- A family history of psychiatric disorders was the variable with the single greatest magnitude of impact in predicting athletes requiring additional sports concussion clinic visits.
- The initial symptom score has the potential to be a significant predictor of additional clinic visits if the score is of a large magnitude.
- Understanding factors that predict the need for additional concussion clinic visits versus release to an athletic trainer may optimize care in clinic and minimize value loss.
Concussion diagnosis and management in student-athletes is commonly multidisciplinary. Many athletes are removed from play and initially evaluated by an athletic trainer (AT).1 Depending on state and national laws, a concussion diagnosis can be rendered by a trainer, team physician, primary care provider, emergency department (ED)/urgent care provider, at a specialty concussion clinic, or by other medical professionals.2,3 Broadly speaking, 85% to 90% of student-athletes take up to 3 weeks to clinically recover from their concussion and require minimal medical interventions.4–6 Athletic trainers serve as important touchpoints initially and at all stages during recovery because ATs see the athletes on a regular basis and often implement/monitor treatment recommendations.7 Outpatient medical providers will often discharge an athlete to an AT to execute a return-to-play protocol.7
The time course from an evaluation in concussion clinic to discharging into the care of an AT is variable and depends on the nature and symptoms of the injury.8 Some athletes are immediately released to be managed by ATs, while others will first require multiple follow-up visits with medical providers (eg, a specialty concussion clinic). Identifying the magnitude of these predictors may direct and shape treatment protocols to optimize care in the clinic to reduce recovery duration. Concussion care can take up a significant amount of time and clinic resources in busy sports medicine or other specialty clinics. Identifying athletes who can be safely discharged to an AT versus athletes who benefit from additional clinic visits may reduce the burden of concussion care on the mainstream healthcare system by reducing the need to hold clinic slots and minimizing value loss related to no-show visits.
Concussion recovery is dependent on factors both related to the injury characteristics/subsequent symptoms as well as an individual's demographics, prior health history, and other variables.9,10 Variables for a prolonged concussion recovery have been well-documented.10 These include the initial severity of symptoms, subacute problems with headaches, personal history of mental health issues, prior concussions, family history of psychiatric disorders, and female sex.8,10–12 Ideally, factors associated with worsened concussion recovery should be consistent with the factors that predict additional clinic visits for concussion evaluation. Any discrepancies between these variables may prompt further investigation into the impact on quality of care.
Despite identification of factors that predict concussion recovery, few studies have examined the variables that predict additional follow-up visits in concussion clinics before discharge to ATs. In children from age 6 to 12 years, younger age and greater symptom scores were associated with longer time to discharge from clinic.13 Meanwhile, no association was found between time to clinic discharge and personal history of prior concussions, cognitive ability, symptom severity on initial presentation to clinic, or family history of attention-deficit/hyperactivity disorder (ADHD)/learning disorders.13 No studies to the best of our knowledge have examined the factors associated with releasing student-athletes to their ATs to begin the return-to-play process. Therefore, the purpose of this study was to identify factors at the time of initial assessment that are associated with additional student-athlete clinic visits in a multidisciplinary concussion clinic before discharge to an AT compared with those who were discharged to an AT after an initial visit. We hypothesized that the same factors associated with prolonged recovery (eg, age,13 initial symptom severity,10 and personal history of depression/anxiety10) would be associated with additional clinic visits.
