Introduction
Bilateral landings are common to many sport-specific movements that contain a vertical airborne phase (13) and are characterized by high magnitudes of vertical impact force. A primary goal of strength and conditioning professionals is to provide an athlete with the skills and resources needed to refine their abilities to attenuate impact forces during landing. The safe and efficient attenuation of high-magnitude impact forces requires a strategic sequencing of contributing body segments (5) that occurs successive to the selection of a response strategy, which is defined as a neuromusculoskeletal solution for the performance of a motor task (14). Certain landing strategies, such as a decrease in plantar flexion (31) and/or the rigid joint positioning at ground contact (11), can result in performance error and/or unnecessarily high magnitudes of vertical impact force.
Researchers have presented training programs for use by strength and conditioning professionals to train their athletes to use strategies that decrease the risk of performance error and/or impact force magnitudes during landing (1,2,17,20,28). The studies forming the foundation for such programs largely evaluated unilateral and bilateral step-off style drop landings (10,14,21,24,38) or hanging drop landings (7,23,32) under the assumption that such landings adequately replicate landings performed in sports. However, sport-specific landing maneuvers are almost exclusively preceded by a propulsive vertical jump (4,19). Thus, the performance adaptations that occur after training programs developed from studies in which highly controlled laboratory landings, such as step-off style drop landings (STL), were evaluated might not transfer to the performance of more sport-specific vertical jump landings (VJL).
There is a paucity of available literature (3,8,15) comparing STL to VJL with respect to performance and/or impact force attenuation differences. This lack of evidence raises concern relative to the use of highly controlled STL as a surrogate for VJL in both training and research. Initial findings suggest that STL produce greater peak impact force magnitudes and loading rates, altered muscle activations, and lesser joint extension/plantar flexion at the time of peak impact force compared with VJL from the same height (3,15). These differences have been observed on hard surfaces (3) and sand surfaces (15) for men and women (3,15). More recently, lower extremity kinematic differences between bilateral VJL and STL were not resolved after multiple days of task-specific practice (8). A gap in the literature exists with respect to the contributions of lower-body movement asymmetry to the differences between VJL and STL. It is expected that asymmetrical movements occur at the start of an STL, and it is reasonable to presume that the differences described previously relate to kinematic asymmetry at the instant of ground contact during STL. Because impact force asymmetry is believed to be a precursor for chronic overload injuries (26,32), evaluating the manner in which the lower limbs move to attenuate impact force is necessary to fully understand the mechanical differences between VJL and STL. Lastly, although bilateral assessments are uncommon in studies that do not explicitly evaluate asymmetry, it has been suggested that such assessments are necessary before evaluating data regardless of limb (32).
It is clear that bilateral analyses are needed to present a well-rounded explanation for the previously observed differences between VJL and STL, However, it is also important to analyze potential differences in joint angular displacement between VJL and STL to fully understand differences in the impact force attenuation strategies used during VJL and STL. Although joint flexion angles at ground contact or at peak impact force, which is commonly presented in the literature, provide an understanding of lower extremity positioning at discrete events, joint angular displacement provides a more detailed description of the segmental movements executed about the joints to attenuate impact forces. For instance, when the ankle is positioned at ground contact such that greater displacement is possible during hanging drop landings, peak impact force magnitudes decrease and loading rates become less rapid (31). The decreased impact force magnitudes and reduced loading rates occur because the increased joint angular displacement allows impact forces to be attenuated over a longer duration (17), thereby decreasing overuse injury potential.
In light of the need to evaluate both lower-body asymmetry and joint angular displacement to fully understand the differences between VJL and STL, our purpose was to examine the interactive effects of limb and task on joint kinematics, ground reaction forces, and temporal parameters during VJL and STL from the same height. We chose to divide the impact phase into 2 subphases representative of early impact and late impact to identify combined multijoint strategies (29) relative to these parameters. Based on the findings reviewed and the described gap in the literature, we chose to evaluate both performance (total landing time) and mechanical variables (sagittal plane joint angles and angular displacements, peak impact force magnitudes, impulse, and the time to the peak impact forces) during each task. In accordance with recent findings (3,8,15) and scientific speculation, we hypothesized that (a) greater peak impact forces and more rapid loading times would occur during STL, (b) lesser hip, knee, and ankle joint extension/plantar flexion angles would occur at ground contact during STL, (c) lesser hip, knee, and ankle joint angular displacements would occur throughout the impact phase of STL, and (d) significant asymmetries would be observed throughout impact during STL but not VJL.
Methods
Experimental Approach to the Problem
Step-off landings (STL) are often perceived to be similar to VJL from a mechanical perspective. Although the current literature suggests that these tasks are not similar and produce different mechanical and neurological outcomes (3,8,15), the literature is limited in that performance and/or injury potential differences have yet to be evaluated. To address this literature gap, we obtained data from a sample of male and female participants who performed both VJL and STL from the same height. Paired-samples t-tests and 2-way repeated measures analyses of variance were performed to compare performance and impact force attenuation parameters while accounting for differences between limbs. This study design was used to provide coaches, athletes, and researchers with appropriate evidence relative to changes in both performance and impact force attenuation to determine the appropriateness of VJL or STL for their training and/or investigative purposes.
