Abstract
Objective:
To provide updated American College of Rheumatology (ACR) recommendations on rheumatoid arthritis disease activity measurements to facilitate a treat-to-target approach in routine clinical care.
Methods:
A working group conducted a systematic literature review from the time of the prior ACR recommendations literature search. Properties of disease activity measures were abstracted, and study quality was assessed using the COnsensus-based Standards for the selection of health Measurement INinstruments (COSMIN) 4-point scoring method, allowing for overall level of evidence assessment. Measures that fulfilled a “minimum standard” were identified, and through a modified Delphi process “preferred” measures were selected for regular use in most clinic settings.
Results:
The search identified 5,199 articles, of which 110 were included in the review. This search identified 46 RA disease activity measures which contained patient, provider, lab, and/or imaging data. Descriptions of the measures, properties, study quality, level of evidence, and feasibility were abstracted and scored. Following a modified Delphi process, 11 measures fulfilled a “minimum standard” for regular use in most clinic settings, and five measures were recommended: 28-joint Disease Activity Score (DAS28-ESR/CRP), Clinical Disease Activity Index (CDAI), Simplified Disease Activity Index (SDAI), Routine Assessment of Patient Index Data 3 (RAPID3), and Patient Activity Scale-II (PAS-II).
Conclusion:
We have updated prior ACR recommendations for preferred RA disease activity measures, identifying 11 measures that met a minimum standard for regular use and 5 measures that are preferred for regular use in most clinic settings.
Keywords: rheumatoid arthritis, disease activity measures, psychometrics
Introduction
A treat-to-target strategy in rheumatoid arthritis (RA) is recommended by the 2015 American College of Rheumatology (ACR) RA Treatment Guidelines (1). In order to adhere to these recommendations, regular RA disease activity assessments must be made during routine care. While the severity of chronic diseases such as diabetes mellitus or hypertension can be directly measured, no equivalent measurement exists in RA. Numerous RA disease activity measures have been proposed for this purpose, most incorporating data gathered from a combination of sources including patient reported measures, provider assessments, laboratory values, and/or imaging modalities. These measures may vary in terms of their performance (e.g. validity, reliability, responsiveness) and feasibility for regular use.
Recognizing the challenge that clinicians face selecting a disease activity measure due to multiple options and varying performance, the ACR convened a working group in 2008 to review the literature and provide recommendations on which RA disease activity measures were best suited for regular use (2). RA disease activity measures were identified through a literature review (3), which were narrowed by an expert advisory panel. Recommendations were drafted after psychometric properties of the measures were compiled and practicing rheumatologists were surveyed. This process resulted in the recommendation of six RA disease activity measures: Clinical Disease Activity Index (CDAI), Disease Activity Score 28-joints (DAS28), Patient Activity Scale (PAS), Patient Activity Scale-II (PAS-II), Routine Assessment of Patient Index Data 3 (RAPID3), and Simplified Disease Activity Index (SDAI) (2).
Since these original recommendations, additional RA disease activity measures have been reported, further studies characterizing the performance of these and other novel measures have been conducted, and imaging modalities have been developed for assessment of disease activity. Therefore, an update to the prior recommendations for selecting a RA disease activity measure was needed, including a critical evaluation of more recent literature. The ACR convened a working group to update these prior recommendations in conjunction with a separate effort to provide initial recommendations on functional status assessment in RA. The objectives of this RA disease activity measures working group were to provide recommendations for 1) RA disease activity measures meeting a “minimum standard” for regular use and 2) preferred RA disease activity measures for regular use. The former objective was added since many measures may be valid, feasibility varies across different practices and healthcare systems, and providers may have comfort and experience with certain disease activity measures.
Methods
Study Design
A working group composed of rheumatologists and rheumatology professionals, including one afflicted with RA, was convened by the ACR to update the recommended RA disease activity measures. A protocol was developed and agreed upon by the working group for providing updated RA disease activity measure recommendations. The recommendation process and preliminary findings were presented in a special session at the 2017 ACR Annual Meeting (San Diego, CA) and opened for public comment (patients, providers, and other stakeholders) following this presentation.
Updated Systematic Literature Review
With a medical librarian (CMS) we updated the prior literature review by searching Ovid MEDLINE, Embase, and Cochrane databases from January 1, 2009 to January 25, 2017 for published original manuscripts on RA disease activity measures using combinations of MESH terms and keywords for rheumatoid arthritis, disease activity measures, and psychometric properties. We did not review components of composite measures individually as prior recommendations selected the composite measures over their individual components (2). A full description of the systematic literature review is available in Appendix 1. Systematic review inclusion criteria were published manuscripts in the English language reporting on a psychometric property of a RA disease activity measure. Exclusion criteria were reports limited to diseases other than RA, reports assessing only cross-cultural validity, radiographic damage, or a single joint area as well as measures not providing numerical values. Titles and abstracts were screened in duplicate (BRE, BKT) for relevance, followed by full text review in duplicate (BRE, BKT) to assess eligibility. Discordance after full text review was settled by a third-party reviewer (KM). Publications retrieved were reviewed to identify additional articles eligible for inclusion. RefWorks was utilized for management of literature search results.
Data Abstraction and Study Quality Assessment
Study details and psychometric properties were abstracted and study quality was assessed from included studies, using the Consensus-based standards for the selection of health measurement instruments (COSMIN) 4-point scoring as the template (4). An abstraction tool was developed and was piloted iteratively for data collection, then applied to the studies by an abstractor (BRE or BKT). To ensure abstraction consistency and quality, regular meetings occurred between the abstractors during the abstraction process.
Items abstracted from studies included those pertaining to the publication (author, year, journal), study (patient characteristics, sample size, setting, patient selection), disease activity measures (measures included, score distributions), and psychometric properties. Psychometric properties abstracted were internal consistency, reliability, measurement error, content validity, structural validity, hypotheses testing, and responsiveness (COSMIN properties (4, 5)). Criterion validity was not abstracted because considering a distinct RA disease activity measure a “gold-standard” would bias recommendations. Rather, studies reporting criterion validity were abstracted as hypothesis testing (i.e. convergent validity).
Study quality assessment for each psychometric property was assessed using the COSMIN checklist with 4-point scale (4). Using this method, each psychometric property reported in each study received a quality rating of Excellent, Good, Fair, or Poor. The score assigned to each property in each study represented the lowest score of all the criteria for that property.
Level of Evidence
Abstracted data on psychometric properties and study quality were synthesized as others have previously reported (6, 7). The psychometric properties for each RA disease activity measure received a level of evidence of Strong (+++ or −−−), Moderate (++ or −−), Limited (+ or −), Conflicting (±), or Unknown (?) (Appendix 2). Assessments of level of evidence was performed in duplicate (BRE, BKT) and discordance was settled by a third party (KM).
Consideration of prior literature
A literature review was previously performed to inform the 2012 ACR RA Disease Activity Recommendations (3). The psychometric properties of RA disease activity measures identified in the prior review were extracted according to the COSMIN groupings utilized in the current systematic review. Additionally, we searched for psychometric properties of studies not previously included in the prior literature review that were published before our search date. As study quality assessment was not part of the prior review, these results were not incorporated into the level of evidence grading with those from the current systematic review. Instead, these prior performance metrics were provided to the working group members for review during the selection (i.e. voting) process.
Feasibility
Validated scoring systems for the feasibility of RA disease activity measures do not currently exist. We scored feasibility from 0–4 (− to ++++) with scores ≥1 (+ to ++++) denoting measures feasible for regular use and scores of 4 representing the most feasible measures. The number of items included in the measure, time to complete, need for provider joint counts, need for laboratory testing, commercial availability, and need for advanced imaging were evaluated as part of the grading (Table 3 legend). All measures not commercially available or requiring advanced imaging (due to additional equipment, training, or consultation being required) were graded as 0 (-, not feasible for regular use). Requirement of provider joint counts or laboratory testing both reduced the maximum score by 1 each. Consideration of number of items and completion time served as final modifiers of the feasibility grade.
