Reliability, construct validity, responsiveness and minimum clinically important difference of the numeric pain rating scale and shoulder pain and disability index in patients with subacromial pain syndrome
Ian Young, James Dunning, James Escaloni, Filippo Maselli, Joshua Prall, Firas Mourad, César Fernández‐de‐las‐Peñas
Abstract
Background The numeric pain rating scale (NPRS) and shoulder pain and disability index (SPADI) are commonly used patient-reported outcome measures (PROMs) in patients with non-specific shoulder pain. To date, there are gaps in the evidence supporting the clinimetric properties of these PROMs for patients with subacromial pain syndrome (SAPS). Methods A clinimetric analysis (n=145) was performed to examine the reliability, construct validity, responsiveness, interpretability, minimal detectable change (MDC 95 ) and minimum clinically important difference (MCID) of the NPRS and SPADI for "improved" (global rating of change from +3 to +7) and "much-improved" (global rating of change from +5 to +7) patients at 3-months follow-up. Results The NPRS (ICC: 0.86; 95%CI, 0.33-0.96) and SPADI (ICC: 0.79; 95%CI 0.12-0.94) exhibited good reliability and excellent responsiveness (NPRS: area under the curve (AUC)=0.96, 95%CI 0.92-0.99; SPADI: AUC=0.90, 95%CI 0.84-0.95) in this patient population. Both outcomes demonstrated strong construct validity (Pearson's r; p<0.001). The MDC 95 was a 1.7- and 20.5-point change for the NPRS and SPADI, respectively. For the NPRS, the MCID was a 1.5-point change in the "improved" group and a 2.5-point change in the "much improved" group. For the SPADI, the MCID was an 18-point or 50% change for the "improved" group, and a 25-point or 70% change in the "much improved" group. Conclusions The NPRS and SPADI demonstrated sound clinimetric properties in patients with SAPS. The MCID exceeded measurement error in the "much improved" groups. Diagnosis, type of intervention, level of improvement, and measurement error should be considered when applying the MCID.