We conducted a cross-sectional study to compare the prevalence and severity of obsessive-compulsive symptoms (OCSs) and obsessive-compulsive disorder (OCD) in patients with schizophrenia treated with clozapine or haloperidol. Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Axis I disorders-patient edition was used to diagnose schizophrenia and OCD. Sixty subjects, 40 of them using clozapine and 20 using haloperidol, completed the Yale-Brown Obsessive-Compulsive Scale, the Positive and Negative Syndrome Scale (PANSS), and the Clinical Global Impression. The prevalence of OCD in patients taking clozapine was 20%, whereas the prevalence of patients taking haloperidol was 10%, although this difference was not statistically significant (P = .540). However, patients using clozapine showed higher severity of OCSs than patients using haloperidol (P = .027) did. When schizophrenia patients were divided according to the presence or absence of OCD or OCSs, patients with schizophrenia and OCD or OCSs showed higher severity of schizophrenia symptoms when compared to those with schizophrenia without OCD and OCSs (P = .002). A PANSS total score higher than 70 and the use of antidepressants were predictors of the presence of OCSs or OCD. Schizophrenia patients taking clozapine had higher severity scores both in obsessive-compulsive and schizophrenia rating scales. These results may support an association between the exacerbation of obsessive-compulsive phenomena and the use of clozapine.