Table 3.
RADAM | # Items | Time | Provider Joint Count | Lab Required | Advanced imaging | Feasibility |
---|---|---|---|---|---|---|
Clinical Disease Activity Index (CDAI) | 3 | 2–5 mins | Yes | No | No | +++ |
Modified CDAI (Baker) | 2 | 5 mins | Yes | No | No | +++ |
Patient derived CDAI | 4 | 5 mins | No | No | No | ++++ |
Disease Activity Score (DAS) | 4 | 10 mins | Yes | Yes | No | + |
Disease Activity Score 28 joints (DAS28-ESR/CRP) | 3 or 4 | 5 mins + lab | Yes | Yes | No | ++ |
Modified DAS28 (Baker) | 3 | 5 mins + lab | Yes | Yes | No | ++ |
Modified DAS28 (no acute phase reactants, Bentley) | 6 | 5 mins | Yes | No | No | +++ |
Patient derived DAS28 | 4 | 5 mins + lab | No | Yes | No | +++ |
Ultrasound derived DAS28 | 4 | − | No | Yes | Yes | − |
Global arthritis score (GAS) | 3 | 5 mins | No | No | No | ++++ |
Hospital Universitario La Princesa Index (HUPI) | 4 | 5 mins + lab | Yes | Yes | No | ++ |
Individualized Ultrasound Score | Up to 7 or 12 | − | No | No | Yes | − |
Individualized Composite Ultrasound Score | Up to 7 or 12 | − | No | No | Yes | − |
Kappa/Lambda hybrid antibody | 1 | Not commercially available | No | Yes | No | − |
Mean Overall Index for RA (MOI-RA) | 7 | 10–20 mins + lab | Yes | Yes | No | + |
Multi-biomarker Disease Activity Score (MBDA, VECTRA) | 12 | Days | No | Yes | No | + |
Optical Spectral Transmission (OST) | 22 | Not commercially available | No | No | Yes | − |
Patient Activity Scale (PAS) | 3 | 5 mins | No | No | No | ++++ |
Patient Activity Scale-II (PAS-II) | 3 | 2 mins | No | No | No | ++++ |
Patient Based Disease Activity Score (PDAS1) | 4 | 5–10 mins + lab | No | Yes | No | +++ |
Patient Based Disease Activity Score (PDAS2) | 4 | 5–10 mins | No | No | No | ++++ |
Patient Reported Clinical Arthritis Activity (PRO-CLARA) | 3 | 5 mins | No | No | No | ++++ |
Rheumatoid Arthritis Disease Activity Index (RADAI) | 5 | 5 mins | No | No | No | ++++ |
Rheumatoid Arthritis Disease Activity Index 5 (RADAI-5) | 5 | 30 sec - 2 mins | No | No | No | ++++ |
Rheumatoid Arthritis MRI scoring (RAMRIS) | 3 | − | No | No | Yes | − |
Routine Assessment of Patient Index Data 3 (RAPID3) | 3 | 30 sec - 2 mins | No | No | No | ++++ |
Routine Assessment of Patient Index Data 4 (RAPID4) | 4 | 5–10 mins | No | No | No | ++++ |
Routine Assessment of Patient Index Data 5 (RAPID5) | 5 | 5–10 mins | No | No | No | ++++ |
Simplified Disease Activity Index (SDAI) | 5 | 2–5 mins + lab | Yes | Yes | No | ++ |
Modified SDAI (Baker) | 3 | 5 mins + lab | Yes | Yes | No | ++ |
Patient derived SDAI | 5 | 5 mins + lab | No | Yes | No | +++ |
Ultrasound derived SDAI | 5 | − | No | Yes | Yes | − |
Simplified RA MRI Score (SAMIS) | 3 | − | No | No | Yes | − |
Swiss Sonography in Arthritis and Rheumatism (SONAR) Score | 22 | 20–30 mins | No | No | Yes | − |
Ultrasound 6 joint (Perricone) | 6 | 14 mins | No | No | Yes | − |
Ultrasound 6 joint (Rosa) | 6 | 5–12 mins | No | No | Yes | − |
Ultrasound 6 joint (Kawashiri) | 6 | − | No | No | Yes | − |
Ultrasound 7 joint (Backhaus) | 7 | 10–20 mins | No | No | Yes | − |
Ultrasound 8 joint (Yoshimi) | 8 | − | No | No | Yes | − |
Ultrasound 12 joint (Naredo) | 12 | 20–25 mins | No | No | Yes | − |
Ultrasound 14 joint (Dale) | 14 | − | No | No | Yes | − |
Ultrasound 20 joint (Dougados) | 20 | − | No | No | Yes | − |
Ultrasound 28 joint (Dougados) | 28 | − | No | No | Yes | − |
Ultrasound 38 joint (Dougados) | 38 | − | No | No | Yes | − |
Ultrasound 78 joint (Hammer) | 78 | − | No | No | Yes | − |
Ultrasound score A and B (Aga) | A. 18, B. 22 | − | No | No | Yes | − |
- Measures started with a score of “++++”
- Any measure not commercially available or requiring advanced imaging was graded “−”
- Requiring a provider joint count reduced feasibility by “+”
- Requiring a laboratory test reduced feasibility by “+”
- Number of items and time to completion were considered and score was reduced by “+” if not feasibile in a routine clinic visit or by “++” if not feasible on the same day as clinic visit.
Selection Process
The RA disease activity measures working group reviewed the literature search, abstracted data, level of evidence for each identified measure, prior literature for each measure, and feasibility scoring as well as their experience with these measures to provide two recommendations on RA disease activity measures feasible for regular use in rheumatology clinics. First, the group identified RA disease activity measures meeting a minimum standard for regular use and second, the group selected measures with the most favorable psychometric properties and feasibility for preferred use.
Fulfilling the minimum standard for a RA disease activity measure in regular use was established by measures 1) providing a numerical value, 2) categorizing to ≥3 disease states which separate low, moderate, and high disease activity, 3) being feasible for regular measurement in clinic and 4) possessing adequate psychometric properties. Items were considered to meet the minimum standard for feasibility in regular use if the aforementioned feasibility score was ≥1. Psychometrics were considered adequate if the level of evidence suggested at least moderate positive results in the COSMIN area of hypothesis testing plus one of the following: level of evidence suggesting at least moderate positive results in another COSMIN area, level of evidence suggesting at least limited positive results in at least two COSMIN areas (one of which must be responsiveness), or a defined minimum important difference/minimum clinically important difference.
A modified Delphi process was utilized to generate the recommendations on RA disease activity measures for preferred use (8). Working group members and an ACR Quality Measures Subcommitee Liason (n=10, Appendix) rated each measure that fulfilled the minimum standard on a scale of 1 to 9, with 9 being “essential this measure be recommended for use”. Ratings of 7–9 constituted a recommended measure for inclusion, while ratings of 4–6 were inconclusive and ratings of 1–3 were recommended measures for exclusion. Measures were recommended if >80% of members (all but 1) rated the measure in the 7–9 range and excluded if >80% of ratings were in the 1–3 range, following best practices (9). The voting process continued iteratively to a maximum of three voting cycles with discussion of RA disease activity measures not fulfilling agreement between voting cycles. Measures not achieving recommendation for inclusion or exclusion were deemed inconclusive. Measures deemed inconclusive remained on the list fulfilling the minimum standard.
The ACR Quality Measures Subcommittee reviewed these recommendations in parallel with the recommendations on functional status assessment, modifying as necessary based upon the goal of identifying preferred tools for regular use in most clinic settings, before voting. The Quality of Care Committee and ACR Board reviewed and approved this manuscript prior to publication.
Results
Systematic Literature Review & Identified Disease Activity Measures
Our systematic literature review identified 5,199 articles (Appendix 3). After screening titles, abstracts, and full texts, 104 manuscripts met criteria for inclusion into the study. Review of retrieved publications identified an additional 6 manuscripts fulfilling eligibility criteria, resulting in a total of 110 included studies. There was 98.2% agreement between reviewers for study inclusion. Characteristics of the individual studies are provided in Appendix 4. The majority of studies were female predominant, with mean age in the 6th decade. Sample sizes, mean DAS28, location, design, and selection varied between studies.
