Evaluating the psychometric properties of the Widespread Pain Index and the Symptom Severity Scale in youth with painful conditions

Abstract

ABSTRACT Background Assessing features of centralized pain may prove to be clinically meaningful in pediatric populations. However, we are currently limited by the lack of validated pediatric measures. Aim We examined the psychometric properties of the Widespread Pain Index (WPI) and Symptom Severity (SS) scale to assess features of centralized pain in youth with painful conditions from three clinical samples: (1) musculoskeletal surgery, (2) headache, and (3) chronic pain. Methods Participants were 240 youth aged 10 to 18 years (M age = 14.8, SD = 1.9) who completed the WPI and SS scale. Subsets of participants also completed additional measures of pain region, pain intensity, quality of life, pain interference, and physical function. Results Increased features of centralized pain by age were seen for the WPI (r = 0.27, P < 0.01) and SS scale (r = 0.29, P < 0.01). Expected differences in sex were seen for the WPI (sex: t132 = −3.62, P < 0.01) but not the SS scale (sex: t 223 = −1.73, P = 0.09). Reliability for the SS scale was adequate (α = 0.70). Construct validity was demonstrated through relationships between the WPI and pain regions (r = 0.57, P < 0.01) and between the SS scale and quality of life (r = −0.59, P < 0.01) and pain interference (r = 0.56, P < 0.01). Criterion validity was demonstrated by differences on the WPI between the surgery sample and the headache and chronic pain samples (F2,237 = 17.55, P < 0.001). Comprehension of the SS scale items was problematic for some youth. Conclusions The WPI showed adequate psychometric properties in youth; however, the SS scale may need to be modified. Our findings support the need to develop psychometrically sound instruments for comprehensive assessment of pain in pediatric samples.