The objective of this study was to determine the frequency of cognitive impairment in patients with rheumatoid arthritis (RA). A cross-sectional study of 40 patients with RA and 40 healthy controls was performed. To assess cognitive impairment, anxiety and depression, the following standardized psychiatric and clinical research methods were used: the Mini-Mental State Examination (MMSE), logic memory tests, short and long memory tests, verbal fluency tests, attention tests, the Brief Psychiatric Rating Scale (BPRS), the Hospital Anxiety and Depression (HAD)/CAGE scale and the Beck Depression Inventory (BDI). Patients and controls with incomplete primary education were excluded from the study. Statistics were performed by chi-square test and by Fisher's exact test. Cognitive impairment was observed in 30% of patients with RA and in 7.5% (p < 0.05) of healthy controls. Patients with RA had a significantly worse outcome in verbal fluency (p < 0.05), logic memory (p < 0.05) and short memory (p < 0.05). No statistical difference was observed among the results obtained in the MMSE, BPRS, HAD/CAGE and BDI. There was no significant relation to the duration of the illness, use of corticotherapy or disability. We observed a high prevalence of cognitive impairment in RA patients. Cognitive impairment was not related to clinical and treatment features or disability. More studies are necessary to determine clinical impact of cognitive impairment in RA.