Validity of the French version of the Core Outcome Measures Index for low back pain patients: a prospective cohort study

Among the many questionnaires available to evaluate low back pain (LBP) patients, the Core Outcome Measures Index (COMI) has the unique advantage to investigate five dimensions using seven short questions. The aim of this study was to explore additional properties of the questionnaire in a French-speaking non-surgical population. This study was conducted on 168 patients suffering from subacute or chronic LBP and followed up for 6 months in three French-speaking countries. In addition to basic psychometric properties (e.g., construct validity, floor and ceiling effect, reproducibility), internal validity was analyzed by a factor analysis using Cronbach's alpha. Responsiveness and sensitivity to change were assessed through minimal detectable change (MDC), effect size, and Minimal Clinically Important Improvement (MCII). We used an anchor-based method with receiver operating characteristic (ROC) curve analysis to assess MCII and the Patient Acceptable Symptom State. Construct validity, reliability (Cronbach's alpha = 0.87), reproducibility and the absence of floor and ceiling effects were confirmed. Factor analysis indicated a one-dimensional construct that validates the use of a sum score. The MDC (2.1) was inferior to the MCII (2.3). The limit below which the patient claims to be in a fair condition (Patient Acceptable Symptom State) was set at 3. The COMI is a self-report questionnaire with the capacity to easily and quickly explore several dimensions in patients with LBP that can be then summarized in a meaningful sum score. Additional knowledge provided by our study should encourage the widespread use of the COMI among the spine community.