Discordance of global assessment by patients and physicians is higher in osteoarthritis than in rheumatoid arthritis: a cross-sectional study from routine care

The study compares patient–physician discordance in global assessment in patients with osteoarthritis (OA) versus patients with rheumatoid arthritis (RA) seen in routine care. This is a cross-sectional study conducted at an academic rheumatology center at which all patients are asked to complete a Multi-Dimensional Health Assessment Questionnaire (MDHAQ), which includes a patient global assessment (PATGL). Rheumatologists are encouraged to complete a physician questionnaire, which includes a physician global assessment (DOCGL). Patients with either OA or RA were identified using ICD9 codes and classified as positive discordance (PATGL-DOCGL ≥ 2), negative discordance (PATGL-DOCGL≤ − 2), and concordance (absolute difference between the two assessments < 2). Discordance was assessed by diagnosis. Agreement between patient and physician global assessments was evaluated using intraclass correlations. Logistic regression was performed to identify explanatory variables for positive discordance. The analysis included 243 OA and 216 RA patients. Mean PATGL was higher in OA versus RA (5.4 versus 4.2, p = 0.005), while mean DOCGL was similar (4.0 versus 3.8, p = 0.23) leading to a higher patient–physician discordance in OA (1.35 versus 0.43, p < 0.001). Positive discordance occurred in 34% of OA versus 18% of RA patients (p < 0.001). Intraclass correlation coefficients were 0.43 in OA versus 0.60 in RA patients. In logistic regressions, pain was the only statistically significant explanatory variable for discordance in both OA (OR 1.34, 95% CI 1.12–1.78) and RA (OR 1.47 95% CI 1.04–2.07). Patients with OA are more likely to be discordant with their rheumatologists than patients with RA because of a higher PATGL. Similarly to RA, the most important explanatory variable for discordance was higher pain.