This article is freely available to all

Article Abstract

Because this piece has no abstract, we have provided for your benefit the first 3 sentences of the full text.

Sir: I read with interest the recent meta-analysis by Sepehryet al.1 of selective serotonin reuptake inhibitor (SSRI) add-ontherapy for negative symptoms of schizophrenia and am concernedthat their negative conclusion largely reflects methodologicallimitations.The small number and heterogeneity of available studiesmake meta-analytic analyses at this time particularly vulnerableto confounds.2 Further, optimal use was not made of availablestudies: all participants in our original study3 received typicalantipsychotics (the only ones available then), a fact easily clarified,and thus were eligible for subanalyses of that group.’ ‹