Background/Purpose: Skin disease is a hallmark of systemic sclerosis (SSc). The modified Rodnan skin score (mRSS) is physician performed measurement that assesses the extent and severity of skin thickness. However, it does not capture all of the clinical features of SSc skin disease. Patient-reported outcomes (PROs) can capture patients’ perspective. The Scleroderma Skin Questionnaire (SSQ) is a novel PRO which assesses skin symptoms. To date, the SSQ has not been validated, and this study sought to assess the psychometric properties of the SSQ to generate evidence to support its further use.

Methods: We used data from CONQUER (Collaborative National Quality and Efficacy Registry), a multi-center, US-based registry of adults with early limited cutaneous (lc) and diffuse cutaneous (dc) SSc. The SSQ uses a 7-day recall period of 6 items (tight, painful, red, hard, itchy, rigid/stiff), which are graded on a 5-point Likert scale and averaged to obtain a score ranging 0-4 (4=worse). The second part of the questionnaire, the SSQ 6-Month, asks respondents to compare these symptoms to 6-months prior. Internal consistency of the SSQ was determined with Cronbach’s α. Baseline response frequencies were used to investigate floor/ceiling effects. The associations between the SSQ and the mRSS and with legacy PROs [Scleroderma Health Assessment Questionnaire (SHAQ), Patient-Reported Outcomes Measurement Information System-29 (PROMIS-29), Patient Global (PtGA), and Physician Global (PGA)] were explored using Pearson’s correlations. The SSQ 6-Month was correlated with 6-month changes in mRSS and in legacy PROs using mixed modeling with a bias-corrected confidence interval determined using cluster bootstrapping.

Results: We included 536 participants who completed the mRSS and SSQ (Table 1). The SSQ was higher (worse) in patients with dcSSc compared with lcSSc (1.5 vs 0.9; p< 0.001). The SSQ demonstrated internal consistency (α=0.90) and likely floor effect (Figure 1). At baseline, the SSQ was moderately correlated with the mRSS (r=0.56). The correlation was stronger among individuals with dcSSc (r=0.56) than lcSSc (r=0.22; all p< 0.05).At baseline, the SSQ was strongly correlated with PROMIS-29’s physical (r=-0.52) and pain interference (r=0.61), and weakly with PROMIS-29’s anxiety (r=0.24) and sleep disturbance (r=0.12; all p< 0.05). For the SHAQ, SSQ correlated strongly to HAQ (r=0.63) and severity score (r=0.61) and weakly to intestinal symptoms, breathing symptoms, Raynaud’s attacks, and finger ulcers symptoms (all p< 0.05). When stratified for SSc subtype, baseline mRSS, and disease duration, similar trends for correlation were seen (Table 2). SSQ 6-Month showed correlation of r=0.21 (p< 0.05) with 6-month change in mRSS and poor correlation with change in legacy PROs (Table 2).

Conclusion: In the CONQUER registry, the SSQ demonstrated internal consistency and moderate correlation with mRSS. The SSQ showed moderate/strong correlations with a subset of legacy PROs. However, the SSQ 6-Month showed weak correlation with change in the mRSS, suggesting either a poor recall for a 6-month time span or a disconnect between the patient perception and the mRSS. This study provides initial characterization of SSQ for future validation.

« Back to ACR Convergence 2023

ACR Meeting Abstracts - https://acrabstracts.org/abstract/initial-characterization-of-a-skin-symptom-questionnaire-for-patients-with-systemic-sclerosis/