OBJECTIVES: To determine whether patient-reported outcomes may differentiate treatment response better than physician-reported outcomes for rheumatoid arthritis (RA) patients being treated with anakinra. METHODS: A meta-analysis was conducted using data obtained from three separate randomized controlled clinical trials (RCTs) (n = 1007). Outcomes from 6-month assessments were grouped into four categories: American College of Rheumatology (ACR) response criteria, patient-reported measures (patient-reported pain, patient global assessment, and assessment of physical function using the Health Assessment Questionnaire), physician-reported measures (tender and swollen joint counts and physician global assessment), and laboratory tests (C-reactive protein and erythrocyte sedimentation rate). Effect sizes were calculated using changes from baseline and pooled standard deviations for each of these types of outcome. RESULTS: Active treatment with anakinra was superior to placebo by ACR(20) responses in all three RCTs. Effect sizes for patient-reported outcomes were greater than for physician-reported outcomes, and also greater than ACR(20) in three of five anakinra cohorts. Across the RCTs, placebo responses were greater with physician-reported than with patient-reported outcomes. In the two studies evaluating patients with longer-standing disease, differences between pooled effect sizes for patient-reported and physician-reported outcomes were even more pronounced. CONCLUSIONS: In three pivotal RCTs, active treatment with anakinra resulted in greater improvements in patient-reported than physician-reported outcomes compared with placebo. These observations confirm those previously reported from RCTs evaluating conventional DMARDs, demonstrating better discrimination of treatment effect with patient-reported outcomes.
OBJECTIVES: To determine whether patient-reported outcomes may differentiate treatment response better than physician-reported outcomes for rheumatoid arthritis (RA) patients being treated with anakinra. METHODS: A meta-analysis was conducted using data obtained from three separate randomized controlled clinical trials (RCTs) (n = 1007). Outcomes from 6-month assessments were grouped into four categories: American College of Rheumatology (ACR) response criteria, patient-reported measures (patient-reported pain, patient global assessment, and assessment of physical function using the Health Assessment Questionnaire), physician-reported measures (tender and swollen joint counts and physician global assessment), and laboratory tests (C-reactive protein and erythrocyte sedimentation rate). Effect sizes were calculated using changes from baseline and pooled standard deviations for each of these types of outcome. RESULTS: Active treatment with anakinra was superior to placebo by ACR(20) responses in all three RCTs. Effect sizes for patient-reported outcomes were greater than for physician-reported outcomes, and also greater than ACR(20) in three of five anakinra cohorts. Across the RCTs, placebo responses were greater with physician-reported than with patient-reported outcomes. In the two studies evaluating patients with longer-standing disease, differences between pooled effect sizes for patient-reported and physician-reported outcomes were even more pronounced. CONCLUSIONS: In three pivotal RCTs, active treatment with anakinra resulted in greater improvements in patient-reported than physician-reported outcomes compared with placebo. These observations confirm those previously reported from RCTs evaluating conventional DMARDs, demonstrating better discrimination of treatment effect with patient-reported outcomes.
Authors: An-Wen Chan; Jennifer M Tetzlaff; Peter C Gøtzsche; Douglas G Altman; Howard Mann; Jesse A Berlin; Kay Dickersin; Asbjørn Hróbjartsson; Kenneth F Schulz; Wendy R Parulekar; Karmela Krleza-Jeric; Andreas Laupacis; David Moher Journal: BMJ Date: 2013-01-08