Return to Learn After Sport-Related Concussion: A Survey of Secondary School and Collegiate Athletic Trainers

Instrument Development

To address the hypotheses of this study, we developed a novel survey instrument that consisted of 4 RTL domains: demographics, current knowledge, current practice, and available resources. The survey instrument consisted of structured items within each domain (eg, true or false, yes or no, multiple choice, Likert scale). For Likert-scale items, responses were reweighted to ensure the validity of each response. For example, within the current practices domain, an item ranking of 5 indicated strongly agree, whereas in a different domain, 5 represented strongly disagree.

Where appropriate, individual items were created, modified, or both to specifically address the secondary school or collegiate setting. Therefore, we created 2 surveys, which consisted of 40 and 41 items, with 7 modified items specific to the secondary school or collegiate setting, respectively. For example, the secondary school version asked about the AT's access to a school counselor, whereas the collegiate version asked about the AT's access to a learning specialist. To assess item construction, comprehension, and ease of completion, a convenience sample of 16 ATs (5 SSATs, 11 CATs) completed a paper-and-pencil version of the survey instrument. The survey was modified based on their written and oral feedback to increase item clarity. Next, the revised survey instrument was sent to an external expert committee comprising a neuropsychologist, a neurologist, and an AT who had expertise in SRC related to RTL and cognitive rest. The expert committee reviewed the survey for content (face) validity and provided written feedback. Based on the expert committee's comments, additional survey modifications were made. After revision, the survey instrument was formatted within SurveyMonkey (Palo Alto, CA) and administered via e-mail link to a small convenience sample (n = 16) of ATs so that we could further refine either the content or SurveyMonkey format if needed. Approximately 68.8% (n = 11) of the ATs worked in the collegiate setting, whereas 31.3% (n = 5) worked in the secondary school setting. Internal consistency was assessed for all Likert-scale items using the Cronbach α and deemed acceptable for the secondary school (α = 0.49) and collegiate (α = 0.63) versions of the survey. The analyzed items were largely subjective; therefore, the lower values of internal consistency observed were deemed acceptable. The final version of the survey took approximately 15 minutes to complete and consisted of demographic (11 items), current knowledge (6 items), current practice (18 items), and available resources (6 items) domains (Tables 1 through 3).

Table 1 Respondent Demographics (Domain 1)^a

View Fullscreen

Table 2 Current Practice (Domain 3)^a Continued on Next Page

View Fullscreen

Table 3 Available Resources (Domain 4)^a

View Fullscreen

Procedures

A convenience sample of ATs employed in the secondary school or collegiate setting (n = 4000) was requested from and selected at random by the NATA. The randomly selected participants represented all 10 NATA districts. Participant anonymity was maintained via the NATA's withholding participants' e-mail addresses. In addition, any participants who had the same e-mail domain as another participant (eg, @virginia.edu) were not selected. This selection process was used to avoid multiple responses from clinicians at the same institution to limit bias. The final survey was distributed online through SurveyMonkey. A cover letter was sent to ATs via e-mail explaining the purpose, importance, and estimated time to complete the survey. They also had the option to enter a drawing for one of two $50 Amazon gift cards upon survey completion. The ATs were allotted 6 weeks to respond (March–April 2015) and were provided with weekly reminders sent by the NATA, which also sent biweekly follow-up e-mails to eligible participants to encourage survey completion. We assumed completion of the survey was acknowledgment of consent. The University of Virginia's Institutional Review Board for Social and Behavioral Sciences approved the study.

Data Processing

Data were exported from SurveyMonkey to an Excel (version 2013; Microsoft Corp, Redmond, WA) spreadsheet and assessed for valid responses. We considered responses valid if all domains of the survey instrument were fully completed. Therefore, respondents who exited early or did not complete the demographics domain (the final domain) were excluded from the final analysis.