Subjects
Ten healthy adults (5 men, ± SD 25.0 ± 1.6 years; 1.7 ± 0.4 m; 79.7 ± 7.1 kg; 5 women, ± SD 20.8 ± 1.6 years; 1.6 ± 0.4 m; 68.5 ± 7.1 kg) participated in this study, all of whom were at least 18 years of age. This sample was determined using an a priori power analysis (G × Power 3.1) according to the nonfatigued knee angle data at ground contact of Edwards et al. (15). This analysis indicated that 10 participants were needed to achieve a proposed effect size (ES) of 1.16, power (1−β) of 0.95, and an alpha level of 0.05. All participants were older than 18 years and were classified as recreationally active because of their participation in exercise or recreational or competitive sports involving jump-landing movements at least 2 times per week for at least 6 months before participating. The participants were free of any condition that would limit their ability to complete the experimental tasks and had no recent history (<1 year) of injury to the lower extremities (fractures, ligament ruptures, etc.). All data were obtained from the participants during the fall semester of the same academic year. Before completing any laboratory tasks, the testing procedures were explained, and written informed consent was provided as approved by the institutional review board at the University of Nevada, Las Vegas (Protocol no. 755011-1). As needed, clarification of the protocol was provided throughout the duration of the investigation.
Procedures
Participants completed 2 laboratory sessions with 24–72 hours between sessions. Participants were asked to maintain consistency relative to the time of food consumption before arriving to the laboratory. We did not control for hydration during and/or between testing sessions, as it has been shown that acute changes in hydration status do not influence jumping performance (22). During session 1, demographic and anthropometric measures were recorded (age, sex, height, and mass), and a demonstration of the procedures was provided. Participants completed a self-selected warm-up (≤10 minutes) consisting of both static and dynamic stretching. Warm-up effects were not expected, as the combination of static and dynamic stretching counterbalances potential warm-up effects that could otherwise alter performance (36). Participants performed up to 5 practice/familiarization attempts before performing 3 maximum effort countermovement vertical jumps. These 3 vertical jump attempts were used to establish the controlled landing height used during session 2. Jump height was measured to the nearest 1.27 cm using a Vertec Jump Trainer (Sports Imports, Hilliard, OH, USA). Jump height was determined by taking the measured height reached on the Vertec subtracting out the participant's 2-handed reach from a plantar flexed position (18). Although the participants reached vertically with both hands, only the hand nearest the Vertec needed to strike the vanes. Because setting the height of the platform during STL to match the vertical displacement of the center of mass during VJL does not create equivalent landing heights (3), the vertical displacement of the center of mass at takeoff due to plantar flexion was accounted for to establish an equal fall height for the STL. This distance was determined by subtracting the participant's 2-handed reach from a flat-footed position from their 2-handed reach from a plantar flexed position. The jump heights (plus the added distance from plantar flexion) from the 3 trials performed in session 1 were averaged to establish the controlled landing height for both tasks during the second session.
During the second session, participants were asked to complete a self-selected warm-up that best replicated the warm-up performed during session 1. Then, reflective markers were applied bilaterally to the lower extremities using the Vicon Plug-in Gait model to obtain 3-dimensional kinematic data using a 10-camera motion capture system (100 Hz; Vicon Motion Systems, Ltd., Oxford, United Kingdom). Three-dimensional kinetic data were acquired synchronously to the kinematic data using a dual force platform system (1,000 Hz; Kistler Instruments, Corp., Amherst, NY, USA). Participants then performed 15 VJL and 15 STL. The tasks were presented to the participants in a counterbalanced order. Up to 3 practice attempts were provided for each task. The VJL were performed with participants starting in a standing, 2-footed position with each foot positioned on a force platform. Similar to session 1, the Vertec was positioned adjacent to the force platforms on the participant's preferred side, and a specific vane of the Vertec was exposed as a target to control the participant's landing height for each VJL trial. Participants were instructed to jump vertically while reaching with both hands to contact the exposed vane on the Vertec set to match their predetermined jump/landing height to the nearest 1.27 cm (±1 vane). Countermovement depth and arm swing were not controlled. After contact with the Vertec, participants landed with each foot contacting its respective force platform and returned to a motionless standing position. If the participants were unable to touch the exposed vane or return to a standing position, the trial was discarded and a subsequent trial was performed. No more than 20 attempts were needed for any participant to complete 15 VJL trials. Rest was provided as needed to mitigate potential fatigue effects.
