Team handball is a popular sport worldwide, and according to the International Handball Federation (IHF), over 30 million athletes in 183 countries currently play the sport. Although the literature examining injury prevalence in team handball players is lacking, there is evidence to support the occurrence of upper extremity injury in this population of athletes (8,14,23). Seil et al. (23) identified the shoulder as the most frequent site to experience overuse symptoms over the course of a year. Shoulder pain in elite German team handball players has been reported to account for 40% of time lost injuries over a 6-month period (8,14). In addition, a recent survey of Norwegian handball players revealed that 36% of team handball players had current shoulder pain (14). These studies indicate shoulder pain is prevalent in team handball players, and this is likely due to the demands placed on the shoulder during throwing. Previous reports have indicated that team handball players make approximately 48,000 throws in a season (1,11). Thus, it is not surprising that shoulder overuse injuries are prevalent due to the number of throws reported.
With the prevalence of upper extremity injury in team handball players, the need to develop effective rehabilitation protocols and return-to-play criteria becomes vital to the care of the athlete. In baseball and softball, it is common following an injury to initiate an interval throwing protocol as the athlete progresses through the rehabilitation program (2–4,6,24,31). The interval throwing protocol is performed during rehabilitation to progressively prepare the athlete for the demands of throwing so that they can successfully return to play (24). Typical interval throwing protocols begin with the athlete throwing from 30 feet or less and gradually increasing the distance up to approximately 180 feet while also incorporating different levels of perceived effort during throwing (24). Although these protocols give general guidelines for clinicians to follow, they are vague on when to increase the effort or distance. Clinicians may choose to focus on effort initially in the early stages of return to throwing. However, clinicians may choose to begin with distance because it is an objective measure, whereas effort is subjective and is difficult to quantify. The integration of an interval throwing protocol is common; however, there is currently no information regarding return-to-play throwing protocols in team handball. One of the first steps in establishing an interval throwing protocol specific to team handball is to gain an understanding of the ability of team handball players to throw at different levels of perceived effort. In a sample of healthy baseball pitchers, Fleisig et al. (7) evaluated the ability of the pitchers to throw at 50 and 75% effort. The results indicated that the velocities were significantly greater than what was suggested by the investigators. Although velocity is just one potential indicator of an athlete's ability to throw at a selected level of perceived effort, examining the kinematics during the throwing motion across trials may be a better indicator of the effectiveness and the demands of throwing at a certain percentage of perceived effort.
From a clinical prospective, concerns exist in team handball about how to effectively implement a return to throwing protocol in players with shoulder injuries due to the differences in ball mass between a team handball (14.99–16.75 oz) and the relevant literature that presents data with a baseball (5.11 oz). With a handball being approximately 3 times the weight of a baseball, the first step in the developmental process of an interval throwing protocol is to examine the kinematics at different levels of perceived effort, at a common distance that these players throw. The set shot involves a stable base of support and contact with the floor and would be implicated early in a return to throwing protocol to ensure that the player is focused on using proper shot mechanics before progressing to the more complex jump shot, which is performed in the air. Early implementation of proper set shot mechanics would allow clinicians to work from a standard distance with the focus being on the perceived effort of players.
Therefore, the purpose of this study was to examine the set shot in team handball players at 50, 75, and 100% perceived effort to determine whether differences in the kinematics exist between these effort levels. The goal of this investigation is to use these data to help sports medicine clinicians assess whether players can successfully alter the mechanics of the set shot at less than maximal effort to assist in the development of a sport-specific interval throwing protocol for team handball players. It was hypothesized that the kinematics of the set shot would be similar with the exception of the segmental angular accelerations, which would be significantly greater for the maximum effort shots.
Experimental Approach to the Problem
We used an exploratory 3-group comparison design to examine whether set shot kinematics in team handball players were different at 50, 75, and 100% perceived effort levels. Each subject performed 2 set shots at each effort level, and the effort levels were randomized for each subject.
