Detection of Flare over the Past 3 Months in Rheumatoid Arthritis: Cross-Cultural Equivalence of the Self-Report Flare-RA Questionnaire

Francis Guillemin¹, Marie-Line Erpelding², Annette de Thurah^3,4,5,6, Elena Myasoedova⁷, Emilce E Schneeberger⁸, Cynthia S. Crowson⁹, Thomas Maribo¹⁰, Gustavo Citera¹¹, Eric L. Matteson¹² and Bruno Fautrel¹³, ¹University of Lorraine, Nancy, France, Nancy, France, ²CIC 1433 Clinical Epidemiology, Inserm, Nancy, France, ³Department of Rheumatology, Aarhus University Hospital, Århus C, Denmark, ⁴Department of Rheumatology, Aarhus University Hospital, Aarhus, DK, Aarhus, Denmark, ⁵Department of Clinical Medicine, Aarhus University, Aarhus, Denmark, ⁶Department of Clinical Medicine, Aarhus University, Aarhus, DK, Aarhus N, Denmark, ⁷Rheumatology, Mayo Clinic, Rochester, MN, ⁸Section of Rheumatology, Instituto de Rehabilitación Psicofísica, Buenos Aires, Argentina, Buenos Aires, Argentina, ⁹Health Sciences Research, Mayo Clinic College of Medicine and Science, Rochester, MN, ¹⁰DEFACTUM, Central Region Denmark, Aarhus, Denmark, ¹¹Section of Rheumatology, Instituto de Rehabilitación Psicofísica, Buenos Aires, Argentina, Buenos Aires, Argentina, ¹²Rheumatology, Mayo Clinic College of Medicine and Science, Rochester, MN, ¹³UPMC University Paris 06, Pitié-Salpétrière Hospital, Paris, France

Meeting: 2017 ACR/ARHP Annual Meeting

Date of first publication: September 18, 2017

Keywords: Disease Activity, measure and rheumatoid arthritis (RA)

Session Information

Date: Monday, November 6, 2017

Title: Patient Outcomes, Preferences, and Attitudes I

Session Type: ACR Concurrent Abstract Session

Session Time: 2:30PM-4:00PM

Background/Purpose:

The Flare-RA questionnaire was developed and validated in French (1) for assessing the occurrence of a flare of rheumatoid arthritis (RA) during a 3-month period, and cross-culturally adapted for the English, Spanish and Danish languages. The objective was to assess its cross-cultural equivalence in dimensionality and scale calibration, conditioning its validity.

Methods: Patients included in studies in each of 4 countries (2 cohort studies and 2 clinical trials) were age 18+ years, had RA according to 2010 ACR criteria, were in routine care in three studies and attended a tele-health consultation in one study. Flare occurrence during the most recent 3 months was assessed at baseline and at 3 months of follow-up using the Flare-RA 11-item questionnaire. Flare was measured in two dimensions, namely arthritis-related (5 items) and general (6 items) subscales scoring from 0 (no flare) to 10 (maximum flare). Statistical analysis used exploratory and confirmatory factor analyses to determine dimensionality. Rasch modelling for scale calibration was used to search for differential item functioning (DIF) for countries, sex, and flare anchor (yes/no) using RUMM 2030®.

Results:

Overall, 571 patients (by center: 75, 105, 138 and 253) with mean±SD age 56.9±13.5 years and 75.3% female were included. Self-perception of a flare was reported by 39.9% of patients. Baseline RA disease activity varied by center. Mean DAS28 scores for each center were 2.1±0.8, 2.5±1.2, 2.9±1.2 and 3.6±1.4. Flare-RA items did not show any ceiling effect; mean score was 2.9±2.7 for arthritis-related and 2.5±2.6 for general subscales.

Principal component analysis showed a consistent first factor (65.7% variance), and demonstrated a strong bi-dimensional structure (75.8% variance)(figure). This dimensionality was stable over time and across countries in confirmatory analysis.

The Rasch partial credit model for subscales showed no local dependency, good person separation index (range 0.76–0.87) and good ordering of scale modalities. Only one item on “pain killer intake” performed less well, with less balanced ordering. Two items presented some misfit (fit residual>|2.5|) for each subscale. It was decided to keep them for clinical relevance. Finally, there was some significant but moderate DIF for countries in 7 items, all being uniform DIF. No interaction (DIF) was seen by sex, and only 2 minor DIF with flare anchor.

Conclusion:

This international effort confirms the solid bi-dimensional structure of the Flare-RA questionnaire, with arthritis-related and general subscales, and its good scale measurement properties. There was a slight DIF that might be investigated in larger samples, but it nevertheless can be used as a quantitative scale. These findings demonstrate that the Flare-RA questionnaire has validity for measurement and for international comparison across cohorts and clinical trials.

1. Fautrel B, et al. Arthritis Rheumatol. 2017;69:309-19.

Disclosure: F. Guillemin, None; M. L. Erpelding, None; A. de Thurah, None; E. Myasoedova, None; E. E. Schneeberger, None; C. S. Crowson, None; T. Maribo, None; G. Citera, Novartis, Pfizer Inc, 2,AbbVie, Bristol-Myers Squibb, Janssen, Novartis, Pfizer Inc, 5; E. L. Matteson, Centrocor, 2,Genentech and Biogen IDEC Inc., 2,UCB, 2,NIH, 2,Mesoblast, 2,Novartis Pharmaceutical Corporation, 2,Up-To-Date, 7,Janssen Pharmaceutica Product, L.P., 2,Amgen, 2,Bristol Myers Squibb, 2; B. Fautrel, AbbVIe, Biogen, BMS, Celgene, Hospira, Janssen, Eli Lilly and Company, Novartis, Pfizer, Roche, SOBI Pharma, UCB, 5.

To cite this abstract in AMA style:

Guillemin F, Erpelding ML, de Thurah A, Myasoedova E, Schneeberger EE, Crowson CS, Maribo T, Citera G, Matteson EL, Fautrel B. Detection of Flare over the Past 3 Months in Rheumatoid Arthritis: Cross-Cultural Equivalence of the Self-Report Flare-RA Questionnaire [abstract]. Arthritis Rheumatol. 2017; 69 (suppl 10). https://acrabstracts.org/abstract/detection-of-flare-over-the-past-3-months-in-rheumatoid-arthritis-cross-cultural-equivalence-of-the-self-report-flare-ra-questionnaire/. Accessed .

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/detection-of-flare-over-the-past-3-months-in-rheumatoid-arthritis-cross-cultural-equivalence-of-the-self-report-flare-ra-questionnaire/