Concussions were first documented by Hippocrates "approximately 2400 years ago" (Mullally, 2017). The 5th International Conference on Concussion in Sport (McCrory et al., 2017) defined sports-related concussion (SRC) as "a traumatic brain injury induced by biochemical forces" that "typically results in the rapid onset of short-lived impairment of neurological function that resolves spontaneously" (McCrory et al., 2017). The consensus statement mentions characterizing symptoms of SRC such as somatic (e.g. headache), cognitive (e.g. feeling like in a fog) and emotional symptoms (e.g. lability), as well as other features like physical signs (e.g. loss of consciousness), cognitive impairment, balance impairment, behavioral changes and sleep-wake disturbances. Furthermore, there are now tools to aid in recognition of SRC e.g. Sport Concussion Assessment Tool 5 (SCAT5) (Echemendia et al., 2017a), and recommendations for management of SRC and subsequent return to play e.g. graduated return to sport (McCrory et al., 2017).
Increased awareness of this condition (Zhang et al., 2016) and understanding of the long-term sequelae of concussion (Kushner, 1998; McKee and Daneshvar, 2015; Tagge et al., 2018; Zhou et al., 2013), has led to some authors labelling it a public health crisis (Harrison, 2014). This increased knowledge has also prompted a number of high profile lawsuits, such as that in the United States National Football League (NFL) (Sirisena et al., 2017) and a possible class-action suit by former Australian National Rugby League players, in light of new evidence of concussion-related injuries (Buckland et al., 2019). Such lawsuits have also increased research in this field.
With governments encouraging increased exercise and sports participation to combat the international physical inactivity epidemic, there is growing interest in SRC (Belechri et al., 2001; Wallace et al., 2017). In the United States of America alone, it is estimated that 3.8 million SRC are reported each year, across multiple competitive sports (Harmon et al., 2013). These numbers may be simply the tip of the iceberg, as many such incidences go unreported (Daneshvar et al., 2011), due to factors such as lack of awareness amongst coaches (Graham et al., 2014), parents (Weerdenburg et al., 2016), match officials (Griffin et al., 2017), sports trainers, and athletes themselves (Taylor and Sanner, 2015). In light of the wide-ranging repercussions of SRC, numerous measures have been put in place, such as changes to legislation (Gibson et al., 2015), game rules (Benson et al., 2013) and increased emphasis on coach education (Emery et al., 2017; Patel et al., 2017).
Immediate removal from play and assessment of players after sustaining SRC is the minimum standard of care for the safety of athletes (McCrory et al., 2017). Even though SRC is a medical diagnosis, healthcare professionals may not always be on-site for all amateur or school sporting events (Kroshus et al., 2017b) leaving coaches and officials to be the first responders to concussed players. While tools like Concussion Recognition Tool 5th Edition (CRT5) also provide guidance to non-medical professionals in decision making for removal of play (Echemendia et al., 2017b), the appropriate use of such tools is contingent on knowledge and awareness of SRC. Inadequate immediate management of SRC has been reported, beckoning the need for better concussion education amongst sports clubs, coaches and other important stakeholders (Haran et al., 2016). Furthermore, given that coaches play an important role in influencing an athlete's decision to report concussions (Baugh et al., 2014), it is pertinent that they demonstrate and communicate knowledge on the consequences SRC and thereby encourage SRC reporting.
The objective of this systematic review was to explore the knowledge level of SRC amongst sports coaches and matches officials to guide further discussion on educational programmes, policy change and guidelines.
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were applied while conducting the review (Liberati et al., 2009). A systematic search was conducted using PubMed, Medline Ovid, Web of Science, CINAHL, SPORTDiscus and Psycinfo for studies published from 1st January 2001 to 5th January 2019. The first International Conference on Concussion in Sport was held in November 2001 (Aubry, 2002), and hence only studies from this year onwards were included, allowing for assessment of current concussion knowledge based on the international consensus. Search terms used included a combination of population and concussion terms (Table 1). The term "Knowledge" was deliberately excluded as a criterion for the search to include more potential studies.
