Shoulder pain and disability index: cross cultural validation and evaluation of psychometric properties of the Spanish version

Miguel David Membrilla-Mesa, Antonio Ignacio Cuesta-Vargas, Rocio Pozuelo-Calvo, Victor Tejero-Fernández, Lydia Martín-Martín, Manuel Arroyo-Morales

The Shoulder Pain Disability Index (SPADI) is a recently published but widely used outcome measure. This study included 136 patients with shoulder disorders. SPADI was first translated and back-translated and then subjected to psychometric validation. Participants completed the Spanish versions of the SPADI, general health (SF-12), the Simple Shoulder Test (SST), Disability of Arm, Shoulder, and Hand (DASH) questionnaires and a pain intensity visual analog scale (VAS). The factors explained 62.8 % of the variance, with an internal consistency of α = 0.916 and 0.860, respectively. The confirmatory factor analysis showed a Comparative Fit Index of 0.82 and a Normed Fit Index of 0.80. The Root Mean Square Error of Aproximation was 0.12. The x (2) test for the 2-factor model was significant (x (2) = 185.41, df = 62, p < 0.01). The test-retest reliability was high, with an item ranging of the interclass correlation coefficient (ICC) from 0.89 to 0.93. The ICC for the total score was 0.91 (95 % CI 0.88 to 0.94). Measurement error by minimal detectable change (MDC)95 was 12.2 %. In the construct validity analysis, strong positive correlations were observed between Spanish Version of the SPADI and DASH (pain: r = 0.80; p < 0.01; disability: r = 0.76; p < 0.01). Moderate positive correlations were observed between Spanish Version of the SPADI and VAS (pain: r = 0.67; p < 0.01; disability: r = 0.65; p < 0.01). Moderate negative correlations were obtained between Spanish Version of the SPADI and SST-Sp (pain: r = -0.71; p < 0.01; disability: r = -0.75; p < 0.01). However, pain total Spanish Version of the SPADI was only weakly correlated with physical and mental components of SF-12 (both r = 0.40; p < 0.01). This Spanish version of SPADI demonstrated satisfactory psychometric properties in a patient sample in the hospital setting.