Preventive Training Program Feedback Complexity, Movement Control, and Performance in Youth Athletes

Youth soccer teams between the under-8 and under-14 age groups from 4 local soccer organizations were invited to participate in this study. We used a subgroup analysis of a cluster-randomized controlled trial combined with a prospective time-series panel study design to measure changes from the intervention (trial) and retention (panel) in movement control and LJ performance.¹⁹  Movement control and LJ performance were assessed in a convenience sample of participants, defined as athletes who were present and able to engage in physical activity on the day of testing at each of the 4 time points: before (PRE) and after (POST) the 8-week intervention period and approximately 6 weeks (R1) and 12 weeks postseason (R2; Figure 1). Postseason testing time points occurred over several days and times at the time points (R1 and R2) and by organization based on participant and team availability. Thus, the postseason testing sessions (R1 and R2) may have occurred at roughly 6 and 12 weeks. The university's institutional review board approved this study before league recruitment.

A total of 28 soccer teams, consisting of both male and female athletes, agreed to participate in the study and be randomized into 1 of the 3 groups. All youth athletes on each team were recruited to participate in the study during team informational meetings. Players and parents or legal guardians read and completed assent and consent forms, respectively, before data collection. All participants were free of any injury or illness that prohibited soccer activity at the times of testing. All athletes on every team completed their team's assigned warmup intervention, but only those who consented to participate in the study were tested. Attendance was not taken at each warmup implementation session, so individual athlete-exposures and program dosages were not recorded. At an organizational level, all teams and athletes had the opportunity to complete the intervention for 8 weeks.

The teams were stratified by age and sex and then randomized into 1 of 3 intervention arms: the PTP with simplified feedback cues (primarily in the sagittal plane; 10 teams), the PTP with traditional feedback cues in all planes of motion (9 teams), or the control (9 teams). Teams assigned to the simplified or traditional program were assigned a research assistant who implemented the 10- to 12-minute PTP (Table 1) before every practice (at least 2 to 3 times per week depending on the team's schedule) for the 8-week intervention. Research assistants were assigned to the simplified and traditional programs to ensure consistency due to the novelty of the feedback strategies.

Research assistants completed training on PTP implementation led by an investigator (L.J.D.) in a single, in-person, 2-hour session. Training included how to physically perform the program exercises as well as how to implement the program and provide appropriate movement cues based on the team's assigned group (simplified or traditional). Research assistants assigned to each team administered the PTP warmup and provided oral feedback and technique instruction to ensure program fidelity. No crossover between research assistants and treatment arms occurred. Teams assigned to the control group performed their normal warmup as determined by their coaches. Control teams were supervised once a week by a research assistant who recorded the warmup components to account for any similarities between the control warmup and the simplified or traditional PTP. Teams were never observed completing any standardized exercises as a team warmup other than warmup cardiovascular laps around the field. The research assistant did not implement the warmups for the teams assigned to the control arm.

The PTP was modified from previous anterior cruciate ligament (ACL) injury-prevention programs that have been successful for youth populations in decreasing injury risk^2,20  and was designed to be a dynamic warmup lasting 10 to 12 minutes (Table 1). The simplified and traditional PTPs consisted of identical flexibility, balance, strengthening, agility, and plyometric exercises and differed only with respect to the oral cues provided by the research assistant during the warmup. For the simplified PTP, research assistants corrected and cued only the movement technique in the sagittal plane (eg, “get low,” “bend your knees!”). The traditional-feedback PTP research assistants provided corrections and cues for all planes of motion (eg, “point your toes straight ahead,” “don't let your knees cave inward”).

Participants completed 3 trials of a standardized jump-landing task during each test session. During the task, participants jumped forward from a 30-cm-high box at a distance half their height and immediately rebounded straight in the air for maximum vertical height. Movement control during the jump-landing task was evaluated using the Landing Error Scoring System (LESS). The LESS is a valid and reliable clinical movement assessment that can identify youth athletes at low risk for sustaining an ACL injury.²¹  All trials were video recorded from the frontal and sagittal planes. A single reliable rater graded all trials at a later time point. The rater was trained by an expert and blinded to the group assignment of the participants. The total scores from each of the 3 trials were averaged for 1 composite LESS score at each time point. The LESS score is a summation of movement errors displayed, in which a higher score indicates more movement errors and a lower score indicates fewer movement errors during the jump-landing task.

