The Multidimensional Symptom Index: A new patient‐reported outcome for pain phenotyping, prognosis and treatment decisions. (21st May 2018)

• Record Type: Journal Article
• Authors: Walton, D.M.; Marsh, J.
• Is Part Of: European journal of pain. Volume 22:Number 7(2018)
• Page Start: 1351
• Page End: 1361
• Publication Date: 2018-05-21
• Subjects: Pain -- Periodicals; Pain -- Treatment -- Periodicals; Pain -- Physiological aspects -- Periodicals; 616.0472
• DOI: 10.1002/ejp.1224
• Languages: English
• ISSNs: 1090-3801

Abstract

Abstract: Background: There are few patient‐reported outcomes routinely used that capture frequency and interference of different pain‐related symptoms on a single scale. The purpose of this study was to describe the development and initial validation of the new Multidimensional Symptom Index (MSI). Methods: Items were generated from patient interviews of the experience of chronic pain. Health valuations were created from rankings of 82 healthy subjects for each of 120 symptom (×10) × frequency (×3) × interference (×4) combinations using preference‐based health valuations (0–100). Ranks for each symptom combination were then used in scale scoring. A sample of 300 patients with acute or chronic pain subsequently completed the MSI and a battery of other tools. Exploratory (EFA) and Confirmatory (CFA) factor analyses were triangulated with theory to arrive at the factor structure. Convergent validity was tested against established measures. Results: Health rankings resulted in scores of 0–12 for each of the 10 symptom types. Factor analyses revealed two factors: MSI Somatic Symptoms and MSI Non‐Somatic Symptoms . The MSI also quantified number of symptoms experienced (/10), mean frequency (/3) and mean interference (/4). The indices showed appropriate associations with the established PROs. Conclusions: The MSI is a new symptom‐focused PRO that allows patient phenotyping and may have value for screening, prognosis and evaluating change. Significance: This article presents the development and psychometric properties of a new measure of pain and related symptom frequency and interference. This measure could aid clinicians in establishing clinically relevant pain phenotypes for screening, prognosis and treatment decisions.