Optimizing Concussion Care Seeking: Using Machine Learning to Predict Delayed Concussion Reporting.

BACKGROUND: Early medical attention after concussion may minimize symptom duration and burden; however, many concussions are undiagnosed or have a delay in diagnosis after injury. Many concussion symptoms (eg, headache, dizziness) are not visible, meaning that early identification is often contingent on individuals reporting their injury to medical staff. A fundamental understanding of the types and levels of factors that explain when concussions are reported can help identify promising directions for intervention. PURPOSE: To identify individual and institutional factors that predict immediate (vs delayed) injury reporting. STUDY DESIGN: Case-control study; Level of evidence, 3. METHODS: This study was a secondary analysis of data from the Concussion Assessment, Research and Education (CARE) Consortium study. The sample included 3213 collegiate athletes and military service academy cadets who were diagnosed with a concussion during the study period. Participants were from 27 civilian institutions and 3 military institutions in the United States. Machine learning techniques were used to build models predicting who would report an injury immediately after a concussive event (measured by an athletic trainer denoting the injury as being reported "immediately" or "at a delay"), including both individual athlete/cadet and institutional characteristics. RESULTS: In the sample as a whole, combining individual factors enabled prediction of reporting immediacy, with mean accuracies between 55.8% and 62.6%, depending on classifier type and sample subset; adding institutional factors improved reporting prediction accuracies by 1 to 6 percentage points. At the individual level, injury-related altered mental status and loss of consciousness were most predictive of immediate reporting, which may be the result of observable signs leading to the injury report being externally mediated. At the institutional level, important attributes included athletic department annual revenue and ratio of athletes to athletic trainers. CONCLUSION: Further study is needed on the pathways through which institutional decisions about resource allocation, including decisions about sports medicine staffing, may contribute to reporting immediacy. More broadly, the relatively low accuracy of the machine learning models tested suggests the importance of continued expansion in how reporting is understood and facilitated.

Migraine epidemiology in collegiate student-athletes: Findings from the Concussion Assessment, Research, and Education (CARE) Consortium.

OBJECTIVE: This cross-sectional study evaluated de-identified data from the National Collegiate Athletic Association-Department of Defense Grand Alliance from 2014 to 2020 to determine the prevalence of migraine and migraine medication and to describe differences in migraine prevalence by sex, race, and sport. BACKGROUND: Epidemiological studies can help identify underdiagnosed and undertreated populations. Understanding migraine prevalence in collegiate student-athletes is essential for positive healthcare outcomes including development of prevention and treatment plans. METHODS: From a concussion baseline assessment, participant's self-reported demographics (e.g., age, sex, sport), migraine diagnosis (i.e., yes/no), and migraine medication usage (e.g., yes/no, type) determined prevalence of migraine and medication use in collegiate student-athletes. RESULTS: Migraine was reported in 5.6% (2617/47,060; 95% confidence interval [CI] 5.4%-5.8%) of the student-athletes, with higher prevalence in females, 7.5% (1319/17,628; 95% CI 7.1%-7.9%), than males, 4.6% (1298/28,116; 95% CI 4.4%-4.9%). Medication usage was reported by 36.2% (947/2617; 95% CI 34.3%-38.0%) of individuals with migraine. Migraine reporting differed by race, with Caucasian reporting highest (5.9%; 1990/33,913; 95% CI 5.6%-6.1%) and Asian the lowest (2.7%; 55/2027; 95% CI 2.1%-3.5%). Women's sports, including golf, gymnastics, and lacrosse, and men's diving and squash had higher migraine reporting than other sports. CONCLUSION: Caucasian females reported higher rates than other groups and sport influenced rates of migraine diagnosis.

Association of Premorbid Anxiety and Depression Symptoms in Concussion Recovery in Collegiate Student-Athletes.

BACKGROUND: Mental health disorders are linked to prolonged concussion symptoms. However, the association of premorbid anxiety/depression symptoms with postconcussion return-to-play timelines and total symptom burden is unclear. OBJECTIVE: To examine the association of self-reported premorbid anxiety/depression symptoms in collegiate student-athletes with (1) recovery times until asymptomatic, (2) return-to-play, and (3) postconcussion symptom burden. STUDY DESIGN: Athletes in the Concussion Assessment, Research and Education Consortium completed baseline concussion assessments (Sport Concussion Assessment Tool [SCAT3] and Brief Symptom Inventory-18 [BSI-18]). Athletes were tested postinjury at <6 hours, 24 to 48 hours, time of asymptomatic and start of return-to-play protocol, unrestricted return-to-play, and 6 months after injury. Injured athletes were categorized into 4 groups based on BSI-18 scores: (1) B-ANX, elevated anxiety symptoms only; (2) B-DEP, elevated depression symptoms only; (3) B-ANX&DEP, elevated anxiety and depression symptoms; and (4) B-NEITHER, no elevated anxiety or depression symptoms. Relationship between age, sex, BSI-18 group, SCAT3 total symptom and severity scores, and time to asymptomatic status and return-to-play was assessed with Pearson's chi-squared test and robust analysis of variance. LEVEL OF EVIDENCE: Level 3. RESULTS: Among 1329 athletes with 1352 concussions, no respondents had a self-reported premorbid diagnosis of anxiety/depression. There was no difference in time until asymptomatic or time until return-to-play between BSI-18 groups (P = 0.15 and P = 0.11, respectively). B-ANX, B-DEP, and B-ANX&DEP groups did not have higher total symptom or severity scores postinjury compared with the B-NEITHER group. CONCLUSION: Baseline anxiety/depression symptoms in collegiate student-athletes without a mental health diagnosis are not associated with longer recovery times until asymptomatic, longer time to return-to-play, or higher postconcussion total symptom and severity scores compared with athletes without baseline symptoms. CLINICAL RELEVANCE: Anxiety and depression symptoms without a clear mental health diagnosis should be considered differently from other comorbidities when discussing prolonged recovery in collegiate student-athletes.

