OBJECTIVE:The aim of this study was to evaluate referral and treatment delays by ethnicity/race in patients with rheumatoid arthritis (RA) treated at an academic rheumatology center. METHODS:We reviewed the medical records of all RA patients evaluated at an outpatient clinic between 2011 and 2016 to identify newly diagnosed and naive-to-treatment patients. We determined the durations between symptom onset and first rheumatology visit and time to initiate treatment. Data extraction included referral source, demographics, treatment, and laboratory tests. Routine use of a multidimensional health assessment questionnaire allowed us to calculate baseline RAPID3 (routine assessment of patient index data 3) scores. Comparisons between self-reported ethnicity/race groups were performed. We used logistic regression models to analyze associations between baseline variables and early referral. RESULTS:Data from 152 disease-modifying antirheumatic drug-naive RA patients were included in the study; 35% were white, 37% black, 20% Hispanic, and 8% other. The range in median time to first rheumatology visit was 6 to 8 months for all patient groups, except Hispanic. This group had a median time of 22.7 months (p = 0.01). The referral pattern was considerably variable between-groups; 40% of Hispanic patients were self-referred (p = 0.01). There were no statistically significant between-group differences for time to treatment initiation according to ethnicity/race. RAPID3 scores (p = 0.04) and erythrocyte sedimentation rates (p = 0.01) were significantly higher in the black and Hispanic groups. A high C-reactive protein value at baseline was associated with earlier referral. CONCLUSIONS:There is significant delay in initial presentation to a rheumatologist that was associated with a higher disease severity at presentation, especially for Hispanic patients.