Abstract
The Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA)-Outcome Measures in Rheumatology (OMERACT) Psoriatic Arthritis (PsA) Working Group provided updates at the 2020 GRAPPA annual meeting on its work toward developing a core outcome set for PsA. Working groups were set up for the 4 prioritized domains: enthesitis, fatigue, structural damage, and physical function. Two instruments for measurement of physical function were provisionally endorsed: (1) the Health Assessment Questionnaire-Disability Index and (2) the physical functioning domain in the Medical Outcomes Study 36-item Short Form survey.
- OMERACT
- outcome assessment
- psoriatic arthritis
The Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA)-Outcome Measures in Rheumatology (OMERACT) Working Group aspires to develop an outcome measurement set for important domains for clinical trials of psoriatic arthritis (PsA).1 In this report, we summarize the updates of the Core Outcome Measures for Psoriatic Arthritis Clinical Trials (COMPACT) projects presented at the GRAPPA 2020 virtual annual meeting and trainee symposium. We focus on describing the processes leading to the endorsement of 2 outcome measures for the measurement of physical function for PsA.
Updates on the 4 Prioritized Domains
Within the COMPACT projects, 4 PsA domains were prioritized2 in 2019 and working groups were established for each of the domains to evaluate and standardize the outcome measures set. These 4 domains are (1) enthesitis (under musculoskeletal disease activity), (2) fatigue, (3) structural damage, and (4) physical function. Each working group was led by a team leader and engaged members across the globe, including at least 2 patient research partners (PRPs).3 The progress by each working group is summarized in Table 1. The enthesitis and fatigue working groups are working on “domain match” and “feasibility” of instruments. Domain match concerns whether stakeholders consider the outcome measures to be a good match with the domain they are supposed to measure. Feasibility addresses whether the outcome measure instruments can be used in randomized controlled trials (RCTs) and longitudinal observational studies. The structural damage working group is completing a systematic literature review (SLR) for radiographic outcomes for PsA and starting the work stream, selecting outcome measures for magnetic resonance imaging. The work on the domain of physical function was the focus at GRAPPA 2020. The physical function working group has completed the necessary evaluation for 2 outcome candidate measures: the Health Assessment Questionnaire-Disability Index (HAQ-DI) and the physical functioning domain in the Medical Outcomes Study 36-item Short Form survey (SF-36 PF).
Physical Function Domain
Physical function is 1 of the 4 common core domains used in clinical trials to measure severity and outcomes for PsA4 and is widely used in RCTs of PsA.5,6,7 A physical function working group composed of 15 members was set up in June 2018.3 The members in the working group represent 3 continents (America, Asia, and Europe) and include 11 rheumatologists, 3 PRPs, and 1 methodologist. The working group discussed and preselected 6 patient-reported outcome measures (PROMs) to undergo vigorous appraisal using the OMERACT Filter 2.18: HAQ and its 4 modifications (HAQ-DI, HAQ-Spondyloarthritis, modified HAQ, multidimensional HAQ), SF-36 PF, and the Patient-Reported Outcomes Measurement Information System physical functioning module (PROMIS PF).3 From 2019-2020, the working group completed the OMERACT instrument selection workbook for HAQ-DI and SF-36 PF.
The OMERACT Methodology
The OMERACT Framework Instrument Selection Algorithm (OFISA) is the methodology OMERACT developed to guide the selection of outcome measures/instruments8 and is based on the 3 pillars of OMERACT: Truth, Discrimination, and Feasibility. The 3 pillars used 4 signalling questions to help evaluate the existing evidence for each measure/instrument: (1) Is it a match with target domain (Truth)? (2) Is it practical to use (Feasibility)? (3) Do numeric scores make sense (Truth)? (4) Can it discriminate between groups of interest (Discrimination)?
Each instrument undergoes evaluation of 7 measurement properties: truth (domain match, construct validity), discrimination (test-retest reliability, longitudinal construct validity [responsiveness], clinical trial discrimination, thresholds of meaning), and feasibility. For each of these properties, it is necessary to answer the question of whether there is adequate evidence to support the instrument in clinical research.
The evidence supporting all 7 measurement properties for each instrument are presented in the summary of measurement property (SOMP) table.8 The ratings for each of the 7 measurement properties are classified as: GREEN (good to go), AMBER (some cautions, but still can use), RED (stop, do not use this), or WHITE (no data). A final rating for whether the instrument has adequate evidence to support its use in clinical research is proposed. The complete appraisal for each instrument is documented in a selection workbook that requires approval by the OMERACT Technical Advisory Group (TAG), to be endorsed as a recommended core instrument.
Evidence Supporting Measurement Properties for HAQ-DI and SF-36 PF
Domain match and feasibility. The assessment of domain match and feasibility requires a consensus from experts and PRPs. Through discussions by webinars and email correspondence, the working group members selected 6 PROMs for physical function.3 The working group members participated in a Delphi exercise to achieve a consensus for domain match and feasibility for all 6 PROMs.
To engage the broader GRAPPA membership audience, a video showing detailed information on the 6 PROMs was developed by the team lead and 2 PRPs in the working group (NG and JC; https://youtu.be/Qd86PwzgvQI). All GRAPPA PRPs were then invited to participate in a separate Delphi exercise on domain match and feasibility for all 6 PROMs.