Statistical Analysis

Descriptive statistics were reported for each item in the 4 survey domains. When appropriate, we performed χ² tests to analyze Likert and dichotomous response items in each domain and compare responses from SSATs and CATs. The Cramér V was used to calculate effect sizes, which were interpreted as small (0–0.29), medium (0.30–0.49), or large (≥0.50).²⁰ Means and standard deviations were reported for respondent demographics. Additional analyses were performed using independent-samples t tests to assess the number of years since certification of respondents and the number of concussions evaluated each year. These analyses were conducted to ensure that respondents were indeed evaluating SRCs in their clinical practices and that, if differences were observed in responses related to knowledge or current practices, they were not because of time since certification. We used descriptive statistics to calculate response rates (%) for each domain for the SSATs and CATs, as well as the overall response rate for each item. The α level was set at ≤.05. All statistical analyses were conducted using SPSS (version 20.0; IBM Corp, Armonk, NY).

Response Rate

A total of 801 SSATs, 435 CATs, and 27 other ATs responded, for an initial response rate of 31.6% (n = 1263). Participant responses from those who did not meet the inclusion criteria, provided incomplete responses, or exited the survey prematurely were excluded (Figure 1). A total of 1083 (27.1%) completed surveys were analyzed; respondents consisted of 729 SSATs (67.3%) and 354 CATs (32.7%) representing all 50 states, the District of Columbia, and each NATA district.

View Figure

• Download as Powerpoint
• Download Figure

Demographics

Data for the demographics domain are presented in Table 1. Most SSATs were primarily employed as head ATs (84.9%, n = 619) in a full-time public school position (63.0%, n = 459) and had been certified for an average of 14.0 ± 9.7 years. The CATs were most often employed as head (43.2%, n = 153) or assistant (41.2%, n = 146) ATs in an NCAA Division I setting (27.4%, n = 97) and had been certified for an average of 13.4 ± 9.7 years. We did not observe a difference in years certified between SSATs and CATs (t₁₀₈₀ = −0.97, P = .33). Men and women composed similar proportions of the sample for SSATs (men = 47.9%, n = 349; women = 51.4%, n = 375) and CATs (men = 49.7%, n = 176; women = 48.9%, n = 173). Overall, SSATs reported evaluating more SRCs than CATs (\({\rm{{\unicode[Times]{x3C7}} }}_4^2\) = 159.96, P < .001, Cramér V = 0.38; Figure 2).

View Figure

• Download as Powerpoint
• Download Figure

Most SSATs (59.5%, n = 434) and CATs (65.8%, n = 233) had been certified through an accredited undergraduate-level athletic training program. The majority of respondents also held a graduate degree (SSATs = 66.8%, n = 487; CATs = 84.7%, n = 300). Most respondents reported providing clinical coverage for high- and low-risk sports.

Current Knowledge

Complete results for the current knowledge domain are reported in Figure 3. For this domain, 41.2% (n = 446) of total respondents correctly indicated the absence of evidence-based guidelines for RTL after a concussion. More specifically, a higher percentage of CATs than SSATs correctly identified the absence of evidence-based RTL guidelines (n = 1083, \({\rm{{\unicode[Times]{x3C7}} }}_1^2\) = 6.40, P = .01, Cramér V = 0.08). No differences were found between SSATs and CATs (Cramér V = 0.05, P = .10) for correctly recognizing that ATs were allowed to recommend academic adjustments for concussed student-athletes during acute recovery. Most respondents (92.0%, n = 996) correctly agreed that NCAA¹⁷ RTL guidelines were posted online. Most SSATs (86.6%, n = 631) correctly noted that the acute concussion evaluation care plan for returning injured athletes to school was available online.²¹ A higher percentage of CATs than SSATs correctly observed that letters of academic accommodations were not legally binding (n = 1083, \({\rm{{\unicode[Times]{x3C7}} }}_1^2\) = 45.19, P < .001, Cramér V = .20). Only 30.6% (n = 331) of total ATs responded correctly to this item.