A within subjects repeated-measures analysis of variance (ANOVA) was performed to determine whether set shot kinematics and kinetics were significantly different at each of the perceived effort shots. If significant differences were observed, further post hoc repeated-measures ANOVAs were performed for each variable by throwing event. Pairwise comparisons were then analyzed to determine which effort levels were significantly different.
Eleven male team handball players (23.09 ± 3.05 years; 185.12 ± 8.33 cm; 89.65 ± 12.17 kg) volunteered to participate in this study. The subjects were all members of the same National Team training program that trained for 2.5 h·d−1 for 5 d·wk−1. This sample included 5 backcourt players (25.2 ± 2.8 years; 188.1 ± 8.1 cm; 92.8 ± 9.0 kg; 4.8 ± 2.8 years of experience), 3 wing players (21.5 ± 2.1 years; 179.1 ± 5.9 cm; 78.5 ± 7.0 kg; 2.3 ± 1.9 years of experience), and 2 pivot players (21.0 ± 0 years; 189.7 ± 9.8 cm; 104.1 ± 6.9 kg; 1.8 ± 1.1 years of experience). The Auburn University Institutional Review Board approved all testing protocols. Before data collection, all testing procedures were explained to each subject, and they were informed of the benefits and risks of the investigation before signing an institutionally approved informed consent document to participate in the study.
The MotionMonitor (Innovative Sports Training, Chicago, IL, USA) synced with electromagnetic tracking system (Track Star; Ascension Technologies, Inc., Burlington, VT, USA) was used to collect data. The electromagnetic tracking system has been validated for tracking humeral movements, producing trial-by trial-interclass correlation coefficients for axial humeral rotation during humeral elevation in the scapular plane in both loaded and nonloaded conditions in excess of 0.96 (12). With electromagnetic tracking systems, field distortion has been shown to be the cause of error in excess of 5° at a distance of 2 m from an extended range transmitter (5), but increases in instrumental sensitivity have reduced this error to near 10° before system calibration and 2° after system calibration (5,13,20). Thus before data collection, the current system was calibrated using previously established techniques (5,10,15,16,18–22). After calibration, magnitude of error in determining the position and orientation of the electromagnetic sensors within the calibrated world axes system was less than 0.01 m and 3°, respectively. The collection rate for all kinematic data describing the position and orientation of electromagnetic sensors was set at 100 Hz (15,17,19,22,30). Raw data were independently filtered along each global axis using a fourth order Butterworth filter with a cutoff frequency of 13.4 Hz (15,17,19,22,30). Force plate data were sampled at a rate of 1,000 Hz.
Subjects had a series of 11 electromagnetic sensors (Track Star; Ascension Technologies, Inc.) attached at the following locations: (1) first thoracic vertebra (T1) spinous process; (2) pelvis at sacral vertebrae 1 (S1); (3) deltoid tuberosity of the throwing arm humerus; (4) throwing arm wrist, between the radial and ulnar styloid processes; (5) acromioclavicular joint of the throwing arm; (6) third metacarpal of the throwing hand; (7–8) bilateral shank centered between the head of the fibula and lateral malleolus; (9–10) bilateral lateral aspect of the femur (15,17,18,32); and (11) third metatarsal of the stride leg foot. Student researchers, who were trained in the application techniques, applied the sensors. Sensors were affixed to the skin using PowerFlex cohesive tape (Andover Healthcare, Inc., Salisbury, MA) to ensure the sensors remained secure throughout testing. Following the application of the sensors, an additional sensor was attached to a stylus and used for digitization following previously established guidelines (15,17,18,32). Subjects stood in anatomical position during digitization to guarantee accurate bony landmark identification. The medial and lateral aspect of each joint was digitized, and the midpoint of the 2 points was calculated to determine the joint center (15,17,18,21,22,32). A link segment model was developed through digitization of joint centers for the ankle, knee, hip, shoulder, thoracic vertebrae 12 (T12) to lumbar vertebrae 1 (L1), and C7 to thoracic vertebrae 1 (T1). The spinal column was defined as the digitized space between the associated spinous processes, whereas the ankle and knee were defined as the midpoints of the digitized medial and lateral malleoli, medial and lateral femoral condyles, respectively. The shoulder and hip joint centers were estimated using the rotation method. This method of calculating a joint center has been reported as providing accurate positional data (9,25). The shoulder joint center was calculated from the rotation between the humerus relative to the scapula, and the hip joint center was from the rotation of the femur relative to the pelvis. The rotation method was implemented with the joint stabilized and then passively moved in 10 positions in a small circular pattern (9). The variation in the measurement of the joint center had to have a root mean square error of less than 0.003 m to be accepted.
