Factors Associated with Persisting Post-Concussion Symptoms Among Collegiate Athletes and Military Cadets: Findings from the NCAA-DoD CARE Consortium

Abstract

Background

Persisting post-concussion symptoms (PPCS) is a condition characterized by prolonged recovery from a mild traumatic brain injury (mTBI) and compromised quality of life. Previous literature, on the basis of small sample sizes, concludes that there are several risk factors for the development of PPCS.

Objective

We seek to identify protective and risk factors for developing slow recovery or persisting post-concussion symptoms (PPCS) by analyzing medical history, contact sport level, setting, and the Sport Concussion Assessment Tool (SCAT) and Brief Symptom Inventory (BSI-18) assessments at baseline and post-injury.

Patients and Methods

We studied 47,860 unique collegiate athletes and Military Service Academy (MSA) cadets enrolled in the 30-site National Collegiate Athletic Association and Department of Defense (NCAA-DoD) Concussion Assessment Research and Education (CARE) Consortium prospective cohort study of baseline and post-injury data from 2014 to 2019. Medical histories and preseason baselines (n = 60,720), along with SCAT and BSI-18 examinations (n = 5379) conducted within 48 h post-injury, were analyzed. PPCS is defined as cleared for return to play (RTP) > 29 days.

Results

Of 5073 concussions, the median (95% CI) and mean (SD) RTP were 15.2 (4.8–82.6) days and 22.2 (23.0) days, respectively, with 891 developing PPCS. Sex, high SCAT score, high BSI-18 score, and delayed reporting produced small effect sizes on RTP (d = 0.22–0.44). Adjusted odds ratios (OR) of developing PPCS indicated the following risk factors: SCAT total score > 45 (OR = 1.91, 95% CI: 1.58–2.30), female sex (OR = 1.80, 95% CI: 1.53–2.13), concussion history (OR = 1.80, 95% CI: 1.29–2.52), and delayed reporting (OR = 1.42, 95% CI: 1.20–1.67). In contrast, protective factors against developing PPCS were: being a limited-contact (OR = 0.34, 95% CI: 0.25–0.47), noncontact (OR = 0.35, 95% CI: 0.24–0.51) or contact sport athlete (OR = 0.34, 95% CI: 0.28–0.41); and receiving the concussion at practice (OR = 0.64, 95% CI: 0.53–0.77) or in competition (OR = 0.44, 95% CI: 0.34–0.56). Athletes diagnosed with a learning disability and taking a neurostimulant were more protected than those who were not (OR = 0.44, 95% CI: 0.26–0.76 versus OR = 0.72, 95% CI: 0.49–1.05, respectively). A prognostic model using these variables offers poor sensitivity (9%) but high specificity (98%) in identifying PPCS (AUC = 0.72).

Conclusions

This study of 5073 concussions indicates that female sex, high symptom burden, and prior concussion are risk factors for slow recovery. In addition, learning disability, contact sport, and concussion in competition are protective against slow recovery. The neurostimulant results suggest that clinicians should keep their attention deficit disorder (ADD) and attention-deficit/hyperactivity disorder (ADHD) patients on their current neurostimulant medication after a concussion. A prognostic logistic regression model based on behavioral clinical findings did a poor job of identifying PPCS.