Secondary Logo

Journal Logo

Special Communications

Sports Medicine Fellowship Training Improves Sport-related Concussion Evaluation

Northam, Weston T. MD1,2; Cools, Michael J. MD1,2; Chandran, Avinash PhD2,3; Alexander, Andrew BS2,4; Mihalik, Jason P. PhD, CAT(C), ATC, FACSM1,2; Guskiewicz, Kevin M. PhD, ATC, FACSM2; Carneiro, Kevin A. DO1,2,5

Author Information
Current Sports Medicine Reports: July 2020 - Volume 19 - Issue 7 - p 272-276
doi: 10.1249/JSR.0000000000000730
  • Free


Sport-related concussion (SRC) is a growing public health problem, representing significant morbidity for athletes and burden for the health care system at large (1,2). International consensus and national medical guidelines have evolved with the evidence base, informing providers on how to evaluate and manage concussions (3,4). These resources enumerate each of the symptom domains involved in comprehensive assessment of a concussed athlete, including headache, vision, vestibular, cervical spine, sleep, mood, and others. Physical examination assessing several important physiologic systems is discussed in these publications and addresses cognitive, neurologic, balance, and musculoskeletal (e.g., evaluating the cervical spine) domains. Little is known regarding the use of consensus and guideline-based symptom and physical examination items in clinical practice, and there are no reports directly examining the effect of sports medicine fellowship (SMF) training on SRC assessments.

Currently, the literature lacks a robust evaluation of the efficacy of physician concussion assessments, with most published studies limited to survey designs (5–7). Studies seeking to review provider clinical documentation have been limited in scope by investigating a small subset of the various consensus and guideline-based symptom domains and physical examination techniques (8,9). None of these studies compared providers with SMF training to those without it (non-SMFs). This is compounded by recent evidence demonstrating most providers evaluating concussed athletes in clinic may be inconsistently applying consensus and guideline-based techniques (7) and may have insufficient training to appropriately address the complexities this condition may present in their patients (8,10). Recent attention has turned toward providing educational initiatives and decision support tools for providers to more effectively evaluate and manage concussion (9,11,12).

This study reviews concussed athletes seen by all outpatient providers in a single academic health care system, to determine whether board-certified SMF training affected their evaluation of concussed athletes. This was determined through the provider's clinical documentation of concussion-specific symptom history and physical examination items. We hypothesized that providers with SMF training would assess SRC history and physical items in higher proportions compared with providers without SMF training (non-SMFs).


Data Collection

We conducted a retrospective medical record review for patients aged 7 to 18 years, diagnosed with concussion between 2014 and 2017 in a large university medical system. The mechanism of injury was examined, and patients were excluded if their concussion was not sport-related. We only reviewed the most recent concussion for any patients who had reported multiple concussions during this time period. For each incident concussion in our data set, we only reviewed the clinical documentation for the first clinical visit with the diagnosing provider. Thus, the first clinical visit for the most recent concussion for each patient was reviewed to synthesize the analytical data set. The data collection procedures were reviewed and approved by our institutional review board.

Statistical Analyses

We examined patient and injury characteristics (sex, insurance status, sport-based contact level) and provider characteristics (medical specialty, credentials, board-certified SMF) using frequencies (percentages). We then examined the proportion of all SRCs seen by SMFs and non-SMFs documenting use of concussion-specific history and physical examination items. Concussion-specific histories were reviewed for documentation addressing prior concussions, headache, dizziness, nausea, vision symptoms, cognitive symptoms, emotional symptoms, sleep symptoms, neck pain, and use of a symptom checklist. Physical examination documentation was reviewed for clinical cognitive examination, basic neurological examination, detailed neurological examination, balance examination, and a cervical spine examination (Table 1).

Table 1:
Concussion-focused physical examination items.

We assessed associations between board-certified SMF training status (yes/no) and patient charactaristics for SRCs examined (sex, insurance, contact level), using χ2 tests. Similarly, we evaluated differential practices in SRC assessment using χ2 tests, namely, in how the history and physical examination items were documented between SMFs and non-SMFs. Multivariable logistic regression models were used to examine differential odds of assessing each history and physical examination item between SMFs and non-SMFs, after adjusting for patient characteristics (sex, insurance status), sport contact level (contact, limited contact, or noncontact) (13), and physician residency training (family medicine, internal medicine, pediatrics, physical medicine and rehabilitation, orthopedics). Statistical significance was evaluated at the 0.05 level, and all analyses were conducted in SAS 9.4 (SAS, Cary, NC).


