Effects of adding dual-task or sport-specific task constrains to jump-landing tests on biomechanical parameters related to injury risk factors in team sports: a systematic review

Introduction

Performance in team sports is a multifactorial phenomenon that depends on physical, coordinative, and cognitive factors, as athletes must adapt their movements and actions to various unpredictable and changing game situations, generated, among others, by opponents, teammates, or the ball (Abernethy, Thomas & Thomas, 1993). This involves performing motor actions such as jumps, landings, and changes of direction while simultaneously controlling and responding to multiple stimuli, therefore processing large amounts of information within a limited time frame (Hughes & Dai, 2023). Furthermore, on certain occasions, multiple motor actions must be executed concurrently  (Moreira et al., 2021; Evans & Stanovich, 2013). Consequently, the ability to allocate attention among the various perceptual demands arising from the environment and the action itself is crucial for successful decision-making (DM).

It is widely accepted that the information processing capacity of the central nervous system is limited, thereby restricting the ability to allocate attention to different inputs and the capacity to prepare for and perform multiple tasks simultaneously (Marois & Ivanoff, 2005). The term ’Dual-Task’ (DT) is the prevailing expression in scientific literature for denoting the concurrent execution of two or more tasks (Moreira et al., 2021; McIsaac, Lamberg & Muratori, 2015). This phenomenon has been studied from various research fields using different paradigms, such as executive control, allocation of resources, task prioritization, task switching, and task type  (McIsaac, Lamberg & Muratori, 2015; Mas, Naranjo & Mollá, 2023). Despite the diverse range of theoretical models, there is consensus that during DT, the neurophysiological process of allocating attention across different activities results in increased challenge, making the management and performance of one or both tasks more difficult (McIsaac, Lamberg & Muratori, 2015; Akin et al., 2021). The effect on task performance during DT can be quantified using various methods, often expressed as cost or the effect one task has on the other: DT Effect = [(dual − single)/single ⋅ ±100] (Moreira et al., 2021; McIsaac, Lamberg & Muratori, 2015). Various models of DT have traditionally been studied, including Motor-Motor, in which two coordinative or motor tasks are combined, and Motor-Cognitive, in which a motor task is combined with cognitive processing (McIsaac, Lamberg & Muratori, 2015; Mas, Naranjo & Mollá, 2023; Akin et al., 2021; O’Shea, Morris & Iansek, 2002). Furthermore, task constraints such DM which implies reacting to a stimulus or modulating the response based on the stimulus, also influence task performance and are dependent on factors such as time available for reaction, the number and type of stimuli, and the number and complexity of response options available (Hughes & Dai, 2023).

The assessment of physical performance is crucial and an integral part of both the sports and rehabilitation domains. Jump-landing tests are frequently used as a tool to assess lower limb muscle function due to their ease of implementation, high reliability, validity, and sensitivity (Meylan et al., 2009; García-Ramos et al., 2023). Jumping is a motor skill that demands significant coordination of both upper and lower extremities, making it fundamental in a wide variety of sports (Zhou et al., 2020; Rodríguez-Rosell et al., 2017). There is a wide range of jump tests, which are commonly classified as horizontal jumps (HJ) or vertical jumps (VJ) based on the force application vector. Some of the most frequently assessed HJ tests include the single-leg hop test (SLHT), double-leg hop test (DLHT), triple single-leg hop test (TSLHT), triple double-leg hop test (TDLHT), and the crossover hop test (CHT) (Meylan et al., 2009). On the other hand, among the commonly used VJ tests are the squat jump (SJ), countermovement jump (CMJ), Abalakov jump (AJ), and the drop jump (DJ) (Markovic et al., 2004; Sattler et al., 2012; Karatrantou et al., 2019). Jump tests, both VJ and HJ, are also regularly employed for the assessment of biomechanical variables associated with higher risk lower-limb injury patterns (e.g., increased knee valgus, increased internal rotation of the hip, and increased external rotation of the tibia) (Read et al., 2016; Redler et al., 2016; Munro & Herrington, 2011; Reid et al., 2007). Additionally, they are frequently used to evaluate athletes after lower limb injuries as part of the criteria for returning to sports practice  (Dutaillis et al., 2023). For instance, a shorter distance in HJ tests is associated with lower isokinetic knee extension torque (Järvelä et al., 2002) and a higher risk of re-injury (Müller et al., 2015). In triple hop tests, prolonged amortization indicates reduced capacity to absorb and regenerate ground reaction forces upon landing (Lloyd et al., 2020). Moreover, several studies have confirmed that evaluating landing during VJ provides valid and reliable indicators of anterior cruciate ligament (ACL) injury risk factors, such as knee valgus and excessive knee abduction torques (Hewett et al., 2005; Leppänen et al., 2017; Padua et al., 2009; Hewett & Myer, 2011). Neuromuscular control deficits can lead to excessive strain on passive ligamentous structures, exceeding their load-bearing capacity and increasing the risk of injury or compromising structural integrity (Read et al., 2016).

