Over 85% of the annual 1.5 million Traumatic Brain Injuries (TBIs) that occur in the U.S. are medically classified as mild. Despite the label of mild, these brain injuries have been associated with significant long term disability. Early detection of mild traumatic brain injury (mTBI) patients at risk of developing persistent post-concussion symptoms (PCS) is critical.

A retrospective case-control study of patients with discharge diagnosis of mTBI at a Level I trauma center was conducted to determine if there are factors that can predict which patients are likely to develop persistent symptoms and therefore require follow-up care. Analyses were conducted to describe this population of emergency department (ED) patients that have been identified as having persistent PCS through a concussion management program (CMP) and to identify any differences between the ED and CMP presentation of patients that could provide further evidence in identifying patients at risk for persistent PCS.

Patients presented to the CMP were 56.5% male, mean age of 25.4 years (SD = 14.3), and were a median of 4.57 weeks post-injury (range .43–59 weeks). Their symptoms persisted a median of 19 weeks (range .6–174 weeks). On average patients endorsed statistically significantly more post-injury symptoms, 7.5 (SD = 3.2) at presentation to the CMP then when in the ED, 2.6 (SD = 1.4, paired samples t-test, t[84] = 12.41, p < .001).

Hierarchical multivariate logistic regression of predicting variables resulted in a final model which included: prior mTBI (OR = 18.36, p < .001), forgetfulness/poor memory (OR = 8.46, p < .001), light sensitivity (OR = 5.94, p = .02) and history of anxiety (OR = 4.74, p = .01) The final model has a specificity of 87.9% and a sensitivity of 69.5%.

A strong prediction model for persistent PCS in ED mTBI patients was developed that can be easily implemented in the ED to identify at risk patients for development of persistent PCS and therefore may require close follow-up care.