Evaluation of the Functional Assessment of Concussion Tool in a Pediatric Sports Concussion Clinic

Background: Mild Traumatic Brain Injury (mTBI) affects nearly two million youth annually, but duration and presentation of mTBI symptoms varies greatly. We developed a tool which provides a functional assessment of concussion symptoms by focusing on how symptoms affect normal daily activities rather than rating symptom severity. This study sought to determine the correlation of the Functional Assessment of Concussion Tool App (FACT) with post-concussion symptom scales and to evaluate functional outcomes at home in pediatric patients following acute mTBI. Methods: 27 subjects ages 8-18 completed a SCAT3 Symptom Scale (SCAT), Post-Concussion Symptom Scale (PCSS), FACT app assessment, Standardized Assessment of Concussion, Somatization Subscale of Child Behavior Checklist, and Pain Catastrophizing Scale in the concussion clinic upon enrollment in the study. Subjects then reported the number of symptoms they experienced and rated the symptoms’ impact on normal activity in the FACT app. Subjects completed a PCSS assessment at each subsequent clinic visit. Linear regression models (simple and multiple) were used to determine the predictive value of the number of FACT symptoms reported and the FACT rating on SCAT3 and PCSS scores across four domains: physical, mood, sleep, and thinking and remembering. Results: The total FACT symptom number was associated with the total SCAT score and PCSS score, and the FACT domain symptom number was associated with the SCAT domain scores and the PCSS domain scores in the four domains (p<0.03). The FACT rating score was associated with the total SCAT score and PCSS score (p<0.01) and SCAT score and PCSS score in the mood and thinking and remembering domains (p<0.02). In the multiple linear regression models controlling for the FACT rating score, the FACT symptom number was a significant predictor of the total SCAT score and PCSS score and SCAT scores in all four domains (p<0.01) and PCSS scores in the physical, mood, and thinking and remembering domains (p<0.01). However, the FACT rating score was not a significant predictor of any SCAT scores or PCSS scores in these multiple linear regression models. FACT symptom scores and clinical PCSS scores trended together, but neither was a significant predictor of recovery. Conclusion: The FACT symptoms are correlated with SCAT scores and PCSS scores across all domains of symptoms, suggesting construct validity. The FACT rating score is a significant predictor of SCAT scores and PCSS scores across some but not all domains, which suggests that FACT may be assessing different elements of symptoms. Although FACT symptom scores and clinical PCSS scores trended together, neither was a significant predictor of recovery. Further analysis with a larger sample size may be needed to more accurately assess the predictive value of FACT.