Standing LJ performance is considered a measurement of power.²²  Participants began the LJ by standing behind a marked line. They had to remain standing but could move into a self-determined semi-squat position to jump horizontally as far as possible. Each person performed 1 practice and 2 recorded trials. Trials were repeated if the individual fell or slipped during the trial. Distance jumped was recorded to the nearest centimeter from the starting line to the closest body part on the ground manually using a standard measuring tape. The 2 trials were averaged for 1 composite score at each time point.

A repeated-measures design in which the same participants are tracked longitudinally over time is not ideal in large youth sport or educational settings due to both the reality of multiple testing time points in a youth population and the concerns associated with the ability of such a design to account for heterogeneous variability.^23–25  Instead, we used a time-series panel design, in which the groups were compared using a cross-sectional approach at each assessment point.^19,25  The panel design is logistically simpler to implement with large samples of children, eliminates bias from loss to follow-up, maximizes power, and reduces the potential for a learning effect. This design is statistically efficient when within-subject correlations are low and has been used in studies conducted in educational and military settings.¹⁹ 

Potential differences in age and sex among groups at PRE were evaluated using a 1-way analysis of variance and χ² test of association, respectively. We calculated change scores for the LESS and LJ variables for each group (simplified, traditional, and control) for each time point compared with baseline (POST-PRE, R1-PRE, and R2-PRE). This baseline comparison among groups ensured that any group effects observed during the follow-up time points were not due to preexisting group differences.

For the panel design, data from all participants who completed testing at each time point were included in each individual time-point analysis. We performed separate univariate analyses of variance with a Bonferroni correction to determine whether group differences (simplified, traditional, and control) existed for LESS scores and LJ performance at each of the 3 follow-up time points (POST, R1, and R2). If a significant group effect was observed, we conducted post hoc pairwise comparisons by evaluating 95% CIs. All data were analyzed using SPSS (version 21.0; IBM Corp) with a prior α level of .05.

All 28 teams completed the interventions as assigned. A total of 420 athletes (simplified group = 173, traditional group = 118, and control group = 129) volunteered to participate in the test sessions. All groups were similar at baseline for sex, age, height, and mass (P values > .05; Table 2).

No group differences were present at POST (P = .07), R1 (P = .557), or R2 (P = .483). The change-score analysis revealed improvements in the LESS score at POST for all 3 groups compared with PRE. Improvements from PRE were also present at R1 and R2 for both intervention groups (simplified and traditional; Table 3; Figure 2).

The traditional group jumped further on the LJ than both the control and simplified groups at POST (P < .001) and further than the control group at R1 (P = .049; Table 3; Figure 3). The traditional group demonstrated improved LJ performance from PRE to POST. Long-jump performance decreased for the simplified and control groups at R1 compared with PRE and for all 3 groups at R2 compared with PRE.

This study was a comparative efficacy trial to evaluate whether a simplified approach to PTPs elicited greater immediate and retained improvements in movement control and LJ performance in youth athletes. The population was youth soccer athletes between the ages of 10 and 12 years. These ages reflect a critical window of opportunity for promoting appropriate motor development.²⁶  In addition, establishing PTPs as a norm in sport participation at a young age may greatly promote long-term acceptance of these programs when injury risk increases in the later ages of adolescence. Our results showed that PTPs, delivered as either a traditional or simplified approach, were effective in improving movement control, as measured by the LESS, with effects persisting as long as 12 weeks after completion of the program. Although these improvements were not different than those from a warmup of the coaches' choosing, PTP use at this young age may be critical for building buy-in and acceptance of these types of warmup strategies given that the benefits of PTPs in adolescence for reducing injury risk are clearly established.^27–29 

Coaches represent the best option for consistent, long-term use of PTPs, particularly at the youth level; however, high program complexity can be a barrier to implementation,³⁰  and coaches have reported that knowing how to give adequate feedback to athletes on injury-prevention techniques is a challenge.³¹  Motor learning can be affected by the type of feedback provided, which can differ based on the focus of attention (eg, internal or external) and complexity.³²  Feedback that is too complex may be challenging for children and counterproductive to learning.¹²  Our findings indicated that a PTP implemented with simplified feedback cues to athletes was as successful as a PTP with a wide range of feedback cues in eliciting movement-control improvements, as measured by the LESS. Youth sport coaches are capable of effectively implementing PTPs after a workshop,^33,34  and authors of future studies should evaluate the training of coaches to use a streamlined set of feedback cues, which would simplify both the coaches' training as well as daily program implementation for the athletes.