Our search identified 47 RA disease activity measures. The components, number of items, scoring method, score range, disease activity category cut-offs, method of administration, and minimum important difference/minimum clinically important difference of each RA disease activity measure are listed in Table 1. A Venn diagram illustrating the components (e.g. patient reported, provider assessment, laboratory values, imaging modalities) of the identified RA disease activity measures is shown in Figure 1.
Table 1.
RADAM | Components | # Items | Formula | Range | Rem | LDA | MDA | High | Method of Administration | MID, MCID |
---|---|---|---|---|---|---|---|---|---|---|
Clinical Disease Activity Index (CDAI) | 28TJC: 0–28, 28SJC: 0–28, Pt Global VAS: 0–10, Pr Global VAS: 0–10 | 4 | 28SJC + 28TJC + PtGA + PrGA | 0–76 | ≤2.8 | >2.8 to 10 | >10 to 22 | >22 | Patient item, Provider assessment | MCID: 12; 12 (HDA), 6 (MDA), 1 (LDA) |
Modified CDAI (Baker) | 28SJC, Pr Global | 2 | 28SJC + 2xPrGA | 0–48 | - | - | - | - | Provider assessment | - |
Patient derived CDAI | Patient 28TJC, Patient 28SJC, Pt Global, Pr Global | 4 | 28SJC + 28TJC + PtGA + PrGA | 0–76 | ≤2.8 | >2.8 to 10 | >10 to 22 | >22 | Patient items, Provider assessment | - |
Disease Activity Score (DAS) | Ritchie Articular Index, SJC44, ESR, Pt Global | 4 | 0.53938xSqrt(RAI) + 0.06465xSJC44 + 0.33ln(ESR) + 0.00722(PtGA) | 0–10 | <1.6 | 1.6 to <2.4 | 2.5 to <3.7 | ≥3.7 | Patient item, Provider assessment, Lab | 1.2 |
Disease Activity Score 28 joints (DAS28) | 28TJC: 0–28, 28SJC: 0–28, Pt Global VAS: 0–10, ESR or CRP | 3 or 4 | 0.56xSqrt(28TJC) + 0.28xSqrt(28SJC) + 0.70xln(ESR) + 0.014xPtGA -OR- 0.56xSqrt(28TJC) + 0.28xSqrt(28SJC) + 0.36xln(CRP + 1) + 0.014xPtGA + 0.96 |
0–9.4 | <2.6 | 2.6 to <3.2 | 3.2 to ≤5.1 | >5.1 | Patient item, Provider assessment, Lab | MID 1.2; MCID 1.2 (DAS28-ESR), 1.0 (DAS28-CRP) |
Modified DAS28 (Baker) | 28SJC, Pr Global, acute phase reactant (ESR or CRP) | 3 | 0.40xln(ESR) + 0.17xSJC28 + 0.26xPrGA -OR- 0.49xln(CRP) + 0.15×28SJC + 0.22PrGA + 1 |
- | - | - | - | - | Provider assessment, Lab | - |
Modified DAS28 (no acute phase reactants, Bentley) | 28TJC, 28SJC, mHAQ, Pain, Pr Global, Pt Global | 6 | 0.53xsqrt(28TJC) + 0.31xsqrt(28SJC) + 0.25xmHAQ + 0.001xPain + 0.005xPrGA + 0.014xPtGA + 1.694 | - | - | - | - | - | Patient items, Provider assessment | - |
Patient derived DAS28 | Patient 28TJC, Patient 28SJC, Patient Global, ESR or CRP | 4 | 0.56xSqrt(28TJC) + 0.28xSqrt(28SJC) + 0.7xln(ESR) + 0.014xPtGA -OR- 0.56xSqrt(28TJC) + 0.28xSqrt(28SJC) + 0.36xln(CRP + 1) + 0.014xPtGA + 0.96 |
0–9.4 | <2.6 | 2.6 to <3.2 | 3.2 to ≤5.1 | >5.1 | Patient items, Lab | - |
Ultrasound derived DAS28 | Pt/Pr 28TJC, Ultrasound 28SJC, Pt Global, ESR or CRP | 4 | 0.56xSqrt(28TJC) + 0.28xSqrt(28SJC) + 0.7xln(ESR) + 0.014xPtGA -OR- 0.56xSqrt(28TJC) + 0.28xSqrt(28SJC) + 0.36xln(CRP + 1) + 0.014xPtGA + 0.96 |
0–9.4 | <2.6 | 2.6 to <3.2 | 3.2 to ≤5.1 | >5.1 | Patient items, Provider assessment, Lab, Imaging modality | - |
Global arthritis score (GAS) | Pain VAS (0–10), mHAQ (0–24), patient reported 28TJC (0–28) | 3 | Pain + mHAQ + Pt 28TJC | 0–62 | - | - | - | - | Patient items | - |
Hospital Universitario La Princesa Index (HUPI) | 28TJC, 28SJC, Pt global, acute phase reactant (ESR or CRP) | 4 | Each component scored 0–3 based on cut-off values | 0–12 | - | ≤2 | >2 to ≤5 | >5 | Patient items, Provider assessment, Lab | 4 (~DAS1.2) |
Individualized Ultrasound Score | Selects up to 7 or 12 of most affected joints for MSUS | Up to 7 or 12 | Sum of individual joint scores (sum of US subscores divided by max score at the joint - gray scale synovial hypertrophy, PD vascularity, tenosynovitis GS & PD) | - | - | - | - | - | Imaging modality | - |
Individualized Composite Ultrasound Score | Selects up to 7 or 12 of most affected joints for MSUS and clinical exam | Up to 7 or 12 | Sum of individual joint scores (sum of US subscores and joint subscores divided by max score at the joint) | - | - | - | - | - | Provider assessment, Imaging modality | - |
Kappa/Lambda hybrid antibody | Antibody measurement | 1 | Values generated referent to a standard curve | - | - | - | - | - | Lab | - |
Mean Overall Index for RA (MOI-RA) | 28SJC, 28TJC, Pt Global, Pr Global, Pain, HAQ, ESR | 7 | Mean of standardized values of individual components (each 0–100) | 0–100 | - | - | - | - | Patient item, Provider assessment, Lab | - |
Multi-biomarker Disease Activity (MBDA) score | CRP, EGF, IL-6, Leptin, MMP-1, MMP-3, Resistin, SAA, TNF-RI, VCAM-1, VEGF-A, YKL-40 | 12 | (0.56sqrt(TJC) + 0.28SJC + 0.14PtGA + 0.36log(CRP+1) + 0.96) x 10.53 + 1 (Biomarker scores to predict above components) | 1–100 | ≤25 | 26–29 | 30–44 | >44 | Lab test | - |
Optical Spectral Transmission (OST) | Bilateral PIP1–5, MCP1–5, wrist | 22 | Not reported | - | - | - | - | - | Imaging modality | - |
Patient Activity Scale (PAS) | HAQ: 0–3, Pain VAS: 0–10, Pt Global VAS: 0–10 | 3 | (HAQx3.33 + Pain VAS + PtGA VAS)/3 | 0–10 | ≤0.25 | >0.26 to 3.70 | 3.71 to <8.0 | ≥8.0 | Patient items | - |
Patient Activity Scale-II (PAS-II) | HAQ-II: 0–3, Pain VAS: 0–10, Pt Global VAS: 0–10 | 3 | (HAQ-IIx3.33 + Pain VAS + PtGA VAS)/3 | 0–10 | ≤0.25 | >0.26 to 3.70 | 3.71 to <8.0 | ≥8.0 | Patient items | - |
Patient Based Disease Activity Score (PDAS1) | Patient global, Patient 50TJC, HAQ, ESR | 4 | 0.19xPtGA + 0.842ln(ESR+2) + 0.432xln(PtTJC +2) + 0.271xHAQ | - | <3.5 | 3.5–4.5 | 4.5–4.8 | >4.8 | Patient items, Lab | 0.8 good response |
Patient Based Disease Activity Score (PDAS2) | Patient global, Patient 28SJC, HAQ, morning stiffness duration | 4 | 2.667 + 0.021xPtGA + 0.483xHAQ + 0.033xPtSJC + 0.002xAM stiffness | - | <3.8 | 3.8–4.6 | 4.6–5.0 | >5.0 | Patient items | 1.2 good response |
Patient Reported Clinical Arthritis Activity (PRO-CLARA) | Recent Onset Arthritis Disability questionnaire (0–10), Patient 16TJC (0–10), Patient global (0–10) | 3 | ROAD + Pt TJC + PtGA /3 | 0–10 | - | - | - | - | Patient items | - |