Raw data regarding sensor orientation and position were transformed to locally based coordinate systems for each of the respective body segments. Two points described the longitudinal axis of each segment, and the third point defined the plane of the segment (15). The second axis was perpendicular to the plane, and the third axis was defined as perpendicular to the first and second axes. The world axis was defined as the y-axis in the vertical direction, horizontal and to the right of y was the x-axis, and posterior was the z-axis (15,17–19,22). Euler angle decomposition sequences were used to describe both the position and orientation of the body segments (15,17–19,22).
After digitization, subjects were allotted an unlimited time to warm up and become familiar with the testing protocols. A standardized warm-up was not used because the investigators wanted to ensure that each subject felt that they were properly ready to perform the requested shots. Once a subject deemed himself ready, the testing protocols began. Subjects performed 2 maximal perceived effort set shots, 2 set shots at 75% perceived effort, and 2 shots at 50% perceived effort from a distance of 8 m, using an IHF size 3 team handball (26–28). The number of set shots at each level of perceived effort was arbitrarily chosen. The order in which each subject completed each shot was randomized. Only accurate shots that hit the center of a 1 × 1 m2 target at a height of 1.75 m were saved (26–28).
The throwing motion was divided into the events of stride foot contact (FC), maximum shoulder external rotation (MER), ball release (BR), and maximum shoulder internal rotation (MIR) (Figure 1). The mean data were compiled for the 2 trials at each perceived effort level. All kinematic and kinetic data for this study were reduced using MotionMonitor (Innovative Sports Training) software. All statistical analyses were performed using SPSS software (version 20.0; SPSS, Inc., Chicago, IL, USA), with an alpha level set a priori at p ≤ 0.05. Next, a within-subjects repeated-measures ANOVA was used to determine whether set shot kinematics and kinetics were significantly different at each of the perceived effort shots. If significant differences were observed, further post hoc repeated-measures ANOVAs were performed for each variable by throwing event. Pairwise comparisons were then analyzed to determine which effort levels were significantly different.
Set shot ball speed data are presented in Table 1. Significant differences in ball speed were observed between all 3 effort level throws. Descriptive kinematic and kinetic data are presented in Tables 2–4.
Shoulder elevation was significantly different at MIR (p < 0.01). Also at the shoulder, plane of elevation was significantly different at BR (p < 0.01).
Trunk flexion was significantly different between the 3 effort levels at BR (p < 0.01) and MIR (p = 0.001). Also, trunk lateral flexion was significantly different at BR (p = 0.001) and MIR (p < 0.001), and pelvis axial rotation (p < 0.001) was significantly different across the 3 shots. Pelvis lateral flexion was significantly different at FC (p < 0.001) and MER (p = 0.010). At MER, there was significantly less pelvis lateral flexion in the 50% shots compared with the 75% shots (p = 0.010).
When examining segmental velocities, significant differences were observed for trunk rotational velocity, humerus rotational velocity, and forearm rotational velocity. Trunk segmental velocity was significantly different at MER (p < 0.001) and BR (p < 0.001). Humerus velocity was significantly different at the events of MER (p < 0.001), BR (p < 0.001), and MIR (p < 0.001) during the 3 set shots. Humerus velocity was significantly different across 3 effort levels at MIR. Lastly, forearm velocity was significantly different at MER (p = 0.006) and MIR (p < 0.001) of the throwing motion.
