Article Text

Download PDFPDF

Evaluation of World Rugby's concussion management process: results from Rugby World Cup 2015
  1. Colin W Fuller1,
  2. Gordon W Fuller2,
  3. Simon P T Kemp3,
  4. Martin Raftery4
  1. 1Colin Fuller Consultancy Ltd, Sutton Bonington, UK
  2. 2Emergency Medicine Research in Sheffield Group, School of Health and Related Research, University of Sheffield, Sheffield, UK
  3. 3Rugby Football Union, London, UK
  4. 4World Rugby, Dublin, Ireland
  1. Correspondence to Dr Colin Fuller, Colin Fuller Consultancy , Sutton Bonington, UK; ColinFullerConsultancy{at}


Objective To evaluate World Rugby's concussion management process during Rugby World Cup (RWC) 2015.

Design A prospective, whole population study.

Population 639 international rugby players representing 20 countries.

Method The concussion management process consisted of 3 time-based, multifaceted stages: an initial on-pitch and/or pitch-side assessment of the injury, a follow-up assessment within 3 hours and an assessment at 36–48 hours. The initial on-pitch assessment targeted obvious signs of concussion, which, if identified, lead to a ‘permanent removal from play’ decision and a diagnosis of concussion. If the on-pitch diagnosis was unclear, a 10-min off-pitch assessment was undertaken for signs and symptoms of concussion leading to a ‘suspected concussion with permanent removal from play’ or a ‘no indication of concussion with return to play’ decision. Evaluations at 3 and 36–48 hours postmatch lead to diagnoses of ‘confirmed concussion’ or ‘no concussion’. Medical staff's decision-making was supported during each stage by real-time video review of events. Players diagnosed with confirmed concussion followed a 5-stage graduated-return-to-play protocol before being allowed to return to training and/or competition.

Results Players were evaluated for concussion on 49 occasions, of which 24 resulted in diagnoses of concussion. Fourteen players showing on-pitch signs of concussion were permanently removed from play: 4 of the 5 players removed from play following off-pitch medical room evaluation were later diagnosed with a confirmed concussion. Five players not exhibiting in-match signs or symptoms of concussion were later diagnosed with concussion. The overall incidence of concussion during RWC 2015 was 12.5 concussions/1000 player-match-hours.

Conclusions This study supports the implementation of a multimodal, multitime-based concussion evaluation process to ensure that immediate and late developing concussions are captured.

  • Concussion
  • Contact sports
  • Epidemiology
  • Rugby
  • Risk factor

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Concussion is a traumatic brain injury that may present with transient, fluctuating, evolving or delayed signs and symptoms. Recently, concussion has attracted growing attention among sports physicians, athletes, media and the public.1 ,2 Interest has focused, in particular, on questions about consistency in the identification of concussed athletes and their subsequent management and rehabilitation.3 ,4 Despite worldwide interest and the intervention of several national and international concussion groups, limited progress has been made in answering these questions.4–9 Advances in the practical, immediate management of concussion have been constrained by a lack of progress in the development of an accepted operational definition of concussion that can be used to identify concussed athletes during match play.10–12 While the immediate effects of most concussions are short-lived, typically resolving within days, athletes who continue to play following a concussive event may increase their risk of developing short-term, medium-term and long-term consequences. These consequences may include reduced playing performance, non-concussion injuries and an increased risk of developing subsequent concussions and long-term conditions such as chronic traumatic encephalopathy and amyotrophic lateral sclerosis.13–15

The reported incidence of concussion in elite 15-a-side rugby union (rugby) has increased from ∼4 to ∼13 injuries/1000 player-match-hours over the past 10 years.16 ,17 It is not possible to determine conclusively from the published data whether this represents a real increase in the incidence, a secondary effect arising from greater clarity and expansion of diagnostic criteria for concussion or a greater tendency by players and physicians to identify and report concussions due to increased awareness of the potential adverse health effects. This latter viewpoint is supported by the fact that overall injury rates in rugby have remained unchanged over the same period.17 In the absence of universal agreement on an operational protocol for evaluating concussion,9 World Rugby introduced a pitch-side protocol specifically for elite rugby in 2012.18 This protocol was subsequently developed further from 2013 to 2015 with the focus on the need to recognise and remove players suspected of sustaining concussion.12 These developments led World Rugby to integrate the match-day concussion management protocol, referred to as the Head Injury Assessment (HIA) process, into the Laws of the Game in August 2015.19 The primary objectives of this paper are to describe the HIA process with enhanced capabilities provided by access to pitch-side real-time and medical room review video facilities and to evaluate the process during Rugby World Cup (RWC) 2015.