Study Design and Patient Selection
This retrospective cohort study was based on chart extractions from a multidisciplinary sports concussion clinic that includes a variety of providers (ie, neuropsychology, neurology, neurosurgery, pediatrics, sports medicine, physical therapy, occupational therapy, speech therapy, and psychiatry). The Institutional Review Board (No. 192033) determined the study to be exempt, and the requirement for consent was waived. Patients were eligible if they were age 12 to 23 years, sustained a sport-related concussion (SRC) and were seen by a clinic provider between January 11, 2017, and January 10, 2020, for concussion (n = 1504). Patients were excluded if they sustained a concussion from a nonsport mechanism, outside the study age range, sustained an injury before the study period, were not assessed in the concussion clinic, presented to concussion clinic more than 3 months after concussion, or had positive acute head imaging. The diagnosis of concussion was based on ICD-9 and ICD-10 codes (eg, S06.0X) for concussion, postconcussion syndrome (eg, 310.2), and variants of these diagnoses (eg, F07.81). All providers used the definition of SRC from the most recent Concussion in Sport Group guidelines.14 Most patients (84.5%) have their initial visit to our program with an orthopedic/sports medicine physician, who decides whether they need further management within our program or they are able to be managed by their AT, limiting the variability across providers regarding the primary outcome variable in our study. The practitioners within the group rendering the initial decision consisted of mainly orthopedic/sports medicine physicians, with a small proportion of neurosurgeons and neuropsychologists. A detailed patient flow diagram with complete exclusion criteria is shown in Figure 1.
Manual chart review was performed with data extraction from provider notes into a REDCap (Research Electronic Data Capture) database hosted at Vanderbilt University.15,16 The first health system contact was defined as the location of the first, non-AT healthcare provider seen after the time of injury (eg, ED/urgent care, sports concussion clinic, and other clinic). During the initial clinic visit, severity of symptoms was quantitatively recorded using self-report by patients on the postconcussion symptom scale (PCSS). This scale is widely administered to patients after a concussion.17 Patients rate each of 22 symptoms on a scale from 0 (no symptoms) to 6 (severe).17 Individual symptom scores are summed for an overall severity score. Internal consistency is estimated to be 0.93.18
The primary outcome was dichotomous (ie, discharged to AT after initial clinic visit; attended additional clinic visits before being discharged to AT). Discharge to AT occurred if the provider believed that the athlete could be released to a school-based AT for oversight in the return-to-play process. In the descriptive analysis (that is, frequencies and proportions for categorical data and concussion history; mean [M], median [Md], [SD], and interquartile ranges [IQR] for continuous data), χ2 statistics were used to assess differences in the proportion of nominal variables between the initial discharge to AT versus addition visits groups (eg, sex, personal and family histories of medical issues). Furthermore, χ2 statistics were used to evaluate the proportions of patients who underwent specific healthcare evaluations before entering our clinic. The distribution of age and PCSS scores was visually inspected for normality and skewness. Both variables were skewed, so we used nonparametric Mann–Whitney U tests to assess group differences. A binary logistic regression evaluated the independent variables that predicted being initially discharged to an AT versus needing additional clinic visits. Based on previous studies, the following predictor variables were selected a priori for the model: age (in years), PCSS severity score, number of prior concussions (0-4), sex (1 = female, 0 = male), family history of psychiatric disorder (1 = yes, 0 = no), personal history of psychiatric disorder (1 = yes, 0 = no), family history of migraines (1 = yes, 0 = no), personal history of migraines (1 = yes, 0 = no), head imaging performed (1 = yes, 0 = no), history of ADHD (1 = yes, 0 = no), history of learning disabilities (1 = yes, 0 = no), and ED or urgent care visit (1 = yes, 0 = no). All statistical analyses were performed using SPSS 27.0 (IBM, Armonk, NY).
Patient Demographic Data
From a total of 1504 visits, 868 (57.7%) were not excluded and further screened (Figure 1). To ensure independent observations, only the first concussion within the study period was analyzed for any athlete (n = 90 excluded for repeat concussions within study period). Because the study tracked the progression of care of athletes, those asked to follow-up to the clinic and subsequently did not (ie, lost to follow-up) were excluded (n = 60). Of the remaining population (n = 718), 194 did not have any indication of discharge to AT regardless of the number of clinic visits, leading to their exclusion. The final sample included 524 patients, where 236 (45%) were discharged to AT after initial clinic visit and 288 (55%) patients attended additional clinic visits before discharge to AT. Demographic data and medical history for all patients are presented in Tables 1 and 2. Most of both groups were male (68.6% and 63.9%); differences in sex between the groups were not significant (χ2 = 1.10, P = 0.25). On average, those in the group who were referred to an AT after one visit were older (Md = 16.35, IQR, 14.98-17.50) than those in the group who were received additional clinic visits (Md = 15.60, IQR, 14.67-16.86, U = 29 735.0, P = 0.014).
