Season Timing and Concussion Outcomes in Division I Football
Key Takeaways
- Preseason accounted for the largest share of concussions and had the highest incidence, with a statistically significant seasonal distribution.
- Time to symptom resolution differed by season period, with slower recovery observed after early- and late-season concussions than after preseason injuries.
- Time to return to play did not differ across season periods by log-rank testing, although early-season concussions showed a lower return-to-play hazard in Cox regression.
The analysis included 682 NCAA athletes from 21 institutions who sustained concussion during practice or competition, within a broader CARE database spanning 30 universities from 2014 to 2020. Preseason also accounted for the largest share of concussions across the season periods examined. Recovery timing differed across the season as well, extending the seasonal signal beyond incidence alone. In this collegiate cohort, preseason had both the highest incidence and the largest share of concussions.
Investigators examined whether seasonality was associated with concussion incidence and recovery outcomes, defined as time to symptom resolution and return to play. This prospective cohort analysis focused on Division I football players with concussion in the multi-institution CARE Consortium rather than a single-program sample. Researchers compared preseason, early season, middle season, and late season among 682 athletes across 21 institutions after concussions sustained in practice or competition. The abstract described chi-square testing, Kaplan-Meier curves, and Cox proportional hazards regression as part of the analysis of how recovery varied by season timing. The study question centered on whether season timing tracked with both injury frequency and the pace of symptom resolution and return to play.
Incidence rates were 4.73 per 100 players in preseason, 2.19 in early season, 2.31 in middle season, and 2.17 in late season. The distribution of concussions was 41.9% in preseason, 19.2% in early season, 20.1% in middle season, and 18.8% in late season. That seasonal distribution was significant, with χ2 = 104.6 and P < .001. Time to symptom resolution differed by season period, with log-rank P = .012. Early-season concussions had delayed symptom resolution versus preseason, with HR = 0.582 and P < .001, and late-season concussions had HR = 0.634 and P = .002.
Time to return to play did not differ across season periods, with log-rank P = .300. Early-season concussions were nevertheless associated with a lower return-to-play hazard, with HR = 0.703 and P = .040.
The authors concluded that contextual factors may influence concussion outcomes and warrant mechanistic investigation to inform future management strategies. Overall, the seasonal pattern was more apparent for symptom resolution than for overall return-to-play timing in the findings.