While bilateral jump-landing tests in a controlled and stable environment are commonly used to evaluate team athletes, the external validity of these tests in measuring lower limb muscle functionality in situations of greater specificity has been questioned (Rodríguez-Rosell et al., 2017; Currell & Jeukendrup, 2008). It is plausible that in a highly controlled environment devoid of uncertainty, such tests may not reveal any bilateral asymmetries or patterns indicative of injury risk. Yet, these factors might become evident when the tests are conducted under conditions with elevated cognitive or coordinative demands (Dutaillis et al., 2023; Millikan et al., 2019). The reduction of conscious attention directed towards the control of the primary task affects the integration of essential visual, auditory, and somatosensory stimuli necessary for neuromuscular control, coordination, and stability (Masters & Maxwell, 2008; Lohse, Sherwood & Healy, 2014; Lohse et al., 2014; Schnittjer et al., 2021). Thus, DM and divided attention could significantly influence some biomechanical variables related to a higher risk of ACL injury, such as reduced knee flexion at initial contact, increased knee valgus angles, greater knee extension, and valgus torques (Hughes & Dai, 2023).

The impact of cognitive factors on the occurrence of injuries related to team sports is increasingly being investigated (Shultz et al., 2019). Traditionally, more emphasis has been placed on anatomical, biomechanical, and hormonal factors in lower-limb injury prevention programs (Zamankhanpour et al., 2023). However, these programs do not incorporate key aspects of injury risk, such as reaction time, processing speed, and visual and verbal memory, which are highly relevant in team sports (Swanik et al., 2007; Wilkerson, 2012). Additionally, the physical tests commonly used as part of the criteria for returning to sports practice after an injury are often conducted with little to no cognitive load and low coordinative demand compared to the specific game actions (Millikan et al., 2019). Hence, new proposals for return-to-sport criteria suggest incorporating DT assessments and coordinative and cognitive constraints into the tasks to simulate the unpredictability and divided attention encountered in competition (Wilk et al., 2023; Grooms et al., 2023; Schweizer et al., 2022). 

Prior to this study, two systematic reviews published in 2014 (Almonroeder, Garcia & Kurt, 2015) and 2015 (Brown, Brughelli & Hume, 2014) investigated the effect of task anticipation (planned and unplanned) on lower-limb movement mechanics during single-limb changes of direction (sidestepping) in healthy individuals. Although these reviews were not focused on jump-landing tasks and were restricted to a specific type of task constraint, they suggest a possible link between task uncertainty and injury risk, as unplanned movements promoted knee mechanics that may increase the risk of injury (Brown, Brughelli & Hume, 2014; Ness et al., 2020). Additionally, another review in 2020 examined the evidence regarding the influence of dual-task (DT) conditions in individuals with anterior cruciate ligament (ACL) injuries (Ness et al., 2020). While the authors highlighted the usefulness of DT tests for assessing injured athletes, they also acknowledged the need for more sport-specific tests (such as cutting or jumping), as their review primarily covered studies using DT in static standing and walking tests. In recent years, several primary studies have been published on this topic. However, to date, no reviews have thoroughly analyzed the effect of incorporating DT during jumping or landing actions. Therefore, the objective of the present review is to examine the influence of performing a DT or sport-specific task constraints during jump-landing tests (vertical jump or horizontal jump) on biomechanical variables related to the risk of lower-limb injury in team sports.