This was a prospective, whole population (teams: 20; players: 639; matches: 48) cohort study describing the management of head impact events at RWC 2015. The HIA process is included within the Laws of the Game of Rugby Union and the injury audit was approved by World Rugby's Institutional Ethics Committee: all players provided signed informed consent for their injury data to be recorded.

World Rugby HIA process

The World Rugby HIA process is a multifaceted, multipoint-in-time assessment process aimed at identifying players with a confirmed or suspected concussion and informing players' subsequent injury management. The overall process consists of a number of inter-related elements.

Education and training

  1. All players and team management completed a concussion education session within the previous 12 months.

  2. All players completed a baseline concussion assessment within the previous 12 months, which involved an SCAT3 assessment as the minimum requirement for the HIA process; however, team physicians could, at their discretion, include more comprehensive tests.

  3. All team medical and tournament match-day medical staff completed the World Rugby training modules for ‘Concussion Management’ and ‘Medical Protocols for Match-day Medical Staff’.

  4. All team doctors, match-day doctors and match-day immediate care leads attended a face-to-face pre-tournament HIA process and concussion management training session.

HIA process

The HIA process consists of three time-based assessment stages designed to include the full range of potential concussion presentation formats, as summarised in figure 1.

  1. HIA-1 stage: consists of two subsections

    A. Section 1: a player is immediately and permanently removed from further participation in the current match if the player's team doctor or the match-day doctor identifies any sign(s) of concussion (table 1).7 Match-day doctors had access to pitch-side real-time video facilities to assess the event and to support their decision-making. Players removed from a match with any of these criteria are deemed to have a confirmed concussion and despite the confirmed concussion diagnosis they subsequently undergo postmatch HIA-2 and HIA-3 assessments.

    B.  Section 2: team medical staff, match-day doctor or referee must request an off-pitch assessment (lasting up to 10 min) for any player who has a head impact event where the injury diagnosis is not immediately clear. Players must be led unaided to a pre-designated medical room for an assessment that includes three sequential domains comprised of the Maddocks' and SAC questions, a tandem gait test and an assessment and recognition of any symptoms and clinical signs of concussion (see table 1). Match-day doctors and team doctors were supported in their clinical decision-making by a real-time video review system, which provided instant replays of match events from up to 30 camera angles/feeds pitch-side and in the medical room. The HIA-1 (Section 2) assessment is not intended to provide a clinical diagnosis of concussion but a triage to determine whether a player may have a suspected concussion and therefore whether they may return to play or not. An abnormal test result in any domain or a clinical judgement of suspected concussion mandates the removal of the player from the match. A ‘normal’ HIA-1 (Section 2) assessment enables the team doctor to allow the player to return to the match. All players undergoing an HIA-1 (Section 2) assessment, irrespective of the outcome, are required to undertake follow-up HIA-2 and HIA-3 assessments to determine whether a player has a confirmed concussion.

  2. HIA-2 stage: consists of an SCAT 3 evaluation (excluding the Glasgow Coma Scale and Maddocks' questions) within 3 hours of injury to assess clinical progress and to provide an early diagnosis of concussion for (a) players removed from the pitch following an HIA-1 (Section 1 or Section 2) assessment, (b) if a postmatch video review of match events indicates a player may have sustained a head injury and (c) players who did not undergo an HIA-1 assessment but developed sign(s) or symptom(s) of concussion within 3 hours of the match. Any abnormal outcome from a player's HIA-2 examination is considered a diagnosis of concussion unless proven otherwise. All players undergoing an HIA-2 assessment are required to undertake an HIA-3 assessment.

  3. HIA-3 stage: consists of a clinical assessment of a player supported by an expanded SCAT 3 symptom checklist that identifies the number, maximum severity and duration of concussion symptoms. HIA-3 also includes a cognitive assessment, preferably using a computerised neuro-cognitive tool such as CogSport (or as a minimum the SAC assessment) and a balance assessment using the balance error scoring system and tandem gait balance tests. The HIA-3 assessment is completed (a) after two nights rest (typically 36–48 hours postinjury), (b) if a player, who did not undergo an HIA-2 assessment, develops sign(s) or symptom(s) of concussion between 3 and 48 hours postmatch or (c) if a postmatch video review of match events indicates that a player may have sustained a head injury. Any abnormal outcome from a player's HIA-3 examination is considered a diagnosis of concussion unless proven otherwise.

