Background/Purpose

In RA the target for treatment is clinical remission or minimal disease activity. Patient self- monitoring of disease activity may enhance treatment by providing early warnings when targets are not met, indicating the need to re-evaluate treatment. The objective of this study is to compare agreement between patient and rheumatologist (MD) derived disease activity states using Clinical Disease Activity Index (CDAI), Simplified Disease Activity Index (SDAI) and Disease Activity Score 28 (DAS28).

Methods

Consecutive RA patients presenting for follow-up to seven rheumatologists were invited to participate. Consenting patients completed a questionnaire and performed a self-report joint count before their visit. MD joint count and lab values were obtained from charts. Disease activity was evaluated using patient and MD versions of CDAI, SDAI and DAS28. In patient versions, patient joint counts were used and MD global assessments were replaced with patient global assessments (for CDAI and SDAI). Disease states were categorized into remission, low, moderate or high, according to published cut points. Because change in treatment is recommended with moderate or high disease activity, we also compared instruments using two categories: remission or low vs. moderate or high. Agreement between patient and MD derived disease states were evaluated using Agreement Coefficient 1 (AC1) for two category comparisons and Agreement Coefficient 2 (AC2), weighted with quadratic weights, for four category comparisons. AC values > 0.62 were considered good agreement. Z tests were used to evaluate the statistical significance of the difference between pairs of ACs. All p values were 2 tailed.

Results

We recruited 150 RA patients [mean (SD) RA duration: 11.9 (11.3) y; age: 57.8 (16.3) y; 81% female]. Table 1 shows agreement between patient and MD derived disease activity measures (all p values < 0.001). There was good agreement between patient and MD derived disease activity states from the same measure, except when comparing patient and MD DAS28 across four categories, which had slightly lower agreement. There was no significant difference in the agreement between the patient and MD versions of the three measures when using four categories [CDAI vs. SDAI AC2: p = 0.480; CDAI vs. DAS28 AC2: p = 0.633; SDAI vs. DAS28 AC2: p = 0.915] or two categories [CDAI vs. SDAI AC1: p = 0.580; CDAI vs. DAS28 AC1: p = 0.052; SDAI vs. DAS28 AC1: p = 0.062].

Conclusion

Our results suggest that patients can self-monitor disease activity using patient derived CDAI, SDAI or DAS28 measures. Good agreement was found between the disease activity states derived from patient data and MD data from the corresponding measure, except for DAS28 when using four categories. There was no statistically significant difference in the agreement across measures.

Table 1. Agreement between patient and MD derived indices measured across four and two disease activity categories.

A	Comparison across four categories (remission vs. low vs. moderate vs. high)
PATIENT MEASURES	RHEUMATOLOGIST MEASURES
	CDAI-MD	SDAI-MD	DAS28-MD
	AC2 [95% CI]	AC2 [95% CI]	AC2 [95% CI]
CDAI-Pt	0.67 [0.55, 0.79]	0.67 [0.54, 0.79]	0.43 [0.29, 0.58]
SDAI-Pt	0.62 [0.49, 0.75]	0.68 [0.56, 0.81]	0.39 [0.24, 0.55]
DAS28-Pt	0.55 [0.42, 0.69]	0.50 [0.35, 0.64]	0.58 [0.45, 0.71]
B	Comparison across two categories (remission or low vs. moderate or high)
PATIENT MEASURES	RHEUMATOLOGIST MEASURES
	CDAI-MD	SDAI-MD	DAS28-MD
	AC1 [95% CI]	AC1 [95% CI]	AC1 [95% CI]
CDAI-Pt	0.80 [0.75, 0.85]	0.80 [0.75, 0.86]	0.43 [0.29, 0.57]
SDAI-Pt	0.77 [0.72, 0.82]	0.79 [0.74, 0.85]	0.38 [0.24, 0.53]
DAS28-Pt	0.71 [0.64, 0.79]	0.67 [0.58, 0.75]	0.79 [0.72, 0.86]
AC1 = agreement coefficient 1; AC2 = quadratic weighted agreement coefficient 2 All AC values have 2 tailed p-values of p < 0.001 Bolded values (AC > 0.62) are considered good agreement

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/using-patient-reported-outcome-measures-to-classify-disease-activity-states-in-rheumatoid-arthritis-a-comparison-of-patient-derived-versions-of-clinical-disease-activity-index-cdai-simplified-dise/