When selecting an outcome measurement instrument for a randomized controlled trial (RCT), evidence is needed to ensure that the instrument will be capable of truthfully measuring change, i.e., responsiveness. More importantly, it should be able to distinguish the change in one treatment arm (assuming the treatment works) relative to the change in a comparator arm. Most recently, RCTs for psoriatic arthritis (PsA) include an active comparator (in addition to placebo) to the new intervention, often head-to-head against an approved biological (b-) disease modifying antirheumatic drug (DMARD), such as a TNF inhibitor (TNFi). These RCTs require the outcome measurement instruments to have even better discriminatory capacity such that an experimental intervention could demonstrate superiority not only against placebo comparator but also an active comparator. Building an argument that the scores of a given outcome measurement instrument can make that distinction in the target population is essential in instrument selection for trials.

Ideally, all candidate outcome measurement instruments would be studied simultaneously in RCTs evaluating the same target populations and interventions, particularly for those in the core domain set [1]. However, that is neither practical nor feasible. As a result, tradition and common practice often drive the instrument selection. Another issue is that when seeking evidence for clinical trial performance of an outcome measurement instrument, there is often no indication of what was expected in that trial for a “good” or “perfect” outcome measurement instrument evaluating a domain. If the trial’s experimental intervention(s) did not demonstrate superiority over the comparator group in a domain of interest utilizing the outcome measurement instrument, it is uncertain if the outcome measurement instrument was not sensitive enough or if it actually performed as intended to show the lack of difference in efficacy of the intervention.

PsA is a chronic inflammatory disease involving peripheral arthritis, enthesitis, dactylitis, axial involvement, psoriasis, and nail disease. It has a profound impact on the psychological, social, and physical well-being of the affected people [2]. Physical functioning is a key concern from both patients’ and doctors’ perspectives, and it is recognized as one of the core domains to be measured in every RCT and longitudinal observational studies [1,3]. There have been several patient-reported outcome measurement instrument that assess physical functioning in PsA [4], with the most commonly used in PsA RCTs being the Health Assessment Questionnaire Disability Index (HAQ-DI) [5] and the physical functioning domain within the Medical Outcomes Study 36-item Short Form Survey (SF36-PF) [6]. The SF-36 physical component summary score (SF36-PCS) also has been commonly reported in RCTs but does not match the physical function domain well [7]. The GRAPPA community has prioritized 6 outcome measurement instruments for physical functioning [7]. Using a qualitative synthesis based on a rigorous systematic review protocol evaluating outcome measurement instrument in PsA, we concluded that clinical trial discrimination was supported for HAQ-DI and SF36-PCS in PsA with a low risk of bias. In contrast, the SF36-PF was supported but with some concerns due to fewer available trials reporting data [6]. As part of the concerted effort of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) - Outcome Measures in Rheumatology (OMERACT) initiative to standardize the outcome measurement set for PsA [8], two outcome measurement instruments, namely the HAQ-DI and SF36-PF were provisionally endorsed by both GRAPPA and OMERACT organizations [9,10]. A research agenda was set to evaluate the clinical trial discrimination of these instruments quantitatively. Therefore, in this study, we aimed to quantitatively compare the sensitivity of these three outcome measurement instruments via a network meta-analysis in a representative sample of RCTs for PsA.