TABLE 1. -
Frequencies of Variables of Interest Across Concussion Clinic Patient Groups
||No F/u Yes AT
||Yes F/u Yes AT
| Sex (female)
| 0 prior
| 1 prior
| 2 prior
| 3 prior
| 4 prior
| Family history of psychiatric disorder
| History of psychiatric disorders
| Family history of migraines
| History of migraines
| Head imaging performed
| History of ADHD
| History of learning disabilities
| ED or urgent care visit
Total Sample N = 524, No F/u Yes AT Sample n = 236, Yes F/u Yes AT Sample n = 288. Two participants (n = 2) had missing family history of psychiatric disorders, family history of migraines, history of migraines, history of ADHD, and history of learning disabilities. Six participants (n = 6) had a missing history of psychiatric disorders.
F/u, follow-up visits.
TABLE 2. -
Age, Initial PCSS, and Number of Prior Concussions Across Concussion Clinic Patient Groups
||No F/u Yes AT
||Yes F/u Yes AT
Total Sample N = 524, No F/u Yes AT Sample n = 236, Yes F/u Yes AT Sample n = 288. Sixty-seven participants (n = 67) had missing initial total PCSS in clinic.
F/u, follow-up visits; M, mean; Md, median.
Injury Characteristics and Course of Management
Patients who received additional clinic visits reported significantly higher PCSS scores at their initial clinic visit (Md = 24, IQR, 13.00-42.00 vs Md = 6, IQR, 0-22.00, respectively, U = 13 074.0, P = 0.002) compared with patients initially discharged to an AT. There was no difference in the proportion of patients who presented to the ED/urgent care before our clinic between the 2 discharge groups (29.2% vs 25.0%, χ2 = 1.14, P = 0.29). However, significantly more patients who received additional visits before AT discharge, compared with patients with only an initial visit, received head imaging (25.7% vs 15.3%, respectively, P < 0.010), despite all head imaging being negative.
Patient History and Course of Management
Compared with those who were referred to an AT after their initial visit, a greater proportion of patients who received additional visits had sustained a prior concussion (36.8% vs 25.0%, χ2 = 11.52, P = 0.02), had a family history for psychiatric disorders (18.5% vs 6.0%, χ2 = 18.07, P < 0.001), and had a family history of migraines (35.5% vs 19.1%, χ2 = 17.16, P < 0.001). There was no significant difference between the groups in the proportion who reported a personal history of psychiatric disorders (χ2 = 1.12, P = 0.29), migraines (χ2 = 0.81, P = 0.78), ADHD (χ2 = 0.02, P = 0.90), and learning disabilities (χ2 = 2.07, P = 0.29).
Modeling the Significant Factors in Predicting Follow-Up
The overall model for the logistic regression was statistically significant (χ2 = 98.118, P < 0.001) and explained 26.3% of the variance (Nagelkerke R2; Table 3). Listed in the order of magnitude from greatest to least, the significant predictors of patients being seen for additional clinical visits were family history of psychiatric disorders (odds ratio 3.12 [95% CI, 1.531-6.343], P = 0.002), total number of prior concussions (odds ratio 1.39 [95% CI, 1.020-1.892], P = 0.037), PCSS score at first clinic visit (odds ratio 1.05 [95% CI, 1.031-1.058], P < 0.001), and age (odds ratio 0.87 [95% CI, 0.773-0.979], P = 0.021). Family history of migraines (P = 0.218), history of psychiatric disorders (P = 0.484), history of migraines (P = 0.677), history of ADHD (P = 0.251), history of learning disabilities (P = 0.568), head imaging (P = 0.141), and ED or urgent care visits (P = 0.208) were not significant predictors.