View Figure

• Download as Powerpoint
• Download Figure

Current Practice

For the current practices domain, 76.8% (n = 832) of all respondents agreed or strongly agreed that managing the RTL process was one of their roles as an AT (Table 2). Similarly, equal proportions of SSATs (88.3%, n = 644) and CATs (90.1%, n = 319) believed they should be involved in the RTL process (P = .38). A total of 73.9% (n = 800) of respondents had an RTL protocol in their current concussion-management plan. However, only 38.1% (n = 413) of total respondents followed the NCAA RTL guidelines. A higher percentage of CATs (57.3%, n = 203) than SSATs (28.8%, n = 210) agreed or strongly agreed with adhering to the NCAA¹⁷ RTL guidelines (n = 549, \({\rm{{\unicode[Times]{x3C7}} }}_1^2\) = 41.9, P < .001, Cramér V = 0.28). Whereas nearly one-third of total respondents did not include an RTL protocol, most (84.6%, n = 916) did address academic adjustments after an SRC in their concussion-management plan, and a greater percentage of SSATs (86.1%, n = 628) did so than CATs (81.4%, n = 288, χ² = 4.19, P = .04, Cramér V = 0.06). A similar proportion of the total number of surveyed ATs (87.4%, n = 947) had provided academic accommodations for a concussed student-athlete in their clinical practice. Interestingly, a higher percentage of CATs (88.7%, n = 314) than SSATs (77.1%, n = 562) had had a student-athlete request academic accommodations after a diagnosed SRC (χ² = 20.86, P < .001, Cramér V = 0.14). The complete item responses for the current practice domain are reported in Table 2.

Of the total respondents, 98.1% (n = 1062) indicated that concussed student-athletes should have access to academic accommodations, and 81.2% (n = 879) noted that full participation in classroom activities should precede a physical progression when recovering from SRC. Most respondents (89.6%, n = 970) were comfortable delaying the RTP progression until a concussed student-athlete was symptom free after a full day of unrestricted classroom activity. Most SSATs (60.2%, n = 439) reported not feeling pressured to return student-athletes to classroom activities. The most commonly reported sources of pressure are presented in Figure 4.

View Figure

• Download as Powerpoint
• Download Figure

Available Resources

Complete item responses for the available resources domain are reported in Table 3. Most SSATs (76.4%, n = 557) and CATs (82.2%, n = 291) had a disability resource center or secondary school equivalent located on campus. Nearly all SSATs (97.1%, n = 708) had access to a school counselor, whereas more than half (54.8%, n = 194) of CATs had access to a learning specialist to assist concussed student-athletes with academic adjustments or accommodations after an SRC. A minority of respondents (38.5%, n = 417) had access to a neuropsychologist, which was similar between employment settings (χ² = 4.42, P = .11, Cramér V = 0.06). In contrast, fewer SSATs (55.4%, n = 404) than CATs (77.7%, n = 275) had access to a school psychologist to whom they could refer student-athletes experiencing academic difficulty after an SRC (χ² = 51.58, P < .001, Cramér V = 0.22). Despite their reported availability, 21.4% (n = 156) of SSATs and 37.0% (n = 131) of CATs had never accessed a school counselor, learning specialist, neuropsychologist, or psychologist when managing a student-athlete with an SRC. When seeking assistance in managing a concussed student-athlete, SSATs most commonly accessed school counselors (71.2%, n = 519), neuropsychologists (32.0%, n = 233), or school psychologists (26.7%, n = 195). In contrast, CATs most commonly accessed neuropsychologists (37.6%, n = 133), psychologists (35.9%, n = 127), or learning specialists (29.9%, n = 106).

Current Knowledge

Our participants self-identified as being knowledgeable about academic accommodations after SRC. However, our results suggested that respondents were not fully aware of the concepts associated with academic accommodations. Yet most agreed that concussed student-athletes should have access to such accommodations. Williams et al¹¹ surveyed SSATs to assess familiarity, beliefs, and attitudes about academic accommodations for student-athletes after SRC. When referencing the Americans with Disabilities Act, the authors¹¹ reported that 82% of SSATs believed concussed student-athletes were eligible for academic accommodations after an SRC, consistent with our findings. Most surveyed SSATs were also aware of existing RTL guidelines and correctly identified their role in facilitating academic accommodations, such as via a 504 plan or IEP. Our results were similar to those of Williams et al,¹¹ as 81.3% of our respondents correctly identified a concussed student-athlete as eligible for a 504 plan. Nevertheless, up to approximately 30% of respondents were unaware that this academic accommodation was an option available to assist injured student-athletes.

