New American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) criteria for the classification of rheumatoid arthritis (RA) have recently been proposed. The aim of this cohort study was to examine whether fulfilling these 2010 ACR/EULAR criteria at the first visit has an impact on the clinical course and on the radiographic progression of the disease. For this observational cohort study, we included patients from the Swiss RA registry SCQM with early RA or undifferentiated arthritis (UA, disease duration ≤1 year), as defined by the treating rheumatologist, who had not received any previous disease modifying anti-rheumatic drugs (DMARDs). Patients were categorized into two groups depending on whether or not they fulfilled the 2010 ACR/EULAR criteria (≥6 points vs <6 points) at the first visit. The primary outcome measures were the evolution of the DAS 28 and of radiographic erosions as measured by the Ratingen score over time. Of the 592 patients fulfilling the inclusion criteria, 352 satisfied the 2010 ACR/EULAR criteria at baseline, whereas 240 were not classifiable as definite RA. The ACR/EULAR criteria scores correlated with disease activity at disease onset (R (2) = 0.31). DMARD treatment was subsequently initiated in all patients, mostly with methotrexate (MTX). There were no significant differences in the therapeutic strategies between patients fulfilling or not fulfilling the classification criteria. Six months after inclusion, patients fulfilling the ACR/EULAR criteria developed a 39.1 % reduction of DAS 28 scores, as compared to a 33.6 % reduction in patients not fulfilling the ACR/EULAR criteria (p = 0.0002), independently of their respective treatment strategy. Importantly, the DAS 28 scores were higher in those patients fulfilling the ACR/EULAR criteria (ACR/EULAR positive patients) throughout the observation, as compared to patients not fulfilling those (ACR/EULAR negative patients). Average radiographic progression was higher among ACR/EULAR positive than negative patients (progression of Ratingen score/year 0.50 vs 0.32, resp., p = 0.03) after 3 years of follow-up. Among early RA/UA patients, a score of the 2010 ACR/EULAR criteria sufficient to classify RA selects patients with worse clinical outcome and more radiographic progression.