Rheumatoid Arthritis Disease Activity Index (RADAI) | Pt global: 0–10, Current swollen/tender joints: 0–10, Pain: 0–10, Duration morning stiffness: 0–10 (transformed), Tender joint list: 0–10 (transformed) | 5 | (PtGA + S/T + Pain + AM stiff + TJC) / 5 | 0–10 | - | <2.2 | ≥2.2 to ≤4.9 | >4.9 | Patient items | 1–1.4 |
Rheumatoid Arthritis Disease Activity Index 5 (RADAI-5) | Patient global 6 mos: 0–10, Patient active joint S/T today: 0–10, Pain: 0–10, Patient general health: 0–10, AM stiffness: 0–10 | 5 | (PtGA + Pt S/T joints + Pain + Patient GH + AM stiff) / 5 | 0–10 | ≤1.4 | 1.6–3.0 | 3.2–5.4 | ≥5.6 | Patient items | - |
Rheumatoid Arthritis MRI scoring (RAMRIS) | Synovitis (7 areas scored 0–3), Osteitis/Bone Edema (23 areas scored 0–3), Erosion (23 areas scored 0–10) | 3 | Subcomponents or Total score (sum) | - | - | - | - | - | Imaging modality | - |
Routine Assessment of Patient Index Data 3 (RAPID3) | MDHAQ: 0–10, Pain VAS: 0–10, Pt Global VAS: 0–10 | 3 | MDHAQ + Pain VAS + PtGA VAS | 0–30 | ≤3 | 4–6 | 7–12 | ≥13 | Patient items | MID 3.2–3.6 |
Routine Assessment of Patient Index Data 4 (RAPID4) | MDHAQ: 0–10, Pain VAS: 0–10, Pt Global VAS: 0–10, RADAI tender joint list: 0–10 | 4 | MDHAQ + Pain + Pt Global + RADAI-TJL | 0–40 | ≤4 | 5–8 | 9–16 | ≥17 | Patient items | - |
Routine Assessment of Patient Index Data 5 (RAPID5) | MDHAQ: 0–10, Pain VAS: 0–10, Pt Global VAS: 0–10, RADAI tender joint list:0–10, Pr Global: 0–10 | 5 | MDHAQ + Pain + Pt Global + RADAI-TJL + Pr Global | 0–50 | ≤5 | 6–10 | 11–20 | ≥21 | Patient items, Provider assessment | - |
Simplified Disease Activity Index (SDAI) | 28TJC: 0–28, 28SJC: 0–28, Pt Global VAS: 0–10, Pr Global VAS: 0–10, CRP: 0–10 | 5 | 28SJC + 28TJC + PtGA + PrGA + CRP | 0–86 | ≤3.3 | >3.3 to ≤11.0 | >11.0 to ≤26 | >26 | Patient item, Provider assessment, Lab | 16 ~ DAS MID1.2; MCID 13 |
Modified SDAI (Baker) | 28SJC, Pr Global, CRP | 3 | CRP + 28SJC + PrGA | - | - | - | - | - | Provider assessment, Lab | - |
Patient derived SDAI | Patient 28TJC, Patient 28SJC, Pt Global VAS, Pr Global VAS, CRP | 5 | 28SJC + 28TJC + PtGA + PrGA + CRP | 0–86 | ≤3.3 | >3.3 to ≤11.0 | >11.0 to ≤26 | >26 | Patient items, Provider assessment, Lab | - |
Ultrasound derived SDAI | 28TJC, Ultrasound 28SJC, Pt Global VAS, Pr Global VAS, CRP | 5 | 28SJC + 28TJC + PtGA + PrGA + CRP | 0–86 | ≤3.3 | >3.3 to ≤11.0 | >11.0 to ≤26 | >26 | Patient item, Provider assessment, Lab, Imaging modality | - |
Simplified RA MRI Score (SAMIS) | Synovitis (7 areas scored 0–2), Osteitisis/Bone Edema (15 areas scored 0–1), Erosion (15 areas scored 0–10) | 3 | Subcomponents or Total score (sum) | - | - | - | - | - | Imaging modality | - |
Swiss Sonography in Arthritis and Rheumatism (SONAR) Score | Bilateral elbow, wrist, MCP2–5, PIP 2–5, knee. B mode (0–3) and PD (0–3) for each joint. | 22 | Sum of individual scores (PD and B mode) | 0–66 B-mode, 0–66 PD | - | - | - | - | Imaging modality | - |
Ultrasound 6 joint – Perricone | Bilateral wrist, MCP2, knee. Synovial effusion (0–3), synovial proliferation (0–3), PD (0–3) | 6 | Sum of individual scores | 0–54 | - | - | - | - | Imaging modality | - |
Ultrasound 6 joint – Rosa | Bilateral wrist, MCP2, MCP3 (0–2) | 6 | Sum of individual scores | 0–12 | - | - | - | - | Imaging modality | - |
Ultrasound 6 joint – Kawashiri | Bilateral wrist, MCP2, MCP3 (0–3) | 6 | Sum of individual scores | 0–18 | - | - | - | - | Imaging modality | - |
Ultrasound 7 joint (US-7, Backhaus) | Unilateral (dominant side) wrist, MCP2, MCP3, PIP2, PIP3, MTP2, MTP5. Synovitis PDUS (0–3), Synovitis GSUS (0–3), Tenosynovitis GSUS (0–1), Tenosynovitis PDUS (0–3), erosion. | 7 | Sum of individual scores | S-GSUS 0–27, S-PDUS 0–39, TS-GSUS 0–7, TS-PDUS 0–21, E 0–14 | - | - | - | - | Imaging modality | - |
Ultrasound 8 joint (US-8, Yoshimi) | Bilateral wrist, MCP2, MCP3, knees (0–3 PDUS) | 8 | Sum of individual scores | 0–24 | - | - | - | - | Imaging modality | - |
Ultrasound 12 joint (US-12, Naredo) | Bilateral elbow, wrist, MCP2, MCP3, knee, ankle (synovitis 0–3, PD 0–3) | 12 | Sum of individual scores, *alternative: sum of # joints | - | - | - | - | - | Imaging modality | - |
Ultrasound 14 joint (US-14, Dale) | Bilateral wrist, MCP2, MCP3, PIP2, PIP3, MTP, MTP5. GSUS 0–3, PDUS 0–3. | 14 | - | - | - | - | - | - | Imaging modality | - |
Ultrasound 20 joint (Dougados) | Bilateral MCP1–5, MTP1–5 | 20 | 0–3 each joint for B-mode and PD | - | - | - | - | - | Imaging modality | - |
Ultrasound 28 joint (Dougados) | Bilateral shoulders, elbows, wrists, MCPs, PIPs, knees | 28 | 0–3 each joint for B-mode and PD | - | - | - | - | - | Imaging modality | - |
Ultrasound 38 joint (Dougados) | Bilateral shoulders, elbows, wrists, MCPs, PIPs, knees, bilateral MTPs | 38 | 0–3 each joint for B-mode and PD | - | - | - | - | - | Imaging modality | - |
Ultrasound 78 joint (Hammer) | B-mode and power doppler: Bilateral PIP1–5, MCP1–5, CMC1–5, wrist (radiocarpal, intercarpal, radioulnar), elbow (ant, post), shoulder (GHu, AC), hip, knee, ankle, foot (talonavicular, subtalar, calcaneocuboidal, cuneonaviicular), TMT 1–5, MTP 1–5, first IP | 78 | 0–3 each joint for B-mode and PD | B-Mode: 0–234, PD: 0–234 | - | - | - | - | Imaging modality | - |
Ultrasound score A and B (Aga) | Bilateral GSUS and PDUS of (A). MCP1, MCP2, PIP3, radiocarpal, elbow, MTP1, MTP2, TP tendon and ECU tendon. (B). A joints + MCP5, MTP5. Each location graded 0–3. | A. 18, B. 22 | Sum of individual scores | A. 0–54, B. 0–66 | - | - | - | - | Imaging modality | - |