TABLE 3. -
Logistic Regression Predicting Follow-Up to Concussion Clinic Before Discharge to Athletic Trainer
||95% CI for OR
|Sex (female = 1)
|FHx of Psych.
|FHx of migraines
|Hx of Psych.
|Hx of migraines
|Hx of ADHD
|Hx of LD
|ED or urgent care
FHx, family history; Hx, history; LD, learning disabilities; Psych., psychiatric disorder.
This retrospective study evaluated factors associated with multiple visits in a multidisciplinary concussion clinic compared with athletes who were discharged to their AT after their initial clinic visit. Among athletes with access to athletic trainers, variables such as a family history of psychiatric disorders and initial symptom score on first presentation to clinic were found to have significant relationships with additional clinic visits. Other significant variables included younger age and a history of prior concussions. These variables are consistent with those that predict worsened recovery after sustaining concussions.10,12 Identification of these associations may help shape concussion care by setting patient and family expectations, maximizing clinic attendance, and avoiding unnecessary clinic appointments.
A positive family history of psychiatric disorders had the largest magnitude of impact (odds ratio [OR], 3.12) out of the significant variables between the groups. This is consistent with the finding that high school athletes with a family history of psychiatric disorders were 2.5 times more likely than controls to suffer a protracted recovery after SRC.12 The results of this study also indicate that athletes who required follow-up concussion clinic visits reported significantly higher initial symptom severity scores in clinic than athletes without follow-up before discharge (median, 24 vs 6, respectively). This finding is consistent with the overwhelming evidence that athletes initially presenting in clinic with greater symptom scores show prolonged recoveries compared with athletes with lower symptom scores.10 These athletes likely require increased levels of care to rehabilitate specific symptom clusters. The magnitude of impact for a 1-point increase in PCSS score reflects an increased odds ratio of 1.045, which is lower than the magnitude of impact for the family history of psychiatric disorders variable (OR, 3.12). However, although family history of psychiatric disorder is a binary variable, the PCSS score is continuous. As an example, the difference in median PCSS score between the 2 groups was 18, which corresponds to a potential odds ratio of 18.81 (18 × 1.045). Hence, the continuous nature of the PCSS score may reflect additional clinic visits.
Our data showed that the group with additional visits in the clinic was younger than those initially discharged to an AT (median age of 15.60 vs 16.35 years, respectively). Although statistically significant, the quantitative difference may not be clinically significant. Classically, age has been considered a risk modifier for prolonged recovery from SRC, with younger athletes requiring more time to recover.19 Many studies have showed this effect in athletes stratified into collegiate and high school status, finding that high school athletes sustained symptoms lasting significantly longer than their collegiate peers.20–22 One proposed explanation is that a lack of full-time ATs in many high schools, which could translate to lower quality initial health care, may predispose high school athletes to worsened outcomes.20 The lack of full-time high school ATs may potentially affect provider decisions at the time of initial assessment to prospectively schedule follow-up visits before discharge to AT. Notably, a substantial proportion of collegiate athletes in this study are students at a SEC Division 1 athletic program, where they may receive more intensive care and oversight compared with other US collegiate and high school athletes given the resources at their disposal. Further studies have shown that high school aged patients are most sensitive to consequences of concussions compared with both children age 9 to 12 years and adults.23 However, more research needs to focus on younger participants age younger than 14 years.21 Our study consisted of patient's age 12 to 23 years, although most of our sample included high school athletes age 14 to 17 years. The finding that younger age was associated with longer times to AT discharge reflects clinical practice consistent with the current knowledge of SRC outcomes between high school and collegiate athletes. There were no other demographical differences. These findings emphasize the relationship between age stratified into high school and collegiate years and concussion outcomes and additional clinic visits before discharge to AT.
The association between the number of prior concussions and worsened clinical outcomes is well-established.10 The history of concussion has been shown to increase the risk of future concussions24 and is associated with higher values of preinjury concussion symptoms,25 which may be mistaken for symptoms caused by injury. The results of our study showed that a history of concussions is significantly associated with additional clinic visits. The odds of receiving additional clinic visits are associated with an increased number of prior concussions (1 prior concussion OR, 1.39 and 2 prior concussions OR, 2.78), suggesting that those who have multiple prior concussions have a higher likelihood of receiving more than one visit in a specialty concussion center before being discharged to their AT.