When developing return-to-play criteria for throwing athletes, it is important to understand the demands that are placed on the upper extremity during throwing. Interval throwing protocols are typically integrated into an athlete's rehabilitation program to progressively prepare the athlete for the demands of throwing (24). Although defined interval throwing protocols have been established for the sport of baseball, the literature is scarce for the sport of team handball. The aim of this study was to examine kinematic data for the set shot at 3 progressively increasing effort levels to build a foundation for future research in a sport-specific interval throwing protocol for team handball athletes. Thus, to the authors' knowledge, this study is the first attempt to quantitatively describe set shot kinematics at 3 different effort levels in team handball athletes. These data are paramount because throwing distance parameters have been established in baseball literature, but effort level is subjective and may affect the kinematics that a player uses during a shot.
Kinematic differences at the shoulder, trunk, and pelvis were observed across effort levels throughout the set shot throwing motion. Shoulder elevation at MIR was significantly less for 50 and 75% effort shots compared with the 100% effort set shots. This was the only event of the throwing motion where shoulder elevation differences were observed. The event of MIR is during the follow-through phase of throwing. It is speculated that when one tries to throw at less effort, they may place more emphasis in attempting to slow their arm after BR. This may result in a player not following through toward the target as they would if throwing at 100% effort. A follow-through toward the desired target would result in lower humeral elevation. Plane of shoulder elevation can also be an indicator of follow-through direction. It was found that at BR there was a significant difference in plane of shoulder elevation from 100 to 50% effort. Meaning, when asked to throw at 50%, there was an obvious lack of follow-through across the body and in the direction of the target. The trunk was positioned with less flexion and lateral flexion in the lower effort level shots compared with the 100% effort shots at BR and MIR. These results indicate that the trunk was in a more upright and neutral position for the submaximal effort shots. The follow-through phase of any overhead throwing motion is critical for decelerating the body after the release of the ball. By shooting at a lower effort level, the team handball athlete may place greater emphasis on their throwing mechanics resulting in significantly different kinematics compared with when they shoot at greater effort levels.
Segmental rotational velocities were most affected by throwing at 3 different effort levels, as expected. When a player attempts to throw a ball with less effort, it can be speculated that they will try to slow their entire throwing motion down, which will result in segmental velocities throughout the kinetic chain being decreased. In addition, the slower segmental velocities observed are in agreement with the kinematics observed in the upper extremity. When examining the differences in ball speed, statistical significance was observed between all 3 shots. Although significant changes were observed, the speed at which the team handball players threw did not equal 50 or 75% of their maximum. The results are similar to what was reported in baseball pitchers (31). The average set shot speeds were 37.5 mph (16.8 m·s−1), 44.1 mph (19.7 m·s−1), and 48.7 mph (21.8 m·s−1) for the 3 effort levels, respectively. These results indicate that team handball players are able to gauge the effort at which they shoot. However, it cannot be assumed that these speeds will be at a certain percentage of their maximum.
Based on the current data, when specifically examining effort levels, set shot kinematic differences were isolated from BR to MIR at 100–50% effort. Whether or not these differences are advantageous are unknown. These observed kinematic differences might indicate that there is more focus at BR and follow-through vs. altering total throwing mechanics. This could be a positive finding in that the players remained consistent getting into the throwing position of MER. The follow-through, BR to MIR, is an important aspect of any overhand throwing motion because it functions to safely decelerate the body after BR. Therefore, if kinematics are altered at the proximal segments (pelvis and trunk) of the kinetic chain, then compensations can occur distally at the upper extremity.
This study aimed to examine kinematic differences in the set shot at 3 different effort levels; however, it is speculated that similar results would be observed for the jump shot as well. The jump shot is a commonly performed shot in team handball practice and competition (26). Previous literature has observed similar upper extremity kinematics in maximal effort set shots and jump shots in experienced team handball players (29). Because similar upper kinematics have been observed between the 2 shot types, it may be a safe progression to advance from set shots to jump shots in an interval throwing protocol.
