Phrasing of the patient global assessment in the rheumatoid arthritis ACR/EULAR remission criteria: an analysis of 967 patients from two databases of early and established rheumatoid arthritis patients

The ACR/EULAR Boolean remission criteria for rheumatoid arthritis (RA) include a strict cutoff for patient global assessment (PGA, value ≤ 1/10). Near-remission corresponds to remission for joint counts and C-reactive protein but with PGA > 1. The objective was to explore whether the contribution of PGA to remission and near-remission varied according to the wording of the PGA and in relation to disease duration. In patients with early arthritis (N = 731, French ESPOIR cohort) or established RA (N = 236 patients from across Europe), frequency of remission versus near-remission was assessed according to the phrasing used for PGA (global health versus disease activity). In 967 patients (mean [standard deviation] age 49.7 [12.7] years, 76.7% women), remission was infrequent: range 12.9–16.7% (according to wording of PGA) in early RA and 6.8–7.2% in established RA. Near-remission was more frequent: 13.0–16.8% in early RA and 13.1–13.6% in established RA. The ratio of remission to near-remission was higher in the early arthritis cohort (0.8–1.3 versus 0.5–0.5 in established RA). Using the disease activity PGA led to more remission and less near-remission than the global health PGA in the early arthritis cohort (12.9 vs 16.7% near-remission, respectively, p = 0.047) but not in established RA. The proportion of patients who can be classified as remission or near-remission differs in early RA compared to establish RA and depends upon the formulation of the PGA question. PGA referring to disease activity and not global health may be preferred in early disease, if the objective is more alignment with inflammation assessment.