Background: Substance use disorders (SUDs) are present in more than 50% of subjects diagnosed with schizophrenia. However, there are no controlled studies assessing the efficacy of antipsychotic drugs in this subgroup of patients. The aim of the present work was to compare the efficacy of risperidone and zuclopenthixol in a sample of schizophrenic subjects with dual diagnosis. Method: Thirty-three male were selected for treatment with risperidone, while another 33 were treated with zuclopenthixol. Substances most commonly used were alcohol, cannabis (both 82%) and cocaine (32%). Patients were randomized and treated for the first six months with one antipsychotic and the second six months with the other antipsychotic. Psychopathological and clinical scales were used every two months. Participants received training on how to reduce their consumption of substances (Substance Abuse Management Module, SAMM). Results: During the first six months risperidone group patients presented fewer positive urine tests and showed better compliance with the SAMM programme. In the second period the patients treated with risperidone significantly improved their scores on the PANSS-negative subscale. Differences between the CGIs indicated that the subjects who moved from risperidone to zuclopenthixol worsened, while those who moved from zuclopenthixol to risperidone significantly improved. Conclusions: Risperidone was more effective than zuclopenthixol in improving the symptoms of schizophrenia and substance use.