Although this study provides valuable data on set shot kinematics, it is important to note some limitations exist. Recruiting male participants who play team handball on a regular basis proved difficult in the United States, which resulted in a small sample size. Because of this, it is important to note that concerns regarding the external validity of the present results exist and should be interpreted with caution. In addition, the variability in playing experience (1–8 years) of the participants who volunteered is a limitation. Because of the variability in team handball playing experience, it is possible that the observed differences cannot be generalized to more experienced players. Only 2 accurate shots at each effort level were examined, and it is possible that intraindividual variation may have affected the results of this study. It is currently unknown the ideal number of throwing trials that need to be completed to limit intraindividual variation. In a similar study in baseball pitchers, Slenker et al. (2014) examined kinetics during 60, 80, and 100% effort level pitches and only collected 3 trials for each condition.
These results indicate that team handball players are able to gauge the effort at which they shoot. However, it cannot be assumed that these speeds are at a certain percentage of their maximum. Caution should be observed during the later stages of rehabilitation to prescribe throwing at different effort levels while using a radar gun as an assessment tool for effort. Future research should aim to examine the kinetics about the shoulder and elbow when team handball players attempt to shoot at different effort levels because this could ultimately relate to the risk of reinjury. Also, examining the effect of effort level on other commonly used team handball throwing motions such as passing and jump shots should be examined in the future.
The results of this study provide a solid foundation for future research into the development of an interval throwing program for team handball players. By establishing the role that different effort level shooting has on mechanics, clinicians can use these results when gauging a player's readiness to progress into sport-specific throwing activities. Future research should examine the role of shot distance on the kinematics and kinetics associated with the throwing motion. The distance that a player shoots may be a better indicator than effort level when developing return-to-play throwing criteria in team handball, as effort level is very subjective between individuals. When beginning throwing after an upper extremity injury, it is likely best to focus more on using proper throwing mechanics from a short distance and then slowly progressing to greater distances as an athlete adapts to the demands of throwing.
1. Almeida GP, Silveira PF, Rosseto NP, Barbosa G, Ejnisman B, Cohen M. Glenohumeral range of motion in handball players with and without throwing
pain. J Shoulder
Elbow Surg 22: 602–607, 2013.
2. Axe M, Hurd W, Snyder-Mackler L. Data-based interval throwing
programs for baseball players. Sports Health 1: 145–153, 2009.
3. Axe MJ, Wickham R, Snyder-Mackler L. Data-based interval throwing
programs for little league, high school, college, and professional baseball pitchers. Sports Med Arthrosc 9: 24–34, 2001.
4. Axe M, Windley TC, Snyder-Mackler L. Data-based interval throwing
programs for collegiate softball players. J Athl Train 37: 194–203, 2002.
5. Day J, Murdoch D, Dumas G. Calibration of position and angular data from a magnetic tracking device. J Biomech 33: 1039–1045, 2000.
6. Fleisig GS, Bolt B, Fortenbaugh D, Wilk KE, Andrews JR. Biomechanical comparison of baseball pitching and long-toss: Implications for training and rehabilitation
. J Orthop Sports Phys Ther 41: 296–303, 2011.
7. Fleisig GS, Zheng N, Barrentine SW. Kinematic and Kinetic Comparison of Full Effort and Partial Effort Baseball Pitching. Atlanta, GA: American Society of Biomechanics, 1996.
8. Gohlke F, Lippert MJ, Keck O. Instability and impingement of the shoulder
of the high performance athlete in overhead stress [in German]. Sportverletz Sportschaden 7: 115–121, 1993.
9. Huang YH, Wu TY, Learman KE, Tsai YS. A comparison of throwing
kinematics between youth baseball players with and without a history of medial elbow pain. Chin J Physiol 53: 160–166, 2010.
10. Keeley DW, Oliver GD, Dougherty CP. Shoulder
kinematics during pitching: Comparing the slide step and traditional stretch deliveries. Hum Mov Sci 31: 1191–1199, 2012.
11. Langevoort G. Glenohumeral Instability. In: Langenvoort G (ed). Sports Medicine and Handball. Vol 7. Basel, Switzerland: Beckmann, 1996. pp. 39–44.
12. Ludewig PM, Cook TM. Alterations in shoulder
kinematics and associated muscle activation in people with symptoms of shoulder
impingement. Phys Ther 80: 276–291, 2000.
