Epidemiology of injuries during the Wheelchair Basketball World Championships 2018: A prospective cohort study

Several international sports federations have implemented a standardized injury reporting system during their championships. However, very few studies have investigated athletes with disabilities during major championships apart from the Paralympic Games. Therefore, the aim of this study was to assess the rate and characteristics of injuries during the Wheelchair Basketball World Championships 2018 (WBWC). This prospective cohort study was conducted during the WBWC held in Hamburg, Germany, from August 16 to August 26, 2018. Physicians or physiotherapists of all 28 participating teams (total 336 players) were asked to report all newly incurred injuries (with location, diagnosis, cause, and estimated duration of absence) daily on a standardized injury report form. Prevalence and incidence rates were calculated. Medical staff of 11 teams (132 players) reported 100 injuries, equivalent to 75.8 per 100 players (95% CI: 60.9‐90.7) or 68.9 per 1000 player‐days (55.4‐82.4). Eight time‐loss injuries were reported (6.1 injuries per 100 players [95% CI: 1.9‐10.3] or 5.5 injuries per 1000 player‐days [1.7‐9.3]). More injuries were incurred during matches (n = 68) than during training. Most injuries affected the neck/cervical spine (16%), thoracic spine/upper back (15%), and shoulder (14%). The most frequent diagnosis was muscle spasms (25%), the most frequent cause was overuse (52%). A high rate of non–time‐loss injuries compared to Paralympic Games was reported. Future studies should focus on the etiology of muscle spasms and further identify injury mechanisms of traumatic and overuse injuries in wheelchair basketball players to develop adequate preventive measures.

examined in such detail. The first published study reports that 79% of British basketball players suffered an injury during the 1992 POG. 5 Curtis and Dillon 6 found that basketball players were the second most injured athletes among wheelchair sports (after track and field). Systematic injury surveillance has been implemented and published for the Summer POG in 2012 7 and 2016 8 and the Winter POG in 2018. 9 Wheelchair basketball was reported to have an injury incidence rate of 12.0 injuries per 1000 athlete-days (95% CI: 8.3-16.8) in 2012 and 12.8 injuries per 1000 athlete-days (95% CI: 9.5-17.4) in 2016. More traumatic than overuse injuries were reported from wheelchair basketball during the 2012 POG. 7 Only little data exist on details of injury, such as injury locations and risk factors, suggesting a higher injury risk for upper extremities in all wheelchair sports. 10,11 Shoulder pain has been retrospectively documented to be prevalent among around half of female players during a major tournament, 12 and prevention strategies for shoulder problems have already been implemented in a pilot study. 13 In summary, only little is known about injury rates and characteristics in wheelchair basketball. Since such information is the first of several steps needed to develop adequate injury prevention strategies, the aim of this study was to analyze the rate and characteristics of injuries during the Wheelchair Basketball World Championships 2018 (WBWC).

| Study design, setting, and participants
A prospective injury surveillance study on acute and overuse injuries newly incurred during the WBWC was conducted. The overall study population comprised 28 teams from 19 different countries with a total of 336 players (male: 16 teams with 192 players; female: 12 teams with 144 players) participating in the WBWC held from August 16 to August 26, 2018, in Hamburg, Germany. During the 11 days of the championships, 94 matches were played, 48 by male and 46 by female teams. The total exposure was 940 player-matches and 3696 player-days.

| Injury surveillance
For the prospective injury surveillance during the WBWC, the standardized system of the International Olympic Committee 14 was used, which is also well established in international tournaments of other sports federations. 15,16 An injury was defined as any newly incurred musculo-skeletal complaint (traumatic or overuse) and/or concussion during the tournament receiving medical attention regardless of the consequences for participation. 14 The teams' medical staff reported daily details of all newly incurred injuries and player details (age, sex, disability classification) on a standardized injury report form. The injury report form was an adapted version of the ones used by IOC, IAAF, and FINA with predefined categories for location, type, cause, match/training, and estimated duration of time loss. [14][15][16] The disability scores (1-4.5) are according to the grade of disability and physical basketball-relevant functionality. 17 Low rates represent a high grade of disability. The injury report form was paper-based and available in English and German.