The STL started with participants standing on a custom-built, height-adjustable platform set to match their controlled landing height established during session 1. The participants identified their preferred limb (referred to here as the lead limb) to initiate the STL, and the lead limb remained consistent across all STL trials. Trials began with the participants stepping off the platform by reaching their lead limb forward followed immediately by the trail limb while attempting to not change the vertical position of the center of mass (21). The participants attempted to drop as vertically as possible (i.e., minimal horizontal movement) with each foot contacting a force platform before returning to a motionless standing position. Similar to VJL, arm position was not controlled, and rest was provided as needed.
Data Reduction
Data were processed in Vicon Nexus (version 1.8.5). Raw data were smoothed with a fourth-order low-pass Butterworth digital filter with cut-off frequencies of 12 and 50 Hz for the kinematic and kinetic data (3,30). Sagittal plane hip, knee, and ankle joint angles were calculated in the Nexus software for the lead and trail limbs and expressed such that positive values indicated flexion/dorsiflexion. A custom Matlab (The MathWorks, Inc., Natick, MA, USA) script was used to extract variables of interest. Specifically, the vertical ground reaction force (vGRF) profiles of the lead and trail limbs were summed to represent a total vGRF profile used to identify the ground contact, first vGRF peak (F1), second vGRF peak (F2), and end of impact events. Ground contact was defined as the time when the vGRF profile exceeded 20 N (after takeoff during VJL). F1 and F2 were identified as the first and second peak magnitudes in the vGRF time history. Both peaks were evaluated because vertical landings reveal a bimodal vGRF profile of which both peaks are examined to evaluate injury potential relative to forefoot (F1) and rearfoot (F2) impact (12). The F1 and F2 magnitudes were normalized to Newton's per kilogram of body mass (N·kg−1). The end of the impact phase was defined as the time when the center of mass changed direction from downward to upward after ground contact. Specifically, the change in vertical position of the center of mass was identified as the first event at which vertical velocity crossed zero after ground contact. To obtain velocity, acceleration was first computed from the vGRF time history using Newton's law of acceleration accounting for gravity. The time integral of acceleration was then calculated to obtain velocity. For both tasks, the last frame of the velocity profile was selected to estimate zero velocity because it represented motionless standing (zero vertical velocity of the center of mass). The velocity profiles were then adjusted such that the first time the end velocity value was reached after touchdown defined the end of the impact phase. To verify that the landing heights were equal between tasks, the velocity magnitudes at ground contact (impact velocity) were extracted for comparison. We used impact velocity rather than calculating jump/landing height due to potential errors within jump height computations when using either the impulse-momentum relationship or the time in the air equation (27).
Statistical Analyses
Data are presented as mean ± 1 SD. Two statistical tests were performed in IBM SPSS Statistics Software (Version 24; IBM Corp, Armonk, NY, USA) to test our hypotheses. Before performing the first statistical test, mean values were computed across trials per participant, per task, for the following variables: impact velocity, time to F1 (tF1), time to F2 (tF2), and time to the end of impact. Mean values were then calculated across participants from the individual mean values for each parameter. Then, paired-samples t-tests (α = 0.05) were used to examine differences between tasks in impact velocity, tF1, tF2, and the time to the end of impact. An intraclass correlation coefficient (ICC) and the associated 95% confidence interval (CI) were used to ensure adequate test-retest reliability for our jump/landing height control during VJL and STL. If the ICC was greater than 0.8, the jump/landing height control procedures were considered reliable (34).
Before performing the second statistical test, mean values were calculated across trials per participant for the following variables: the contributions of the lead and trail limbs to F1 and F2, the hip, knee, and ankle joint angular positions of the lead and trail limbs at ground contact, at F1, at F2, and at the end of impact, and the hip, knee, and ankle joint angular displacements of the lead and trail limbs between ground contact and F1, between F1 and F2, and between F2 and the end of impact. Mean values were then calculated across participants from the individual mean values for each parameter. Two-way analyses of variance (α = 0.05) were used to compare limb and/or task differences for these bilateral variables. If a significant limb by task interaction was detected, paired-samples t-tests (α = 0.05) were used both for unilateral comparisons between-tasks (each limb evaluated between tasks) and between-limb (asymmetry) comparisons within tasks (lead and trail limbs each evaluated during both tasks). If no limb by task interaction was detected, main effect comparisons both between tasks (regardless of limb) and between limbs (regardless of task) were performed applying the Sidak adjustment.
For the 2 statistical tests, histograms and probability-probability plots (P-P plots) were used to inspect the data relative to the assumption of normality, and scatterplots (zpred vs. zresid) were used to test the data relative to the assumption of homogeneity/homoscedasticity (16). If the assumptions for normality and/or homogeneity of variance were violated, Bonferroni corrections were used. These corrections were used to create a more conservative test to compensate for a violated assumption (34). To provide the meaningfulness of the differences, ES were computed for all significant pairwise and main effect comparisons. Effect size values were computed as the mean difference divided by the pooled SD (34). Effect size magnitudes were evaluated using the interpretation of Cohen (9), of which magnitudes of 0.2, 0.5, and 0.8 are the lower thresholds for small, moderate, and large mean differences, respectively. In addition, the ES values, the lower, and upper bounds for a 95% CI were presented for the sample mean for each task when a significant pairwise or main effect comparison was detected. The CI values were calculated as described by Vincent and Weir (33).