Materials & Methods

Results

Selection of the studies

A total of 656 potentially relevant publications were identified as eligible from the selected databases. After screening the titles and abstracts, 610 (92.99%) were excluded for not meeting the inclusion criteria and 46 (7.01%) were selected for full-text review. One of the 46 potentially relevant publications (2.17%) could not be retrieved, therefore, 45 full-text articles were checked for eligibility by inclusion and exclusion criteria. After peer-review, 13 articles were included from database searching. Two additional articles were included via manually screened references from included articles and previous related systematic reviews. 15 studies were ultimately included for analysis (Lin et al., 2020; Kajiwara et al., 2019; Stephenson et al., 2018; Wilke et al., 2021; Almonroeder et al., 2018; Akbari, Kuwano & Shimokochi, 2023; Fílter et al., 2022; Imai et al., 2022; Fischer et al., 2021; DiCesare et al., 2020; Alanazi et al., 2020; Beardt et al., 2018; Ren et al., 2022; Ford et al., 2017; Richwalski et al., 2018) (Fig. 1). All the selected studies focus on jump-landing tests in team sports and biomechanics related to increased lower-limb injury risk.

Discussion

The present article addresses a review of the effects of incorporating DT or sport-specific task constrains during the jumping or landing action, both coordinative (motor-motor), cognitive (motor-cognitive) or conditioned by DM, on the biomechanical variables linked to lower limb injury risk factors in team sports. The reviewed studies demonstrate that the inclusion of coordinative DT, cognitive DT, or tasks involving DM in jump or landing actions can significantly alter various biomechanical variables such as joint torques, as well as variations in stability and postural control, which are directly related to the risk of lower limb injuries in team sports. This section has been organized based on the classification of the dual-task (DT) constrains presented in the materials & methods section. Thus, in the following paragraphs, the effects of coordinative DT, cognitive DT, or tasks involving DM on jump-landing tests are analysed.

Regarding studies that evaluated the effects of introducing constraints or coordinative DT (motor-motor) (Almonroeder et al., 2018; Akbari, Kuwano & Shimokochi, 2023; Fílter et al., 2022; DiCesare et al., 2020; Alanazi et al., 2020; Beardt et al., 2018; Ford et al., 2017), significant changes in biomechanical parameters related to injury risk were recorded in all of them. From a kinetic perspective, increases in peak vertical ground reaction forces (vGRF) during landing (Almonroeder et al., 2018), rate of force development, and mechanical impulse (Fílter et al., 2022), along with increases in peak knee abduction torque (Almonroeder et al., 2018), were reported. From a kinematic perspective, a decrease in hip, knee, and ankle flexion, hip abduction, and ankle inversion at initial contact was observed (Akbari, Kuwano & Shimokochi, 2023; DiCesare et al., 2020; Alanazi et al., 2020; Beardt et al., 2018), along with an increase in temporal differences in initial contact between both feet  (Beardt et al., 2018). Additionally, the peak knee flexion angle (Almonroeder et al., 2018; Akbari, Kuwano & Shimokochi, 2023), vertical displacement of the center of mass (Akbari, Kuwano & Shimokochi, 2023) and contact times with the platform (Almonroeder, Garcia & Kurt, 2015; Fílter et al., 2022; Beardt et al., 2018) decreased. This suggests that the inclusion of a constraint or coordinative difficulty tends to increase the stiffness of jump-landing. Unlike tests with cognitive DT (motor-cognitive), coordinative DT (motor-motor) were generally more specific to the sports modality; some studies even used virtual reality to simulate action in a real sports context (DiCesare et al., 2020; Ford et al., 2017). A large number of studies that have introduced a coordinative (motor-motor) DT have included a sport-specific mobile in the task, for example heading (Akbari, Kuwano & Shimokochi, 2023; Fílter et al., 2022; DiCesare et al., 2020; Alanazi et al., 2020); grabbing (Almonroeder et al., 2018; DiCesare et al., 2020; Ford et al., 2017); or spiking a ball (Beardt et al., 2018).