Abbreviations: AC, acromioclavicular joint; CMC, carpometacarpal joint; CRP, C-reactive protein; ECU, extensor carpi ulnaris; ESR, erythrocyte sedimentation rate; GH, global health; GHu, glenohumeral joint; GS, gray scale; GSUS, gray scale ultrasound; HAQ, health assessment questionnaire; HAQ-II, health assessment questionnaire II; HDA, high disease activity; LDA, low disease activity; MCID, minimum clinically important difference; MCP, metacarpophalangeal joint; MDA, moderate disease activity; MDHAQ, multidimensional health assessment questionnaire; mHAQ, modified health assessment questionnaire; MID, minimum important difference; MSUS, musculoskeletal ultrasound; MTP, metatarsophalangeal joint; PD, power Doppler; PDUS, power Doppler ultrasound; PIP, proximal interphalangeal joint; PrGA, provider global assessment; PtGA, patient global assessment; RADAM, rheumatoid arthritis disease activity measure; RAI, Ritchie articular index; Rem, remission; ROAD, recent onset arthritis disability questionnaire; SJC, swollen joint count; TJC, tender joint count; TMT, tarsometatarsal joint; TP, tibialis posterior; VAS, visual analogue scale
Properties of RA Disease Activity Measures
Individual performance of RA disease activity measures in each study are provided in Appendix 5 and study quality assessment using the COSMIN checklist with 4-point scale is provided in Appendix 6. Based on both the measure performance and study quality, overall level of evidence was generated for each psychometric property for each RA disease activity measure (Table 2). This process was completed in duplicate with 96.6% agreement between raters in assessing overall level of evidence for RA disease activity measures.
Table 2.
RADAM | Internal Consistency | Reliability | Measurement Error | Content Validity | Structural Validity | Hypotheses testing | Responsiveness |
---|---|---|---|---|---|---|---|
Clinical Disease Activity Index (CDAI) | + | ++ | + | ++ | +++ | +++ | |
Modified CDAI (Baker) | ++ | ||||||
Patient derived CDAI | + | + | + | ||||
Disease Activity Score (DAS) | ++ | ||||||
Disease Activity Score 28 joints (DAS28) | ++ | ++ | ++ | −−− | ++ | +++ | +++ |
Modified DAS28 (Baker) | ++ | ||||||
Modified DAS28 (Bentley) | ? | ++ | ++ | ||||
Patient derived DAS28 | ++ | + | ++ | + | |||
Ultrasound derived DAS28 | + | ++ | + | ||||
Global Arthritis Score (GAS) | ++ | ++ | |||||
Hospital Universitario La Princesa Index (HUPI) | ? | ++ | ++ | ||||
Individualized Ultrasound Score | ? | ||||||
Individualized Composite Ultrasound Score | ? | ||||||
Kappa/Lambda hybrid antibody | + | ||||||
Mean Overall Index for RA (MOI-RA) | + | + | |||||
Multi-biomarker Disease Activity Score (MBDA) | ? | +++ | +++ | ++ | ++ | ||
Optical Spectral Transmission (OST) | ++ | ||||||
Patient Activity Scale (PAS) | + | + | |||||
Patient Activity Scale-II (PAS-II) | |||||||
Patient Based Disease Activity Score (PDAS1) | + | + | |||||
Patient Based Disease Activity Score (PDAS2) | + | + | |||||
Patient Reported Clinical Arthritis Activity (PRO-CLARA) | + | ? | ++ | + | |||
Rheumatoid Arthritis Disease Activity Index (RADAI) | ? | ? | ++ | ++ | |||
Rheumatoid Arthritis Disease Activity Index 5 (RADAI-5) | ? | ++ | ++ | ||||
Rheumatoid Arthritis MRI scoring (RAMRIS) | −− | − | |||||
Routine Assessment of Patient Index Data 3 (RAPID-3) | ? | ? | ? | +++ | +++ | +++ | |
Routine Assessment of Patient Index Data 4 (RAPID-4) | ++ | + | |||||
Routine Assessment of Patient Index Data 5 (RAPID-5) | ++ | ++ | |||||
Simplified Disease Activity Index (SDAI) | + | ++ | + | +++ | +++ | ||
Modified SDAI (Baker) | ++ | ||||||
Patient derived SDAI | + | + | ++ | ||||
Ultrasound derived SDAI | + | ++ | + | ||||
Simplified RA MRI Score (SAMIS) | + | ? | |||||
Swiss Sonography in Arthritis and Rheumatism (SONAR) Score | ? | ? | ++ | ++ | |||
Ultrasound 6 joint (Perricone) | +++ | + | + | ||||
Ultrasound 6 joint (Rosa) | + | ||||||
Ultrasound 6 joint (Kawashiri) | ? | ||||||
Ultrasound 7 joint (Backhaus) | ++ | +++ | ++ | ++ | |||
Ultrasound 8 joint (Yoshimi) | +++ | ++ | |||||
Ultrasound 12 joint (Naredo) | + | +++ | + | + | |||
Ultrasound 14 joint (Dale) | ? | ||||||
Ultrasound 20 joint (Dougados) | ? | + | ++ | ||||
Ultrasound 28 joint (Dougados) | ? | + | ++ | ||||
Ultrasound 38 joint (Dougados) | ? | + | ++ | ||||
Ultrasound 78 joint (Hammer) | ? | ? | ? | ||||
Ultrasound Score A and B (Aga) | +++ | + | + |
Level of evidence grading:
Strong, +++ or −−−, Consistent findings in multiple studies of good methodological quality –OR- in one study of excellent methodological quality
Moderate, ++ or −−, Consistent findings in multiple studies of fair methodological quality –OR- in one study of good methodological quality
Limited, + or −, One study of fair methodological quality
Conflicting, ±, Conflicting findings
Unknown, ?, Studies only of poor methodological quality
Hypothesis testing (testing hypotheses regarding relationships to other instruments measuring similar constructs, i.e. content validity) was the most frequently assessed psychometric property. Reliability and responsiveness were also frequently assessed for RA disease activity measures. CDAI, DAS28, Multibiomarker Disease Activity (MBDA) score, RAPID3, and SDAI were the most frequently studied RA disease activity measures. While negative content validity was reported for the DAS28, it should be noted this was based on one study of excellent quality which showed underestimation of radiographic progression in the feet, joints not included in the 28-joint count (10).
Properties of RA disease activity measures from before the current search period were collected from the prior review (3) and hand searches for measures not previously included (Appendix 7). A full reference list of all articles identified and abstracted in the systematic literature review as well as searches for earlier time periods is available in Appendix 8.
Feasibility of RA Disease Activity Measures
Feasibility scoring of the RA disease activity measures is shown in Table 3. Twenty-five measures were scored to be feasible for regular use in most clinics. Of these measures 44% (n=11) were scored 4 (++++), 25% (n=6) were scored 3 (+++), 20% (n=5) were scored 2 (++), and 12% (n=3) were scored 1 (+).