Approximately one-third of surveyed SSATs and CATs incorrectly identified academic-accommodations letters as legally binding. According to the Children's Healthcare of Atlanta concussion consensus statement,¹⁵ letters of academic accommodations from a physician are merely a tool intended to prompt the use of established recommendations and are not legally binding. Whereas a letter from a treating physician may help facilitate academic accommodations for a concussed student-athlete, ATs should not rely on this documentation alone. A survey²⁷ of secondary school nurses indicated that letters requesting academic accommodations were received primarily from parents. A minority of letters were received from physicians, coaches, and teachers.²⁷ This finding emphasizes that ATs should not rely solely on a physician's request to make academic accommodations for a concussed student-athlete. In the absence of a physician's note, ATs may rely on a variety of resources, including letter templates from the American Academy of Pediatrics,²⁸ the Centers for Disease Control and Prevention,²⁹ and others,³⁰ to facilitate education and communication with teachers or professors, counselors, psychologists, nurses, social workers, school administrators or physicians (or both), neuropsychologists, faculty athletics representatives, college administrators, office of disability services representatives, and coaches.^2,31

More than half of surveyed ATs, regardless of setting, incorrectly thought standardized, evidence-based RTL guidelines existed. To date, standardized, evidence-based protocols or guidelines that detail the management of cognitive recovery and reintegration into the classroom of a concussed athlete do not exist.^{1,13,15,18,32} Evidence-based practice has been defined as the “integration of the best research evidence with clinical expertise and patient values to make clinical decisions.”³³ Whereas evidence-based cognitive recovery guidelines are lacking and their implementation is highly variable,³⁴ a 5-step RTL plan based on clinical experience is available to assist ATs at all levels of sport.⁸ Given that most SSATs and CATs agreed or strongly agreed that a stepwise cognitive progression should be followed when returning a student-athlete to the classroom after SRC, this 5-step RTL plan may be an appropriate starting point for ATs to develop an RTL plan as empirical evidence continues to emerge.²⁶

Current Practices

Our results suggested that substantial progress has been made in RTL clinical practices compared with previously published studies. Most respondents (73.9%) had an established RTL policy or protocol. For approximately 84.6% of participants, regardless of setting, academic modifications (eg, IEPs) or accommodations (eg, modification to the classroom environment or homework assignments) were a part of this plan. More specifically, 79.9% of SSATs had an established RTL policy or protocol in their concussion-management plan. In a similar 2016 survey,²⁶ 43.7% of SSATs reported having an RTL policy. These data demonstrate an increased emphasis on RTL concepts after SRC despite a limited number of available resources to help ATs develop protocols. The existence of a template or sample RTL protocol may facilitate the development of an RTL policy. The NCAA's¹⁷ RTL guidelines provide a sample RTL protocol. Despite these recommendations, two-thirds of our total sample were neutral, disagreed, or strongly disagreed with adhering to the NCAA's¹⁷ RTL guidelines. When isolating our results to CATs, more than half (57.3%) agreed or strongly agreed with adhering to the NCAA's¹⁷ RTL recommendations. The remaining respondents (42.7%) were neutral about or disagreed with adhering to the NCAA's¹⁷ recommendations. Whereas a limitation of our study was excluding open-ended responses to address items such as this, one plausible explanation for CATs not adhering to the NCAA's¹⁷ RTL recommendations is access to a greater variety of health care team members.

Regarding additional support, 20.6% of SSATs and 30.8% of CATs reported the absence of a health care team member to assist in providing academic accommodations. In the absence of an additional health care team member, most SSATs (84.9%) and CATs (92.7%) were responsible for making return-to-classroom decisions, which was consistent with the literature.²⁶ Additional health care team members would most likely benefit the CAT in designing an RTL policy and facilitating academic modifications based on available resources. Related to this point, up to 17.4% of SSATs and 29.1% of CATs were unsure of the presence of 1 or more additional support personnel on their respective campuses to assist with an RTL policy. In agreement with this finding, fewer SSATs than CATs reported not having accessed a health care professional or department similar to a disability resource department. These CATs and SSATs should explore their available resources, such as a disability resource department or similar option, to potentially expand their RTL team and provide better care for concussed student-athletes. In addition to supplementary professional items, letter templates, videos, and other resources can help ATs communicate with and educate stakeholders about SRC and better facilitate RTL guidelines. Even when ATs have the appropriate health care team support, they may perceive a lack of authority to implement the changes related to recommended guidelines.²⁴ This may be especially true within the collegiate setting, where ATs may perceive they have limited authority or control, or both, to institute academic adjustments. Given that incorporation of RTL policies into a concussion-management protocol is now commonly recommended,^1,12 ATs who self-identified in our study as leaders of RTL protocol and policy development should use this evidence to advocate for their concussed secondary school or collegiate student-athletes.

