Although discouraged in available treatment guidelines, combination antipsychotic prescribing (CAP) is a common practice in the treatment of schizophrenia. Patient characteristics may be associated with this type of treatment. A dataset (N = 363) derived from parallel randomised controlled trials was interrogated to identify factors associated with the receipt of CAP, and a logistic regression analysis was used to predict the occurrence of CAP. Significant predictors of CAP were longer illness, low global functioning score and high treatment adherence rating. Co-prescribed patients received a higher combined dose. © 2010 British Association for Psychopharmacology.