Oral versus depot antipsychotic drugs for schizophrenia—A critical systematic review and meta-analysis of randomised long-term trials

Abstract

Objective

Non-adherence is a major problem in the treatment of schizophrenia. Depot antipsychotic drugs are thought to reduce relapse rates by improving adherence, but a systematic review of long-term studies in outpatients is not available.

Method

We searched the Cochrane Schizophrenia Group's register, ClinicalTrials.gov, Cochrane reviews on depot medication, and the reference sections of included studies for randomised controlled trials lasting at least 12 months in outpatients that compared depot with oral antipsychotics in schizophrenia. Data on relapse (primary outcome), rehospitalisation, non-adherence, and dropout due to any reason, inefficacy of treatment and adverse events were summarised in a meta-analysis using a random-effects model. Study quality was assessed with the Cochrane collaboration's risk of bias tool, and publication bias with funnel plots.

Results

Ten studies with 1700 participants met the inclusion criteria. Depot formulations significantly reduced relapses with relative and absolute risk reductions of 30% and 10%, respectively (RR 0.70, CI 0.57–0.87, NNT 10, CI 6–25, P = 0.0009), and dropout due to inefficacy (RR 0.71, CI 0.57–0.89). Limited data on non-adherence, rehospitalisation and dropout due to any reason and adverse events revealed no significant differences. There were several potential sources of bias such as limited information on randomisation methods, problems of blinding and different medications in the depot and oral groups. Other studies reduced a potential superiority of depot by excluding non-adherent patients.

Discussion

Depot antipsychotic drugs significantly reduced relapse. Due to a number of methodological problems in the single trials the evidence is, nonetheless, subject to possible bias.

Introduction

Maintenance treatment with antipsychotic drugs is highly effective in reducing the relapse rates of people with schizophrenia (Davis et al., 1980, Gilbert et al., 1995). However, this effect is jeopardized by high rates of non-adherence to oral medication which has been estimated to occur in as many as 42% of the patients (Cramer and Rosenheck, 1998).

The introduction of second-generation antipsychotic drugs generated enormous hope that medication-adherence would be improved due to fewer side effects, in particular extrapyramidal side effects (Borison, 1997, Buckley, 1997). However, a number of epidemiological analyses have shown that non-adherence to second-generation antipsychotics is not much different than conventional compounds (Dolder et al., 2002, Gilmer et al., 2004, Valenstein et al., 2004).

Depot antipsychotic drugs could have clear advantages in relapse prevention, because medication intake is assured and because the doctor immediately knows when a patient stops treatment. Nevertheless due to various reasons such as refusal of patients or reservations by psychiatrists (Hamann et al., 2010, Heres et al., 2007, Heres et al., 2008), in some countries such as the US (Ahn et al. 2008) depot formulations are rarely prescribed, while in others such as the UK (Barnes et al., 2009) they are used quite frequently. According to a survey among German psychiatrists less than 36% of patients have ever been offered an antipsychotic depot treatment (Heres et al., 2006).

Moreover, the evidence whether depot formulations really reduce relapse is unclear. While an early meta-analysis by Davis et al. (1994) suggested a significant superiority compared to oral medication, the Cochrane reviews of the various depots did not find any convincing difference (Abhijnhan et al., 2007, Da Silva Freire Coutinho et al., 1999, David et al., 1999, David et al., 2004, David et al., 2005, Dinesh et al., 2004, Hosalli and Davis, 2003, Quraishi et al., 1999, Wong et al., 2004; summarised by Adams et al., 2000). However, these reviews also included studies in inpatients and short-term trials of just a few weeks duration. The inclusion of inpatient studies might have reduced the “depot-effect”, because in hospitals medication is usually administered by nurses thereby improving adherence. Therefore outpatient settings are more appropriate to investigate the value of depot antipsychotics. Furthermore, short-term studies are not adequate to examine depot medication, because relapses usually do not occur right after stopping the medication, but often occur after a delay of several months (Gilbert et al., 1995).