Session Information
Date: Tuesday, November 10, 2015
Title: Health Services Research Poster III: Patient Reported Outcomes, Patient Education and Preferences
Session Type: ACR Poster Session C
Session Time: 9:00AM-11:00AM
Background/Purpose: Accurate
rheumatoid arthritis (RA) disease activity assessment is required for treatment
consistent with Treat-to-Target (T2T) recommendations. Rheumatologists
currently use a variety of patient-generated, physician-generated, and laboratory
data to assess RA disease activity (DA). Discordance among different clinical
measures, and more recently between clinical measures and a multi-biomarker
test, has been documented. This study investigates discordance among DA
measures in multiple clinical practice RA populations.
Methods: The
Rheumatoid Arthritis Practice Performance (RAPP) Project is a voluntary
collaboration of U.S. clinician rheumatologists whose goal is to provide T2T
care and optimal RA disease outcomes. Of the 168 participants in the RAPP
Project, 86 have enrolled their entire RA population (ICD-9 code 714.0) in a
HIPAA-compliant disease population registry and are tracking DA measures in the
registry. With isolated exceptions, their preferred measures include one or
more of the following: RAPID3, a 0-10 Provider Global Assessment (PGA),
Clinical Disease Activity Index (CDAI), and a multi-biomarker (MB) test. Table
1 reports the number of registries tracking each of these measures and the
distributions of controlled, low, moderate, and high DA for each measure using
the most recent assessment for each patient. Table 2 reports discordance rates
between different date-matched DA results from those registries entering more
than one measure, again using the most recent assessment for each patient.
Results:
|
Table 1. DA Distributions for Different Measures in 86 RAPP Practice Registries
|
||||
|
RAPID3 |
PGA |
CDAI |
MB |
|
|
Registries tracking (N) |
15 |
39 |
15 |
86 |
|
Controlled and Low DA (median %) |
52% |
74% |
55% |
22% |
|
Moderate DA (median %) |
22% |
21% |
27% |
39% |
|
High DA (median %) |
26% |
5% |
19% |
39% |
|
Table 2. Discordance among Paired DA Measures
|
|||
|
MB vs CDAI
|
MB vs PGA
|
MB vs RAPID3
|
|
|
Registries included (N) |
5 |
12 |
10 |
|
Measure pairs (N) |
746 |
2162 |
1845 |
|
Discordance (median %) |
44% |
41% |
33% |
|
Discordance range |
41-49% |
18-59% |
24-42% |
|
|
|
|
|
|
CDAI vs PGA
|
CDAI vs RAPID3
|
PGA vs RAPID3
|
|
|
Registries included (N) |
4 |
5 |
10 |
|
Measure pairs (N) |
1759 |
2239 |
3570 |
|
Discordance (median %) |
17% |
26% |
32% |
|
Discordance range |
15-18% |
15-41% |
20-71% |
Conclusion: 1. Different
measures provide differing DA distributions within RA populations and
discordant results for many individual patients. 2. Treatment decisions will
likely differ based on which measures are used. 3. Factors contributing to
discordance among measures at the population and patient levels appear to
include patients’ and providers’ variable subjective perceptions, variability
of joint examinations, and sub-clinical inflammation detected only by MB
testing.
To cite this abstract in AMA style:
Crump G, Bower J, Foley T, Harrington JT, Hegde N, Johnson D, Khalil R, LaCour E, Perhala R. Different Rheumatoid Arthritis Disease Activity Measures Often Provide Discordant Results in Clinical Practice Populations [abstract]. Arthritis Rheumatol. 2015; 67 (suppl 10). https://acrabstracts.org/abstract/different-rheumatoid-arthritis-disease-activity-measures-often-provide-discordant-results-in-clinical-practice-populations/. Accessed .« Back to 2015 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/different-rheumatoid-arthritis-disease-activity-measures-often-provide-discordant-results-in-clinical-practice-populations/