13. Meskers CGM, Fraterman H, van der Helm FCT, Vermeulen HM, Rozing PM. Calibration of the “Flock of Birds” electromagnetic tracking device and its application in shoulder
motion studies. J Biomech 32: 629–633, 1999.
14. Myklebust G, Hasslan L, Bahr R, Steffen K. High prevalence of shoulder
pain among elite Norwegian female handball players. Scand J Med Sci Sports 23: 288–294, 2013.
15. Oliver GD. Relationship between gluteal muscle activation and upper extremity kinematics and kinetics in softball position players. Med Biol Eng Comput 52: 265–270, 2014.
16. Oliver GD, Dwelly PM, Kwon YH. Kinematic motion of the windmill softball pitch in prepubescent and pubescent girls. J Strength Cond Res 24: 2400–2407, 2010.
17. Oliver GD, Keeley DW. Gluteal muscle group activation and its relationship with pelvis and torso kinematics in high-school baseball pitchers. J Strength Cond Res 24: 3015–3022, 2010.
18. Oliver GD, Keeley DW. Pelvis and torso kinematics and their relationship to shoulder
kinematics in high-school baseball pitchers. J Strength Cond Res 24: 3241–3246, 2010.
19. Oliver GD, Plummer H. Ground reaction forces, kinematics, and muscle activations during the windmill softball pitch. J Sports Sci 29: 1071–1077, 2011.
20. Perie D, Tate AJ, Cheng PL, Dumas GA. Evaluation and calibration of an electromagnetic tracking device for biomechanical analysis of lifting task. J Biomech 35: 293–297, 2002.
21. Plummer HA, Oliver GD. Quantitative analysis of kinematics and kinetics of catchers throwing
to second base. J Sports Sci 31: 1108–1116, 2013.
22. Plummer HA, Oliver GD. The relationship between gluteal muscle activation and throwing
kinematics in baseball and softball catchers. J Strength Cond Res 28: 87–96, 2014.
23. Seil R, Rupp S, Tempelhof S, Kohn D. Sports injuries in team handball. A one-year prospective study of sixteen men's senior teams of a superior nonprofessional level. Am J Sports Med 26: 681–687, 1998.
24. Slenker NR, Limpisvasti O, Mohr K, Aguinaldo A, ElAttrache NS. Biomechanical comparison of the interval throwing
program and baseball pitching: Upper extremity loads in training and rehabilitation
. Am J Sports Med 42: 1226–1232, 2014.
25. Veeger HEJ. The position of the rotation center of the glenohumeral joint. J Biomech 33: 1711–1715, 2000.
26. Wagner H, Buchecker M, von Duvillard SP, Muller E. Kinematic description of elite vs. low level players in team-handball jump throw. J Sports Sci Med 9: 15–23, 2010.
27. Wagner H, Muller E. The effects of differential and variable training on the quality parameters of a handball throw. Sports Biomech 7: 54–71, 2008.
28. Wagner H, Pfusterschmied J, Klous M, von Duvillard SP, Muller E. Movement variability and skill level of various throwing
techniques. Hum Mov Sci 31: 78–90, 2012.
29. Wagner H, Pfusterschmied J, von Duvillard SP, Müller E. Performance and kinematics of various throwing
techniques in team-handball. J Sports Sci Med 10: 73–80, 2011.
30. Wicke J, Keeley DW, Oliver GD. Comparison of pitching kinematics between youth and adult baseball pitchers: A meta-analytic approach. Sports Biomech 12: 315–323, 2013.
31. Wilk KE, Meister K, Andrews JR. Current concepts in rehabilitation
of the overhead throwing
athlete. Am J Sports Med 30: 136–151, 2002.
32. Wu G, Siegler S, Allard P, Kirtley C, Leardini A, Rosenbaum DA, Whittle M, D'Lima DD, Cristofolini L, Witte H, Schmid O, Stokes I. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion-part I: Ankle, hip, and spine. J Biomech 35: 543–548, 2002.