Integrated treatment vs. treatment-as-usual for recent onset schizophrenia; 12 year follow-up on a randomized controlled trial

Increased focus on the importance of early interventions in psychosis, with initiation of treatment during the “critical period”, has led to the development of time-limited specialized early interventions [1, 2]. The aims of these early specialized treatments are to shorten the duration of untreated psychosis, preventing relapses and thereby improving prognosis [3, 4].

There are three different early intervention approaches; 1) interventions focusing on recognizing and helping young individuals at high risk of developing psychosis (i.e., before debut of psychosis); 2) early identification programs where the goal is to identify and treat patients in the very beginning of the first psychotic episode, and 3) specialized services to patients with recent but established psychosis [1, 5].

Randomized controlled trials (RCT) have shown promising effects of early specialized interventions on short term outcomes (reduction in hospital admissions and symptoms and improved function) or as long as the treatment continues [6‐8]. Furthermore, early interventions in psychosis have been found to be cost effective compared to standard care [9].

The long-term effects of these treatments are, however, still unclear. In a review of outcomes at 3–5 years after early psychosis intervention, Bird et al. concluded that it is undetermined if the effects of early interventions are sustainable [2]. Except from indications of improved social outcomes, recent findings from 5-year follow-up studies have shown low effects on admission rate and mean number of bed days [10], level of psychotic and negative symptoms [11, 12], global functioning, substance abuse, depression, and suicidal behaviour [12].

The aim of this study was to compare the long-term effects of 2 years of integrated treatment (IT) for recent-onset psychosis versus treatment as usual (TAU) in a randomized controlled trial. Firstly we hypothesized that integrated treatment would have long-term effects on number of days in hospital during the 12–year follow-up period. Hence, we also hypothesized that patients receiving integrated treatment would have reduced number of psychiatric admissions; number of outpatient psychiatric contacts, use of involuntary admissions, duration of inpatients and outpatient involuntary treatment, and mortality.

To our knowledge this is the first long-term RCT reporting a follow up on outcome 14 years after initiation of treatment.

The present study was a centre in the International Optimal Treatment multi-site project initiated by Dr. Ian R.H. Falloon in 1994. The aim was to evaluate the effects of applying evidence-based integrated biomedical and psycho-social interventions in routine services for patients with recent-onset and chronic schizophrenia and other non-affective psychotic disorders [13]. The present study is based on the original RCT [6].

To our knowledge this is the first long term RCT follow-up (12 years) study on intensive Integrated Treatment for patients with recent-onset schizophrenia.

Significantly fewer patients receiving IT (11 of 28 patients in the IT group and 12 of 17 in the TAU group) were involuntary hospitalized. This is an important finding and should have implications in planning of services for patients with schizophrenia. There were, however, no difference in the number of days they were involuntary admitted. No significant differences were found between groups receiving integrated treatment or treatment-as-usual with respect to number of inpatient days, number of patients admitted to psychiatric wards, number of admissions to psychiatric wards, use of outpatient coercion or number of outpatient contacts over a period of 12 years. Although median time to readmission was considerably longer for the IT group, this difference did not reach statistical significance (possibly because of the sample size). Given that no effects on hospitalization were found in the first two years, these findings might not be surprising and limited conclusions can be drawn given that only 45 patients were included in the analyzes.

The findings are in line with other recent studies on 3–5 year outcomes reporting that time-limited early specialized interventions for schizophrenia do not improve outcome over time [10, 11].

This lack of significant long-term effects on the use of services may be understood in several ways. Time-limited evidence- based psychosocial and pharmacological interventions do alleviate the present suffering from the illness, but may not change or hamper the long-term course of the illness [19]. It is, however, possible that there are long-term effects of such interventions, but that such effects only applies to subgroups of patients responding to elements of the interventions [5]. It is interesting that 10 of the 45 patients in this study counted for 76% of all inpatient days during the 12-year follow-up period. This group of patients with schizophrenia with extensive use of psychiatric services may represent a group which needs special attention and could potentially gain much from long term standardized care.

In the original study, IT had a beneficial impact on negative symptoms, minor psychotic episodes, and stabilizing positive symptoms [6]. In modern psychiatric care, improvements in negative symptoms might reduce the need for community services more than reduce the need for hospital admissions.

In this study the participants had recent-onset but established psychosis. The majority (87%) entered the study after their first hospitalization and had been hospitalized on average over 100 days before inclusion.

It is possible that integrated interventions at an earlier stage could have improved their long-term use of services. Such an assumption is supported by the results from the early Treatment and Intervention in Psychosis study where early detection and intervention of recent onset schizophrenia improved 10 year outcomes [20, 21]. In this study the patients in both the experimental and the control group received early intervention treatment. The difference was that those in the experimental group received this treatment significantly earlier then the controls (shorter duration of untreated psychosis). Support for the effects of even earlier interventions comes from a recent meta-analysis on “risk of transition to psychosis” among persons with high risk for psychosis. This study suggests that transition into psychosis among high risk persons may be reduced by cognitive therapy and antipsychotic medications [22].

Yet another explanation for the lack of long-term effects on the use of services of the early integrated treatment demonstrated in our study is that the treatment was too time-limited. The effects of early interventions during the treatment period have been proven through multiple trials. Long-term continuous interventions may be needed if clinically significant effects are to be sustained. In the Prevention and Early Intervention Program for Psychoses in London, Ontario, long-term symptom improvements were obtained after providing the patients with five years continued, albeit less intensive, specialized treatment [23].

Involuntary admissions remain a controversial medical procedure associated with ethical dilemmas and patients with schizophrenia are among those most often involuntary admitted [24]. Research show that involuntary treatment is associated with more severe illness, reduced treatment motivation and less insight [25].

Reduction in the use of involuntary admissions is an important treatment goal. Our results suggest that long-term reduction in involuntary admissions of patients with schizophrenia is attainable by providing early integrated treatment. Forced admissions are not only a result of symptom exacerbations, but may be associated with other psychosocial variables as insight [26], violence and adherence to psychosocial treatment [27]. It is therefore possible that the group difference in forced admissions could be explained by variety of such other aspects of the illness. Components of the IT such as cognitive–behavioural family communication and problem solving skills training, individual cognitive-behavioural strategies for residual symptoms and disability have many similarities to a “crisis plan” which is a psychiatric intervention aimed specifically at reducing use of coercive measures [28]. A RCT on the use of crisis plans found a significant reduction in use of involuntary admissions [29]. This was not associated with differences in overall admissions, as is the case in our finding.

Use of in- and outpatient services is only one of many factors used as outcome measure in schizophrenia. Recovery in schizophrenia is a complex and multi-dimensional concept encompassing both objective and subjective outcome dimensions [30] and an ongoing change progress [31]. Objective outcome dimensions include symptom remission, employment, housing and relations to others while subjective outcome dimensions include appraisal of life circumstances and self-appraisal. Because of the limited scope of outcome measures in this study the possible long-term effects of IT on these multiple aspects of recovery may be hidden and not revealed.

Data from one patient in the TAU group who died 4 years and 9 months into the follow-up period were included in the statistical analyses and no statistical corrections were made to make up for the lack of data for the remaining period. As this patient died of suicide, which is the worst possible clinical outcome, it would be likely that this patient’s long-term clinical outcome would have been poor if the patient had lived. Therefore the inclusion of this patient’s data in the analyses may somewhat have influenced the results for the TAU group.

The intensive two-year psychosocial intervention seemed to have little long-term effects on use of in- and outpatient services. Significantly fewer patients receiving IT were involuntary hospitalized during the follow up period.