Not all predictors of the decision to schedule additional clinic visits were consistent with risk modifiers of slower concussion recovery. Although a personal history of psychiatric disorders has been consistently associated with slower recovery in previous studies,10 this variable did not significantly predict additional clinic visits in this study. In addition, female sex and subacute problems with headaches have been associated with a worse/slower clinical recovery in previous studies10,11 but did not have a substantial relationship with additional clinic visits in our analyses. These discrepancies can serve as targets for reducing the burden of concussion care on the mainstream healthcare system. Consistent with previous studies,10,26 history of ADHD, history of learning disabilities, and an ED visit after the injury were not significantly associated with additional clinic visits in our data. Significant and nonsignificant variables are presented in Figure 2.
The limitations of this study are important to acknowledge. This study was retrospective and only analyzed the medical records of youth/young adults who presented to a specialty sports concussion clinic that clearly indicated discharge to an athletic trainer; those whose charts did not indicate discharge to an AT (patient was lost to follow-up, provider did not indicate discharge to AT, no athletic trainer to discharge the patient to, etc.) were not included. We recognize the limitations of a retrospective study in making causal statements about factors that definitively predict our outcomes. Future prospective cohort studies may elaborate on the associations that we found in this study with a common outcome of interest, such as request by AT for additional visits or complications during recovery process. We were also limited in knowing when the athlete returned to play. We hope future prospective studies that involve athletic trainers are completed in the future. This may allow us to gather data on additional visits for a concussion episode that were requested by an AT after initial discharge. We recognize that the request of additional required visits related to an athlete's course of recovery under the supervision of an AT may provide information about the care model around early versus late discharge to AT. These data should be prospectively collected from ATs and clinic visits. In addition, there may be some differences in management because of different providers seeing patients; however, our clinic has worked to standardize the initial visit and referral pathway. Given that our clinic is located in an urban setting with more access to resources and with schools with the presence of ATs, we cannot comment on geographical disparities of patients located in underresourced areas. We also cannot comment on the portion of athletes that were not discharged to AT because their schools did not have an AT. Furthermore, among the collegiate athlete subset within the study population, a significant portion attended a Division 1 SEC program. Finally, we did not obtain data on athletes who were released to AT but did not have a planned next appointment. Future directions involve analyzing predictive factors of more concussion care in areas that are not represented by the cohort in this study.
Although there are multiple risk factors that predict worsened outcomes after a SRC, there is a limited knowledge regarding variables that predict the length of care. In this study, we determined that the initial symptom score had the potential to be the variable with the largest magnitude associated with the clinical decision to schedule additional clinic visits. Other significant factors included family history of psychiatric disorders, number of prior concussions, and younger age. Optimal concussion care is ideally efficient if it reduces the chances of suffering long-term adverse consequences or of entering premature return-to-play protocols. Knowledge of significant variables associated with additional clinic visits may contribute to better expectations understood by patients and their families during the recovery process.
1. Lempke LB, Schmidt JD, Lynall RC. Athletic trainers' concussion-assessment and concussion-management practices: an update. J Athletic Train. 2020;55:17–26.
2. Silverberg ND, Iaccarino MA, Panenka WJ, et al.; American Congress of Rehabilitation Medicine Brain Injury Interdisciplinary Special Interest Group Mild TBI Task Force. Management of concussion and mild traumatic brain injury: a synthesis of practice guidelines. Arch Phys Med Rehabil. 2020;101:382–393.
3. Arbogast KB, Curry AE, Pfeiffer MR, et al. Point of health care entry for youth with concussion within a large pediatric care network. JAMA Pediatr. 2016;170:e160294.
4. McKeon JMM, Livingston SC, Reed A, et al. Trends in concussion return-to-play timelines among high school athletes from 2007 through 2009. J Athletic Train. 2013;48:836–843.