Results
Impact Velocity, Ground Reaction Forces, and Temporal Parameters
The comparison of impact velocity detected no difference between VJL and STL (2.69 ± 0.48 m·s−1, VJL; 2.65 ± 0.51 m·s−1, STL; p = 0.483), and the test-retest reliability was very strong (ICC = 0.970; CI = 0.881–0.993). No task differences were detected for tF1 (0.042 ± 0.014 seconds, VJL; 0.047 ± 0.013 seconds, STL; p = 0.234) or the time to the end of the impact phase (0.322 ± 0.180 seconds, VJL; 0.328 ± 0.146 seconds, STL; p = 0.898). However, a significant, large magnitude task difference was detected for tF2, revealing a more rapid loading time during STL compared with VJL (0.062 ± 0.016 seconds, CI = 0.051–0.073, STL; 0.086 ± 0.018 seconds, CI = 0.073–0.100, VJL; p < 0.001; ES = 1.00). Individual participant values for tF2 during VJL and STL are shown in Figure 1. No significant interaction or limb main effect was detected for the contributions of the lead and trail limbs to F1 (p = 0.954, p = 0.137), although significant task main effect identified a significant, moderate magnitude difference for F1 between VJL (13.98 ± 4.45, CI = 10.80–17.16) and STL (19.38 ± 9.17, CI = 12.83–25.93) regardless of limb (p = 0.002; ES = 0.57; Table 1). Lastly, no significant interaction (p = 0.956) task main effect (p = 0.832) or limb main effect (p = 0.419) was detected for F2 (Table 1).
;)
Figure 1.: Individual participant values for tF2 during VJL and STL. Data are presented as mean values across trials collapsed by limb; VJL = vertical jump landing; STL = step-off landing; male participants are identified by the solid line; female participants are identified by the dashed line.
Table 1.: Contributions of the lead and trail legs to the F1 and F2 magnitudes.*
Hip, Knee, and Ankle Joint Angles
Hip, knee, and ankle joint angles at ground contact, F1, F2, and the end of impact are documented in Table 2. No significant interactions or limb main effects were detected for hip joint angle at ground contact (p = 0.101, p = 0.297), F1 (p = 0.376, p = 0.205), or the end of the impact phase (p = 0.923, p = 0.809). In addition, no significant interaction (p = 0.857), task main effect (p = 0.218), or limb main effect (p = 0.868) was detected for hip joint angles at F2. However, significant task main effects revealed moderate-to-large magnitude differences in hip joint angles between tasks at ground contact (15.75 ± 4.52, CI = 12.52–18.98, VJL; 24.73 ± 9.52, CI = 17.93–31.53, STL; p < 0.001; ES = 0.85), F1 (24.72 ± 8.01, CI = 19.00–30.44, VJL; 31.23 ± 10.17, CI = 23.96–38.50, STL; p = 0.025; ES = 0.50), and the end of the impact phase (42.56 ± 27.28, CI = 23.06–62.06, VJL; 67.70 ± 29.93, CI = 46.31–89.09, STL; p = 0.001; ES = 0.62) regardless of limb (Table 2).
Table 2.: Bilateral hip joint positions at discrete impact phase events.*
A significant interaction was detected for knee joint angles at ground contact (p = 0.011; Table 3). Post hoc analyses revealed a significant, large magnitude difference between the lead limb and trail limb knees during STL (12.32 ± 7.72, CI = 6.80–17.84, lead; 22.34 ± 9.78, CI = 15.35–29.33, trail; p = 0.019; ES = 0.81; Table 3) but not VJL (p = 0.555). Individual participant values for the lead and trail limb knee joint angles during STL are shown in Figure 2. Post hoc analyses also revealed a significant, moderate magnitude difference at ground contact between VJL (14.73 ± 5.81, CI = 10.58–18.88) and STL (22.34 ± 9.78, CI = 15.35–29.33) for the trail leg knee (p = 0.005; ES = 0.67; Table 3) but not the lead leg knee (p = 0.110). No significant interaction (p = 0.911), task main effect (p = 0.261), or limb main effect (p = 0.436) was detected for knee joint angles at F1 (Table 3). Finally, no significant interaction or limb main effect was detected for knee joint angles at F2 (p = 0.831; p = 0.650) or the end of the impact phase (p = 0.931; p = 0.700). However, significant task main effects revealed moderate-to-large magnitude differences between tasks for knee joint angles at both F2 (46.72 ± 9.65, CI = 39.82–53.62, VJL; 38.84 ± 11.10, CI = 30.91–46.77, STL; p = 0.018; ES = 0.54) and the end of the impact phase (52.96 ± 20.81, CI = 38.90–67.83, VJL; 80.38 ± 20.89, CI = 65.45–95.31, STL; p = 0.001; ES = 0.93) regardless of limb (Table 3).