Despite the heterogeneity regarding cognitive dual-tasking and assessment systems employed across studies, all included studies that utilized cognitive DT during jump and landing tasks (motor-cognitive) reported significant changes in some key biomechanical variables related to injury risk factors (Kajiwara et al., 2019; Wilke et al., 2021; Imai et al., 2022; Fischer et al., 2021; Ren et al., 2022; Ford et al., 2017). Specifically, an increase in peak vGRF during landing was observed (Imai et al., 2022), as well as increases in TTS and Stability Indexes in different axes (Ren et al., 2022), and an increase in the COP (Wilke et al., 2021; Ren et al., 2022). Additionally, an increase in hip extension  (Imai et al., 2022; Ford et al., 2017), knee abduction (Imai et al., 2022), and ankle plantar flexion torques were observed (Imai et al., 2022). Kinematically, increases in peak tibial internal rotation angle were described (Kajiwara et al., 2019), along with a reduction in peak knee flexion angle (Fischer et al., 2021) and trunk flexion angle (Ford et al., 2017). Regarding the difficulty level of the cognitive task employed, it is important to note that only one of the included articles studied different levels of complexity in this task (Wilke et al., 2021). This could be considered a relevant factor to consider, as previous studies have suggested that postural control in easy cognitive tasks tends to increase due to a shift in attention focus, while more demanding tasks tend to impair postural control, showing an inverted “U” relationship between task difficulty and control (Huxhold et al., 2006). As for the type of cognitive tasks proposed in the included studies, they are generally nonspecific (see Table 4) with respect to the type of tasks, level of unpredictability, and cognitive demands that can be encountered in team sports practice  (Fuster, Caparrós & Capdevila, 2021; Gonçalves et al., 2016). The most specific tasks within the included studies are those related to identification and memorization of opponents’ numbers  (Wilke et al., 2021), and visual tracking of spheres interacting in a three-dimensional space (Ren et al., 2022). In future research in this area, it would be interesting to incorporate tasks that more accurately reflect the cognitive demands inherent in team sports or based on the player’s position and role  (Viñas et al., 2023). Additionally, proposing variations in the difficulty levels of these tasks to adapt them to different contexts and needs would be beneficial.

Another factor to consider in the review is the effect that DM has on the biomechanical changes of the lower limb, when an athlete chooses between different motor responses based on external stimuli and the evaluation of information in a limited time. Of the studies included in this review, there are seven (Lin et al., 2020; Kajiwara et al., 2019; Stephenson et al., 2018; Almonroeder et al., 2018; Fischer et al., 2021; Beardt et al., 2018; Richwalski et al., 2018) that have specifically analyzed the effect of including DM (conditioned discriminative response with reaction time) in comparison to the same action without DM as a control. Overall, significant changes in various biomechanical variables of the lower limbs were observed in all studies due to the inclusion of DM. Specifically, several studies found lower peak angles of hip flexion (Beardt et al., 2018), and knee flexion (Lin et al., 2020; Almonroeder et al., 2018; Fischer et al., 2021; Beardt et al., 2018), along with increased knee abduction (Almonroeder, Garcia & Kurt, 2015) and tibial internal rotation (Kajiwara et al., 2019). Additionally, increases in vGRF were also reported (Almonroeder, Garcia & Kurt, 2015; Lin et al., 2020). In contrast to these results, one of the included studies showed an increase in knee internal rotation torques and peak vGRF in the control condition compared to the DM condition when the task was pivoting on the dominant limb upon landing (Richwalski et al., 2018). The results of this study do not align with previous studies that have investigated the effects of anticipated vs. unanticipated actions (Brown, Brughelli & Hume, 2014; Brown, Palmieri-Smith & McLean, 2009; Borotikar et al., 2008; Houck, Duncan & Kenneth, 2006). The authors of the study suggested that the observed differences may be due to the movement following landing that required the participant to move 90° laterally, whereas other studies look at anterior translational movement (e.g., running, long jump) that were controlled prior to landing (Akbari, Kuwano & Shimokochi, 2023; Fílter et al., 2022; Richwalski et al., 2018). In any case, it is shown that including a DM condition produced significant changes in joint mechanics, altering the preplanned movement pattern. Lastly, only one of the included studies evaluated the effect of modifying the anticipation time of the motor response to be performed (Stephenson et al., 2018). They found that the kinematic and kinetic parameters of the knee upon landing vary depending on whether the response is anticipated or not, as long as there is enough time for the reaction to be adequate and evoke the preplanned movement pattern (probable threshold at 300 ms). It is noteworthy that the type of stimulus and the possible motor responses in most of the DM studies included are non-specific to the sport (Table 4), with the exception of the study by  Beardt et al. (2018), in which the specific actions to be performed by the participants were conditioned by an opponent.