Table 1

HIA-1 stage: signs and symptoms of concussion leading to a player's immediate and permanent removal from play7

Figure 1

Flowchart of HIA Concussion management protocol.

The three-stage process, therefore, leads to a diagnosis of concussion if one or more of the following outcomes are reported:

  1. The player had any Criteria 1 sign(s) of concussion (see table 1) during the on-pitch evaluation that resulted in his removal from play (HIA-1, Section 1).

  2. Any Criteria 1 sign(s) of concussion were identified during the medical room video reviews of the player's head impact event.

  3. The 3-hour, postinjury (HIA-2) assessment is ‘abnormal’.

  4. The 36–48-hour, postinjury (HIA-3) assessment is ‘abnormal’.

  5. The treating doctor deemed the player to have sustained a concussion.

World Rugby injury surveillance study

All teams were required to take part in the routine RWC Injury Surveillance Study, which followed the international consensus statement for rugby injury surveillance studies:20 a full description and the results obtained from this study were presented in a separate report.21 Within this broader injury audit, team doctors were required to report players' anthropometric data (age, stature, body mass) and details of all confirmed concussions resulting from match and training activities: the injury surveillance study therefore, provides a cross-check on the HIA process and provides an additional time-based assessment for concussions that develop more than 48 hours postmatch.

World Rugby graduated-return-to-play protocol

All players sustaining a confirmed concussion completed a five-stage stepwise graduated-return-to-play protocol as described in previous publications.4–9

Untoward incident review

All players and team management involved in RWC 2015 were required to sign a Terms of Participation document, which included a section specifically related to player welfare matters. Of relevance to this study was the requirement that players and team management must comply with the HIA process and stepwise graduated-return-to-play protocol and failure to do so was regarded as misconduct and the relevant parties would be subject to an ‘Untoward Incident Review’ lead by World Rugby's Chief Executive Officer, Head of Legal Services and Chief Medical Officer.

Data collection and statistical analysis

The HIA process assessments, including the influence of video reviews on the identification of head impact events and concussion diagnoses, were recorded on a tablet computer-based system (CSx, Auckland, New Zealand), pre-loaded with every player's SCAT 3 baseline data. All test results were automatically uploaded onto a secure World Rugby database.

Player flow through the HIA process was evaluated descriptively. Assessments of differences in players' anthropometric data were made using unpaired t-tests, in the numbers of backs and forwards sustaining injuries using χ2 tests, in incidence and mean severity of injuries using z-tests and in median severities of injury using Mann-Whitney U tests:22 statistical significance was accepted at p<0.05 values. Statistical analyses were conducted in Stata V.13.1 (StataCorp, College Station, USA).


Sample characteristics

Figure 2 summarises the player flow through each stage of the HIA concussion management process. There were 49 cases when players were evaluated for concussion (43 players once; 3 players twice); of these, 24 cases resulted in diagnoses of confirmed concussion (22 players once; one player twice), which corresponds to a concussion period prevalence among the players of 3.8% and an incidence of 12.5 concussions/1000 player-match-hours. Of the 24 confirmed concussions, 15 were sustained by backs and 9 by forwards (p=0.120): there were no statistically significant differences in the mean age (backs: p=0.234, forwards: p=0.697), stature (backs: p=0.078, forwards: p=0.115) or body mass (backs: p=0.968, forwards: p=0.123) of backs or forwards sustaining concussions compared to all backs and forwards competing at RWC 2015 (table 2).

Table 2

Anthropometric data for the total and the concussed populations

Figure 2

Data flow for players undergoing assessments for concussion. n/a, not assessed.

Player outcomes in the HIA management process

Of the 49 cases evaluated for concussion, 39 were first assessed during matches. Of these 39 cases, players were removed from play immediately on 14 occasions following the identification of signs of concussion (table 1) during the on-pitch HIA-1 (Section 1) evaluation and on 5 occasions because there was an abnormal off-pitch HIA-1 (Section 2) assessment indicating a suspected concussion. Four of the five players suspected of sustaining a concussion were later diagnosed with confirmed concussions following an abnormal HIA-2 (three players) or HIA-3 (one player) evaluation. Of the 20 players not suspected of sustaining a concussion following their HIA-1 (Section 2) evaluation, one player was later diagnosed with a concussion following a normal HIA-2 evaluation and an abnormal HIA-3 evaluation. Eight players presented within 3 hours postmatch either with a suspicion of concussion or had experienced a head impact event identified on postmatch video review; two of these players were diagnosed with concussion following abnormal HIA-2 evaluations and one was diagnosed with a concussion following a normal HIA-2 but an abnormal HIA-3 evaluation. Two players presented 36–48 hours postmatch, both of whom were diagnosed with concussion following abnormal HIA-3 evaluations. Of the 19 players removed from the field of play during the game, one player did not complete an HIA-2 or an HIA-3 examination (due to a fractured mandible injury sustained in the same incident) and two players completed HIA-2 examinations (abnormal) but did not complete HIA-3 examinations. Of the 20 players with a normal HIA-1 (Section 2) evaluation, all completed HIA-2 and 18 completed HIA-3 examinations.