Whereas SSATs and CATs described being primarily responsible for RTL decision making, respondents also identified a variety of other individuals (eg, the student-athlete, parents, academic advisor, teachers or faculty, other ATs, physicians, neuropsychologists, school psychologists) who assisted in returning students to the classroom. Creating and implementing an RTL policy or protocol is critical for managing concussed student-athletes and facilitating communication among all stakeholders. McGrath⁸ suggested that academic support during a concussed student-athlete's recovery was necessary to balance rest with academics and encouraged an individualized plan for each injured athlete using a team-based approach. In a retrospective study, Baker et al³⁵ investigated factors likely to be associated with difficulty in reintegrating concussed pediatric student-athletes into the classroom. The authors observed that symptom burden, which varies highly from one student-athlete to another, was weighted heavily as being problematic when returning to the classroom. An increased awareness of a student's specific symptom phenotype by those involved in his or her recovery (eg, school nurse, staff, guidance counselors, teachers, and parents) may facilitate a more rapid and full recovery. As with any medical care, to improve patient outcomes, the RTL and RTP teams should consider a team “huddle”^36,37 or “time-out”³⁸ to facilitate effective communication about athletes who experience prolonged symptoms after their injuries. Team huddles or time-outs are designed to increase communication among stakeholders before or in response to patient care and have been demonstrated to improve patient outcomes.^36,37 A meeting of the RTL team before and after implementing academic modifications, accommodations, or both may help ATs refine the RTL policy and improve patient outcomes.

Unique to our study was that most of the surveyed ATs (80.9%) noted that student-athletes had requested academic accommodations (eg, temporary classroom environment or coursework alterations) after an SRC. This finding suggests a cultural change in the expectations of concussed student-athletes. These different expectations may result from increased concussion education and awareness of the injury and potential consequences (eg, prolonged recovery) associated with mismanagement.^39–41 Ransom et al¹⁰ proposed that student-athletes with a higher scholastic load (eg, academic credit hours) need more support than those with a lower academic load. This may explain the observed 11.6% difference between the requests for academic accommodations in the secondary and collegiate settings. Student-athletes with a greater number of symptoms after an SRC are also most likely to need additional support to facilitate recovery from their injury.³⁵ Again, this finding emphasizes a potential disconnect between creating and implementing an RTL policy. For example, whereas student-athletes may start an RTL progression that has been suggested by multiple governing bodies,^17,28 they may face challenges when asking educators for academic accommodations. Our results indicated that student-athletes were limited in obtaining academic accommodations via communication with their educators; therefore, they requested assistance from ATs to facilitate the adjustments, regardless of the academic setting. Athletic trainers may consider writing a letter (Figure 5) to each of the injured student's faculty members to facilitate an individualized approach to reintegration into the classroom. Because the SSAT or CAT may be considered the “point person” for developing and implementing a concussion policy that includes the parent or guardian, student-athlete, and stakeholder education, ATs should theoretically be able to provide written communication requesting academic accommodations on the concussed student-athlete's behalf. This practice is also in line with more recent supplementary recommendations for complementing state legislation.²⁰ Furthermore, ATs should explore various mechanisms for educating school staff and faculty about concussion with an emphasis on RTL guidelines. Future researchers should examine the use of AT-recommended academic adjustments in the classroom by secondary school and collegiate faculty.

