1410: Examining the Interchangeability of Two Different Patient-Reported Global Assessment Measures in an Observational Axial Spondyloarthritis Cohort

Background/Purpose: In axial spondyloarthritis cohorts, patient reported outcomes are vital for tracking disease progression and response to therapy. However, many cohorts enrolled prior to the development and validation of the ASDAS-CRP as a generally accepted standard measurement of disease activity. Our goal is to formally assess if patient global assessment (PGA) activity prior to ASDAS can be used when the modern questionnaire is unavailable.

Methods: Data from the Prospective Study of Outcomes in Ankylosing Spondylitis (PSOAS) observational cohort were analyzed. Patients met Modified New York (mNY) criteria for AS and assessed their global disease activity during study visits. Before the development of ASDAS-CRP, global disease activity was rated on a numeric rating scale with the prompt: “Considering all the ways that your ankylosing spondylitis affects you, rate how you are doing.” The currently used ASDAS global prompt instead asks: “How active was your spondylitis on average during the last week?”

We analyzed patients with visits that included both the older global score and the ASDAS global score to assess agreement between the two measures; ASDAS-CRP was calculated using both the older global assessment compared to the standard ASDAS global assessment. We also compared the standard ASDAS-CRP and ASDAS-ESR. We compared ASDAS disease activity categories between the comparisons and calculated weighted Cohen’s Kappa coefficients. We also compared the older vs. newer global single-item scores with intraclass correlation coefficients (ICC) to evaluate agreement.

Results: 610 unique patient visits were analyzed. The proportion of agreement of ASDAS-CRP calculated with the old PGA compared to the newer PGA was 86.7% based on disease activity categories with Cohen’s Kappa Coefficient of 0.818 (values of 0.81 – 1.00 are considered perfect agreement). The ICC was calculated using a two-way random effects model and was 0.980, consistent with excellent agreement. In other studies, the proportion of agreement of ASDAS-CRP to ASDAS-ESR has been calculated based on disease severity categories. In our study, this proportion was 72.6% with Kappa Coefficient of 0.62, which is similar to other studies.

Conclusion: Prior measures of PGA utilized before the validation of ASDAS-CRP are a valid substitution for the PGA in the ASDAS-CRP based on calculation of a Kappa Coefficient showing perfect agreement. This is important in scenarios where investigators are calculating the ASDAS-CRP in patients who were a part of observational cohorts prior to its development.