The systematic review conducted has allowed for a detailed exploration of different studies on the biomechanical performance of the lower extremities in contexts of DT or jump-landing tasks conditioned by added constraints. Focusing on performing a secondary task correctly potentially diverts attentional resources, adversely affecting the coordinative motor control necessary for optimal execution of the primary task (McIsaac, Lamberg & Muratori, 2015). This phenomenon resembles real-life situations in sports, where attention is divided among multiple simultaneous stimuli, such as making accurate passes, eluding opponents, or performing decoy movements; and where attention is focused precisely on performing the secondary task during jump-landing (passing, receiving, decision making) as correctly as possible, without errors. In this context, a recent review highlights measuring errors in the secondary task as a critical factor that could significantly influence performance and biomechanical execution of the primary task in DT conditioned tests  (Chaaban, Turner & Padua, 2023). However, it is notable that of the fifteen studies included in this review, only four have specifically addressed trial-error measurement or performance in the secondary task, through recording the number of attempts (Almonroeder et al., 2018), errors, time, reaction times, and accuracy scores (Fischer et al., 2021; Stephenson et al., 2018), as well as landing errors and memory errors (Wilke et al., 2021). The absence of recording these errors in most of the reviewed studies may introduce bias in the results, as they focus exclusively on evaluating the primary task without considering the efficiency in managing simultaneous tasks. This omission not only deviates from real competition conditions but also may lead to biased conclusions regarding the effect of such tasks on biomechanical variables and injury risk factors in team sports, emphasizing the importance of incorporating these parameters in future research.

Another notable aspect of the systematic review is the remarkable shortage of comparative analysis between lower limbs in the examined literature. Most studies included in this review gather biomechanical data solely from one limb in unilateral jumps performed with the subjectively dominant limb (Lin et al., 2020; Kajiwara et al., 2019; Stephenson et al., 2018; Akbari, Kuwano & Shimokochi, 2023; Richwalski et al., 2018) or non-dominant limb (Almonroeder et al., 2018), or in bilateral jumps (Wilke et al., 2021; Fílter et al., 2022; DiCesare et al., 2020; Alanazi et al., 2020; Ren et al., 2022; Ford et al., 2017). In two of the studies included in the review, the effects are measured biomechanically in both limbs, but no statistical comparisons are made between them (Fischer et al., 2021; Beardt et al., 2018). Only one study conducts the analysis of different variables between the dominant and non-dominant limb (Imai et al., 2022). The scarcity of direct comparisons significantly limits our understanding of biomechanical symmetry between limbs during task execution in DT conditions, crucial for identifying potential imbalances that may increase injury risk. This review underscores the need to incorporate comparative analyses between the dominant and non-dominant limb, as well as between previously injured and uninjured limbs, in future research to advance toward a more precise evaluation of biomechanical function in sports and clinical contexts.

The clinical implication derived from the observed biomechanical changes in the reviewed studies underscores a direct relationship between the performance of DT and the increased risk of lower extremity injuries in team sports. These studies focus their analysis on kinetic and kinematic alterations during DT in jump-landing tests, identifying variables associated with the risk of anterior cruciate ligament (ACL) injury as the primary focus of attention. It has been observed that team sports athletes exhibiting significant knee abduction torque during landing have a heightened risk of ACL injuries (Hewett & Myer, 2011). Evidence suggests that a decrease in trunk neuromuscular control contributes to an increase in knee abduction torque, thereby elevating the risk of ACL injury (Hewett & Myer, 2011). Furthermore, it is highlighted that a reduction in knee flexion leads to an increase in the angle between the patellar tendon and the tibial axis, generating greater anterior tibial loading and consequently a higher risk of ACL injury (Norcross et al., 2013). When knee flexion angle is insufficient, the ability of the flexor musculature to provide posterior force to the tibia is compromised, increasing shear force on the anterior tibia (Shimokochi & Shultz, 2008). In relation to the above, elevated vGRF, coupled with reduced knee and hip flexion, indicates a relatively rigid landing pattern, which may increase forces acting on the ACL. This could lead to joint instability, especially when neuromuscular control is deficient (Lin et al., 2020). Therefore, the collective results of the various studies included in the review suggest a biomechanical profile of the lower extremity with an increased risk of injury in team sports.

