The Symptom Inventory Disability-Specific Short Forms for Multiple Sclerosis: Construct Validity, Responsiveness, and Interpretation

Schwartz CE, Bode RK, Quaranto BR, Vollmer T. The Symptom Inventory Disability-Specific Short Forms for multiple sclerosis: construct validity, responsiveness, and interpretation. Arch Phys Med Rehabil 2012;93: 1617-28. Objectives: To test the cmss-sectional and longitudinal construct validity of the disability-specific short forms of the Symptom Inventory for multiple sclerosis, to compare its internal consistency reliability and construct validity with those of the original (1999) 29-item short form of the Symptom Inventory, and to provide for the new disability-specific short forms interpretation metrics for use in sample size calculation for future research. Design: A Web-based longitudinal study, with data collected at baseline and 6 months after baseline. Correlations evaluated the overlap among disease-specific and generic patient-reported outcome measures. Responsiveness was assessed by using symptom transition scores and modified standardized response means. Interpretation guidelines were provided by using Cohen's effect size and crosswalks to the disease-specific and generic quality-of-life measures. Setting: National Multiple Sclerosis Registry. Participants: People with multiple sclerosis (N=1142) who participated in the North American Research Committee on Multiple Sclerosis Registry. Interventions: Not applicable. Main Outcome Measures: The Symptom Inventory; the disease-specific Performance Scales and the Patient-Determined Disease Steps; the generic Short Form 12. Results: The Symptom Inventory evidenced convergent and divergent validity, and the disability-specific short forms evidenced similar psychometric performance as the 1999 short form but had slightly better alpha reliability. They also evidenced moderate responsiveness to clinically important change, with more responsiveness among those with mild and moderate disabilities than among those with severe disabilities. Effect sizes were larger among patients who reported symptom improvement, rather than deterioration, suggesting that the tool would be more useful in clinical research focused on testing whether an intervention improves symptom experience, particularly for patients with mild and moderate disabilities. Crosswalks provided graphic and numeric links between the Symptom Inventory and other patient-reported outcomes. Conclusions: The Symptom Inventory can be useful for elucidating the patient's experience, but it varies considerably across and within disability groupings. Directions for future research are discussed.