Recommended RA Disease Activity Measures
Eleven measures fulfilled the minimum standard defined for RA disease activity measures for regular use (Table 4). Four were part of the prior ACR RA disease activity measure recommendations: CDAI, DAS28-ESR/CRP, RAPID3, and SDAI. Of the seven not in the original recommendations, the Disease Activity Score (DAS) was a predecessor to the DAS28, the patient derived DAS28 was derived from the DAS28, and the Routine Assessment of Patient Index Data 5 (RAPID5) was related to the RAPID3. The remaining measures were the Hospital Universitario La Princesa Index (HUPI), MBDA score, Rheumatoid Arthritis Disease Activity Index (RADAI), and RADAI-5. Of the 36 measures not fulfilling the minimum standard, 75% (n=27) did not categorize into disease activity states, 78% (n=28) did not have adequate psychometrics, and 61% (n=22) were not scored as feasible for regular use (Table 4).
Table 4.
RA Disease Activity Measure | Numeric | Categorizes 3–4 states | Feasible | Adequate psychometrics | Meet Minimum Standard |
---|---|---|---|---|---|
Fulfilled minimum standard for regular use criteria | |||||
Clinical Disease Activity Index (CDAI) | + | + | + | + | + |
Disease Activity Score (DAS) | + | + | + | + | + |
Disease Activity Score 28 joints (DAS28-ESR/CRP) | + | + | + | + | + |
Patient derived DAS28 | + | + | + | + | + |
Hospital Universitario La Princesa Index (HUPI) | + | + | + | + | + |
Multi-biomarker Disease Activity Score (MBDA score, VECTRA DA) | + | + | + | + | + |
Rheumatoid Arthritis Disease Activity Index (RADAI) | + | + | + | + | + |
Rheumatoid Arthritis Disease Activity Index 5 (RADAI-5) | + | + | + | + | + |
Routine Assessment of Patient Index Data 3 (RAPID3) | + | + | + | + | + |
Routine Assessment of Patient Index Data 5 (RAPID5) | + | + | + | + | + |
Simplified Disease Activity Index (SDAI) | + | + | + | + | + |
Did NOT fulfill minimum standard for regular use criteria | |||||
Modified CDAI (Baker) | + | − | + | − | − |
Patient derived CDAI | + | + | + | − | − |
Modified DAS28 (Baker) | + | − | + | − | − |
Modified DAS28 (Bentley) | + | − | + | + | − |
Ultrasound derived DAS28 | + | + | − | + | − |
Global arthritis score (GAS) | + | − | + | + | − |
Individualized Ultrasound Score | + | − | − | − | − |
Individualized Composite Ultrasound Score | + | − | − | − | − |
Kappa/Lambda hybrid antibody | + | − | − | − | − |
Mean Overall Index for RA (MOI-RA) | + | − | + | − | − |
Optical Spectral Transmission (OST) | + | − | − | − | − |
Patient Activity Scale (PAS) | + | + | + | − | − |
Patient Activity Scale-II (PAS-II) | + | + | + | − | − |
Patient Based Disease Activity Score (PDAS1) | + | + | + | − | − |
Patient Based Disease Activity Score (PDAS2) | + | + | + | − | − |
Patient Reported Clinical Arthritis Activity (PRO-CLARA) | + | − | + | + | − |
Rheumatoid Arthritis MRI scoring (RAMRIS) | + | − | − | − | − |
Routine Assessment of Patient Index Data 4 (RAPID4) | + | + | + | − | − |
Modified SDAI (Baker) | + | − | + | − | − |
Patient derived SDAI | + | + | + | − | − |
Ultrasound derived SDAI | + | + | − | + | − |
Simplified RA MRI Score (SAMIS) | + | − | − | − | − |
Swiss Sonography in Arthritis and Rheumatism (SONAR) Score | + | − | − | + | − |
Ultrasound 6 joint (Perricone) | + | − | − | − | − |
Ultrasound 6 joint (Rosa) | + | − | − | − | − |
Ultrasound 6 joint (Kawashiri) | + | − | − | − | − |
Ultrasound 7 joint (Backhaus) | + | − | − | + | − |
Ultrasound 8 joint (Yoshimi) | + | − | − | + | − |
Ultrasound 12 joint (Naredo) | + | − | − | − | − |
Ultrasound 14 joint (Dale) | + | − | − | − | − |
Ultrasound 20 joint (Dougados) | + | − | − | − | − |
Ultrasound 28 joint (Dougados) | + | − | − | − | − |
Ultrasound 38 joint (Dougados) | + | − | − | − | − |
Ultrasound 78 joint (Hammer) | + | − | − | − | − |
Ultrasound score A and B (Aga) | + | − | − | − | − |
Measures deemed feasible if feasibility scoring was ≥1 (Table 3). Measures were considered to have adequate psychometrics if the level of evidence suggested at least moderate positive results in the COSMIN area of hypothesis testing plus had ≥1 of the following: level of evidence suggesting at least moderate positive results in another COSMIN area, level of evidence suggesting at least limited positive results in ≥2 COSMIN areas (one of which must be responsiveness), or a defined minimum important difference/minimum clinically important difference.
Results of the modified Delphi voting process are shown in Table 5. Four measures achieved consensus for preferred use: CDAI, DAS28, RAPID3, SDAI. CDAI (mean score 8.8) and SDAI (mean score 7.6) achieved consensus during the first round of voting, RAPID3 (mean score 7.6) during the second round of voting, and DAS28 (mean score 7.6) during the third round of voting. The remaining 7 RA disease activity measures (mean score range 2.6–5.6) did not achieve consensus after the third round of voting and were deemed “inconclusive” for preferred use.
Table 5.
Round 1 | Round 2† | Round 3†† | Final Recommendation | ||||
---|---|---|---|---|---|---|---|
Mean | n 1–3/4–6/7–9* |
Mean | n 1–3/4–6/7–9* |
Mean | n 1–3/4–6/7–9* |
||
Clinical Disease Activity Index (CDAI) | 8.8 | 0/0/10 | N/A | Recommended | |||
Simplified Disease Activity Index (SDAI) | 7.6 | 0/1/9 | N/A | Recommended | |||
Routine Assessment of Patient Index Data 3 (RAPID3) | 7.4 | 0/3/7 | 7.6 | 0/1/7 | N/A | Recommended | |
28-joint Disease Activity Score (DAS28-ESR/CRP) | 7.6 | 0/2/8 | 7.1 | 0/2/6 | 7.6 | 1/0/9 | Recommended |
Rheumatoid Arthritis Disease Activity Index-5 (RADAI-5) | 6.1 | 4/2/4 | 5.3 | 2/4/2 | 5.6 | 2/4/3 | Inconclusive |
Disease Activity Score (DAS) | 5.0 | 3/4/3 | 3.8 | 5/2/1 | 4.2 | 4/5/1 | Inconclusive |
Patient Derived-DAS28 | 4.9 | 4/2/4 | 4.5 | 2/6/0 | 4.2 | 4/6/0 | Inconclusive |
Rheumatoid Arthritis Disease Activity Index (RADAI) | 5.1 | 4/3/3 | 4.2 | 5/2/1 | 4.4 | 4/5/1 | Inconclusive |
Routine Assessment of Patient Index Data 5 (RAPID5) | 5.2 | 4/1/5 | 4.5 | 2/5/1 | 3.8 | 5/3/1 | Inconclusive |
Multibiomarker Disease Activity (MBDA) score | 4.2 | 7/1/2 | 3.5 | 5/2/1 | 3.2 | 7/1/1 | Inconclusive |
Hospital Universitario La Princesa Index (HUPI) | 4.0 | 6/1/3 | 3.5 | 5/3/0 | 2.6 | 8/2/0 | Inconclusive |
Ratings on the 1–9 Likert scale correspond to the following 1–3 “not recommended”; 4–6 “sometimes recommended”; 7–9 “essential to have” and >80% agreement required for recommendation
Eight voters participated in round 2 voting
One missing vote for RADAI-5, RAPID5, MBDA score.
