Handball is an Olympic team sport played worldwide and is especially popular in Europe (1). It is played in 199 countries and has approximately 19 million registered players (2). It is a team sport with high physical demands, and body contact actions are an integral part of the game (3, 4). These body contacts mainly happen during throwing attempts by offensive players and tackles by defending players (5). To interrupt the flow of play, body actions against the opponent are highly important (6, 7). Thus, it comes as no surprise that the International Olympic Committee ranks handball as one of the Olympic sports with the highest injury rates (8, 9), with adolescent handball players ranging from 8.3 injuries/1,000 match hours in males to 14.5 injuries/1,000 match hours in female U17 players (3). Møller et al. (4) reported an injury incidence of 11.5 in male and 10.8 in female U16 players, as well as 13.0 in female and 17.2 in male U18 players per 1,000 match hours. Asai et al. (10) even reported a higher number of match injuries in 13- to 14-year-old players, with 26.5 injuries/1,000 match hours, further differentiating between boys (32.7 injuries/1,000 match hours) and girls (20.1 injuries / 1,000 match hours). However, it must be kept in mind that different injury definitions exist (6, 9), which makes comparisons between studies difficult.

Injuries have adverse effects, as injured players cannot develop individually and cannot contribute to their team's success (11). In addition, the subsequent costs of injuries are a high economic burden (12), with professional handball players reporting mean costs of 1,100 € per injury, and injuries in amateur players often resulting in a temporary inability to work (13). Therefore, preventing injuries in youth handball is a key priority for practitioners and scientists alike (11, 14).

In conjunction with the high injury rates due to the game’s physical nature, handball is also characterized by unilateral movements. Throwing and passing techniques are predominantly executed with one side of the body, i.e., the throwing arm (15). This functional specialization may lead to distinct adaptations, such as differences between the throwing and non-throwing arm in terms of stability, mobility, or throwing velocity (15), but also motor control, performance outcome, and skill acquisition (16). These diverging adaptations may result from the asymmetrical distribution of load and the associated neuromuscular stimuli. In this context, exceeding a specific cut-off value between the throwing and non-throwing arm in terms of reach differences, i.e., upper extremity asymmetry, may pose a potential risk factor for injuries (17, 18). In addition, pronounced asymmetries have also been reported as having detrimental effects on physical performance (19). Preseason assessments may therefore be reasonable to identify injury risk factors (14). One suitable and highly reliable test that has the potential to assess upper extremity mobility and stability, as well as asymmetries, is the upper quarter Y-balance test (YBT-UQ) (20, 21). Different authors have already examined whether preseason YBT-UQ performance is related to injuries during the competitive season. Teyhen et al. (18) investigated the risk of suffering a time-loss injury in 922 US Army soldiers/warrior athletes (age: 24.7 ± 5.2 years). A maximum superolateral (SL) reach distance of ≤80.1% arm length (AL) as well as an inferolateral (IL) reach asymmetry of ≥7.75% AL was found to increase the likelihood of a future time-loss injury. Further, specific cut-off values for military personnel (N = 494, m = 454, f = 40; age: 28.6 ± 6.8 years) were reported by Campbell et al. (22). In this study, the authors identified an SL reach distance of ≤57.75 cm and a composite score (CS) of less than ≤81.1% AL in the YBT-UQ as risk factors for suffering an upper quadrant musculoskeletal injury.

In addition, Cosio-Lima et al. (23) assessed the association between YBT-UQ performance and physical training-related injuries in male Coast Guard Maritime Security Response Team candidates (N = 31, age: 24 ± 4 years). The odds ratio (OR) to suffer an injury was 5.4 in the case of a CS of 81.8–89.3% AL, and an OR = 3.6 in the case of a CS of 89.0–99.0% AL for the left arm. For the right arm, an OR = 5.4 was present in the case of a CS 77–88.9% AL, and an OR = 3.6 in the case of a CS of 89.0–102.7% AL. The authors concluded that lower scores in the YBT-UQ were associated with an increased injury risk. In the most similar group to handball players, Bennett et al. (24) assessed preseason YBT-UQ performance in 257 elite adolescent male Australian Football players (age: 17.1 ± 0.8 years) to determine the relationship with in-season injuries. The authors did not detect significant associations between prospective upper quarter body injury risk and the YBT-UQ values.

