Background Patient-reported outcomes (PRO) are increasingly used to assess responsiveness after treatment.
Objectives To determine the ability of different PRO in assessing responsiveness in psoriatic arthritis (PsA) after the initiation of biological treatment.
Methods This is a retrospective analysis conducted on all PsA patients followed at a single centre PsA clinic who were started and continued a biological treatment and had 1 year of follow-up were identified.
PRO were administered on baseline and follow-up visits: patient global assessment (PGA) [very good, good, fair, poor and very poor], Short-Form 36 questionnaire (SF-36), Health Assessment Questionnaire (HAQ), Functional Assessment of Chronic Illness Therapy (FACIT), Fatigue Severity Scale (FSS), Euroqol (EQ-5D) and Dermatology Life Quality Index (DLQI) (high scores reflect better outcome for SF-36, FACIT and EQ-5D, and lowe scores reflect better outcome in HAQ, FSS and DLQI).
On follow-up visit and based on the PGA, patients were stratified as improved (very good/good), same (fair) and worsened (poor/very poor).
Descriptive statistics were used to define the patients’ characteristics and disease activity. The mean change (Δ) in the PRO scores was determined by subtracting the scores of the follow-up visits from the baseline visit. Spearman correlations of the change in scores of PRO were studied in patients who improved, same and worsened. Responsiveness of PRO was evaluated with Effect Size (ES) and Standardized Response Mean (SRM).
Results 233 patients (62.8% male and 90% Caucasian) were identified. The age at diagnosis of PsA was 35.2 (12.4) and the duration of PsA was 12.9 (9.2). 73% had active joints with a joint count of 10.0 (9.9). All patients had active PsA (peripheral, peripheral and axial, and peripheral or axial) at the baseline visit and qualified for the initiation of a biologic. The mean time between baseline and follow-up visit was 1.18 (0.48) year.
At the follow-up visit the following correlations were identified among PRO: 1) in patients who improved (n= 70) [Δ FACIT and Δ FSS (r= -0.82, p= 0.007), Δ FSS and Δ SF-36 MCS (r= -0.52; p= 0.06)], 2) in patients who worsened (n= 27) Δ FACIT and Δ HAQ (r= -0.91, p= 0.004), Δ FACIT and Δ SF-36 PCS (r= -0.69, p= 0.04), Δ EQ-5D and Δ SF-36 MCS (0.80, p= 0.02), Δ SF-36 PCS and Δ SF-36 MCS (r= -0.64, p= 0.003) and 3) in patients who stayed the same (n= 126) Δ FSS and Δ HAQ (r=0.27, p= 0.02), Δ FACIT and Δ FSS (r= -0.56, p= 0.0002), Δ FSS and Δ SF-36 MCS (r= -0.35, p= 0.0008), Δ EQ5-D and Δ SF-36 PCS (r= 0.55, p= 0.0003), Δ SF-36 PCS and Δ SF-36 MCS (r= -0.35, p= 0.0004) and SF3-6 PCS and Δ DLQI (r= -0.32, p= 0.04).
Responsiveness (SRM) was remarkably noticeable in the patients who improved ranging from large to small effect (DLQI -1.07, FACIT 0.79, EQ-5D 0.69, FSS -0.67, HAQ -0.63, SF-36 PCS 0.43 and SF-36 MCS 0.36) and patients who worsened (FSS 0.80, EQ-5D -0.54, FACIT -0.47, SF-36 PCS -0.38, DLQI 0.35 and SF-36 MCS -0.27).
Conclusions PRO are responsive and valid measures and have the ability to capture improvement and worsening in response to treatment based on patient judgment. PRO can measure the same concept as demonstrated with the presence of moderate to high correlation among scores, e.g FSS and FACIT, thus there is no need to administer them simultaneously.
Disclosure of Interest None Declared