A total of 889 patient records were included in this analysis. The sample included a higher proportion of male patients (n = 590, 66.4%) than female patients (n = 299, 33.6%), and a higher proportion of SRCs from contact sports (n = 721, 81.1%) than limited or noncontact (n = 168, 18.9%). A considerable proportion of patients reported no medical insurance (n = 367, 42.2%). Patients were primarily seen by medical doctors (n = 730, 82.1%), with family medicine (n = 298, 33.5%) or pediatric residency training (n = 294, 33.1%). Furthermore, 399 (44.9%) patients were examined by SMFs, while 490 (55.1%) patients were examined by non-SMFs. Statistically significant associations were observed between SMF training status and profiles of SRCs examined (by patient characteristics and injury mechanism). Table 2 and Table 3 provide these data in addition to group-stratified frequencies and proportional distributions of patient characteristics, injury mechanism, insurance status, and provider specialty and credentials.

Table 2:
Group-stratified (by SMF, non-SMF) frequencies and proportional distributions of patient characteristics and injury mechanism for SRCs.
Table 3:
Specialty and credentials of clinicians evaluating sport-related concussion.

Each of the history items were assessed in higher proportions of SRCs evaluated by SMFs, as compared with non-SMFs (Table 4). There were no associations observed between SMF training status and headache assessment due to the high prevalence (>93.0%) of headache assessment observed in both groups. However, SMF training was significantly associated with assessing each of the remaining tested history examination items. Notably, a symptom checklist was used in 99.5% of SRCs evaluated by SMFs, and in only 6.1% of SRCs evaluated by non-SMFs (χ2 = 768.2; P < 0.001). Considerable differences between SMFs and non-SMFs also were observed in asking about prior concussions (SMFs, 69.9%; non-SMFs, 28.8%; χ2 = 149.4; P < 0.001), dizziness (SMFs, 84.7%; non-SMFs, 51.8%; χ2 = 106.8; P < 0.001), and emotional symptoms (SMFs, 57.6%; non-SMFs, 17.1% χ2 = 157.9; P < 0.001). Each of the physical examination items also were assessed in higher proportions of SRCs evaluated by SMFs, as compared with non-SMFs (Table 4). Once again, SMF training was associated with assessment of each of the tested physical examination items, except for the basic neurologic examination. Importantly, while 88.5% of SRCs evaluated by SMFs involved a detailed neurological examination, only 26.9% of those evaluated by non-SMFs included this item (χ2 = 335.9; P < 0.001). Notable differences were observed in cognitive testing practices between SMFs and non-SMFs (SMFs, 69.9%; non-SMFs, 13.5%; χ2 = 295.2; P < 0.001). Table 4 summarizes the total and group-stratified data and statistical outcomes for these variables, in addition to the multivariable logistic regression modeling results found below.

Table 4:
Group-stratified (by SMF, non-SMF) frequencies of concussion-specific history and physical examination items, and multivariable logistic regression modeling of the odds of assessing each item.

In multivariable logistic regression modeling, sports medicine training status emerged as a significant determinant of the odds of assessing each of the history items, except for nausea and sleep, and each tested physical examination item, even after adjusting for patient and injury characteristics and physician residency training type (family medicine, internal medicine, pediatrics, physical medicine and rehabilitation, orthopedics). In particular, the odds of neck pain assessment were higher in SRCs assessed by SMFs as compared with non-SMFs (odds ratio [OR], 4.42; 95% confidence interval [CI], 2.50–7.81). Similarly, odds of assessing emotional symptoms (OR, 1.78; 95% CI, 1.09–2.89), and vision symptoms (OR, 2.95; 95% CI, 1.86–4.69) were higher in SRCs assessed by SMFs as compared with non-SMFs. Within the physical examination, we observed higher odds of detailed neurological examinations (OR, 14.74; 95% CI, 8.66–25.06), and clinical cognitive examination (OR, 11.71; 95% CI, 6.73–20.38) in SRCs assessed by SMFs as compared with non-SMFs.


Medical providers with previous SMF training demonstrated significantly higher odds of assessing and documenting more thorough and comprehensive acute concussion evaluations than peers without SMF training in a single university health care system. These findings held true even when accounting for patient sex, insurance status, sport-based contact level, and residency training background. Excluding headache history and performing a basic neurologic examination, there were large discrepancies in the way concussed athletes were evaluated based on whether their provider had SMF training. This is the first effort, to our knowledge, to utilize medical provider documentation to study the effects of SMF training on evaluating acute sport-related concussion.

Contemporary return-to-play (RTP) and return-to-learn (RTL) algorithms utilize symptom inventories (4,14,15). Patients may prematurely RTP or RTL without comprehensive evaluation and documentation of these symptoms. Missing important concussion-specific signs and symptoms could result in increased morbidity for patients, protracted recoveries, and lost opportunities for early targeted therapy such as vision, vestibular, or cervical spine rehabilitation (16–20).