Mechanisms of Injury Leading to Concussions in Collegiate Soccer Players: A CARE Consortium Study.

BACKGROUND: Few previous studies have investigated how different injury mechanisms leading to sport-related concussion (SRC) in soccer may affect outcomes. PURPOSE: To describe injury mechanisms and evaluate injury mechanisms as predictors of symptom severity, return to play (RTP) initiation, and unrestricted RTP (URTP) in a cohort of collegiate soccer players. STUDY DESIGN: Cohort study; Level of evidence, 2. METHODS: The Concussion Assessment, Research and Education (CARE) Consortium database was used. The mechanism of injury was categorized into head-to-ball, head-to-head, head-to-body, and head-to-ground/equipment. Baseline/acute injury characteristics-including Sports Concussion Assessment Tool-3 total symptom severity (TSS), loss of consciousness (LOC), and altered mental status (AMS); descriptive data; and recovery (RTP and URTP)-were compared. Multivariable regression and Weibull models were used to assess the predictive value of the mechanism of injury on TSS and RTP/URTP, respectively. RESULTS: Among 391 soccer SRCs, 32.7% were attributed to a head-to-ball mechanism, 27.9% to a head-to-body mechanism, 21.7% to a head-to-head mechanism, and 17.6% to a head-to-ground/equipment mechanism. Event type was significantly associated with injury mechanism [χ2(3) = 63; P < .001), such that more head-to-ball concussions occurred in practice sessions (n = 92 [51.1%] vs n = 36 [17.1%]) and more head-to-head (n = 65 [30.8%] vs n = 20 [11.1]) and head-to-body (n = 76 [36%] vs n = 33 [18.3%]) concussions occurred in competition. The primary position was significantly associated with injury mechanism [χ2(3) = 24; P < .004], with goalkeepers having no SRCs from the head-to-head mechanism (n = 0 [0%]) and forward players having the least head-to-body mechanism (n = 15 [19.2%]). LOC was also associated with injury mechanism (P = .034), with LOC being most prevalent in head-to-ground/equipment. Finally, AMS was most prevalent in head-to-ball (n = 54 [34.2%]) and head-to-body (n = 48 [30.4%]) mechanisms [χ2(3) = 9; P = .029]. In our multivariable models, the mechanism was not a predictor of TSS or RTP; however, it was associated with URTP (P = .044), with head-to-equipment/ground injuries resulting in the shortest mean number of days (14 ± 9.1 days) to URTP and the head-to-ball mechanism the longest (18.6 ± 21.6 days). CONCLUSION: The mechanism of injury differed by event type and primary position, and LOC and AMS were different across mechanisms. Even though the mechanism of injury was not a significant predictor of acute symptom burden or time until RTP initiation, those with head-to-equipment/ground injuries spent the shortest time until URTP, and those with head-to-ball injuries had the longest time until URTP.

Sex Differences in Recovery Trajectories of Assessments for Sport-Related Concussion Among NCAA Athletes: A CARE Consortium Study.

OBJECTIVES: The purpose of this study was to determine sex differences in recovery trajectories of assessments for sport-related concussion using Concussion Assessment, Research and Education (CARE) Consortium data. METHODS: National Collegiate Athletic Association athletes (N = 906; 61% female) from sex-comparable sports completed a pre-season baseline assessment and post-sport-related concussion assessments within 6 h of injury, 24-48 h, when they initiated their return to play progression, when they were cleared for unrestricted return to play, and 6 months post-injury. Assessments included the Standardized Assessment of Concussion, Balance Error Scoring System, Brief Symptom Inventory-18, Immediate Post-concussion Assessment and Cognitive Testing (ImPACT), Sport Concussion Assessment Tool-3 symptom evaluation, Clinical Reaction Time, King-Devick test, Vestibular Ocular Motor Screen, 12-item Short-Form Health Survey, Hospital Anxiety and Depression Scale, and Satisfaction with Life Scale. RESULTS: Only the Vestibular Ocular Motor Screen Total Symptom Score at the 24-48 h timepoint (p = 0.005) was statistically significantly different between sexes. Specifically, female athletes (mean = 60.2, 95% confidence interval [CI] 51.5-70.4) had higher Vestibular Ocular Motor Screen Total Symptom Scores than male athletes (mean = 36.9, 95% CI 27.6-49.3), but this difference resolved by the time of return-to-play initiation (female athletes, mean = 1.8, 95% CI 1.1-2.9; male athletes, mean = 4.1, 95% CI 1.5-10.9). CONCLUSIONS: Sport-related concussion recovery trajectories for most assessments were similar for female and male National Collegiate Athletic Association athletes except for Vestibular Ocular Motor Screen symptoms within 48 h of sport-related concussion, which was greater in female athletes. Female athletes had a greater symptom burden across all timepoints, suggesting that cross-sectional observations may indicate sex differences despite similar recovery trajectories.