Screening and review process
The results of all searches were entered into the Veritas's Covidence software (Innovation, 2017) where title and abstract screening for eligible studies was performed. This entailed a 2-stage process, initially by independent reviewers PC and QY, and subsequently by reviewers DS, JP and SS. Full texts of remaining studies were then assessed for eligibility by all 5 reviewers. Disagreements were resolved by consensus discussion at each stage. The following eligibility criteria were used in the screening and reviewing process.
Primary studies published in English in peer-reviewed journals from 1st Jan 2001 till 5th Jan 2019 were included. The studies that were included assessed level of concussion knowledge and/or level of education on concussion amongst coaches and/or match officials in all sporting levels (professional, amateur, college or school sports).
Studies that were not related to the level of concussion knowledge and/or level of education on concussion amongst coaches and/or match officials as well or studies assessing concussion knowledge amongst athletic trainers instead of coaches were excluded. Studies with no abstract available for screening and those without available English translation were also excluded.
Information regarding concussion knowledge was extracted from each study, particularly pertaining to the following domains:
1) Identification of concussion
2) Initial management of concussion
3) Return to play (RTP)
4) Prevention of concussion
5) Consequences of concussion
These domains were frequently assessed amongst the studies reviewed which are relevant to coaches and sports officials according to international consensus (McCrory et al., 2017). Other information such as prior concussion education and desire for further concussion education were also documented. The Appraisal Tool for Cross-sectional Studies (AXIS) was employed to assess the quality and reliability of each study (Downes et al., 2016).
The search identified 20880 studies after exclusion of 2744 duplicates, with 42 full texts assessed for eligibility. 27 studies were eventually included (Figure 1).
Characteristics of studies
There were 26 cross-sectional studies (Refer to Table 2) and one randomised control trial (Glang et al., 2010) of which 20 studies assessed concussion knowledge amongst coaches, one explored knowledge amongst officials while six studies assessed both. The randomised controlled trial by Glang et al. (2010) involved 75 coaches who went through an interactive e-learning program as part of the intervention arm. Pre-test results were used to determine baseline concussion knowledge of this study population.
Fifteen studies took place within the United States (Refer to Table 2), with other studies from Canada (n=3), Australia (n=2), Ireland (n=2), the United Kingdom (n=2), Germany (n=1), Italy (n=1) and New Zealand (n=1). Twelve studies assessed coaches and officials across multiple different sports, while others were specific for Rugby (n=6), Hockey (n=4), American football (n=3), Soccer (n=1) and Wrestling (n=1). Across all studies, a total of 3750 coaches and 1790 officials were included. The smallest study involved eight coaches (Kirk et al., 2018), and the largest study involved 1324 referees (Kroshus et al., 2017a). Two studies included coaches and officials across different tiers (Mathema et al., 2016; Saunders et al., 2013), 12 studies focused on high school or junior school sports, four on college sports, six on amateur sports and three on professional or semi-professional sports.
A large proportion (13 out of 27) of studies crafted original questionnaires to assess concussion knowledge (Refer to Table 2), many of which were utilised and adapted in subsequent studies. There was a variety of questionnaire types including open-ended, multi-choice, true-false and scenario-based questions.
The AXIS guideline was employed to assess a study's quality and risk of bias based on 20 criteria, as summarised in Table 3. Studies were found to be unsatisfactory in an average of five criteria. All except one study (Bramley et al., 2011) did not manage to take measures to limit or characterise non-responders, and most studies were at risk of selection bias as participants were invited to participate via a voluntary opt-in system. Response rates were low, with only seven (Bramley et al., 2011; Broglio et al., 2010; Faure and Pemberton, 2011; Fraas et al., 2014; Mathema et al., 2016; Niederer et al., 2018; Valovich McLeod et al., 2007) having a response rate of more than 60%. Only one study (Bramley et al., 2011) satisfied all 20 criteria.
Results of individual studies
Identification of concussion: Twenty-four of the 27 studies (Refer to Table 2) assessed for knowledge on identification of SRC. Most studies employing scored questionnaires reported concussion identification scores of above 50% (38.9% to 90.5%) amongst both coaches and officials.
The exception to this was one study based in Italy where identification of SRC-associated symptoms was poor...