Bénédicte Delcoigne1, Sella Aarrestad Provan2, Hilde Berner Hammer2,3, Daniela Di Giuseppe1, Thomas Frisell1, Bente Glintborg4,5, Gerdur Grondal6, Bjorn Gudbjornson6, Merete Lund Hetland4,5, Brigitte Michelsen2,7, Dan Nordström8,9, Heikki Relas8, Johan Askling1. 1. Department of Medicine Solna, Clinical Epidemiology Division, Karolinska Institutet, Stockholm, Sweden. 2. Division of Rheumatology and Research, Diakonhjemmet Hospital, Oslo, Norway. 3. Faculty of Medicine, University of Oslo, Norway. 4. The DANBIO registry and Copenhagen Center for Arthritis Research (COPECARE), Center for Rheumatology and Spine Diseases, Centre of Head and Orthopaedics, Copenhagen University Hospital Rigshospitalet, Glostrup, Denmark. 5. Department of Clinical Medicine, University of Copenhagen, Faculty of Health and Medical Sciences, Copenhagen, Denmark. 6. Centre for Rheumatology Research (ICEBIO), Landspitali University Hospital and Faculty of Medicine University of Iceland, Reykjavik, Iceland. 7. Division of Rheumatology, Department of Medicine, Hospital of Southern Norway Trust, Kristiansand, Norway. 8. Division of Medicine and Rheumatology, Helsinki University Hospital, Helsinki, Finland. 9. Department of Medicine, University of Helsinki, Finland.
Abstract
OBJECTIVES: To investigate whether patient-reported outcomes vary across countries and are influenced by cultural/contextual factors. Specifically, we aimed to assess inter-country differences in tender joint count (TJC), pain and patient's global health assessment (PGA), and their impact on disease activity (DAS28-CRP) in rheumatoid arthritis (RA) patients from five Nordic countries. METHODS: We collected data (baseline, 3- and 12-months) from rheumatology registers in the five countries comprising RA patients starting a first-ever methotrexate or a first-ever tumor necrosis factor inhibitor (TNFi). In order to assess the role of context (=country), we separately modelled TJC, pain and PGA as functions of objective variables (C-reactive protein, swollen joint count, age, sex, calendar period and disease duration) with linear models. Analyses were performed at each time point and for both treatments. We further assessed the impact of inter-country differences on DAS28-CRP. RESULTS: 27 645 RA patients started methotrexate and 19 733 started a TNFi. Crude inter-country differences at methotrexate start amounted to up to 4 points (28 points scale) for TJC, 10 and 27 points (0-100 scale) for pain and PGA, respectively. Corresponding numbers at TNFi start were 3 (TJC), 27 (pain) and 24 (PGA) points. All differences were reduced at 3- and 12-months, and attenuated when adjusting for the objective variables. The variation in predicted DAS28-CRP across countries amounted to < 0.5 units. CONCLUSIONS: Inter-country differences in TJC, pain and PGA are greater than expected based on differences in objective measures, but have a small clinical impact on DAS28-CRP across countries.
OBJECTIVES: To investigate whether patient-reported outcomes vary across countries and are influenced by cultural/contextual factors. Specifically, we aimed to assess inter-country differences in tender joint count (TJC), pain and patient's global health assessment (PGA), and their impact on disease activity (DAS28-CRP) in rheumatoid arthritis (RA) patients from five Nordic countries. METHODS: We collected data (baseline, 3- and 12-months) from rheumatology registers in the five countries comprising RA patients starting a first-ever methotrexate or a first-ever tumor necrosis factor inhibitor (TNFi). In order to assess the role of context (=country), we separately modelled TJC, pain and PGA as functions of objective variables (C-reactive protein, swollen joint count, age, sex, calendar period and disease duration) with linear models. Analyses were performed at each time point and for both treatments. We further assessed the impact of inter-country differences on DAS28-CRP. RESULTS: 27 645 RA patients started methotrexate and 19 733 started a TNFi. Crude inter-country differences at methotrexate start amounted to up to 4 points (28 points scale) for TJC, 10 and 27 points (0-100 scale) for pain and PGA, respectively. Corresponding numbers at TNFi start were 3 (TJC), 27 (pain) and 24 (PGA) points. All differences were reduced at 3- and 12-months, and attenuated when adjusting for the objective variables. The variation in predicted DAS28-CRP across countries amounted to < 0.5 units. CONCLUSIONS: Inter-country differences in TJC, pain and PGA are greater than expected based on differences in objective measures, but have a small clinical impact on DAS28-CRP across countries.