Background The ultimate goal of RA treatment is to achieve disease remission. Studies have shown that objective, measure-based treatment decision-making leads to improved patient outcomes. In addition, the National Quality Forum (NQF) recently endorsed three measures for RA, including assessment of disease activity using a standardized tool. However, in clinical practice physicians may still rely on subjective assessment of disease activity.

Objectives To compare physician-reported and DAS28-based clinical assessments of RA disease remission.

Methods Data were drawn from the Adelphi RA-DSP, a cross-sectional survey of United States rheumatologists and their RA patients between January and March 2014. Clinical remission status (DAS28 ESR <2.6) was compared with physician-reported remission and categorized as “match”, “physician over-reported” (physician-reported remission, but no-remission by DAS28 criteria) and “physician under-reported” (physician-reported no-remission, but remission by DAS28 criteria). Multinomial logistic regressions were conducted to evaluate patient and physician characteristics associated with differences between DAS28-based and physician-reported remission.

Results Included in the analysis were 531 RA patients (75% female, mean age 56.4 years, 7.7 years since RA diagnosis, 53% treated with biologic) in the care of 78 rheumatologists (31% female, 37% practice only in office). 56% of physicians reported using only subjective criteria to assess remission; 30% of patients were evaluated using a standardized disease activity measure during last visit. While physicians reported 50% of patients were in remission, 32% were in remission by DAS28 criteria. Remission status was over-reported by physicians in 25% and underreported in 7% of patients.

Regression analyses indicated that physician over-reporting was significantly higher among patients treated with biologics (Relative Risk Ratio (RRR)=2.01, P=0.003), those in the care of physicians with high RA workload (RRR=2.51, P=0.011), and patients considered by their physicians to have “satisfactory” RA control (RRR=6.09, P<0.001) or with average general health level (RRR=1.87, P=0.049); but significantly lower for patients with medium to high current pain level (RRR=0.23, P=0.004) or considered by their physicians to be highly involved in treatment (RRR=0.60, P=0.035).

Physician under-reporting was significantly higher among obese patients (RRR=2.95, P=0.002), patients with average general health level (RRR=2.03, P=0.029), and those with more than 5 tender or swollen joints (RRR=2.72, P=0.019), but significantly lower among patients with medium to high current pain level (RRR=0.18, P=0.004) or those who saw physicians with high RA workload (RRR=0.43, P=0.029).

Conclusions This study points out the need to increase the use of standardized measures of RA disease activity, and ensure consistency of use across clinical care. Addressing these gaps in care may help optimize treatment decisions, reduce variability in delivery of patient care and ultimately improve RA patient outcomes.

Disclosure of Interest W. Wei Shareholder of: Sanofi, Employee of: Sanofi, C. Chen: None declared, E. Sullivan: None declared, S. Blackburn: None declared, J. Curtis Grant/research support from: Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie, Consultant for: Roche/Genentech, UCB, Janssen, CORRONA, Amgen, Pfizer, BMS, Crescendo, AbbVie