Is the Rivermead Post-Concussion Symptoms Questionnaire a Reliable and Valid Measure to Assess Long-Term Symptoms in Traumatic Brain Injury and Orthopedic Injury Patients? A Novel Investigation Using Rasch Analysis

Persistent post-concussion syndrome (PCS) symptoms are known to last years after traumatic brain injury (TBI), and similar symptoms are increasingly being documented among those who have not experienced a TBI. There remains however, a dearth of empirical evidence on the structural composition of symptoms beyond the post-acute symptom phase after TBI, and little is known about the potential use of PCS symptom scales to measure PCS-like symptoms in non-TBI individuals. Our objective was therefore to examine the psychometric performance and dimensionality of the Rivermead Post-Concussion Symptoms Questionnaire (RPQ) as a measure of long-term PCS symptoms among a TBI and non-TBI sample. A case-control sample of 223 patients with injury, consisting of age- and sex-matched TBI participants (<i>n</i> = 109) and orthopedic participants (<i>n</i> = 114) were recruited from a regional trauma registry in New Zealand (NZ), and assessed at mean 2.5 years post-injury. Results from the Rasch analysis showed that the RPQ achieved fit to the Rasch model, demonstrating very good reliability (Person Separation Index [PSI] = 0.87), thereby indicating that the measure can be used reliably for individual and group assessment of symptoms among both TBI and orthopedic patients. In this study we demonstrated evidence of a unidimensional construct of PCS symptoms in both groups, which helps alleviate previous uncertainty about factor structure, and permits the calculation of a total RPQ score. Conversion of ordinal to interval total scores presented within are recommended for clinicians and researchers, to improve instrument precision, and to facilitate the interpretation of change scores and use of parametric methods in data analysis.