Table 3.: Bilateral knee joint positions at discrete impact phase events.*
Figure 2.: Individual participant values for ground contact knee joint angles during STL. Data are presented as mean values across trials; STL = step-off landing; Lead = lead limb determined during STL; Trail = trail limb determined during STL; male participants are identified by the solid line; female participants are identified by the dashed line.
No significant interaction, task main effect, or limb main effect was detected for ankle joint angles at F1 (p = 0.904, p = 0.132, p = 0.975; Table 4). In addition, no significant interaction or limb main effect was detected for ankle joint angles at ground contact (p = 0.379, p = 0.886), F2 (p = 0.629, p = 0.947), or the end of the impact phase (p = 0.781, p = 0.297; Table 2). However, significant task main effects detected very large magnitude differences in ankle joint angles at ground contact (−18.39 ± 11.05, CI = −26.29 to −10.49, VJL; −11.52 ± 13.50, CI = −21.17 to 1.87, STL; p = 0.006; ES = 1.71) and F2 (28.21 ± 4.29, CI = 25.14–31.28, VJL; 16.73 ± 4.97, CI = 13.18–20.28, STL; p < 0.001; ES = 1.75), whereas a moderate magnitude difference was detected at the end of the impact phase (24.08 ± 5.05, CI = 20.47–27.69, VJL; 28.74 ± 7.08, CI = 23.68–33.80, STL; p = 0.046; ES = 0.54) regardless of limb (Table 4).
Table 4.: Bilateral ankle joint positions at discrete impact phase events.*
Hip, Knee, and Ankle Joint Angular Displacements
No significant interactions, task main effects, or limb main effects were identified for joint displacement between ground contact and F1 at the hip (p = 0.131, p = 0.830, p = 0.144) or ankle (p = 0.350, p = 0.294, p = 0.838). However, a significant interaction was detected for knee joint displacement between ground contact and F1 (p = 0.031). Post hoc analyses revealed a significant, large magnitude difference in knee joint displacement between the lead limb and trail limb from ground contact to F1 during STL (17.80 ± 6.50, CI = 13.15–22.45, lead; 10.82 ± 5.32, CI = 7.02–14.62, trail; p = 0.017; ES = 0.83; Figure 3A) but not VJL (p = 0.732; Figure 3A). A significant, moderate magnitude difference in knee joint displacement was also detected between VJL and STL from ground contact to F1 in the trail limb (15.14 ± 5.90, CI = 10.92–19.36, VJL; 10.82 ± 5.36, CI = 6.99–14.65, STL; p = 0.036; ES = 0.54; Figure 3A) but not the lead limb (p = 0.246: Figure 3A). No significant interactions or limb main effects were detected for joint displacement from F1 to F2 at the hip (p = 0.548, p = 0.680), knee (p = 0.764, p = 0.741), or ankle (p = 0.908, p = 0.958). However, significant task main effects revealed large magnitude differences (Figure 3B) in joint displacement from F1 to F2 between VJL and STL regardless of limb at the hip (p = 0.001; ES = 0.97) and ankle (p < 0.001; ES = 0.97), with a significant, moderate magnitude difference at the knee (p = 0.003; ES = 0.76). No significant interactions or limb main effects were detected for joint displacement from F2 to the end of the impact phase at the hip (p = 0.469, p = 0.563), knee (p = 0.366, p = 0.742), or ankle (p = 0.346, p = 0.336). However, significant task main effects revealed a small magnitude difference between VJL and STL in joint displacement from F2 to the end of the impact phase (Figure 3C) at the hip (8.91 ± 4.78, CI = 5.49–12.33, VJL; 2.84 ± 4.03, CI = −0.04 to 5.72, STL; p = 0.002; ES = 0.48), whereas large magnitude differences were detected at the knee (18.11 ± 10.24, CI = 10.79–25.43, VJL; 7.24 ± 9.91, CI = 0.16–14.32, STL; p < 0.001; ES = 0.96) and ankle (24.84 ± 11.45, CI = 16.66–33.02, VJL; 9.12 ± 11.51, CI = 0.89–17.35, STL; p < 0.001; ES = 1.17). Individual participant values for hip, knee, and ankle joint displacements between VJL and STL are shown in Figure 4.