The ACR Quality Measures Subcommittee approved the aforementioned recommendations with a single modification, the additional recommendation of PAS-II. This recommendation was based upon PAS-II feasibility, current use, strength of its inclusion in prior ACR recommendations that included evidence not captured in this current work, and alignment with the concurrent functional status assessment project (2).
Discussion
Patient outcomes in RA, including physical function, quality of life, and achieving remission/low disease activity, have improved as a result of treatment advances including the early initiation of treatment, treating-to-target, and novel therapeutics (11, 12). Critical to adhering to a treat-to-target approach is the regular integration of disease activity measurement as part of routine care, a practice included in ACR RA treatment guidelines (1) and selected as a quality measure by the Centers for Medicare and Medicaid Services (Quality ID #177: Rheumatoid Arthritis: Periodic Assessment of Disease Activity). In this study, we have updated the initial ACR 2012 recommendations for RA disease activity measures (2) through an updated systematic literature review, RA disease activity measure performance assessment, study quality assessment, level of evidence synthesis, and a modified Delphi voting process. Five preferred RA disease activity measures for regular clinical use were selected: CDAI, DAS28-ESR/CRP, PAS-II, RAPID3, and SDAI. Seven additional RA disease activity measures that met a “minimum standard” for regular use were identified: DAS, patient derived DAS28, HUPI, MBDA score, RADAI, RADAI-5, and RAPID5. Preferred measures represent those with the most support for their performance and feasibility as assessed by the working group, while those fulfilling the minimum standard have adequate performance and feasibility for regular use. Clinicians can utilize these recommendations when selecting a RA disease activity measure for integration into their care for RA patients, and any of the 11 measures shown in Table 4 that meet the minimum standard reasonably satisfy quality measures for assessing RA disease activity.
The purpose of these recommendations was to assist clinicians in the care of RA patients by identifying RA disease activity measures and evaluating their performance and feasibility for regular use. These recommendations are not meant to dictate the specific RA disease activity measure a clinician utilizes. The working group recognizes that feasibility varies based on practice and provider. Furthermore, providers may have comfort and experience with specific RA disease activity measures. Therefore, we aimed to identify not only preferred RA disease activity measures, but also RA disease activity measures that met a minimum standard by categorizing into disease activity states, possessing adequate psychometric properties, and being feasible for regular clinical use. For providers adopting a RA disease activity measure or aiming to integrate disease activity measurement into care through a standardized fashion (i.e. integration into the electronic health record), we recommend selecting a preferred RA disease activity measure (CDAI, DAS28, PAS-II, RAPID3, SDAI).
In addition to not precluding the use of other RA disease activity measures, these recommendations importantly do not provide recommendations on disease activity measures in special circumstances. An example might include the use of musculoskeletal ultrasound or magnetic resonance imaging in a patient with a difficult or equivocal joint exam who is being considered for treatment escalation or withdrawal. There are certainly specific circumstances or patient populations where alternative disease activity assessments may be clinically indicated. Additionally, there are certain RA subpopulations where the validity of RA disease activity measures may vary. Disease activity scores including patient reported measures are higher in patients with comorbid fibromyalgia (13), and disease activity scores including inflammatory markers are higher in obese patients (14). Providing recommendations for disease activity assessment in these specific situations or patient populations was beyond the scope of these recommendations and are left to the judgement of the treating clinician.
The preferred RA disease activity measures are largely unchanged from those previously recommended (2), with the difference being that the PAS was not recommended for preferred use in these updated recommendations. Both PAS and PAS-II were infrequently studied since the prior recommendations and subsequently did not satisfy the requirement of having demonstrated adequate psychometrics during this period. It is important to note that PAS and PAS-II differ from the RAPID3 only by the functional status component of each composite measure. PAS-II contains the Health Assessment Questionnaire-II (HAQ-II) (15), while PAS contains the Health Assessment Questionnaire (16) and RAPID3 contains the MultiDimensional Health Assessment Questionnaire (MDHAQ) (17). Assessment and recommendation of functional status measures in RA has been conducted in parallel, with recommendations for the use of PROMIS Physical Function 10, MD-HAQ, and HAQ-II. Given the overlap between PAS-II and RAPID3 as well as the results from the parallel functional status assessment project, the Quality Measures Subcommittee additionally recommended the PAS-II as a preferred measure. The consistency in the selection of preferred disease activity measures between the prior and current recommendations provides further support for these measures.
There are limitations to this effort. We conducted a systematic literature review from the time of the prior review. Therefore, generation of overall level of evidence from measure performance and study quality assessment was only able to be completed for studies since the initial review. Properties assessed early in measure development may not have been routinely re-assessed in later literature. Though not included into level of evidence, we synthesized data from the prior literature review as well as additional searches from before our current search period and provided these to working group members to inform the selection process. In contrast to the parallel functional status assessment recommendations, which were limited to patient-reported measures, we assessed RA disease activity measures with several different components – patient reported, provider assessment, laboratory, and imaging. The broad nature of these components makes selecting adequate measure performance and study quality assessment tools challenging. We selected the COSMIN checklist with 4-point scoring system to adapt for our study because it was designed to facilitate selection of health instruments in systematic reviews (18) and could be applied to both the RA disease activity and functional status assessment projects. While COSMIN was designed primarily for patient reported outcomes measures, it has been adapted beyond health-related patient reported instruments (19, 20). An updated COSMIN tool was developed after study inception that penalizes studies less for smaller samples sizes and not reporting handling of missing data, which may affect level of evidence grading (21). Finally, because there are no validated feasibility scoring systems for RA disease activity measures, we developed a scoring system to be used for this effort. Feasibility is inherently subjective based on varying viewpoints of different providers and practice types; therefore, we focused our feasibility scoring on identifying measures that could be regularly used by the majority of providers and practice types. As adoption of, and training in, the advanced imaging modalities continues to increase, the feasibility will need to be re-assessed in future efforts (22). While advanced imaging modalities were all deemed not feasible for regular use, all measures solely based on advanced imaging also did not fulfill the minimum standard by the absence of categorizing into 3–4 disease activity states.
There are several strengths to this effort. The working group was composed of content experts and practicing rheumatologists. The process and preliminary results were presented at the 2017 ACR Annual Meeting and underwent public comment. A systematic literature review with duplicate screening of articles for inclusion and standardized data abstraction was performed. Study quality was assessed using a standardized approach with a widely accepted tool and combined with the performance of RA disease activity measures to generate an overall level of evidence. A modified Delphi process was used to obtain final recommendations and incorporated the prior literature search as well as additional hand searches over the period before the current literature review.
In conclusion, we updated prior ACR recommendations for RA disease activity measures providing recommendations for both measures that meet a “minimum standard” for regular use and preferred measures for regular use – CDAI, DAS28-ESR/CRP, PAS-II, RAPID3, and SDAI. These recommendations can assist clinicians with adhering to a treat-to-target approach for the management of RA, but should not be interpreted as dictating the “proper” measure to be used in individual circumstances or clinical practices. As additional measures are developed and performance of measures is further characterized, these recommendations should again be re-evaluated.
Supplementary Material
Significance & Innovations.
This is the first update to the American College of Rheumatology’s recommended rheumatoid arthritis disease activity measures for regular clinical use.
We used a systematic approach to identify and evaluate measures meeting a minimum standard that can be repeated in future updates and provides a path for research on existing or new measures.
Acknowledgements
We thank Cynthia Schmidt, MD, MLS (University of Nebraska Medical Center McGoogan Library of Medicine) for her assistance performing the systematic literature review searches. We thank the American College of Rheumatology staff members Amy Turner and Regina Parker for their support and assistance through the recommendation process.
