Related Experiment Video
Updated: Aug 28, 2026
The Multiple Sclerosis Performance Test (MSPT): An iPad-Based Disability Assessment Tool
Published on: June 30, 2014
Validity of patient-reported outcome measures for pain in multiple sclerosis
Cigdem Yilmazer1, Miguel D'haeseleer2, Bernardita Soler3
1REVAL Rehabilitation Research Center, Faculty of Rehabilitation Sciences, Hasselt University, Hasselt, Belgium; UMSC, Hasselt/Pelt, Hasselt, Belgium.
Background:
Pain is a frequent and disabling symptom of multiple sclerosis (MS), highlighting the need for valid patient-reported outcome measures (PROMs) to assess pain in both clinical practice and research. Although pain PROMs demonstrated good reliability in MS, their validity remains unknown.
Objectives:
This study investigated the construct validity of Douleur Neuropathique en 4 (DN4), painDETECT, Neuropathic Pain Scale (NPS), Neuropathic Pain Symptom Inventory (NPSI), and Brief Pain Inventory-Short Form (BPI-SF).
Methods:
110 persons with MS participated (mean age 45±13 years, Expanded Disability Status Scale (EDSS) = 3 [Interquartile range (IQR):2.6]. was performed to evaluate. Construct validity was evaluated by structural validity using confirmatory factor analysis (CFA) and by hypothesis testing.
Results:
Confirmatory factor analysis (CFA) indicated that almost all pain PROMs achieved an acceptable limit of goodness of fit. DN4 and painDETECT, as well as NPS and NPSI, showed a strong correlation with each other (Spearman rho= 0.657, 0.788, respectively). Weak-to-moderate correlations were found with self-report questionnaires (perceived stress, depression, anxiety, sleep, and fatigue) (Spearman rho from 0.109 to 0.427), and weak to nonsignificant correlations with cognitive information processing speed, manual dexterity, walking speed, spasticity, muscle strength, and activity level (Spearman rho ranging from -0.028 to 0.248).
Conclusions:
Patient-reported outcome measures for pain (PROMs) are valid for assessing pain in MS. They are strongly correlated with each other but show weak to moderate correlations with psychosocial factors and non-significant correlations with physical and functional measures.