Current literature regarding physician evaluation of SRC suggests many providers may lack adequate training and decision-support tools to effectively evaluate and manage these patients (10). Surveys indicate lack of awareness is a common reason that physicians do not apply published concussion guidelines (5), with only 14.6% of providers in one study reporting usage of concussion consensus guidelines in their practice (21). A survey of emergency medicine providers showed only 23% incorporated national guidelines into evaluation of concussed athletes (22). This deficit of training and knowledge may be a factor early in the medical education process, as resident physicians in the same study had a similarly poor application of national guidelines to their practice (23%). Further, almost one quarter of graduating medical students, neurology, and neurosurgery residents reported they could not remember if their medical curriculum included material related to concussion in one survey study (23). Investigation into specialist training has yielded similar findings, with one survey of pediatric neurologists showing the majority had not completed any continuing medical education about concussion, and only 35.9% reporting they had adequate training to address this common clinical problem (24).

Board-certified fellowship training in sports medicine involves a wide spectrum of care for competitive and recreational athletes, and concussion is an implicit part of this care (25). Although literature studying the practice patterns of SMF-trained providers is limited, one survey study indicated that 96.7% reported basing their practice on published criteria or guidelines (6). The majority (91.9%) also reported using the most up-to-date guidelines available. These data are compatible with the results of our study, which show that the majority of SMFs asked patients about important concussion symptoms and completed physical examination tasks supported by international consensus and guidelines. In our study, non-SMF providers frequently performed a basic neurologic examination, but often did not perform a detailed neurologic examination that could reveal vestibulo-ocular deficits, or balance testing. This is corroborated by a survey of general family practitioners, showing that 88.1% of practitioners performed a clinical examination, but only 26.7% performed balance testing, and only 19.8% used neurocognitive testing (26). These data, when taken together, suggest non-SMF evaluation of SRC may be limited by concussion-specific education and adherence to published guidelines and consensus.

Although our study illuminates areas for improvement in the assessment of SRCs, our study has limitations. First, it is a retrospective design at a single academic medical system, which could limit generalizability. However, the large sample size (n = 889) and range of clinical practices and specialties across the medical system help to address this. The advantage of our medical chart review is in providing a direct assessment of the provider's own documentation and management and capturing all providers treating concussion across a large academic medical center, as opposed to a survey, which only includes limited respondents. Despite this, we realize that some providers may have performed history items or physical examination tests but did not document these items. Given the large and consistent discrepancy between SMF and non-SMF documentation in our study, and how frequently SMFs documented important items, we feel this is unlikely. There were several significant differences in the patient characteristics seen by SMFs versus non-SMFs, but these were taken into account in the multivariable logistic regression modeling and are not thought to compromise the inferential capacity of the observed results. We suspected that physicians with physical medicine and rehabilitation residency training could potentially have more of a background in brain injury evaluation and management compared with other providers (27), and therefore, controlled for this variable in the multivariable statistical modeling. Even after controlling for residency training background, our data indicate that SMFs perform more thorough and comprehensive acute concussion evaluations compared with non-SMFs.


SMF-trained providers utilized important concussion-specific history and physical examinations significantly more frequently in evaluating SRC than providers without fellowship training. The odds of concussion-specific history and physical examination items being utilized were higher in fellowship-trained providers, even when accounting for patient sex, insurance status, sport-based contact level, and the provider's residency training background. Further educational and decision-support efforts are needed for nonfellowship-trained providers, who make up the majority of those caring for SRC. Future educational projects should focus on the content of the training process that underlies SMF to optimize the evaluation of concussion across all practitioners and throughout the medical training continuum.

The authors would like to thank Dr. Kathryn Pietrosimone for her work in editing and revising the manuscript.

W.T.N. and M.J.C. contributed equally to the manuscript.

The authors declare no conflicts of interest and do not have any financial disclosures.


