Objective: The aim was to compare the cognitive ability of people with RA with healthy controls (HCs). Methods: People with RA were recruited from the Norfolk Arthritis Register (NOAR), a population-based cohort study of people with inflammatory arthritis. Data on aged-matched HCs (people with no cognitive impairment) came from the comparison arm of The Dementia Research and Care Clinic Study (TRACC). People with RA and HCs performed a range of cognitive ability tasks to assess attention, memory, verbal fluency, language, visuospatial skills, emotional recognition, executive function and theory of mind. A score of <88 on the Addenbrooke’s Cognitive Examination III was considered cognitive impairment. Scores were compared using linear regression adjusting for age, sex, smoking status, education, BMI, anxiety and depression. Results: Thirty-eight people with RA [mean (S.D.) age: 69.1 (8.0) years; 25 (65.8%) women] were matched with 28 HCs [mean (S.D.) age: 68.2 (6.4) years; 15 (53.6%) women]. Twenty-three (60.5%) people with RA were considered to have mild cognitive impairment [mean (S.D.) Addenbrooke’s Cognitive Examination III: RA = 85.2 (7.4), HC = 96.0 (2.5)]. People with RA had impairments in memory, verbal fluency, visuospatial functioning, executive function and emotional recognition in faces compared with HCs, after adjustment for confounders. Conclusion: People with RA had cognitive impairments in a range of domains. People with RA might benefit from cognitive impairment screening to allow for early administration of appropriate interventions.