Patient-reported outcomes in multiple sclerosis: Relationships among existing scales and the development of a brief measure

Abstract

Several patient-reported outcome (PRO) measures are commonly used in multiple sclerosis (MS) research, but the relationship among items across measures is uncertain. We proposed to evaluate the associations between items from a standard battery of PRO measures used in MS research and to develop a brief, reliable and valid instrument measure by combining these items into a single measure. Subjects (N=537) enrolled in CLIMB complete a PRO battery that includes the Center for Epidemiologic Studies Depression Scale, Medical Outcomes Study Modified Social Support Survey, Modified Fatigue Impact Scale and Multiple Sclerosis Quality of Life-54. Subjects were randomly divided into two samples: calibration (n=269) and validation (n=268). In the calibration sample, an Exploratory Factor Analysis (EFA) was used to identify latent constructs within the battery. The model constructed based on the EFA was evaluated in the validation sample using Confirmatory Factor Analysis (CFA), and reliability and validity were assessed for the final measure. The EFA in the calibration sample revealed an eight factor solution, and a final model with one second-order factor along with the eight first-order factors provided the best fit. The model combined items from each of the four parent measures, showing important relationships among the parent measures. When the model was fit using the validation sample, the results confirmed the validity and reliability of the model. A brief PRO for MS (BPRO-MS) that combines MS-related psychosocial and quality of life domains can be used to assess overall functioning in mildly disabled MS patients.