Background/Purpose: Achieving tight control of RA with reliable methods to monitor and assess disease activity in a more objective way is imperative in clinical practice to obtain treat to target results. We conducted a longitudinal study to evaluate utility, adoption, and analysis of the correlation and agreement of the Clinical Disease Activity Disease Index (CDAI) and Simple Disease Activity Index (SDAI) scores and categories of disease. We adapted our electronic health record (EHR) (Centricity) to standardize real time assessment with the CDAI in hopes to change physicians’ therapeutic behavior, in order to provide better care and afford patients with a single number to understand their disease activity.

Methods:  In the first seven months a total of 153 cases were included in the analysis. Spearman rank and intraclass correlation (ICC) were computed between CDAI and SDAI scores. The Kappa agreement and McNemar’s test was conducted for the agreement of CDAI and SDAI disease categories. Finally one-way ANOVA was used to test the association between categorized SDAI and CDAI and selected lab variables. The quality analysis of the project was performed with quarterly meetings with rheumatologists and the information technology (IT) department to discuss implementation and data collection.

Results:  The “CDAI calculator” was launched on October 2014 during this month we had a 48% documentation rate for RA patient clinical encounters, but by December we achieved a 100% EHR documentation rate. Our study was in an 87.32% Hispanic population with 84.31% being female. Based on CDAI disease activity categories, the distribution for Remission, Low, Moderate, and High activity was 43.79%, 31.37%, 15.03%, and 9.8% respectively. The association between CDAI and SDAI scores was found to be very high with a spearman value of 0.95 (p-value <0.0001) and further confirmed with the ICC of 0.987 (CI: 0.982-0.991). There was no significant difference in proportion of discordant pairs of the SDAI and CDAI categories, showing that scores that switched classification when converted were negligible (p-value 0.4815). The agreement between the CDAI and SDAI categories was high with a Kappa of 0.87 (CI: 0.81-0.94). Of note, there was a difference between the categorized SDAI versus platelets ANOVA analysis (p-value 0.053) with platelets in Remission being 272.4 (± 91.0), Low 239.9 (± 71.1), Moderate 286.8 (± 84.9), and High 293.5 (± 72.0).

Conclusion:  Incorporation of the CDAI into point-of-care visits was possible with coordination of an interdisciplinary team and the continual reassessment methodology of the Plan Study Do Act approach. On analysis, the SDAI and CDAI showed that even when a CRP was not included in composite scoring the CDAI categorization was utile for proper assessment and sensitive to change. When seeing association between lab values and the indices, the SDAI categorization was able to show the theory of thrombocytosis in higher inflammation states confirming that the index can differentiate disease activity accurately. At this time, we need more follow up CDAI scores in order to objectively see treat to target changes in both patient population and physician management behavior.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/implementation-of-the-clinical-disease-activity-index-to-treat-to-target-rheumatoid-arthritis-in-the-ambulatory-setting-a-plan-do-study-act-quality-analysis/