Based on the combined results of the studies included in the review, it is evident that conducting standard bilateral jump-landing tests under controlled and stable conditions to assess team sports athletes entails deficiencies in external validity to understand what occurs in situations of greater specificity or competition. Such assessment tests appear not to reveal biomechanical patterns indicative of high injury risk, which may emerge when tests are performed under high cognitive, coordinative, or temporally conditioned demands and/or DM. Therefore, the evaluation tests used to determine the risk of injury or recurrence, or to support the return to play in team sports, should incorporate specific cognitive, coordinative, or DM elements that simulate the specific demands inherent in team sports.

Finally, this systematic review is not without limitations. Numerous publications exist regarding the execution of motor tasks with varying levels of complexity, either in isolation or in combination with other tasks. A comprehensive search strategy, detailed in the materials and methods section, was implemented. However, it is possible that some relevant studies addressing the aims of this review may exist, albeit without explicit reference to the specific terms employed in the search strategy. A notable challenge pertains to the diverse approaches and paradigms applied in the interpretation and examination of DT conditions. Notably, discrepancies arise in defining what constitutes a DT, with some definitions or authors categorizing certain scenarios as such, while others characterize them as singular tasks influenced by overlaid constraints. Consequently, not all studies incorporated into this review may align with every conceptual definition of DT. Another aspect to consider in this systematic review is that the majority of the research on this topic has been conducted with female athletes. We believe this might be due to the higher prevalence of non-contact ACL injuries in female team sports athletes compared to males (Chia et al., 2022), which may have driven research interest in this population. The overrepresentation of female participants in the included studies in this systematic review may constrain the extrapolation of the findings to male athletes.

Conclusions

This systematic review contributes to the existing body of knowledge by specifically focusing on the effects of incorporating DT during jump-landing tests on biomechanical variables related to lower-limb injury risk in team sports. The inclusion of DT or constraints in jump-landing tests significantly alters biomechanical variables related to lower extremity injury risk in team sports. Generally, in the included studies, a stiffer landing at the joint level was observed when cognitive, coordinative, or DM tasks were included, resulting in increased vGRF and peak joint torques, a decrease in peak knee and hip flexion angles, and overall reduced stability and postural control. These biomechanical changes suggest a higher risk of injury when performing the task under such conditions.

Regarding the types of tasks used in the studies included in the present review, coordinative tasks generally involved incorporating equipment from the sports discipline into jump-landing or using simulated situations with virtual reality, thus having a certain specificity with respect to the sport. In contrast, most studies that introduced cognitive or DM aspects into the task used stimuli and constraints that were highly nonspecific to the sport. Additionally, the majority of studies focused on bilateral tests or the specific evaluation of one of the two limbs, without conducting comparative analyses between limbs (dominant vs. non-dominant or injured vs. uninjured). Lastly, few studies have focused on recording errors in secondary tasks, which could bias the results obtained with the inclusion of the task or constraint.

Practical Applications

Looking ahead, research in the field of sports science should focus on implementing studies that more faithfully reflect the complex demands inherent in real sports practice in team sports. This implies, first and foremost, designing and utilizing tasks and constraints that specifically emulate sporting situations, with particular emphasis on tasks related to DM and cognitive processes. Additionally, it is imperative to analyze the effects of including DT and constraints on both limbs, conducting detailed comparisons between the dominant and non-dominant or injured limbs with respect to the uninjured limb. This approach would allow for a deeper understanding of motor and cognitive dynamics in sports, especially regarding injury prevention and recurrence reduction. Furthermore, it is crucial to record errors in secondary tasks to control potential biases and ensure the validity and applicability of the results obtained.

The findings of research on DT and constraints in jump-landing tasks have direct application in the professional practice of physical trainers and sports rehabilitators. The use of classic jump-landing tests, both in screening for injury risk factors and in assessment batteries for return to play, has shown significant limitations by not adequately simulating the specific demands of competition. These limitations result in an incomplete representation of the coordinative and motor patterns that, under real competition conditions, pose a higher risk of injury. Therefore, there is a need to develop and validate specific tests, adapted to each sport and situation, that faithfully reproduce the characteristics and demands of sports practice. The implementation of these tests, which should include different levels of difficulty, will facilitate a more precise evaluation of the athlete’s functional status, promoting more effective and safer intervention and rehabilitation strategies aligned with the true demands and levels of unpredictability of team sports.

Supplemental Information