Background/Purpose: Patient reported outcomes are integral to measuring patient response to treatment for rheumatoid arthritis (RA). RAPID3 is a patient reported outcome metric that consists of three categories: pain, global assessment of health, and physical functional assessment. Here we investigate the relationship between the components of RAPID3 and serum biomarkers as measures of disease activity among RA patients within a private rheumatology practice.

Methods: Unstructured and structured de-identified data from 2019 of patients who satisfied ACR classification criteria for rheumatoid arthritis were extracted from the electronic medical record, including medication history, laboratory values, and patient reported outcomes. The last available RAPID3 composite score, RAPID3 component variable scores, and multi-biomarker disease activity (MBDA) values were recorded for each patient in 2019. Stable RA patients were defined as those with no change of DMARD therapy for at least 6 months prior to their last visit in 2019. Relationships between variable and aggregate scores were determined based on standard t-test and correlation analysis.

Results: Among a total of 516 patients with stable RA, 477 patients had available RAPID3 composite and component variable scores, and 407 patients had MBDA values. There was a weak, though significant correlation between RAPID3 composite score and MBDA score (Pearson’s correlation coefficient 0.16; p=0.0012). Considering the component RAPID3 variables, pain scale and patient global score were most highly correlated (r=0.83; p< 0.0001), while physical functional was the least correlated, with r=0.61 (p< 0.0001) and r = 0.59 (p< 0.0001) between patient global and pain, respectively. Interestingly, patient global and pain scores were not significantly different (p=0.36), but physical function assessment scores were significantly different from both patient global and pain assessments (p< 0.0001; p< 0.0001, respectively). The distribution of component variable scores is shown in Figure 1. Comparing the components of RAPID3 with the composite RAPID3 score, physical function was the least correlated of the three (r=0.77; p< 0.0001). All three component variables of RAPID3 had significant correlations with MBDA, with physical function assessment score as the most significantly correlated (r=0.18; p< 0.0001).

Conclusion: Here we identify a strong positive correlation between patient reported outcomes for the categories of pain and global assessment in the RAPID3 assessment among a large rheumatoid arthritis patient panel within a community rheumatology practice. While pain and patient global assessment appear to be the primary drivers of overall RAPID3 scores, the functional component of RAPID3 more closely correlates with serum biomarkers reflecting inflammation such as the MBDA. While RAPID3 reflects patient pain and response to pain, it fails to correlate with markers of inflammation like the MBDA score, which drive changes in medication regimen in our own practice. Further research is needed to understand the relationship between RAPID3 and serum inflammatory markers in various patient phenotypes within the rheumatoid arthritis patient population.

Table 1: Correlation Matrix Between Each RAPID3 Component, RAPID3, and MBDA (Pearson’s correlation coefficient [p-value])

Figure 1: Distribution of RAPID3 Component Scores in a Stable RA Population

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/assessment-of-the-components-of-rapid3-patient-reported-outcomes-in-an-community-rheumatology-practice/