Rate and predictors of service disengagement in an epidemiological first-episode psychosis cohort

Abstract

Objectives

To assess the prevalence and predictors of service disengagement in a treated epidemiological cohort of first-episode psychosis (FEP) patients.

Methods

The Early Psychosis Prevention and Intervention Centre (EPPIC) in Australia admitted 786 FEP patients from January 1998 to December 2000. Treatment at EPPIC is scheduled for 18 months. Data were collected from patients' files using a standardized questionnaire. Seven hundred four files were available; 44 were excluded, because of a non-psychotic diagnosis at endpoint (n = 43) or missing data on service disengagement (n = 1). Rate of service disengagement was the outcome of interest, as well as pre-treatment, baseline, and treatment predictors of service disengagement, which were examined via Cox proportional hazards models.

Results

154 patients (23.3%) disengaged from service. A past forensic history (Hazard ratio [HR] = 1.69; 95%CI 1.17–2.45), lower severity of illness at baseline (HR = 0.59; 95%CI 0.48–0.72), living without family at discharge (HR = 1.75; 95%CI 1.22–2.50) and persistence of substance use disorder during treatment (HR = 2.30; 95%CI 1.45–3.66) were significant predictors of disengagement from service.

Conclusions

While engagement strategies are a core element in the treatment of first-episode psychosis, particular attention should be paid to these factors associated with disengagement. Involvement of the family in the treatment process, and focusing on reduction of substance use, need to be pursued in early intervention services.

Introduction

The early phase of psychotic disorders has received increasing attention over the last 20 years, and the efficacy of specialised early intervention programs has now been established through randomised controlled trials (Garety et al., 2006, Petersen et al., 2005). One of the main aims of such programs has been to reduce the duration of untreated psychosis (DUP) (Melle et al., 2004, Marshall et al., 2005, Schimmelmann et al., 2008, Polari et al., 2009) in order to provide treatment at an earlier stage of the illness, considering findings of a link between longer DUP and poorer outcome (Marshall et al., 2005). However, if early identification of patients is an essential element to early intervention, the benefits of these efforts are likely to be mitigated by various factors, among which poor adherence to treatment and disengagement from services are particularly critical. Previous studies have shown that rates of disengagement from outpatient psychiatric services range from 17% to 60% (Garety et al., 2006) and that this prevalence is also relatively high in programs specialised in early intervention in psychosis, ranging for the majority of them between 18% and 25% (Turner et al., 2007, Malla and Norman, 2001, Craig et al., 2004) after 12 months of treatment. Petersen et al. (2005) reported a remarkably lower disengagement rate (of 7% over 2 years) in the OPUS project, and showed that the development of integrated early intervention programs offers better chances to keep patients in treatment; the disengagement rates reported by other programs providing such type of treatment approach (Turner et al., 2007, Malla and Norman, 2001., 2001; Craig et al., 2004) suggest however that such strategies are not sufficient to resolve the issue in every context.

A better knowledge of the characteristics of first-episode psychosis (FEP) patients who are more likely to disengage from treatment may allow the development of preventive strategies and treatment adaptations that could improve this matter. To our knowledge, only two studies conducted in FEP samples have so far attempted to identify predictors of service disengagement defined as “treatment discontinued in spite of need”. First, in a sample of 288 FEP patients followed up for 2 years, Turner et al (2007) found a 24.6% rate of disengagement that was associated with longer DUP, lower levels of psychotic symptoms, presence of a co-morbid substance use disorder (SUD) and low level of insight at program entry, and with a diagnosis of a non-affective psychosis. However, due to the relatively low number of patients included in the study, the regression model constructed on these variables did not allow a good prediction of later disengagement, calling for replication in larger samples. Second, in a sample of 157 adolescents (aged 15 to 18) drawn from the First Episode Psychosis Outcome Study cohort (FEPOS) (Conus et al., 2007), service disengagement occurred in 28% of patients, and was more likely in patients who had lower levels of psychopathology at baseline, persistence of SUD during treatment and were living away from family (Schimmelmann et al., 2006). The focus on such a specific and relatively small sub-group is however likely to have generated information that cannot be generalised to the vast majority of patients, considering onset of FEP usually occurs at a later age.

Taking these elements and limitations into account, we planned the current study in order to explore the issue of service disengagement in a representative cohort of subjects with FEP and to assess (a) the rate of service disengagement in all patients treated at Early Psychosis Prevention and Intervention Centre (EPPIC) between 1998 and 2000 and (b) the predictors of service disengagement from the EPPIC program. Predictors were grouped chronologically in (i) demographic and pre-treatment variables, (ii) baseline variables (i.e. baseline illness characteristics as well as diagnoses) and (iii) variables during the course of treatment.