Video review outcomes

Pitch-side, real-time videos and medical room video reviews were used to identify or confirm 11 of the 19 head contact events that led to the immediate and permanent removal of players from the field of play. Video review was also used to assist team doctors in their evaluations in five of these incidents. In addition, pitch-side videos were used to identify 13 of the 20 head injury incidents that returned ‘normal’ HIA-1 results and were used to assist team doctors in their evaluations in five of these incidents. Postmatch video analysis was used to identify a head contact incident that subsequently resulted in an abnormal HIA-3 and a diagnosis of confirmed concussion.

Concussion severity and return-to-play

The mean and median duration of all concussions were 7.7 (95% CI 6.1 to 9.3) and 6 (95% CI 5 to 8) days, respectively. There were no significant differences in the mean or median severities as a function of the stage within the HIA process that the concussions were first confirmed (table 3). Of the 24 players with a confirmed concussion (71%), 17 returned to play in ≤7 days and 7 (29%) returned to play in ≥8 days.

Table 3

Severity of concussion as a function of the stage within the HIA process that concussion was first confirmed

Every player with a confirmed concussion followed World Rugby's graduated-return-to-play protocol and returned within the recommended timeframe of the protocol.7


Each stage of the HIA concussion management process made a contribution to the diagnosis of at least one of the 24 confirmed concussions reported in this study. The proportions of these 24 cases where contributions were made were: HIA-1 (Section 1)—58%; HIA-1 (Section 2)—17%; HIA-2–54%; HIA-3–38%; pitch-side video (incident identification)—46%; pitch-side video (signs of concussion)—21%; postmatch video analysis (incident identification)—4%; postmatch video analysis (signs of concussion)—4%. Postmatch video analysis identified a head impact event missed on match day that resulted in a confirmed concussion. Only one player (4.2%) who returned a normal HIA-1 (Section 2) evaluation (and consequently was not removed from play) was later diagnosed with a confirmed concussion following a normal HIA-2 evaluation and an abnormal HIA-3 evaluation. Importantly, diagnoses of confirmed concussion were reached via seven of the eight possible HIA pathways (figure 2).

The summary above highlights a major advantage of the three-time-point HIA process, as it identified five ‘delayed concussions’ that might otherwise have been missed if the process had relied on in-match assessments alone. This is an important result that could influence future recommendations about how and when athletes should be evaluated for concussion. At the present time, concussion consensus statements provide guidance for on-pitch/side-line evaluation of players and make recommendations for follow-up assessments only for those players diagnosed with suspected or confirmed concussion.4–9 None of the consensus statements present recommendations for further postmatch assessments for those players not diagnosed with a confirmed or suspected concussion during the on-pitch/side-line evaluation. In this study, 25% of all confirmed concussions were ‘delayed’ concussions, which were only identified postmatch. Furthermore, our findings support the role of real-time pitch-side video and medical room video reviews in managing head impact events during play, as the use of video contributed to the majority of concussion cases identified during play.

Development of concussion

While this study evaluated the contribution made by each stage of the HIA process, it did not set out to evaluate the reliability of individual tests for concussion, as each test has been independently studied previously or is recommended by one or more concussion consensus groups.4–9 It is recognised that the sensitivity and specificity associated with each test means that some test results may be false positives and others may be false negatives. It is also recognised that there will be an assessment threshold level above which a diagnosis of concussion will be recognised and below which a diagnosis will not be recognised. If, for the purposes of the following discussion, it is assumed that the outcomes from the HIA-1, HIA-2 and HIA-3 evaluations presented in figure 2 represent valid assessments of a player's condition with regard to concussion at the moment in time when the test was carried out, the study provides important new insights into the complexity and the variations observed in the presentation of concussion in the elite tournament setting.