Although the PTP programs for movement control produced similar outcomes, the traditional program elicited greater changes in LJ performance than the simplified program and the control group. The LJ is a measure of power and muscular fitness.²²  Thus, to achieve performance benefits, training programs must incorporate the progressive loading needed to achieve strength gains.³⁵  As performance gains declined more quickly for the simplified and control groups than the traditional group, it is likely that the athletes in these groups did not incur enough overloading to cause gains in power performance. Athletes in the traditional program saw declines but over a longer period, indicating that this program elicited more sustainable changes in performance than the other programs. Past researchers^36,37  demonstrated that traditional injury-prevention programs result in improved performance measures, including strength, power, coordination, posture, balance, and agility. The traditional and simplified programs we used consisted of similar types of exercises; the former were more challenging, with demands across all planes of motion. These challenges appear to be important, even with a short dynamic warmup exercise program, for the neuromuscular improvements that produce power improvements in youth athletes. Only the traditional PTP group improved from PRE to POST and then steadily declined at R1 and R2. Earlier authors^38,39  evaluated the effect of PTPs on injury risk or sport performance from preseason to postseason. We offer additional evidence that implementation strategies need to include maintenance plans beyond a single sport season to maintain performance benefits. Further, coaches may be more interested in performance improvements than injury-prevention benefits.⁴⁰  Therefore, to generate initial buy-in for long-term implementation, it may be important to consider which program strategy facilitates greater performance changes.

We applied an innovative time-series panel design for sports medicine. This design involves a cross-sectional approach at each follow-up test point and includes all available participant data. This approach has been advocated for use in educational and military settings, as it reduces follow-up bias, maximizes power, and provides a snapshot of the population at each test point.¹⁹  A classic approach in sports medicine for this type of study would be a repeated-measures design in which only the participants who completed all 4 test points would be included in the analyses. With this method, significant data (approximately 36% to 50%) would have been lost to attrition. Field studies of large samples of children are greatly needed in sports medicine to understand the true effects of various interventions. However, field testing carries scheduling challenges at the organization, team, and individual level, particularly in the context of volunteer coaches and young athletes who may have variable levels of attendance. Tracking individual athletes' program dosages and recording the specific duration between 2 testing time points are important considerations for understanding effectiveness. We encourage investigators to consider using a time-series panel approach in future studies of large samples of children.

To our knowledge, this is the first investigation to evaluate the effect of simplified cues during a PTP on movement control and sport performance. Given the novelty of the research question, we controlled program adoption, fidelity, and compliance by relying on trained research assistants to implement the program. A limitation of this work was that we did not track athlete-level attendance for each warmup session, so the individual dosages for each program are unknown. Further, multiple testing time points were possible. For each of the 4 time points, testing was conducted over multiple (1 to 4) days to provide ample opportunity for athletes to participate. For example, retention testing for 1 organization took place over 2 different days, which could have influenced the number of days between the last program exposure and testing of participants within the same warmup group. This should be considered in the context of our findings and the absence of large group differences. Even so, our results still demonstrated improvements across time, regardless of the slight deterioration from R1 to R2. This indicates that, at the simplest level, PTP programming at the team level improved movement control. Program maintenance is important, and future researchers must determine a threshold dosage for individual athletes to maximize long-term PTP effectiveness. Subsequent researchers can now evaluate the effectiveness of simplified movement cues on movement control and jump performance. In addition, although having 2 retention time points was a strength of the design necessary to measure true motor learning, participant attrition occurred at each time point, as follow-up communication with athletes was more challenging after the season ended. Future directions for this work should address mechanisms to reduce attrition between time points. Moreover, the PTPs were overseen by investigators, and thus, it will be important to identify whether these results can be replicated when the program is implemented by the end users (eg, coaching staff).

The delivery of PTPs using simplified sagittal-plane cueing strategies provided movement-quality improvements similar to those of a traditional PTP that supplied feedback in the frontal, sagittal, and transverse planes. These findings highlight the capability of simplified PTPs to directly influence movement control in younger athletes who may be overwhelmed by too many feedback cues. Authors of future studies should evaluate the effectiveness of such programs when they are delivered autonomously by coaches as well as the effects of other PTP adaptations in a real-world context.