A COMPARISON OF THE EFFECTIVENESS OF TNF ALPHA BLOCKING AGENTS IN THE TREATMENT OF RHEUMATOID ARTHRITIS BETWEEN RANDOMIZED CONTROLLED TRIALS AND DUTCH OBSERVATIONAL DATA

Background: On basis of the evidence from RCTs, TNFalpha blocking agents have been prescribed in Dutch clinical practice since January 2000 (etanercept and infliximab) and November 2003 (adalimumab). However, to what extent can the results from RCTs be generalized to patients in clinical practice?Objectives: In the present study, both data from RCTs as well as observational data were used to evaluate the difference between patient characteristics, treatment and mean effectiveness found in RCTs as opposed to clinical practice in the case of anti-TNFalpha treatment for RA patients.Methods: In April 2003 a registry was started to monitor and evaluate the use of anti-TNF in patients with RA in 12 centres in the Netherlands, the Dutch Rheumatoid Arthritis Anti-TNFalpha Monitoring (Dream) study. All patients starting on TNFalpha blocking agents have been assessed by a trained research nurse every three months Disease activity was measured using core set measures: 28 joint count for tenderness (TJC) and for swelling (SJC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level and visual analogue scales (VAS) for general health, disease activity and pain. Further, functionality was measured by use of the Health Assessment Questionnaire (HAQ).Performing a systematic review, all randomized placebo controlled trials for etanercept, infliximab and adalimumab were selected, regardless of the concomitant use of methotrexaat. For every selected article, a dataset of patients with an equal follow-up time was created from the TNF-registration database. A comparison was made on basis of the patient characteristics, baseline values, therapeutic setting and effectiveness. The difference in effectiveness was evaluated as the absolute improvements on the HAQ, SJC, VAS pain and CRP in the RCT treatment group, the RCT net group (treatment minus placebo) and the observational group.Results: The mean absolute improvements on the HAQ and the VAS pain in the observational group were smaller than in the RCT treatment group yet they ranged around the RCT net improvements. This finding was less clear on the CRP and the SJC but the improvements in the observational group tended to be smaller than in the RCT treatment group. Concerning patients' characteristics, patients from the clinical practice reported a lesser percentage of tender joints and for several RCTs a smaller mean CRP. Next, only 34% to 79% of all patients starting with TNF alpha blocking agents in clinical practice would be eligible for the RCTs concerning their RA activity, indicating that they have a lower RA activity.Conclusion: It can be concluded that the effectiveness of TNFalpha for RA in daily clinical practice ranged around the RCT net (treatment minus placebo) effectiveness as found in RCTs. Further, patients in daily clinical practice do have a lower RA activity, which might explain the difference in response in RCTs and daily clinical practice.Citation: Ann Rheum Dis, volume 65, supplement II, year 2006, page 282Session: Health service and outcome research