Page 140 - THE EVOLUTION OF EARLY ARTHRITIS AND CARDIOVASCULAR RISK Samina A. Turk

CHAPTER 7
(2.5-9.5) versus 13.0 (7.0-34.0) (p<0.01), respectively. Differences at baseline were seen between patients in and not in self-perceived remission after 12 weeks. Patients in self- perceived remission scored significantly lower at baseline on the VAS global, the HAQ, and on the RAID questions about functional disability assessment, fatigue and physical well-being. Of the clinical outcomes, only ESR was significantly lower in patients in self-perceived remission compared to those who were not: 15.0 (7.0-30.0) versus 23.0 (15.0-40.0) mm/hour (p<0.01; Table 1), respectively.
Agreement between physician, patient and clinical response and remission definitions
The agreement between physician-perceived remission and patient-perceived remission was 67% (kappa 0.32, p<0.01; Table 2).
The physician-perceived remission had the best agreement with EULAR good response: 79% agreement, with a kappa of 0.48; p<0.01(Table 2).
The agreement with patient-perceived remission was highest for EULAR good response as well as ACR70 response: both 69% (kappa 0.36 (p<0.01) and 0.40 (p<0.01), respectively). The agreement with Boolean-based remission was slightly lower and the lowest agreement was seen with DAS44 remission.
Concordance between the different clinical response and remission definitions differed enormously. For example, the agreement between EULAR good response and DAS44 remission was 83% (kappa 0.62, p<0.01), but the agreement between EULAR good response and ACR70 response was 52% (kappa 0.22, p<0.01; Table 2). Agreement between physician- and patient perceived remission differed as well within the different response and remission criteria (Additional file 1).
138