A - 23 Confirmatory Factor Analysis of the Concussion Profile (CP) Screen
Abstract Objective The Concussion Profile (CP)-screen is intended to assess symptom domains of concussion, including migraine, vestibular, ocular, mood/anxiety, and cognitive/fatigue domains. The CP-screen was developed to develop more targeted management by identifying presentation profiles. The hypothesized profile structure of the CP-screen has not yet been psychometrically validated. An a priori five-factor structure (migraine; vestibular; ocular; mood/anxiety; cognitive/fatigue) was hypothesized. Methods Three-hundred twenty-three patients (age M = 15.62 ± 4.03 years, 59% male; 74% sport-related) seen within 30 days of concussion (M = 7.58 ± 6.10 days) completed a medical history clinical interview and the CP-screen at their first clinical visit. Multiple confirmatory factor analyses (CFAs) of CP-screen items were estimated (using robust weighted least squares estimation). Results A CFA was completed for each factor separately and then a five-factor CFA was estimated. Model fit was evaluated using the Comparative Fit Index [CFI], Tucker-Lewis Index [TLI], Root Mean Square Error of Approximation [RMSEA], and inspection of model parameters (i.e., loadings, residuals, & factor correlations). The five-factor model had excellent fit, but factor correlations indicated a second-order model was more appropriate. Thus, the final CFA included a five-factor first-order structure, with migraine, ocular, and cognitive/fatigue factors loading onto a second-order factor. All models had good fit (CFI >0.965; TLI > 0.96; RMSEA <0.072), and all loadings were significant and substantial (>0.63). Conclusions This is the first study to provide psychometric support for the profile structure of the CP-screen. These results also provide support for using these profiles to help target management of those with concussions.