View Figure

• Download as Powerpoint
• Download Figure

Most (88.9%) of our surveyed ATs believed they should be involved in the RTL process. With regard to making academic adjustments, Kasamatsu et al²⁶ reported that SSATs more often (28.9%) identified themselves as the “academic point person” than school counselors (17.2%) or school nurses or health clerks (11.9%). However, the manner in which they acted varied from published recommendations. For example, only 18.8% of respondents were neutral or disagreed that “a student-athlete must complete a full day within the classroom with no restrictions prior to beginning the exercise component of a[n] RTP protocol.” This finding contradicts RTP progression recommendations after an SRC, which suggest an athlete must be symptom free before completing the latter stages of a graded exercise protocol.¹ If a student-athlete reports SRC symptoms after a full day at school, then he or she should not progress to the next step in the RTP protocol. Thus, standardized, evidence-based guidelines are needed. However, our study was conducted before the 5th International Conference on Concussion in Sport,¹ which expanded on this recommendation by suggesting that symptomatic student-athletes should not be engaged in active practice or returned to competition but may participate in activities specified in the earlier stages of an RTP protocol.

Another interesting result of our study was that ATs believed they received less pressure to return student-athletes to the classroom than to physical activity. This finding may reflect the relatively recent implementation of state legislation mandating the care of secondary school student-athletes suspected of having an SRC. It may also be associated with negative media coverage associated with premature RTP decisions. Our SSATs and CATs perceived pressure primarily from the student-athletes to make RTP decisions. Whereas this finding was not entirely unexpected based on previous literature⁴² addressing the rationale for the underreporting of concussions, it emphasizes the need for education and cultural change related to the expectations associated with SRC management. As with the aforementioned survey items, this result highlights the need for the AT and the entire RTL team to work with all stakeholders (via communication and education) to reduce any anxiety associated with time missed from the classroom and participation, which has been demonstrated to negatively affect recovery.³⁵

Available Resources

The SSATs and CATs had varying degrees of access to supporting health care team members for assistance in managing RTL protocols for concussed student-athletes. Our results demonstrated that 38.5% of ATs had access to a neuropsychologist, which was consistent with the findings of similar studies.²² Approximately one-third of CATs had accessed a learning specialist, a neuropsychologist, or a psychologist to assist in managing a concussed student-athlete. However, an additional one-third of CATs had never sought the services of these professionals to care for a student-athlete with an SRC. Buckley et al²⁴ investigated concussion-management–practice patterns of NCAA Division II and III ATs and reported major differences between Division I and Division II and III ATs that reflected staffing and budget variations. Our SSATs reported less access to various health care professionals, which most likely was due to budgetary concerns and limited resources. Most surveyed SSATs (71.2%) had accessed a school counselor; a minority stated they had accessed a neuropsychologist (32.0%) or school psychologist (26.7%). Given that school counselors are typical staff members in the secondary school setting, they theoretically would be the most likely candidates to assist an AT with an RTL protocol. In addition, approximately 10% fewer SSATs than CATs commented that they had not accessed these professionals to assist in managing SRC specific to the RTL process. Athletic trainers, regardless of academic setting, should investigate their access to all potential resources (eg, neuropsychologists, school counselors, and others) to assist in creating and implementing an RTL protocol.

Limitations

Our study had limitations. Whereas our respondents represented the entire United States and the District of Columbia, equal representation per state was not achieved. Therefore, geographic bias may have been introduced to obtain a greater total sample. Because ATs self-reported their roles and training, we assumed participants were familiar with the concept of RTL. If unfamiliar, respondents may have answered items based on their own definitions rather than established definitions from the literature. Given that 1 aspect of our study was to investigate ATs' knowledge related to RTL protocols, we believed that providing definitions of concepts, such as cognitive rest, might have skewed participant responses. Future researchers should aim to identify the most efficient mode and duration of cognitive rest and its interaction with physical activity during recovery. Finally, whereas several differences were observed between SSATs and CATs, most of our significant findings had small effect sizes. However, small effect sizes did not decrease the clinical relevance of our results because we identified several areas related to RTL policy development and implementation that should be considered by SSATs and CATs alike. By understanding the current state of RTL in the context of concussion management, clinicians and investigators can make strides toward implementing strategies to improve best practices. Taking these steps will inherently lead to creating evidence-based RTL consensus statements to be used by ATs and other health care professionals.