| Calculation of exposure time and injury rates
Team sizes and team match exposure were determined based on the publicly available roster and schedule. 18 All teams had training sessions on days without a match before their elimination. Player-matches were calculated by multiplying the number of players in a match by number of matches, 15,19 and player training days by multiplying number of players per team multiplied by the number of training sessions. Playerdays were calculated by multiplying the number of registered players by number of days of the WBWC. 20 Injury rates were calculated as number of injuries per 100 players, per 1000 player-days, as training injuries per 1000 player training days, and as match injuries per match and per 1000 player-matches, and reported with 95% CIs.

| Data collection
The study was introduced at the educational meeting for IWBF team manager 1 day prior to the WBWC. All teams willing to participate in the injury surveillance study received information regarding the purpose and logistics of the study. During the championships, one member of the study group (FG) was present at the venue to encourage and assist with participation in cooperation with the local organizing committee. Response rates and quality of data were analyzed daily. One country was excluded completely due to insufficient and implausible reporting (ie, returning report forms without any data). Duplication of data entries (such as reporting of the same injury of the same player on more than 1 day) was resolved by consensus of AJ and KH. Injuries not caused by Wheelchair Basketball were excluded from the analysis. Confidentiality of all information was ensured, and no individual athlete or team could be identified. Ethical approval was granted by the Ethics Committee of the University of Hamburg (protocol number AZ 2018_198). All authors followed the rules of the Helsinki Declaration. The study reports according to the STROBE guidelines for reporting observational studies. 21

| Statistical analysis
Results are described as means with standard deviation or frequencies with percentage. Differences in injury location, type, and mechanism between groups (match vs. training and female vs. male) were analyzed using chi-square tests. All data were processed using Excel (for Mac 11, version 14.7.1, Microsoft Cooperation) and SPSS (V.23).

| Injuries during the championships
Six female and five male teams with a total of 132 players (mean ± SD age 29.7 ± 6.1) from nine different countries participated in the study ( Table 1). The response rate of participating teams was 57.9%, and coverage of all teams and players was 39.3% (for details, see Figure 1). These 11 teams played 74 matches and completed 34 training sessions during a total of 1452 player-days. Exposure time, and number and rates of injuries during the WBWC are presented in Table 2.
No significant differences between training and match injuries were found regarding location and type but for mechanism (χ 2 = 11.1, P = .05; Table 3). Injuries of female and male players differed significantly in mechanism (χ 2 = 15.4, P < .01) but not for location or type (Table 3). Male players had a higher percentage of contact injuries with players, while female players had more non-contact injuries (eg, falls). No sex difference was observed in the proportion of match and training injuries. Distribution of injury numbers and rates shows a wide range over disability classification (Table 5).

| DISCUSSION
This is the first prospective epidemiological study on injuries in wheelchair basketball during a major tournament. The overall injury rate was 75.8 injuries per 100 players or 68.9 injuries per 1000 player-days. About half of the injuries were incurred due to overuse, and a quarter were classified as muscle spasms, mainly at the cervical and thoracic spine. Two-thirds of the injuries occurred during matches, and eight injuries lead to time loss with a maximum of two days. Injury mechanisms differed between training and match, and between female and male players.

