Lead Investigator: Désirée van der Heijde, Leiden University Medical Center
Title of Proposal Research: Measurement properties of the instruments used in the outcome assessment of axial spondyloarthritis
Vivli Data Request: 5292
Funding Source: None.
Potential Conflicts of Interest: Consulting fees AbbVie, Amgen, Astellas, AstraZeneca, BMS, Boehringer Ingelheim, Celgene, Daiichi, Eli-Lilly, Galapagos, Gilead, Glaxo-Smith-Kline, Janssen, Merck, Novartis, Pfizer, Regeneron, Roche, Sanofi, Takeda, UCB and Director of Imaging Rheumatology. Management thereof: The reported conflicts of interest will not be an issue for this work.
Summary of the Proposed Research:
Axial spondyloarthritis (axSpA) is a form of inflammatory arthritis that most commonly affects the joints of the spine and pelvis. Common features of spondyloarthritis include inflammatory back pain, as well as pain and inflammation in the peripheral joints, the digestive tract or eyes. Nowadays, patients with axSpA are usually classified in two groups: i) Patients with radiographic axSpA also known as ankylosing spondylitis (AS), who have developed substantial damage of the pelvis visible on radiographs, and ii) Patients with non-radiographic axSpA, who do not have damage visible on radiographs but do have other symptoms specific for axSpA.
During the last decades, new efficacious drugs have been developed to treat patients with axSpA. Nevertheless, these drugs are not efficacious in all patients with axSpA and therefore it is necessary to continue doing research and developing new drugs for this disease. Anytime that a new drug is developed, its efficacy and safety need to be investigated through specific studies named as clinical trials. If these studies do not report on the same outcome measures, they cannot be compared. Moreover, if there is no need to report a minimum of outcomes, selective reporting of (positive) outcomes may occur. In order to improve comparability between studies and prevent selective reporting, there is a need for standardized measurements (e.g. the same questionnaire is used to measure pain in all trials). The minimum set of measurements to be evaluated in all trials are usually combined into a core outcome set. This set ensures every study includes the same measurements.
The Assessment of SpondyloArthritis international Society (ASAS) developed such a core outcome set for ankylosing spondylitis in 1999. This core set describes the domains and instruments that should be measured in each study including patients with ankylosing spondylitis. Currently, ASAS is updating this core set to make it valid also for patients with non-radiographic axSpA and to evaluate whether or not new instruments that have been developed in the last two decades need to be added to this core set. In order to select the best set of instruments available, the measurement properties of each of the instruments need to be evaluated. Measurement properties include several types of validity and reliability, one of which is test-retest reliability. Test-retest reliability provides information on the stability and reliability of an instrument in stable situations. E.g. the result of a questionnaire should provide very similar results when assessed twice while nothing changed; a lab test should provide the same result if tested twice or the number of swollen joints should be the same if assessed by two people or by the same person on a different occasion. However, results are never identical. In order to calculate this ‘measurement error’ (test-retest reliability) accurately, two measurements (e.g. pain) taken closely together without a change in the situation should be used.
In clinical trials, there is a screening visit to determine whether a participant is eligible to partake in the trial, followed by a baseline visit shortly thereafter (usually within the next 7-10 days) without any intervention in between. On a group level, we do not expect there to be any changes in the level of pain reported by the patients between these two measurements, since there was no extra treatment for the pain provided during this period. For the purpose of this study we therefore would like to use data measured at the screening visit and the baseline visit of trials, to calculate the test-retest reliability of the different measurement instruments which are candidates to be included in the updated ASAS core set for axSpA. Knowledge about this measurement property of the candidate instruments will help to select the best instrument for the core set, ensuring an appropriate core set for all clinical trials.
Statistical Analysis Plan
The aim of this work is to calculate the test-retest reliability of instruments used to evaluate efficacy in clinical trials. We will use intraclass correlation coefficients to assess reliability of continuous measures. In addition, we will generate Bland & Altman plots and judge the scedasticity of the data. We will also determine the systematic error (i.e. the mean difference between the assessment at screening and baseline) and random error (i.e. 95% limits of agreement and Smallest Detectable Change (SDC)). Subgroup analyses will be performed in order to determine whether the measurement instrument is valid both for AS and non- radiographic axSpA. In the event of missing data the particular participant will be excluded from analysis for that particular outcome measure.
Phase 3, Multicenter, Randomized, Double-Blind, Placebo-Controlled Study to Evaluate Efficacy and Safety of Certolizumab Pegol in Subjects With Active Axial Spondyloarthritis
Study ID: NCT01087762
Sponsor ID: AS001
Boel A, Navarro-Compán V, van der Heijde D. Test–retest reliability of outcome measures: data from three trials in radiographic and non-radiographic axial spondyloarthritis. RMD Open 2021;7:e001839. DOI: 10.1136/rmdopen-2021-001839