Background Though continuous composite measures of disease activity for psoriatic arthritis (PsA) assessment exist, abbreviated measures that are more feasible for screening in routine clinical practice are needed. The 3 Visual Analogue Scale (VAS) and 4VAS scores were developed by abridging the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) Composite Exercise (GRACE) measure to be the first short multidimensional composite measures specifically for use in PsA routine clinical care. The measures were shown to have superior performance than several established composite measures using small datasets.

Objectives To further test 3VAS/4VAS in observational and trial datasets, as GRAPPA members recommended.

Methods This post-hoc analysis of the DISCOVER 1/2 and COSMOS studies used pooled data through week (W) 24 from all treatment groups across studies. The correlation of 3VAS/4VAS with Disease Activity Index for PsA (DAPSA), PsA Disease Activity Score (PASDAS), physician global assessment (PhGA), and patient GA (PtGA) was assessed with Pearson’s correlation coefficient. Minimal Important Difference (MID) was assessed with 4 distribution-based methods (based on standard error of the measurement, effect size, reliable change index [RCI], and RCIdiff). Minimal detectable change (MDC) was assessed with the standard formula: MDC = 1.96*sqrt(2)sdsqrt(1-ICC). Clinically relevant thresholds for low, moderate and high disease activity were estimated with receiver operating characteristic analysis and DAPSA (≤4, >4 − ≤14, >14 − ≤28, >28), PASDAS (≤1.9, >1.9 − ≤3.2, >3.2 − <5.4, ≥5.4), and PhGA/PtGA (≤1, >1 − ≤3, >3 − ≤6, >6 cm) as anchors.

Results 1405 patients were included, of whom 51.3% were male, with a mean (SD) age of 47.1 (11.8) and PsA duration of 6.4 (6.5) years. At baseline, the mean (SD) 3VAS, 4VAS, DAPSA, PASDAS, PhGA, and PtGA scores of 6.4 (1.6), 6.3 (1.6), 45.8 (20.2), 6.5 (1.1), 6.5 (1.6), and 6.7 (2.0), respectively, reflected high levels of disease activity. Through W24, both 3VAS and 4VAS showed very strong correlation with PtGA (r3VAS=0.92, r4VAS=0.94) and PASDAS (r3VAS=0.81, r4VAS=0.82), strong with PhGA (r3VAS=0.77, r4VAS=0.74), and moderate to strong with DAPSA (r3VAS=0.59, r4VAS=0.61). Calculated MIDs were 0.9 (range: 0.7-1.3 across methodologies) for 3VAS and 0.9 (range: 0.6-1.3) for 4VAS (Table 1). MDC estimates were 3.3 (range 2.1-4.2 across follow-up intervals) for 3VAS and 3.2 (range: 2.0-4.0) for 4VAS. Cut-off values for low, moderate, and high disease activity were 2.0, 3.4, and 4.9 for 3VAS and 2.1, 3.5 and 5.1 for 4VAS, respectively (Figure 1).

Conclusion Using a large pooled clinical trial dataset of patients with active PsA, we have calculated clinically relevant thresholds for improvement and disease activity for 3VAS and 4VAS. These estimates are generally comparable to those previously reported and can be used to set treatment targets and screen disease activity in routine care when resources are limited or during remote patient monitoring.