Characteristics and outcomes of athletes with slow recovery from sport-related concussion

Key points

1. 0122.8% of 1,751 concussed collegiate athletes (n = 399) met the slow-recovery definition (more than 14 days to asymptomatic and/or more than 24 days to return to play)
2. 02Slow-recovery athletes were more likely to be female, injured during practice, and had higher initial symptom severity and worse balance and cognition scores within 48 hours
3. 0377.6% of slow-recovery athletes returned to play within 60 days and 83.4% within 88 days of injury
4. 04Only 10.6% of the slow-recovery group had not returned to play by 6 months post-injury, representing just 2.4% of the entire concussed cohort
5. 05No clinical or demographic variable reliably predicted which slow-recovery athlete would eventually return to play, suggesting early markers lose predictive value after the initial recovery window

How it was conducted

Design

Prospective multisite observational cohort study (NCAA-DoD CARE Consortium, 30 sites, Fall 2014 to Spring 2018)

Participants

1,751 concussed collegiate varsity athletes (mean age 19.2 years, 63% male, 37% female; contact, limited-contact, and non-contact sports)

Slow-recovery definition

Time to asymptomatic status of 14 or more days and/or time to unrestricted return to play of 24 or more days (the slowest 20% of the broader cohort)

Primary outcomes

Time from injury to asymptomatic status and time to unrestricted return to play, analyzed with Kaplan-Meier survival curves and Cox proportional hazards models

Assessments

SCAT-3 symptom severity, Standardized Assessment of Concussion (SAC), Balance Error Scoring System (BESS), Brief Symptom Index (BSI), and daily symptom reporting

What they found

• 22.8% of concussed athletes (399/1,751) met the slow-recovery definition
• Slow-recovery athletes were more likely to be female (41.1% vs 35.6%, p = 0.05) and injured in practice (70.7% vs 65.1%, p = 0.04)
• Initial postinjury SCAT symptom severity was higher in slow-recovery athletes (mean 36.62 [SD 23.42] vs 25.41 [SD 19.85], p < 0.001)
• Initial postinjury SAC scores were lower in slow-recovery athletes (mean 25.74 [SD 2.98] vs 26.26 [SD 2.85], p = 0.004)
• Initial postinjury BESS total score was worse in slow-recovery athletes (mean 17.81 [SD 8.91] vs 15.91 [SD 8.50], p = 0.002)
• Median time to return to play in the slow-recovery group was 34.7 days (Q25-Q75: 32.6-36.7) vs 12.8 days (Q25-Q75: 8.7-20.1) in the overall concussed cohort
• 77.6% of the slow-recovery group returned to play within 60 days; 83.4% within 88 days
• Of those unable to return to play at 90 days (n = 50), the median return to play was 187.1 days
• 10.6% of the slow-recovery group (2.4% of the entire cohort) had not returned to play by 180 days
• History of prior concussion was the only significant predictor of return-to-play timing within the slow-recovery group (hazard ratio 0.90, 95% CI 0.80-1.00, p = 0.05); no variable showed strong predictive ability (all C-statistics near 0.50)
• Slow-recovery athletes had significantly fewer postinjury assessment contacts in the first 14 days (mean 48.84% [SD 29.70] vs 67.93% [SD 24.64] compliance, p < 0.001)

Limitations

• All participants were collegiate varsity athletes, limiting generalizability to younger athletes, recreational players, or non-sport mild traumatic brain injury populations
• 73 of 1,751 injuries were right-censored (season ended or athlete left campus before recovery endpoints were captured), so final recovery status is unknown for those individuals
• Brief Symptom Index data were missing for 46% of the slow-recovery group, making it impossible to draw firm conclusions about the role of psychological factors
• The CARE protocol did not systematically capture autonomic, vestibular, or oculomotor function, which may be important drivers of prolonged recovery

Why it matters

For patients

Athletes who are still not cleared after two to three weeks can be reassured that the large majority in similar situations do eventually return to play, typically within two to three months.

For clinicians

Early clinical markers (symptom severity, balance, cognition) identify athletes at risk for slow recovery, but once an athlete is in the slow-recovery category those same markers do not reliably predict eventual return-to-play timing, warranting individualized management.

For readers

This large prospective study provides the first data-driven threshold and prognosis for slow concussion recovery in college athletes, supporting a measured, time-extended approach rather than early medical disqualification.