrenia has not significantly improved despite extensive research. There is often a relatively long delay between onset of symptoms and treatment initiation. Lately, duration of untreated psychosis (DUP), the time between the onset of psychosis and initiation of treatment, has been one of the most studied variables in patients presenting for a first psychotic episode in order to evaluate the impact of early intervention on the prognosis of schizophrenia. In the literature, a variety of criteria have been used to define both transition to psychosis and initiation of treatment. Furthermore, the dating of both of these variables is usually retrospective, further complicating the measurement of DUP.

We conducted a comprehensive review about DUP using Pubmed and Google Scholar databases up to January 2015 using the following keywords “schizophrenia”, “duration of untreated psychosis”, “duration of untreated illness” and “early intervention”. Papers were included if they were published in French or English.

The mean DUP was found to be 2 years but it can vary according to multiple factors such as denial of illness by the patient and family, withdrawal and isolation from friends and relatives, diagnostic errors, paranoid views of the mental health treatment systems, or negative symptoms. Long DUP may also be a correlate of poor premorbid functioning or of an insidiously unfolding psychosis. Considerable discrepancies exist in the way that DUP is estimated in different studies. Although the clinical interview remains the most common way of measuring DUP, so far there is no evidence for favoring one method over another. Regardless of measurement method, a longer DUP is found to be associated with poorer outcome in schizophrenia in both the short and long-term across a number of domains: symptoms severity, remission rates, the risk of relapse, global functioning and quality of life. Its role in functional outcome appears to be mediated largely by negative symptoms, for which there is still no effective treatment. A recent meta-analysis has shown that shorter DUP is associated with less severe negative symptoms at short and long-term follow-up, especially when DUP is shorter than 9 months. The mechanism of the relationship between DUP and outcome is still undefined. A hypothesis is that the shorter the DUP, the more likely the intervention is being applied during the period in which neurobiological deficit processes in schizophrenia are most active.

A study of the duration of untreated illness (DUI), which is defined as the DUP and the prodromal phase, seems necessary because results of studies evaluating the effect of early detection and intervention in individuals with clinical high risk for psychosis are promising. A number of interventions such as omega 3 fatty acids and integrated psychosocial interventions seem to delay transition in the at-risk population. However, replication studies are lacking, and a great proportion of at high-risk individuals will spontaneously remit or develop diseases other than chronic psychosis, making us question the advantages and disadvantages of a treatment. Taking into consideration the high prevalence of comorbidities in individuals referred for clinical high-risk state and their effect on the individual's functioning, future interventions in the field need to address not only the preventative efficacy on psychosis transition but also their effectiveness in improving the functioning of this population and their effect on the outcome of schizophrenia when transition to psychosis has occurred.

Despite the huge advances in the field of schizophrenia, many questions remain unanswered and huge efforts are still necessary to understand the pathophysiology of this illness in order to improve its outcome.