Background Patient-reported outcomes (PRO) provide an important perspective on patient's perception of their clinical well-being. A better understanding of the association between improvements in PROs and rheumatoid arthritis (RA) disease measures could assist clinicians in managing symptoms and treatment expectations in patients with rheumatoid arthritis (RA).
Methods We used methodology based on an analysis of variance model in patients with stable disease on stable therapy (Behrens et al., Arthritis Care Res 2013;65:1608) to establish critical differences (dcrit) defining the minimum change that constitutes a significant individual patient response in the PRO measures of pain (10cm visual analog scale [VAS]), fatigue (10cm VAS), and function (Funktionsfragebogen Hannover [FFbH] questionnaire; range of 0 to 100% remaining functional capacity). We then evaluated significant individual PRO responses in patients with an individual significant DAS28 response (DAS28-dcrit; decrease of ≥1.8 from baseline).
Results In an exploratory cohort of approximately 600 patients with stable disease, the long-term retest reliability was 0.79 for pain, 0.79 for fatigue, and 0.94 for function; these values compared favorably with the retest reliability of 0.83 for DAS28. The critical difference defining the minimum change that exceeded random fluctuation was conservatively determined to be 3 points for pain, 4 points for fatigue, and 18 points for FFbH. These measures were then applied to a treatment cohort of 2788 patients. Of the 1483 patients with a significant individual DAS28-dcrit response (≥1.8 points) after 12 months of adalimumab treatment, 1130 (76.2%) achieved a significant response in at least 1 of the 3 PROs. Almost 70% achieved a significant improvement in pain while approximately 40% achieved a significant improvement in fatigue or function (Figure). Significant improvements in all 3 PROs occurred in 20.7% of patients; 23.8% did not have any significant PRO responses. In contrast, significant improvements in all 3 PROs occurred in only 3.3% of 1305 patients who did not achieve a DAS28-dcrit response at month 12, while 60.8% of these patients had no significant PRO responses at all.
Conclusions Using a statistical method to calculate significant individual responses, we found that pain was the PRO most closely associated with a DAS28-dcrit response; almost 70% of patients with a DAS28-dcrit response had a pain response, but fewer than half the patients showed significant improvements in fatigue or function. The DAS28-dcrit criterion effectively identified therapeutic responses that are meaningful to patients, as 76.2% of patients with a DAS28-dcrit response achieved at least one significant PRO response while 60.8% of patients who did not have a DAS28-dcrit response failed to achieve any significant PRO responses.
Disclosure of Interest E. Scharbatke Grant/research support from: Chugai, Consultant for: AbbVie, Roche, Chugai, F. Behrens Grant/research support from: Pfizer, Consultant for: AbbVie, M. Schmalzing Consultant for: AbbVie, Chuhgai, Actelion, UCB, Roche, Pfizer, M. Koehm Grant/research support from: Pfizer, Consultant for: AbbVie, Roche, Chugai, G. Greger Employee of: AbbVie Deutschland GmbH & Co.KG, H. Gnann Grant/research support from: AbbVie Deutschland GmbH & Co.KG, H.-P. Tony Consultant for: AbbVie, Roche, Chugai, MSD, Pfizer, UCB, Speakers bureau: AbbVie, H. Burkhardt Grant/research support from: Pfizer, Consultant for: AbbVie, Pfizer, BMS, UCB, Chugai