Figure 3.: Hip, knee, and ankle joint displacements during VJL and STL. Data are presented as mean + 1 SD; VJL = vertical jump landing; STL = step-off landing; Lead = lead limb determined during STL; Trail = trail limb determined during STL; graph A = knee joint displacement for the lead and trail legs between ground contact at F1; graph B = hip, knee, and ankle joint (collapsed by limb) displacements between F1 and F2; graph C = hip, knee, and ankle joint displacements (collapsed by limb) between F2 and the end of the impact phase; **significantly greater than trail (p ≤ 0.05); *significantly greater than VJL/STL for the respective leg/joint (p ≤ 0.05).
Figure 4.: Individual participant values for joint displacement during VJL and STL. Data are presented as mean values across trials collapsed by limb; VJL = vertical jump landing; STL = step-off landing; graph A = hip joint displacement; graph B = knee joint displacement; graph C = ankle joint displacement; male participants are identified by the solid line; female participants are identified by the dashed line.
Discussion
To compare VJL to STL, it was imperative that the landing heights for each task were not different. The current procedure was intentionally different than the methods of both Edwards and colleagues (15) and Afifi and Hinrichs (3), as those methods were limited in that they could not precisely match landing heights (15) or required extensive amounts of time to match trials (3). The decision to use a novel approach to match landing height was supported by the fact that this procedure was successful in matching landing heights with very strong reliability across 15 trials.
The higher magnitude F1 detected during STL indicates that the forefoot experienced a greater amount of stress compared with VJL shortly after contact with the ground. Although the greater stress on the forefoot suggests that there is a greater risk of overuse injury (12) during STL, the greater F1 also leads to a more rapid tF2 during STL. The more rapid tF2 is consistent with both recent research (3,15) and our expectations and indicates that athletes could be further exposed an increased risk of overuse injury during chronic STL performance in comparison with chronic VJL performance. Although the current result for tF2 is consistent with previous findings, the cause of the more rapid tF2 is not well understood. Previously, it was concluded that the more rapid tF2/greater loading rate was related to less preparedness for impact during STL (3), as that conclusion was supported by lesser hip, knee, and ankle flexion/dorsiflexion at the time of F2. However, we conjecture that these participants were more prepared (had greater anticipation) for impact during STL. Our conclusion is evidenced by the lack of a prelanding jump (15) and the utilization of greater hip flexion and ankle dorsiflexion at ground contact, which was not expected. The fact that this sample of individuals frequently performed jump landings for exercise or sport suggests that the greater hip and ankle flexion/dorsiflexion was the result of different/modified strategy for impact during STL compared with VJL. Thus, these participants likely used less refined motor programs while emphasizing voluntary lower-body control during STL, thereby disrupting the automaticity of motor control processes (37) governing lower-body movements. Disrupting automatic control processes is a known performance limiter during landing (25,35), and these data further suggest that such disruptions can increase overuse injury risk during STL by increasing F1 and leading to a more rapid tF2.
In addition to the greater magnitudes of hip and ankle flexion/dorsiflexion at ground contact during STL, the decreased magnitudes of joint displacement between F1 and F2 during STL (an unexpected result) might further explain the more rapid tF2 observed in comparison with VJL (17). The reduced displacements of the hip, knee, and ankle joints during early impact attenuation required greater magnitudes of hip, knee, and ankle joint displacements after F2, and also resulted in greater magnitudes of joint flexion at each joint at the end of impact. These greater joint displacements between F2 and the end of impact seem to be compensatory response strategies during STL used to rapidly decelerate the vertical velocity of the center of mass as intended before impact. Although not examined herein, we speculate that the knee joints experienced greater magnitudes of joint loading during the latter portion of the impact phase to stop downward motion while compensating for the limited involvement of the joints during initial impact. We also speculate that the magnitudes of knee joint loading likely differed between limbs between ground contact and F1 during STL due to both the asymmetrical ground contact knee joint angles and asymmetrical joint displacement. Because no asymmetries were detected beyond the instant of F1, it appears as though the compensatory actions discussed were also executed to correct for the initial knee joint asymmetry.
The different effects these tasks can have on an athlete's training outcomes should be strongly considered by strength and conditioning coaches. Relative to landing performance, an athlete's goal is to attenuate impact forces as quickly as possible to perform a subsequent movement if needed (19). From the current data, the similarity in total impact times indicates that both STL and VJL, when performed from equal heights, could elicit the same performance effects after training in which athletes work to safely decrease the total time of impact. However, when dividing the impact phase into early and late impact phases, it seems that training for the rapid performance of sport-specific landings using STL can cause a dependency on compensatory movements (increased magnitudes of joint motion during the latter phase of the impact phase) to decelerate the body after F2. If STL are included in training in place of VJL, the compensatory movements executed during STL could alter an athlete's ability to quickly initiate a secondary task (repeated jump, change of direction, forward sprint, etc.) during competitive sport-specific landings.