5. Manzanero S, Elkington LJ, Praet SF, et al. Post-concussion recovery in children and adolescents: a narrative review. J Concussion. 2017;1:205970021772687.
6. Holmes A, Chen Z, Yahng L, et al. Return to learn: academic effects of concussion in high school and college student-athletes. Front Pediatr. 2020;8:57.
7. Broglio SP, Cantu RC, Gioia GA, et al.; National Athletic Trainer's Association. National athletic trainers' association position statement: management of sport concussion. J Athletic Train. 2014;49:245–265.
8. Leddy JJ, Sandhu H, Sodhi V, et al. Rehabilitation of concussion and post-concussion syndrome. Sports Health. 2012;4:147–154.
9. Langer LK, Alavinia SM, Lawrence DW, et al. Prediction of risk of prolonged post-concussion symptoms: derivation and validation of the TRICORDRR (Toronto rehabilitation institute concussion outcome determination and rehab recommendations) score. PLoS Med. 2021;18:e1003652.
10. Iverson GL, Gardner AJ, Terry DP, et al. Predictors of clinical recovery from concussion: a systematic review. Br J Sports Med. 2017;51:941–948.
11. Neidecker JM, Gealt DB, Luksch JR, et al. First-time sports-related concussion recovery: the role of sex, age, and sport. J Am Osteopathic Assoc. 2017;117:635–642.
12. Legarreta AD, Brett BL, Solomon GS, et al. The role of family and personal psychiatric history in postconcussion syndrome following sport-related concussion: a story of compounding risk. J Neurosurg Pediatr. 2018;22:238–243.
13. Risen SR, Reesman J, Yenokyan G, et al. The course of concussion recovery in children 6-12 years of age: experience from an interdisciplinary rehabilitation clinic. PM R. 2017;9:874–883.
14. McCrory P, Meeuwisse W, Dvořák J, et al. Consensus statement on concussion in sport—the 5th international conference on concussion in sport held in Berlin, October 2016. Br J Sports Med. 2017;51:838–847.
15. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)-A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381.
16. Harris PA, Taylor R, Minor BL, et al.; REDCap Consortium. The REDCap consortium: building an international community of software platform partners. J Biomed Inform. 2019;95:103208.
17. Barlow M, Schlabach D, Peiffer J, et al. Differences in change scores and the predictive validity of three commonly used measures following concussion in the middle school and high school aged population. Int J Sports Phys Ther. 2011;6:150–157.
18. Lovell MR, Iverson GL, Collins MW, et al. Measurement of symptoms following sports-related concussion: reliability and normative data for the post-concussion scale. Appl Neuropsychol. 2006;13:166–174.
19. Foley C, Gregory A, Solomon G. Young age as a modifying factor in sports concussion management: what is the evidence? Curr Sports Med Rep. 2014;13:390–394.
20. Field M, Collins MW, Lovell MR, et al. Does age play a role in recovery from sports-related concussion? A comparison of high school and collegiate athletes. J Pediatr. 2003;142:546–553.
21. Covassin T, Elbin RJ, Harris W, et al. The role of age and sex in symptoms, neurocognitive performance, and postural stability in athletes after concussion. Am J Sports Med. 2012;40:1303–1312.
22. Zuckerman SL, Lee YM, Odom MJ, et al. Recovery from sports-related concussion: days to return to neurocognitive baseline in adolescents versus young adults. Surg Neurol Int. 2012;3:130.
23. Baillargeon A, Lassonde M, Leclerc S, et al. Neuropsychological and neurophysiological assessment of sport concussion in children, adolescents and adults. Brain Inj. 2012;26:211–220.
24. Abrahams S, Fie SM, Patricios J, et al. Risk factors for sports concussion: an evidence-based systematic review. Br J Sports Med. 2014;48:91–97.
25. Iverson GL, Silverberg ND, Mannix R, et al. Factors associated with concussion-like symptom reporting in high school athletes. JAMA Pediatr. 2015;169:1132–1140.
26. Eisenberg MA, Andrea J, Meehan W, et al. Time interval between concussions and symptom duration. Pediatrics. 2013;132:8–17.