Combining results from both Delphi exercises for working group members and PRPs, the final voting for domain match and feasibility were AMBER and GREEN, respectively, for HAQ-DI. As for SF-36 PF, the final ratings were AMBER for both domain match and feasibility.
Systematic literature review and final ratings. The working group has conducted 2 SLRs on the PROMs for physical function. One SLR included articles with a primary aim to evaluate measurement properties of all PROMs in PsA (PROSPERO ID: CRD42016032546),9 and the second SLR was to evaluate data of the physical function PROMs from RCTs (PROSPERO ID: CRD42019129557).10 The first SLR identified 10 relevant articles for HAQ-DI and 6 articles for SF-36 PF. Subsequently, new evidence was generated by working group members to bridge the gap of knowledge for HAQ-DI (n = 4 studies) and SF-36 PF (n = 1 study). For the second SLR, relevant RCTs were reviewed for data on HAQ-DI (n = 31) and for SF-36 PF (n = 4), as shown in Table 2. At least 2 working group members evaluated each article and reached a consensus on the quality of each measurement property using the OMERACT good method checklist.8 For those articles that passed this assessment, study findings and characteristics supporting each measurement property were reviewed and summarized in the instrument selection workbook. Then, the evidence supporting HAQ-DI and SF-36 PF was summarized in the final SOMP tables. Extracts of the SOMP tables are presented in Table 2. At the GRAPPA 2020 annual meeting, the working group proposed a provisional endorsement for both HAQ-DI and SF-36 PF. Further evidence on test-retest reliability, RCT discrimination, and development of threshold of meaning for SF-36 PF should be collected to ensure full endorsement of both instruments.
Endorsement
After presenting the relevant data, live polling was conducted at the GRAPPA 2020 annual meeting to endorse the final ratings for both instruments. Two questions were asked: “Given the evidence, do you agree with the working group to endorse HAQ-DI as provisional core instrument for the measurement of physical function in PsA studies?” The same question was asked for the SF-36 PF. Forty-nine GRAPPA members participated in the live polling, with 93.9% and 86.7% voting positively for HAQ-DI and SF-36 PF, respectively (Table 3). The 2 instrument selection workbooks were then reviewed by the OMERACT TAG, who suggested minor updates, though the final ratings remained unchanged. Both instrument selection workbooks and ratings were endorsed by the OMERACT TAG in September 2020. A subsequent online poll of all GRAPPA members was conducted from October 20, 2020, to November 16, 2020. The original data, updates, and results of the live poll at the GRAPPA 2020 annual meeting were presented. The identical voting questions were used in both polls. GRAPPA members participated (n = 119) and voted for provisional endorsement of HAQ-DI (97.5% positive) and SF-36 PF (77.3% positive) as core instruments for the measurement of physical function in PsA studies.
Future plans for this working group include comparing discrimination in RCTs for the selected instruments using metaanalysis, further evaluation of thresholds of meaning for both the HAQ-DI and SF-36 PF, and appraisal of 4 other preselected PROMs for physical function using OFISA.
Conclusion
This report provides an update for the workstream involved in select core outcome instruments for 4 prioritized domains in PsA. The measurement properties of HAQ-DI and SF-36 PF for measurement of physical function in PsA were evaluated using the OFISA of the OMERACT Filter 2.1. Provisional endorsement for both instruments was obtained from the GRAPPA community.
Footnotes
As part of the supplement series GRAPPA 2020, this report was reviewed internally and approved by the Guest Editors for integrity, accuracy, and consistency with scientific and ethical standards.
The authors are members of the Outcome Measures in Rheumatology (OMERACT), an organization that develops and validates outcome measures in rheumatology randomized controlled trials and longitudinal observational studies and receives arms-length funding from 36 sponsors. YYL is funded by the Clinician Scientist award of the National Medical Research Council, Singapore (NMRC/CSA-INV/0022/2017). The views expressed are those of the author(s) and not necessarily those of the NMRC. AMO is funded by the Jerome L. Greene Foundation Scholar Award, the Staurulakis Family Discovery Award, the Rheumatology Research Foundation, and the National Institutes of Health (NIH) through the Rheumatic Diseases Resource-based Core Center (P30-AR053503, Cores A and D; and P30-AR070254, Cores A and B). All statements in this report, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the NIH, the National Institute of Arthritis Musculoskeletal and Skin Diseases (NIAMS), or the Rheumatology Research Foundation (RRF). AO is funded by the Rheumatology Research Foundation and NIH/NIAMS K23 AR063764 and R01 AR072363. RC (the Parker Institute) is supported by a core grant from the Oak Foundation (OCAY-18-774-OFIL). AJM is funded by the National Psoriasis Foundation Psoriatic Disease Research Fellowship. LCC is funded by a National Institute of Health Research (NIHR) Research Clinician Scientist award. The research was supported by the NIHR Oxford Biomedical Research Centre (BRC). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health.
LE has been on advisory boards or received educational/research grants from Novartis, UCB, Eli Lily, Pfizer, and Janssen. There are no conflicts of interest from other co-authors.
This paper does not require institutional review board approval.
- Copyright © 2021 by The Journal of Rheumatology
References
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.