Strength and conditioning professionals should also consider the differences between VJL and STL with respect to both impact force and kinematic asymmetry, as these data indicate that there may be an increased risk of an overuse injury if STL are included in training (6,32). The inclusion of STL in training could cause an athlete to experience repeated occurrences of increased magnitude forces that are more rapidly experienced while also developing dependence on asymmetrical lower-body kinematics during the time at which impact force is attenuated. Thus, athletes whose performance demands require the efficient performance of VJL, the inclusion of STL in training would likely result in the utilization of inappropriate movement patterns that contribute to increased overuse injury potential. Nonetheless, athletes could also benefit from the controlled inclusion of STL when training for specific landing situations during which the feet are not positioned symmetrically at impact. When an appropriate volume of STL is performed and the athlete is appropriately monitored with respect to overuse injury potential, the ability to quickly and efficiently attenuate impact force during asymmetrical landings could be improved.
A possible limitation of this study was the lack of instruction provided to the participants related to their focus of attention during both tasks. Our participants might have used different attentional foci during each task, influencing the differences in joint angular positioning at ground contact and the subsequent differences in F1, tF2, and joint displacement. However, it would be difficult to use cues that produce the same focus of attention, as these tasks are inherently different because of the lack of a prelanding jump during STL. Another limitation was the fact that these tasks are not purely vertical. Although we attempted to limit any horizontal movements, it is impossible to completely remove horizontal movements. Lastly, our decision not to control the movement of the upper body during both tasks was a limitation, as it is known that permitting arm movements can reduce impact force magnitudes when compared to restricting arm movements (33). However, it is unknown as to which task is more greatly affected by arm movements, and unconstrained movements best characterize landings performed outside laboratory settings. As such, we expected greater generalizability by allowing arm movements during both tasks at the discretion of the participants.
Practical Applications
This analysis identified kinetic, kinematic, and temporal differences between VJL and STL from equal heights. Impactful differences include asymmetrical knee joint angles at ground contact (during STL only), a higher magnitude F1, a more rapid tF2, and compensatory actions after F2 to decelerate downward velocity of the center of mass in STL vs. VJL. In combination with previous studies (3,15), it is clear that VJL and STL are associated with different joint motions and impact attenuation parameters during the impact phase of landing. Although STL seem to increase the risk of overuse injury compared with VJL, STL can also be used to improve an athlete's impact attenuation abilities during vertical landing tasks characterized by asymmetrical lower-body positions at ground contact. Based on the current findings, strength and conditioning professionals are advised to consider the demands of their athlete's sport or competition when including landing training in their protocols. Lastly, strength and conditioning professionals aiming to improve an athlete's performance during sport-specific jump landings should consider the likely impact attenuation outcomes before selecting STL or VJL in training.
Acknowledgments
This project did not receive funding from the public or private sectors. The authors have no conflicts of interest to disclose. Finally, the results of this study do not constitute endorsement by the authors or the NSCA.
References
1. Aerts I, Cumps E, Verhagen E, Meeusen R. Efficacy of a 3 month training program on the jump-landing technique in jump-landing sports. Design of a cluster randomized controlled trial. BMC Musculoskelet Disord 11: 281, 2010. 2474-11-281.
2. Aerts I, Cumps E, Verhagen E, Wuyts B, Gucht VD, Meeusen R. The effect of a 3-month prevention program on the jump-landing technique in basketball: A randomized controlled trial. J Sport Rehab 24: 21–30, 2015.
3. Afifi M, Hinrichs RN. A mechanics comparison between landing from a countermovement jump and landing from stepping off a box. J Appl Biomech 28: 1–9, 2012.
4. Barker LA, Harry JR, Dufek JS, Mercer JA. Aerial rotation effects of vertical jump performance among highly skilled collegiate soccer players. J Strength Cond Res 31: 932–938, 2017.
5. Bates BT. Scientific basis of human movement. J Phys Educ Recreation 48: 70, 1977.
6. Bressel E, Cronin J. The landing phase of a jump: Strategies to minimize injuries. J Phys Educ Recreation Dance 76: 30–47, 2005.
7. Caster BL, Bates BT. The assessment of mechanical and neuromuscular response strategies during landing. Med Sci Sports Exerc 27: 736–744, 1995.
8. Chiu LZ, Moolyk AN. Segment kinematics differ between jump and drop landings regardless of practice. J Appl Biomech 31: 357–362, 2015.
9. Cohen J. A power primer. Psychol Bull 112: 155, 1992.
10. Decker MJ, Torry MR, Wyland DJ, Sterett WI, Steadman JR. Gender differences in lower extremity kinematics, kinetics and energy absorption during landing. Clin Biomech 18: 662–669, 2003.
11. Devita P, Skelly WA. Effect of landing stiffness on joint kinetics and energetics in the lower extremity. Med Sci Sports Exerc 24: 108–115, 1992.
12. Dufek JS, Bates BT. Biomechanical factors associated with injury during landing in jump sports. Sports Med 12: 326–337, 1991.