1. Ledoux AA, Tang K, Yeates KO, et al. Natural progression of symptom change and recovery from concussion in a pediatric population. JAMA Pediatr. 2019; 173:e183820.
2. Fridman L, Scolnik M, Macpherson A, et al. Annual trends in follow-up visits for pediatric concussion in emergency departments and physicians' offices. J. Pediatr. 2018; 192:184–8.
3. Lumba-Brown A, Yeates KO, Sarmiento K, et al. Centers for Disease Control and Prevention guideline on the diagnosis and management of mild traumatic brain injury among children. JAMA Pediatr. 2018; 172:e182853.
4. McCrory P, Meeuwisse W, Dvorák 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; 51:838–47.
5. Pleacher MD, Dexter WW. Concussion management by primary care providers. Br. J. Sports Med. 2006; 40:e2–2.
6. Stache S, Howell D, Meehan WP 3rd. Concussion management practice patterns among sports medicine physicians. Clin. J. Sport Med. 2016; 26:381–5.
7. Carson JD, Rendely A, Garel A, et al. Are Canadian clinicians providing consistent sport-related concussion management advice? Can. Fam. Physician. 2016; 62:494–500.
8. Carson JD, Lawrence DW, Kraft SA, et al. Premature return to play and return to learn after a sport-related concussion: physician’s chart review. Can. Fam. Physician. 2014; 60:e310, e312–15.
9. Arbogast KB, Curry AE, Metzger KB, et al. Improving primary care provider practices in youth concussion management. Clin. Pediatr. (Phila). 2017; 56:854–65.
10. Zonfrillo MR, Master CL, Grady MF, et al. Pediatric providers’ self-reported knowledge, practices, and attitudes about concussion. Pediatrics. 2012; 130:1120–5.
11. Ip IK, Raja AS, Gupta A, et al. Impact of clinical decision support on head computed tomography use in patients with mild traumatic brain injury in the ED. Am. J. Emerg. Med. 2015; 33:320–5.
12. Melnick ER, Lopez K, Hess EP, et al. Back to the bedside: developing a bedside aid for concussion and brain injury decisions in the emergency department. eGEMs (Wash DC). 2015; 3:1136.
13. Rice SG; American Academy of Pediatrics Council on Sports Medicine and Fitness. Medical conditions affecting sports participation. Pediatrics. 2008; 121:841–8.
14. Baker JG, Rieger BP, McAvoy K, et al. Principles for return to learn after concussion. Int. J. Clin. Pract. 2014; 68:1286–8.
15. King D, Brughelli M, Hume P, Gissane C. Assessment, management and knowledge of sport-related concussion: systematic review. Sports Med. 2014; 44:449–71.
16. Guskiewicz KM, McCrea M, Marshall SW, et al. Cumulative effects associated with recurrent concussion in collegiate football players: the NCAA Concussion Study. JAMA. 2003; 290:2549–55.
17. Matuszak JM, McVige J, McPherson J, et al. A practical concussion physical examination toolbox: evidence-based physical examination for concussion. Sports Health Multidiscip Approach. 2016; 8:260–9.
18. Leddy JJ, Baker JG, Merchant A, et al. Brain or strain? Symptoms alone do not distinguish physiologic concussion from cervical/vestibular injury. Clin. J. Sport Med. 2015; 25:237–42.
19. Anzalone AJ, Blueitt D, Case T, et al. A positive vestibular/ocular motor screening (VOMS) is associated with increased recovery time after sports-related concussion in youth and adolescent athletes. Am. J. Sports Med. 2017; 45:474–9.
20. Kontos AP, Deitrick JM, Collins MW, Mucha A. Review of vestibular and oculomotor screening and concussion rehabilitation. J. Athl. Train. 2017; 52:256–61.
21. Carl RL, Kinsella SB. Pediatricians' knowledge of current sports concussion legislation and guidelines and comfort with sports concussion management: a cross-sectional study. Clin. Pediatr. (Phila). 2014; 53:689–97.
22. Giebel S, Kothari R, Koestner A, et al. Factors influencing emergency medicine physicians' management of sports-related concussions: a community-wide study. J. Emerg. Med. 2011; 41:649–54.
23. Boggild M, Tator CH. Concussion knowledge among medical students and neurology/neurosurgery residents. Can. J. Neurol. Sci. 2012; 39:361–8.
24. Broshek DK, Samples H, Beard J, Goodkin HP. Current practices of the child neurologist in managing sports concussion. J. Child Neurol. 2014; 29:17–22.
25. Harmon KG, Drezner JA, Gammons M, et al. American Medical Society for Sports Medicine position statement: concussion in sport. Br. J. Sports Med. 2013; 47:15–26.
26. Lebrun CM, Mrazik M, Prasad AS, et al. Sport concussion knowledge base, clinical practises and needs for continuing medical education: a survey of family physicians and cross-border comparison. Br. J. Sports Med. 2013; 47:54–9.
27. Accreditation Council for Graduate Medical Education. ACGME program requirements for graduate medical education in physical medicine and rehabilitation [Internet]. Accreditation Council for Graduate Medical Education. 2017 [cited 2020 May 16]. Available from:
Copyright © 2020 by the American College of Sports Medicine