A RANDOMIZED CONTROLLED TRIAL OF RHEUMATOLOGIST EDUCATION IMPACTING ON SYSTEMATIC MEASUREMENTS AND TREATMENT DECISIONS IN RHEUMATOID ARTHRITIS (RA): RESULTS OF THE METRIX STUDY

Background: The Metrix study was an investigator initiated pilot study where consenting rheumatologists in Ontario were randomized either to an IRB approved, accredited educational intervention over 6 months or no intervention and involved 2000 RA patient-encounters. Objectives: The objective was to determine if the intervention could result in physician behavioral change. Methods: Twenty rheumatologists participated (all of whom who did a prospective chart audit of 50 consecutive RA patients at the beginning and again 50 consecutive RA patients at end of 6 months) and 10 rheumatologists were randomized to intervention. Only the intervention group was aware of the results of their practice audit including the frequency of measurements they were performing and outcomes of their patients compared to the others in both the intervention and control groups (comparative data of their practice to others). Interventions were monthly web-based conferences on the value of systematic assessments in RA and barriers to care with recent evidence based information, journal club, surveys and improvements on forms used in daily practice to collect data. Results: 1000 serial RA charts were audited at 0 and another at 6 months with no between groups differences in patient characteristics (mean disease of 10 years and 77% women, 74% RF positive and mean DAS 3.7); 68% on current Mtx, 14% on steroids and 27% on biologics. At 6 months, there were significantly different within and between groups changes in how often many variables were measured and changes in treatment. The intervention group collected more patient global assessments (53% pre vs 66% post intervention and MD globals 51% vs 60%; p<0.05) and HAQs collected went from 37% to 42%; whereas control group had no change in outcomes collected. For the intervention group there was a 32% increase in calculable composite scores (such as DAS, CDAI, SDAI) (p<0.05) and no change in the control group. There was more targeting to a low disease state. For those with SDAI between 3.3 and 11, the % receiving a change in Rx (injection, or change in DMARD) was 66% in intervention and 36% in control group (p<0.05); similarly in DAS between 2.4 and 3.6; 57% of intervention and 38% of control group made changes to treatment (p<0.05). Conclusions: Despite a good baseline systematic assessment and many patients in a low disease state, there was improvement in the intervention group that surpassed the control group in both the frequency of performing assessments and the number of treatment changes when patients were not in remission. This is the first RCT of RA rheumatologist education where a result of a change in behavior has been linked directly to an intervention (comparative practice and education) as the control group did not change behavior. Small group learning with feedback from practice audits is an inexpensive way to improve outcomes in RA. Disclosure of Interest: J. Pope Grant/Research support from: Abbott Pharmaceuticals, J. C. Thorne Grant/Research support from: Abbott Pharmaceuticals, A. Cividino Grant/Research support from: Abbott Pharmaceuticals, K. Lucas Grant/Research support from: Abbott PharmaceuticalsCitation: Annals of the Rheumatic Diseases, volume 70, supplement 3, year 2011, page 279Session: RA – prognosis, predictors and outcome (Poster Presentations )