Background/Purpose: Patient involvement and personalization of healthcare have become increasingly important in managing chronic diseases. Knowledge about their disease and its treatment is an important precondition for patients to be able to be involved in their own care. Therefore many healthcare professionals offer patients with Rheumatoid Arthritis (RA) education about RA and its treatment. An instrument to assess patients’ knowledge about their disease and its treatment provides health professionals as well as researchers with a tool to identify and target patients’ information needs. Also, it can be used to evaluate the effect if educational efforts and to monitor progress. Currently available patient knowledge questionnaires were developed before the introduction of biologicals as well as treat to target strategies in rheumatoid arthritis (RA). In fact a general limitation of static questionnaires is that their content may become outdated over time. Therefore we developed the Disease and treatment Associated Knowledge in RA item bank (DataK-RA). This Item Response Theory (IRT) based item bank can be kept up to date by adding or removing items, without changing the common scale. Moreover, the scoring procedure also provides the flexibility to use any subset of items, while retaining comparability with other applications of the item bank.

Methods:  An initial item pool was developed from a systematic review of existing knowledge questionnaires, supplemented by an elaborate qualitative approach in cooperation with health professionals and patients. Consensus was reached on relevant content through a RAND modified Delphi scoring procedure by rheumatology nurses (n=6) and rheumatologists (n=6), and a consensus meeting with rheumatology nurses (n=6) and rheumatologists (n=4). Subsequently, a focus group among RA patients (n=9) was organized to identify additional content areas. Patients and health professionals also rated readability, feasibility and comprehensiveness of the resulting items. Item pool reduction and initial validation of the item pool (n=62) were performed using a cross-sectional sample of 473 patients with RA.

Results:  Twenty items were discarded based on a corrected item-total point biserial correlation <0.30. Confirmatory factor analysis with weighted least squares estimation on the polychoric correlation matrix suggested acceptable fit for a unidimensional model for the remaining 42 items (CFI .97 TLI=.97, RMSEA=0.02, WRMR=0.97), supporting the proposed scoring procedure for the item bank. Scores were highly reliable and normally distributed with minimal ceiling (1.8%) and no floor effects (0.0%).

Conclusion: The DataK-RA item bank of 42 items was developed to comprehensively measure knowledge on all relevant aspects of RA and its treatment. The results of this study suggest it is a promising measure of factual disease related knowledge in RA. Work directed at developing tailored short-forms and a computerized adaptive testing algorithm is currently ongoing. Acknowledgements: We thank all rheumatology nurses, rheumatologists and patients who have helped in the development or validation of the item bank.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/development-of-an-item-bank-on-disease-and-treatment-associated-knowledge-of-rheumatoid-arthritis-to-improve-patient-engagement-in-care/