13. Dufek JS, Bates BT. Lower extremity performance models for landing. Hum Mov Sci 11: 299–318, 1992.
14. Dufek JS, Bates BT, Stergiou N, James CR. Interactive effects between group and single-subject response patterns. Hum Mov Sci 14: 301–323, 1995.
15. Edwards S, Steele JR, McGhee DE. Does a drop landing represent a whole skill landing and is this moderated by fatigue? Scand J Med Sci Sports 20: 516–523, 2010.
16. Field A. The beast of bias. In: Discovering statistics using IBM SPSS statistics. Thousand Oaks, CA: Sage, 2013. pp. 174–180.
17. Fong C, Blackburn JT, Norcross MF, McGrath M, Padua DA. Ankle-dorsiflexion range of motion and landing biomechanics. J Athl Train 46: 5–10, 2011.
18. Harry JR, Paquette MR, Caia J, Townsend RJ, Weiss LW, Schilling BK. Effects of footwear condition on maximal jumping performance. J Strength Cond Res 29: 1657–1665, 2015.
19. Harry JR, Barker LA, Mercer JA, Dufek JS. Vertical and horizontal impact force comparison during jump-landings with and without rotation in NCAA division 1 male soccer players. J Strength Cond Res 31: 1780–1786, 2017.
20. Iida Y, Kanehisa H, Inaba Y, Nakazawa K. Short-term landing training attenuates landing impact and improves jump height in landing-to-jump movement. J Strength Cond Res 27: 1560–1567, 2013.
21. James CR, Bates BT, Dufek JS. Classification and comparison of biomechanical response strategies for accommodating landing impact. J Appl Biomech 19: 106–118, 2003.
22. Judelson DA, Maresh CM, Farrell MJ, Yamamoto LM, Armstrong LE, Kraemer WJ, Volek JS, Spiering BA, Casa DJ, Anderson JM. Effect of hydration state on strength, power, and resistance exercise performance. Med Sci Sports Exerc 39: 1817, 2007.
23. Kernozek TW, Torry MR, Iwasaki M. Gender differences in lower extremity landing mechanics caused by neuromuscular fatigue. Am J Sports Med 36: 554–565, 2008.
24. Lephart SM, Ferris CM, Riemann BL, Myers JB, Fu FH. Gender differences in strength and lower extremity kinematics during landing. Clin Orthop 401: 162–169, 2002.
25. Makaruk H, Porter JM, Czaplicki A, Sadowski J, Sacewicz T. The role of attentional focus in plyometric training. J Sports Med Phys Fitness 52: 319–327, 2012.
26. McCaw ST, Bates BT. Biomechanical implications of mild leg length inequality. Br J Sports Med 25: 10–13, 1991.
27. Moir GL. Three different methods of calculating vertical jump height from force platform data in men and women. Meas Phys Educ Exerc Sci 12: 207–218, 2008.
28. Mothersole G, Cronin JB, Harris NK. Jump-landing program for females: Development of a systematic progression model. Strength Cond J 36: 52–64, 2014.
29. Norcross MF, Lewek MD, Padua DA, Shultz SJ, Weinhold PS, Blackburn JT. Lower extremity energy absorption and biomechanics during landing, part I: Sagittal-plane energy absorption analyses. J Athl Train 48: 748–756, 2013.
30. Puddle DL, Maulder PS. Ground reaction forces and loading rates associated with parkour and traditional drop landing techniques. J Sports Sci Med 12: 122–129, 2013.
31. Rowley KM, Richards JG. Increasing plantarflexion angle during landing reduces vertical ground reaction forces, loading rates and the hip's contribution to support moment within participants. J Sports Sci 33: 1922–1932, 2015.
32. Schot PK, Bates BT, Dufek JS. Bilateral performance symmetry during drop landing: A kinetic analysis. Med Sci Sports Exerc 26: 1153–1159, 1994.
33. Shetty AB, Etnyre BR. Contribution of arm movement to the force components of a maximum vertical jump. J Orthop Sports Phys Ther 11: 198–201, 1989.
34. Vincent W, Weir J. Statistics in Kinesiology. Champaign, IL: Human Kinetics, 2012. pp. 181.
35. Welling W, Benjaminse A, Gokeler A, Otten B. Enhanced retention of drop vertical jump landing technique: A randomized controlled trial. Hum Mov Sci 45: 84–95, 2016.
36. Wong del P, Chaouachi A, Lau PW, Behm DG. Short durations of static stretching when combined with dynamic stretching do not impair repeated sprints and agility. J Sports Sci Med 2: 408–416, 2011.
37. Wulf G, McNevin N, Shea CH. The automaticity of complex motor skill learning as a function of attentional focus. Q J Exp Psychol A 54: 1143–1154, 2001.
38. Zhang SN, Bates BT, Dufek JS. Contributions of lower extremity joints to energy dissipation during landings. Med Sci Sports Exerc 32: 812–819, 2000.