Drawing upon these findings, the YBT-UQ, especially the SL reach direction and a high IL reach asymmetry, may be associated with future time-loss injuries in general. Because studies assessing the injury risk in adolescent handball players based on upper extremity mobility and stability performance are lacking, the present study aims to investigate whether preseason YBT-UQ performance is associated with the occurrence of sport-related injuries during the competitive season in this group. Our original contribution to the field was to determine whether a relationship exist between YBT-UQ values and handball-specific injuries. In this context, it will be assessed whether surpassing the cut-off values in terms of asymmetry or YBT-UQ scores that were reported in other cohorts (18, 22) also poses a risk [increased hazard ratio (HR)] in adolescent handball players. Confirming predefined YBT-UQ cut-off values that increase the likelihood of injuries may help practitioners develop intervention programs that specifically target these weaknesses in terms of total YBT-UQ scores and asymmetries. We hypothesized that injured players would show significantly worse results in the YBT-UQ and a significantly increased HR when (a) the SL reach distance was ≤80.1% AL, (b) the SL reach distance was ≤57.75 cm, (c) the CS was ≤81.1% AL, or (d) the IL reach asymmetry exceeded ≥7.75% AL.

Of the 133 players assessed, all players were tracked over the course of the entire season. There were no dropouts, and only one player changed clubs during the season. However, data on this player were still accessible, as he changed from one club participating in the study to another. Overall, the coaches’ response rate throughout the season was 100%, indicating that all data and weekly reports were provided to the examiner. No significant differences in preseason YBT-UQ performance were found between injured and non-injured players for the throwing and the non-throwing arm reach (Table 5).

Fifty-seven players (43%) incurred a sport-related injury during the competitive season, of which 27 (47%) were upper body injuries and 30 (53%) lower body injuries (Table 6). In terms of asymmetry scores (Table 5), no significant differences were present in either reach score (MD: p = 0.898; IL: p = 0.854; SL: p = 0.312; CS: p = 0.751). In total, 14 out of the 57 injuries occurred during matches, while the remaining 43 occurred during training sessions. Based on the total match hours of all assessed players (3,134 match hours), this corresponds to a match injury rate of 4.5 injuries/1,000 match hours.

The results of the Cox proportional hazard survival regression models are shown in Tables 2–4. The analyses revealed that only the presence of an IL with an asymmetry score ≥7.75% AL was associated with a moderate increase in the risk [HR = 2.18, 95% confidence interval (CI) = 1.02–4.68, p = 0.045] of lower body injuries (i.e., hip, leg, and foot) (Table 4).

The study aims to investigate if preseason YBT-UQ performance is associated with sport-related injuries during the competitive season in adolescent handball players. The main results can be summarized as follows: (a) no significant differences in preseason YBT-UQ scores were found between injured and non-injured players both for the throwing as well as the non-throwing arm; (b) YBT-UQ distance and asymmetry cut-off values (except IL reach asymmetry) could not be confirmed as factors for an increased risk of sport-related injuries in the present target group.

Our results align with those of Bennett et al. (24), who also did not find significant associations between upper quarter body injury risk and the modified YBT-UQ values. However, our findings contradict those of Cosio-Lima et al. (23), Teyhen et al. (18), and Campbell et al. (22), who reported an increased risk of injuries when specific YBT-UQ cut-off values were surpassed. In addition, we did not find significant differences in our sub-analyses which differentiated between upper and lower body injuries. In contrast to Campbell et al. (22), we did not find a significant relationship between the YBT-UQ cut-off values and injury incidence in the upper body. However, Campbell et al. (22) assessed adult (28.6 ± 6.8 years) military personnel who, contrary to adolescent handball players, might be more susceptible to the reported cut-off values, as the duration of the exposure to potential risk factors in terms of years and overall load is much higher in a military cohort compared with an adolescent group of handball players.

