The Spine Functional Index: development and clinimetric validation of a new whole-spine functional outcome measure

Abstract

Background context: Most spine patient-reported outcome measures are divided into neck and back subregions. This prevents their use in the assessment of the whole spine. By contrast, whole-spine patient-reported outcome measures assess the spine from cervical to lumbar as a single kinetic chain. However, existing whole-spine patient-reported outcomes have been critiqued for clinimetric limitations, including concerns with practicality. Purpose: To develop the Spine Functional Index (SFI) as a new whole-spine patient-reported outcome measure that addressed the limitations of existing whole-spine questionnaires; and to determine the SFI's clinimetric and practical characteristics concurrently with a recognized criterion, the Functional Rating Index (FRI). Study design: Observational cohort study within 10 physical therapy outpatient clinics. Patient sample: Spine-injured patients were recruited from a convenience sample referred by a medical practitioner to physical therapy. A pilot study (n=52, 57% female, age 47.6±17.5 years) followed by the main study (n=203, 48% female, age 41.0±17.8 years) that had an average symptom duration of less than 5 weeks. Outcome measures: Spine Functional Index, FRI, and Numerical Rating Scale (NRS). Methods: The SFI was developed through three stages: 1) item generation, 2) item reduction with an expert panel and patient focus group, and 3) pilot field testing to provide provisional clinimetric properties and sample size requirements and to determine suitability for a larger study. Participants completed the SFI, FRI, and NRS every 2 weeks for 6 weeks, then every 4 weeks until discharge or study completion at 6 months. Responses were assessed to provide individual psychometric and practical characteristics for both patient-reported outcomes, with the overall performance evaluated by the Measurement of Outcome Measures and Bot clinimetric assessment scales. Results: The SFI demonstrated a high criterion validity with the FRI (Pearson r=0.87, 95% confidence interval [CI]), equivalent internal consistency (a=0.91), and a single-factor structure. The SFI and FRI demonstrated suitable reliability (intraclass correlation coefficient2,1=0.97:0.95), responsiveness (standardized response mean=1.81:1.68), minimal detectable change with 90% CI (6.4%:9.7%), Flesch scale reading ease (64%:47%), and user errors (1.5%:5.3%). The clinimetric performance was higher for the SFI on the Measurement of Outcome Measures (96%:64%) and on the Bot scale (100%:75%). Conclusions: The SFI demonstrated sound clinimetric properties with lower response errors, efficient completion and scoring, and improved responsiveness and overall clinimetric performance compared with the FRI. These results indicated that the SFI was suitable for functional outcome measurement of the whole spine in both the research and clinical settings. © 2013 Elsevier Inc. All rights reserved.