Introduction: The DAS28 score has now consolidated as a fundamental variable for the assessment of rheumatoid arthritis activity and is the main parameter used to establish therapeutic decisions in this disease, including the start and change of biologic therapies.
Objectives: We have studied the clinimetric properties of DAS28, including ceiling and floor effects and its behavior in several clinical scenarios.
Material and method: Individualized study of the variables included in the DAS28 formula along its possible range. Sensitivity analysis of the results of the DAS28 of 4 variables in four theoretical scenarios corresponding to low (DAS28=2.43), fair (DAS28=4.05), high (DAS28=6.32) or very high (DAS28=8.40) clinical activity.
Results: Tender joint count (NAD) and erithrosedimentation rate (ESR) have a weight of 35- 40% each on the total DAS28 score, while swollen join count (SJC) and global health assessed by the patient (GH) only contribute with 15% each. As tender joints weights double than swollen joints, in the simulation models having one swollen joint needed just 3 tender joints to get the DAS28 above the non remission level (DAS28>2.6), while having one tender joint needed 5 swollen joints to be above remission. Given its logarithmic calculation in the DAS28 formula, ESR contribution is much higher in its lower range, and thus small variations of ESR in the normal range can influence decisively in the final DAS28 score.
Conclusions: The asymmetric weight of each component in the complex DAS28 formula must be taken into account when interpreting changes in the DAS28 lower range as they influence the estimation of clinical remission and thus can be relevant when taking therapeutic decisions.
Copyright © 2008 Elsevier España S.L Barcelona. Published by Elsevier Espana. All rights reserved.