Several reasons may explain the lack of relationships between YBT-UQ performance and sport-related injuries. The study by Campbell et al. (22) used a different assessment technique than the present study, as in their study only upper quadrant musculoskeletal injuries were assessed. Therefore, the proposed risk factors of age, sex, and cohort may be subject-related moderator variables. However, based on our sub-analysis on upper body injuries, we could not confirm the results of Campbell et al. (22), as our results did not show any associations between upper body injuries and YBT-UQ results. Further, adolescent handball players have had lower exposure to the specific demands of handball training (i.e., overall training load) compared with senior players (36). Therefore, they may not be as susceptible to the proposed risk factors as an adult military/combat cohort. Contrary to our expectations, however, the Cox proportional hazard survival regression model revealed a relationship between IL reach asymmetry and sport-related lower body injuries. This finding is in line with those of Ruffe et al. (36), who found adolescent male runners with a YBT-UQ SL reach distance of ≥4.0 cm seven times more likely to incur an injury to the lower body. The findings of Ruffe et al. (36) and our findings may be explained by the fact that the YBT-UQ also demands trunk mobility/stability. Limitations in reach directions or asymmetries may lead to decreases in neuromuscular control for the trunk and the lower extremities. Therefore, our results strengthen the notion that injuries are multifactorial (38). Consequently, a multifactorial assessment in the sense of a complex system approach (29) may be better suited as an injury risk screening tool. Overall, the YBT-UQ in isolation may not be sensitive enough to differentiate between low- and high-risk groups to suffer an injury throughout the competitive season. As an additional reason for the lack of relationships between the YBT-UQ results and in-season injury occurrence, the performance of the present sample in terms of YBT-UQ norm values is only slightly above the reported reference values (39). The difference in total performance of the YBT-UQ between those at high risk and those at low risk remained relatively small, making significant results less likely. Further, the YBT-UQ, which assesses upper extremity mobility and stability at the end range of motion, remains a relatively unspecific test concerning the demands placed on the athletes during training and competition. This is especially true as the subjects have full control over their bodies with no perturbations during the YBT-UQ, while during training and games, balance is often challenged by body contact. It is well documented that body contact actions are an essential cause of injuries in handball (9) due to exposure to external forces (5). In addition, based on the notion that injuries are multicausal (40) and due to the possible interaction of many factors (38), weaknesses in one system (e.g., upper extremity mobility and stability as assessed by the YBT-UQ) may be compensated for by other systems.

The match injury rate of 4.5 injuries/1,000 match hours found in the present study is less than those reported by Olsen et al. (3), who found a match injury rate of 8.3 injuries/1,000 match hours in male and 14.5 injuries/1,000 match hours in female U17 players. It is also lower than the results of Møller et al. (4), who found a higher injury incidence in adolescent male and female players, with an incidence of 10.8 in female and 11.5 in male U16 players, as well as 13.0 in female and 17.2 in male U18 players, all per 1,000 match hours. Asai et al. (10) even reported higher incidence rates of 32.7 injuries/1,000 match hours in boys and 20.1 injuries/1,000 match hours in girls (age: 13–14 years). The low incidence of 4.5 injuries per 1,000 match hours in our study may be related to the rather young age (15.7 ± 1.7 years) of this adolescent cohort. A lower injury incidence in this young age group is in line with different studies (3, 4) and a review by Vila et al. (9) that also reported fewer injuries in younger players. Importantly, no conclusions in terms of causation can be drawn from the present results.

We investigated whether preseason YBT-UQ performance was associated with sport-related injuries during the following competitive season in adolescent handball players. Our results indicate that neither the YBT-UQ reach distances of the throwing or non-throwing arm nor the asymmetry cut-off values (except IL reach asymmetry) were associated with the occurrence of a time-loss injury. Thus, the YBT-UQ seems to have limited benefit as a field-based screening tool for assessment of injury risk in adolescent handball players. Future studies may investigate if the YBT-UQ could be a useful tool when added to a multifactorial test battery. In addition, it would be desirable to include injury mechanisms in terms of contact vs. non-contact, acute vs. chronic, and timing during the games. Moreover, future studies should assess whether the limited association between YBT-UQ results and handball-specific injuries that was demonstrated in our study also applies to other groups of players, i.e., even younger or adult players.