Maintenance antipsychotic treatment versus discontinuation strategies following remission from first episode psychosis: systematic review

Abstract

Background: Understanding the relative risks of maintenance treatment versus discontinuation of antipsychotics following remission in first episode psychosis (FEP) is an important area of practice.

Method: A systematic review and meta-analysis. Prospective experimental studies including a parallel control group were identified to compare maintenance antipsychotic treatment with total discontinuation or medication discontinuation strategies following remission in FEP.

Results: Seven studies were included. Relapse rates were higher in the discontinuation group (53%; 95% CIs: 39%, 68%; N = 290) compared with maintenance treatment group (19%; 95% CIs: 0.05%, 37%; N = 230). In subgroup analyses, risk difference of relapse was lower in studies with a longer follow-up period, a targeted discontinuation strategy, a higher relapse threshold, a larger sample size, and samples with patients excluded for drug or alcohol dependency. Insufficient studies included psychosocial functioning outcomes for a meta-analysis.

Conclusions: There is a higher risk of relapse for those who undergo total or targeted discontinuation strategies compared with maintenance antipsychotics in FEP samples. The effect size is moderate and the risk difference is lower in trials of targeted discontinuation strategies.

Declaration of interest: A.T. has received honoraria and support from Janssen-Cilag and Otsuka Pharmaceuticals for meetings and has been has been an investigator on unrestricted investigator-initiated trials funded by AstraZeneca and Janssen-Cilag. He has also previously held a Pfizer Neurosciences Research Grant. S.M. has received sponsorship from Otsuka and Lundbeck to attend an academic congress and owns shares in GlaxoSmithKline and AstraZeneca. J.H. has attended meetings supported by Sunovion Pharmaceuticals.

Fig. 1

PRISMA flowchart outlining the search and selection strategy.

Fig. 2

(a) Studies reporting risk difference (comparing maintenance to discontinuation groups) for relapse. (b) Studies reporting risk difference (comparing maintenance to discontinuation groups) for hospital admission. Relapse studies included in analysis: Boonstra et al (2011), Chen et al (2010), Crow et al (1986), Gaebel et al (2002, 2011),^, Kane et al (1982) and McCreadie et al (1989). Hospitalisation studies included in analysis: Boonstra et al (2011), Chen et al (2010), Gaebel et al (2002, 2011)^, and McCreadie et al (1989). ES, Effect size

Fig. 3

(a) Subgroup analysis of risk difference of relapse rates according to study characteristics (treatment strategy, follow-up period and relapse threshold). Discontinuation studies: Boonstra et al (2011) , Chen et al (2010), Crow et al (1986), Kane et al (1982) and McCreadie et al (1989). Targeted discontinuation studies: Gaebel et al (2002, 2011).^, Follow-up ≤1 year studies: Chen et al (2010), Gaebel et al (2002), Kane et al (1982) and McCreadie et al (1989). Follow-up >1 year studies: Boonstra et al (2011), Crow et al (1986) and Gaebel et al (2002). Relapse studies with lower threshold: Boonstra et al (2011), Chen et al (2010) and Kane et al (1982). Relapse studies with higher threshold: Crow et al (1986), Gaebel et al (2002, 2011)^, and McCreadie et al (1989). (b) Subgroup analysis of risk difference of relapse rates according to study characteristics (sample size, blinding and risk of bias). Studies with sample size of <40 participants: Boonstra et al (2011),Kane et al (1982) and McCreadie et al (1989). Studies with sample size of >40 participants: Chen et al (2010), Crow et al (1986) and Gaebel et al (2002, 2011).^, Blinded studies: Chen et al (2010), Crow et al (1986), Kane et al (1982) and McCreadie et al (1989). Open studies: Boonstra et al (2011) and Gaebel et al (2002, 2011).^, High risk of bias: Crow et al (1986), Kane et al (1982) and McCreadie et al (1989). Low risk of bias: Boonstra et al (2011), Chen et al (2010) and Gaebel et al (2002, 2011).^,