Factors Associated With Return-to-Learn Duration After Concussion in US Service Academy Cadets and Midshipmen

Objective: Patterns of clinical recovery and factors that may influence return to activity after concussion have been well studied; however, data regarding return-to-learn (RTL) duration after concussion and factors that influence this period have not been adequately explored. Hypothesis: We have hypothesized that sport level, sex, concussion history, and initial symptom burden will be associated with RTL duration after concussion. Setting: The study was conducted at US military service academies. Study Design: This was a prospective cohort study. Participants: A total of 1509 service academy cadets (41% female, 20.2 ± 1.48 y, 174.9 ± 10.3 cm, 73.8 ± 14.0 kg) who sustained a concussion during the study period from December 2014 to March 2020 were included. Main Measures: The primary outcome of interest was time elapsed between an incident injury and self-reported return to normal academic performance. All participants provided baseline demographics, including sex (male, female), sport level (varsity, non-varsity), and concussion history (yes, no). Participants completed standardized concussion assessments at regular time points. Initial symptom burden was collected within 48 hours after injury, and self-reported RTL duration was collected during the unrestricted return-to-activity (URTA) evaluation. Results: On average, RTL duration was faster in men (8.1 ± 11.2 days) than in women (13.6 ± 20.3 days). Univariate models demonstrated that participants with an initial symptom burden ≤6 returned sooner than those endorsing 7–15 symptoms (HR = 0.73, 95% CI = 0.64–0.83, P ≤ .001) and ≥16 symptoms (HR = 0.56, 95% CI = 0.48–0.65, P ≤ .001). Sex (HR = 0.70, 95% CI = 0.62–0.77, P ≤ .001), and sport level (HR = 0.66, 95% CI = 0.59–0.73, P ≤ .001) also contributed to RTL duration in the univariate model. In the multivariable model, participants with an initial symptom burden ≤6 returned 25%–41% sooner than those endorsing 7–15 symptoms (HR = 0.75, 95% CI = 0.65–0.86, P ≤ .001) and ≥16 symptoms (HR = 0.59, 95% CI = 0.51–0.69, P ≤ .001). Sex (HR = 0.71, 95% CI = 0.64–0.80, P ≤ .001) and sport level (HR = 0.75, 95% CI = 0.67–0.85, P < .001) also contributed to RTL duration in the multivariable model. Conclusions: RTL duration for men, varsity athletes, and individuals with fewer initial symptoms after injury was significantly faster than for women, nonvarsity athletes, and individuals with a greater initial symptom burden.