Recently, sex hormones such as estrogens and testosterone or its derivatives have been the focus of interest for treatment of persistent symptoms associated with schizophrenia.To review the effects of dehydroepiandrosterone (DHEA)/testosterone as adjunctive therapy to standard antipsychotic drugs.We searched the Cochrane Schizophrenia Group Trials Register (January 2007).We included all clinical randomised trials comparing DHEA/testosterone plus standard antipsychotic treatment with standard treatment alone.We independently selected studies and extracted data. For dichotomous data we calculated the relative risk (RR) and its 95% confidence interval (CI) on an intention to treat basis, using a fixed effects model. We presented continuous data using the weighted mean difference statistic, with a 95% confidence interval using a fixed effects model.We found three relevant small, short trials (total n=126). Clinical Global Impression data were equivocal (n=27, 1 RCT, WMD -0.43 CI -0.9 to 0.1). Average total PANSS scores were not significantly different between the DHEA plus antipsychotic group and those given antipsychotic drugs and placebo (n=82, 2 RCTs, WMD -4.16 CI -13.8 to 5.5). PANSS positive scores were equivocal (n=55, 1 RCT, WMD -1.00 CI -3.8 to 1.8). For negative symptoms binary SANS scale data favoured the DHEA plus antipsychotic group (n=30, 1 RCT, RR 0.23 CI 0.1 to 0.6, NNT 2 CI 2 to 3) but PANSS negative scores were not significantly different between comparison groups (n=55, 1 RCT, WMD -2.30 CI -6.4 to 1.8). About 17% of people left both groups early (n=64, 2 RCTs, RR 0.80 CI 0.3 to 2.4). St Hans Rating Scale data for extrapyramidal symptoms favoured the DHEA plus antipsychotic group (n=30, 1 RCT, WMD -5.00 CI -8.8 to -1.2) but akathisia ratings were equivocal (n=34, 1 RCT, RR 2.67 CI 0.3 to 23.1). Ratings of parkinsonian movement disorder differed within the same trial depending of the outcome scale used. Quality of life seemed unaffected by use of DHEA (n=55, 1 RCT, WMD 6.20 CI -1.4 to 13.8).Results are inconclusive with most outcomes being either non-significant or producing contradictory findings. Currently, adjunctive DHEA should remain an experimental treatment for people with schizophrenia.