Funding: Internal funding through the American College of Rheumatology
Other Support: BRE is supported by a UNMC Internal Medicine Scientist Development Award, the UNMC Mentored Scholars Program, the UNMC Physician-Scientist Training Program, a Rheumatology Research Foundation (RRF) Scientist Development Award, and the National Institute of General Medical Sciences (1 U54 GM115458). SK is supported by funding from the Dartmouth Institute. JRC is supported by funding from Amgen, Pfizer, PCORI, Radius and NIAMS. TRM is supported by funding from the VA, RRF, NIGMS (1 U54 GM115458). JRO is supported by funding from the VA. VKR is supported by funding from Genentech, Pfizer and RRF. VKR is supported by funding from Genentec, Pfizer and RRF. AL is supported by Glaxo, Janssen, Novartis, Gilead and Pfizer. LGS receives salary support from the Veterans Administration and YNHHS/CORE through a contract to the Centers for Medicare and Medicaid Services to develop and implement accountability outcome measures. KM is supported by funding from Pfizer ASPIRE award and RRF.
Footnotes
Disclosures: BRE, BKT, SK, TRM none. MJB receives consultant fees from BI, Horizon, Novartis, Pfizer, Sanofi, Abbvie and Sandoz, and speaker fees from Novartis, Pfizer, Sanofi, Abbvie, Celgene and Merck. JRC receives consultant fees from BMS, Roche, UCB, Abbvie, Janssen, Radius, Regeneron, Amgen, Lilly, Pfizer, Corrona, Myriad. VKR receives fees from Amgen, ACR and FDA. LGS develops and implements non-rheumatology accountability outcome measures under contract to the Centers for Medicare and Medicaid Services.
References
- 1.Singh JA, Saag KG, Bridges SL Jr., Akl EA, Bannuru RR, Sullivan MC, et al. 2015 American College of Rheumatology Guideline for the Treatment of Rheumatoid Arthritis. Arthritis Rheumatol. 2016;68(1):1–26. [DOI] [PubMed] [Google Scholar]
- 2.Anderson J, Caplan L, Yazdany J, Robbins ML, Neogi T, Michaud K, et al. Rheumatoid arthritis disease activity measures: American College of Rheumatology recommendations for use in clinical practice. Arthritis Care Res (Hoboken). 2012;64(5):640–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Anderson JK, Zimmerman L, Caplan L, Michaud K. Measures of rheumatoid arthritis disease activity: Patient (PtGA) and Provider (PrGA) Global Assessment of Disease Activity, Disease Activity Score (DAS) and Disease Activity Score with 28-Joint Counts (DAS28), Simplified Disease Activity Index (SDAI), Clinical Disease Activity Index (CDAI), Patient Activity Score (PAS) and Patient Activity Score-II (PASII), Routine Assessment of Patient Index Data (RAPID), Rheumatoid Arthritis Disease Activity Index (RADAI) and Rheumatoid Arthritis Disease Activity Index-5 (RADAI-5), Chronic Arthritis Systemic Index (CASI), Patient-Based Disease Activity Score With ESR (PDAS1) and Patient-Based Disease Activity Score without ESR (PDAS2), and Mean Overall Index for Rheumatoid Arthritis (MOI-RA). Arthritis Care Res (Hoboken). 2011;63 Suppl 11:S14–36. [DOI] [PubMed] [Google Scholar]
- 4.Terwee CB, Mokkink LB, Knol DL, Ostelo RW, Bouter LM, de Vet HC. Rating the methodological quality in systematic reviews of studies on measurement properties: a scoring system for the COSMIN checklist. Qual Life Res. 2012;21(4):651–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mokkink LB, Terwee CB, Patrick DL, Alonso J, Stratford PW, Knol DL, et al. The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes. J Clin Epidemiol. 2010;63(7):737–45. [DOI] [PubMed] [Google Scholar]
- 6.Hendrikx J, de Jonge MJ, Fransen J, Kievit W, van Riel PL. Systematic review of patient-reported outcome measures (PROMs) for assessing disease activity in rheumatoid arthritis. RMD Open. 2016;2(2):e000202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lee J, Kim SH, Moon SH, Lee EH. Measurement properties of rheumatoid arthritis-specific quality-of-life questionnaires: systematic review of the literature. Qual Life Res. 2014;23(10):2779–91. [DOI] [PubMed] [Google Scholar]
- 8.Dalkey N, Helmer O. An Experimental Application of the DELPHI Method to the Use of Experts. Management Science. 1963;9(3):458–67. [Google Scholar]
- 9.Diamond IR, Grant RC, Feldman BM, Pencharz PB, Ling SC, Moore AM, et al. Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies. J Clin Epidemiol. 2014;67(4):401–9. [DOI] [PubMed] [Google Scholar]
- 10.Bakker MF, Jacobs JW, Kruize AA, van der Veen MJ, van Booma-Frankfort C, Vreugdenhil SA, et al. Misclassification of disease activity when assessing individual patients with early rheumatoid arthritis using disease activity indices that do not include joints of feet. Ann Rheum Dis. 2012;71(6):830–5. [DOI] [PubMed] [Google Scholar]
- 11.Aga AB, Lie E, Uhlig T, Olsen IC, Wierod A, Kalstad S, et al. Time trends in disease activity, response and remission rates in rheumatoid arthritis during the past decade: results from the NOR-DMARD study 2000–2010. Ann Rheum Dis. 2015;74(2):381–8. [DOI] [PubMed] [Google Scholar]
- 12.Overman CL, Jurgens MS, Bossema ER, Jacobs JW, Bijlsma JW, Geenen R. Change of psychological distress and physical disability in patients with rheumatoid arthritis over the last two decades. Arthritis Care Res (Hoboken). 2014;66(5):671–8. [DOI] [PubMed] [Google Scholar]
- 13.Ranzolin A, Brenol JC, Bredemeier M, Guarienti J, Rizzatti M, Feldman D, et al. Association of concomitant fibromyalgia with worse disease activity score in 28 joints, health assessment questionnaire, and short form 36 scores in patients with rheumatoid arthritis. Arthritis Rheum. 2009;61(6):794–800. [DOI] [PubMed] [Google Scholar]
- 14.George MD, Giles JT, Katz PP, England BR, Mikuls TR, Michaud K, et al. Impact of Obesity and Adiposity on Inflammatory Markers in Patients With Rheumatoid Arthritis. Arthritis Care Res (Hoboken). 2017;69(12):1789–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wolfe F, Michaud K, Pincus T. Development and validation of the health assessment questionnaire II: a revised version of the health assessment questionnaire. Arthritis Rheum. 2004;50(10):3296–305. [DOI] [PubMed] [Google Scholar]
- 16.Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum. 1980;23(2):137–45. [DOI] [PubMed] [Google Scholar]
- 17.Pincus T, Sokka T, Kautiainen H. Further development of a physical function scale on a MDHAQ [corrected] for standard care of patients with rheumatic diseases. J Rheumatol. 2005;32(8):1432–9. [PubMed] [Google Scholar]
- 18.Mokkink LB, Terwee CB, Patrick DL, Alonso J, Stratford PW, Knol DL, et al. The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study. Qual Life Res. 2010;19(4):539–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Larsen CM, Juul-Kristensen B, Lund H, Sogaard K. Measurement properties of existing clinical assessment methods evaluating scapular positioning and function. A systematic review. Physiother Theory Pract. 2014;30(7):453–82. [DOI] [PubMed] [Google Scholar]
- 20.Vrijman C, Linthorst Homan MW, Limpens J, van der Veen W, Wolkerstorfer A, Terwee CB, et al. Measurement properties of outcome measures for vitiligo. A systematic review. Arch Dermatol. 2012;148(11):1302–9. [DOI] [PubMed] [Google Scholar]
- 21.Mokkink LB, de Vet HCW, Prinsen CAC, Patrick DL, Alonso J, Bouter LM, et al. COSMIN Risk of Bias checklist for systematic reviews of Patient-Reported Outcome Measures. Qual Life Res. 2018;27(5):1171–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Torralba KD, Cannella AC, Kissin EY, Bolster MB, Salto LM, Higgs J, et al. Musculoskeletal Ultrasound Instruction in Adult Rheumatology Fellowship Programs. Arthritis Care Res (Hoboken). 2017. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.