| Injury rates and characteristics
From the Barcelona 1992 Paralympic games, 5  No differences in match and training injuries were found regarding location and type but in mechanism. An increase in injury risk during matches can be assumed, which has been reported in pedestrian basketball and other team sports. [22][23][24][25] However, comparing the injury rates to data during a regular season is not possible at the moment and should be investigated in future studies.
Most injuries occurred at the neck, back, and upper extremity, especially at the shoulder. This is consistent with other reports on wheelchair sports 5,26 and could be explained by the fixation of the lower back and lower extremity in the wheelchair. A shoulder injury prevention program has already been investigated in a pilot study and should be further evaluated in future studies. 13 In the present study, 52% of the injuries were due to overuse, while during the 2012 London Paralympic games, acute traumatic injuries have been reported to be more prevalent (65%). 7 Therefore, both injury mechanisms (trauma and overuse) seem to be a problem in wheelchair basketball and should be addressed in injury prevention.
In this study, we provide data on injury numbers and rates according to disability classification. There seem to be differences in the injury rates, even though numbers were too small for a proper analysis. Unfortunately, these data cannot be compared to other research in wheelchair basketball. Future prospective studies in wheelchair basketball should consider including the disability classification as a relevant factor. 27 Eight time-loss injuries (6.1 injuries per 100 players) were reported in our study, which is similar to pedestrian basketball during major tournaments. 3 Furthermore, in contrast to a retrospective study on concussions in wheelchair basketball reporting that 6.2% of players had sustained a concussion during the prior season, 28 no concussion was reported in our study. However, unreported minor concussions have been estimated to be prevalent in about half of the cases. 29

| Strengths and limitations/ methodological considerations
This study was the first that reported details on rates and characteristic of injuries of wheelchair basketball players during a major tournament. Injury surveillance is an important step to develop injury prevention measures and reduce the athlete's injury risk. 30 Nonetheless, this study has some limitations that should be addressed in future studies. Selection bias might have occurred, since teams with more medical personal could have been more willing to participate in the study. Therefore, we do not know whether this population is representative of all elite wheelchair basketball players, and it might not be representative of all nonelite players. Regarding exposure data, training days were documented on a team level, and thus, training injury rates  might be underestimated if single players missed a training session. Female teams had to play more games during the tournament compared to male teams, which increased the match exposure for female players. Furthermore, only newly incurred injuries were monitored, while illnesses were not surveyed, even though illnesses in athletes with disabilities are very relevant 31 ; especially urinary tract infections, decubitus ulcers, and temperature regulation disorders have been reported scientifically and anecdotally by team physicians. 6,31-33 Therefore, prospective injury and illness surveillance should be implemented during the regular season and compared to major tournaments. 2,23,[34][35][36] The Oslo Sports Trauma Research Center (OSTRC) questionnaire could be a potential tool to conduct such studies. 35 The sample size of our study was not high enough for an in-depth analysis of the effect of disability classification on injury risk, and this should be taken into consideration in future prospective studies in wheelchair basketball.

| Perspective
A shoulder injury prevention program focussing on flexibility and strength has been shown to improve shoulder rotational range of motion (ROM) in a pilot study on seven wheelchair basketball players. 13 Internal and external shoulder rotation ROM is considered to be a risk factor for shoulder pain and injury; however, its impact on prevention has not been investigated in wheelchair athletes. In addition, future injury prevention should try to influence the high numbers of muscle spasms by investigating their pathophysiology and etiology.
To date, only little is known about the pathophysiology of muscle spasms in athletes with disabilities, especially spinal cord injured athletes who have problems regarding their thermoregulatory function. 38 Adequate rehydration, supplementation of electrolytes (sodium), or stretching strategies during the competition period have been reported to be possible prevention strategies of exercise-associated muscle spasms. 39 It is known that injuries differ between sport disciplines and that injury surveillance during tournaments is essential to determine sport-specific risk factors for injuries. Since tournaments are just a small part of the athlete´s life, routine monitoring of injuries and health complaints throughout the athletic season is recommended. A more complex approach regarding the nonlinear interaction of risk factors ("web of determinants") and pattern recognition techniques should be implemented in future studies to improve prediction and prevention of injuries as proposed by Bittencourt et al (2016). 40 In conclusion, a higher injury rate was found compared to the rates reported from POG. A more detailed analysis of injuries characteristics revealed a high number of non-timeloss injuries, such as muscle spasms and skin laceration. Most injuries occurred at the neck, upper back, and shoulder. Eight minor time-loss injuries and no moderate or severe injuries were reported. High injury rates could be reduced targeting preventable injuries such as muscle spasms, strains, and sprains.