It is generally accepted that concussions can manifest in a range of presentations , including transient signs such as ataxia or loss of consciousness and/or short-term and medium-term symptoms such as headache or irritability.23–25 These effects, some of which resolve within seconds/minutes while others resolve within hours/days, are generally considered to be caused by changes in various pathophysiological processes, such as ionic shifts, altered metabolism, impaired connectivity and changes in neurotransmission.23–25 The range of concussion presentations observed in this study can all be explained by adopting a simplified two-phase process consisting of the following phases:

Phase 1: a rapidly developing effect leading to transient signs such as tonic posturing and ataxia, which would be observed during the HIA-1 assessment stage but which are likely to have resolved by the time of the HIA-2 and HIA-3 assessments (figure 3A).

Figure 3

Theoretical model for the development and recovery from short-term and long-term signs and symptoms of concussion. (A) Short-term signs and symptoms. (B) Medium-term signs and symptoms. (C) Combined signs and symptoms.

Phase 2: a slower-developing effect leading to longer-lasting symptoms such as lack of concentration and susceptibility to light and noise (figure 3B).

At any point in time, it can be assumed that athletes being assessed for concussion will exhibit the sum of the signs and symptoms derived from these two phases (figure 3C). The manifestation of each player's concussion status during HIA-1 (5–10 min postevent), HIA-2 (3 hours postmatch) and HIA-3 (36–48 hours postmatch) assessments would therefore depend on the relative magnitudes and rates of development of and recovery from these two phases of concussion. While this is a very simplistic model of an athlete's development and recovery from concussion, it serves to describe and explain the results observed in this study, especially how players can be cleared of concussion during on-pitch assessments (absence of phase-1 effects) but develop symptoms of concussion 36–48 hours postmatch (development of phase-2 effects); the model therefore highlights why it is important to incorporate postmatch assessments within a diagnostic protocol for concussion.

Strengths and limitations of study

This study has a number of strengths. In particular, it is a whole population, prospective study with all concussions being evaluated by experienced physicians, who underwent specific training for the assessment and management of concussion prior to the start of the tournament. Data were collected in real-time using a computer-based system with a high level of compliance. In terms of the study limitations, the number of concussions recorded is relatively small, which limits the conclusions that can be drawn from the results. There is also the possibility that a player may sustain a head impact event with the potential for a concussion that is not identified during the match by the team doctor, match-day doctor or referee or postmatch by the video analysts: in these circumstances, concussion reporting is dependent on the player and this could lead to slight under-reporting if players choose not to report their injuries. Although no concussions were identified in the injury surveillance study that had not previously been identified through the HIA process, delayed presentations beyond 48 hours cannot be ruled out. It is also recognised that the concussion management process used at RWC 2015 represents the most comprehensive evaluation process currently available in rugby and it is likely that such a high-level process would only be achievable at the elite professional level of rugby.


The concussion management process used at RWC 2015 and reported in this paper provides a strong case to support the implementation of a multimodal, multitime-based concussion evaluation process incorporating video review. It has also identified the most important elements of the overall process, which can be used at other levels of the game. Ongoing injury surveillance is also recommended at least until the processes involved in the presentation of delayed-onset concussion are more clearly understood.

What are the findings?

  • This is the first study to evaluate a multimodal, multitimeframe head injury management process in the elite adult professional sport setting.

  • Real-time video and match-day doctors provided important elements of the comprehensive concussion management process.

  • Concussions presented in timeframes varying from minutes to at least 48 hours posthead impact.

How might it impact on clinical practice in the future?

  • A multitime-based assessment process is recommended for the identification of transient, developing and delayed concussions.

  • Real-time video review should be integrated, where possible, into the concussion assessment process.

  • Players sustaining head impacts should be monitored for late developing concussions for at least 48 hours postmatch for signs/symptoms of concussion.


The authors acknowledge the invaluable support provided by doctors and physiotherapists from the 20 national teams who provided HIA and injury surveillance concussion data used within this study. They would also like to acknowledge the match-day doctors and medical teams, medical room video reviewers, Hawkeye staff and video operators during the Rugby World Cup, who all contributed to the delivery of the concussion management process.



  • Contributors CWF, GWF, SPTK and MR contributed equally to the design, implementation and evaluation of the study. CWF prepared the first draft of the manuscript and all authors contributed to and approved the final manuscript.

  • Funding The study was funded by World Rugby and supported by England Rugby 2015.

  • Competing interests CWF is an independent research consultant providing support to World Rugby. SPTK is the Chief Medical Officer for the Rugby Football Union. MR is the Chief Medical Officer for World Rugby.

  • Ethics approval World Rugby Institutional Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.