1 Introduction

Approximately 75% of children and adolescents in North America participate in some form of organized sport (The National Federation of State High School Association, 2019). Unfortunately, more sports-related injuries are being reported and account for approximately 66% of all injuries in this population (Werner et al., 2016). Of those injuries, more than 50% occur in the lower extremities, with the knee and ankle accounting for many of these injuries that often lead to rehabilitation and, in some cases, surgical intervention (Hootman et al., 2007). Moreover, the risk for a subsequent injury is higher after a primary injury (Salmon et al., 2005). Despite the impact of these injuries, and considering adolescent females have a 1.5 time higher risk of a season-ending lower extremity injury compared to boys (Fernandez et al., 2007), there is surprisingly very limited sex-specific research on functional testing in an adolescent population.

Single-limb hop testing is one of the most common assessments to determine post-injury, post-surgical, and post-recovery/rehabilitation return-to-activity readiness in the lower extremities (Cristiani et al., 2019; Mattacola et al., 2002; Sharma et al., 2011). Four single-limb hop tests were developed and have been shown to be sensitive indicators of post-injury deficits in adult males (Barber et al., 1990; Noyes et al., 1991). Multiple authors have also established reliability (Dingenen et al., 2019; Ross et al., 2002); however, these single-limb hop tests have yet to be evaluated in an adolescent population and between sexes. It is also crucial to consider the environment in which functional performance is assessed during physical activities. Current testing is based in a closed environment such as a laboratory or clinician’s office. These closed environments may not challenge the patient in the same way as an open environment, which is especially important when considering the environmental context within which actual activities and sports take place (Davies et al., 2017). Several laboratory-based experiments have confirmed an increased injury risk when movements are performed with additional cognitive or visual stimuli (Borotikar et al., 2008; Brown et al., 2009); however, typical functional testing does not incorporate different environments, which may limit a true assessment of the patient’s performance. It is therefore important to ensure these tests can be performed outside a laboratory setting by measuring and establishing the reproducibility due to the varying locations and responsiveness of the scores within adolescent males and females to detect real and potentially clinically important changes over time.

Although limb symmetry indices (LSI) are commonly used to assess the functional capacity in adults with and without injuries (Cristiani et al., 2019; Johnson & Stoneman, 2007); there is limited evidence for its use in adolescent populations. Most studies calculate the LSI by dividing the dominant limb by the non-dominant limb’s score with an accepted threshold of 0.90 indicating symmetry (Barber-Westin & Noyes, 2011; Munro & Herrington, 2011). However, some studies have begun reporting lower thresholds in adolescents (Girard et al., 2020), and differences in asymmetry direction based on the task (Mulrey et al., 2020). This highlights the need to further examine this population’s performance during functional testing.

Injury recovery among adolescents differs from adults, given the complex multifactorial nature of puberty (LaBella et al., 2014). As such, when evaluating an adolescent patient’s recovery from injury, it is important to have insight into the functional capacity with respect to their age, sex, maturation state, and activity-matched cohort since it is with those individuals that they will be returning to full activity once they have recovered from their lower extremity injury. It is also important to ensure these tests can be performed outside the laboratory in a less controlled environment such as a gymnasium during a physical activity class and provide similar results. Therefore, the purpose of this study was to: i) assess the reproducibility and responsiveness of a battery of single-limb hop tests in two environments (open: physical activity class; and, closed: laboratory) for an uninjured adolescent population, and ii) determine whether there are differences in baseline measures between adolescent males and females. Once test-retest reproducibility has been established in an adolescent population, these measures can be used in various environments by clinicians, strength and conditioning professionals, and researchers alike.

2 Methods

3 Results

Participants’ demographics including age, height, body mass and puberty stage did not differ (p > 0.05) between sexes and were (mean and standard deviation [SD]): 12.0 years (SD = 0.3), 1.58 m (SD = 0.08), 50.1 kg (SD = 12.6), and stage 3 (SD = 1), respectively. All participants were physically active (16 [SD = 6] and 23 [SD = 5] for females and males respectively). Male participants scored on average higher than females on the HSS/FR Pedi-FABS questionnaire (p < 0.01) and were more active (p < 0.001) while no other differences existed in the population’s demographics. As such, hop performances (Table 1) were not normalized to the participants’ height.

4 Discussion

The objectives of this study were to: i) determine whether a battery of single-limb hop tasks could be reproducible and responsive in an adolescent population, and ii) determine whether sex differences existed. This study demonstrated that this battery of single-limb hop tasks is a highly reproducible and responsive method to assess the functional capacity of adolescents and their performances in both a laboratory and during a regularly scheduled physical education class. This study also found important sex differences that must be considered when evaluating functional capacity in adolescents.

This is the first study to determine the test-retest reproducibility of the anterior, lateral, triple, cross and 6-meter timed hop tests in an adolescent population. In line with previous studies on adults, the test-retest reproducibility coefficients for each of the single-limb hop tests were moderate to good (> 0.60) (Ross et al., 2002). In addition, the 95% CI for all tests was narrow (Lai & Kelley, 2011), indicating good agreement. The ICC provides information about the reproducibility between two or more locations (Fleiss, 2011), and includes the variance term for individuals and is therefore affected by sample size heterogeneity, meaning that high correlations can still mean unacceptable measurement error (Atkinson & Nevill, 1998). In the present study, we further analyzed the absolute agreement, which is unaffected by the range of measurements using the SEM (Abrams et al., 2014; Lexell & Downham, 2005). The smaller the SEM, the more reliable and reproducible the measurement, often used to interpret the results of a true improvement after an intervention. The SRD also allows clinicians to determine whether a change in hop distance or time reflects a true change, rather than irrelevant variations when repeatedly testing an individual. The SRDs in Table 2 were 18.5-35.9% of the mean of the group, and similar across the different hop tests. This suggests that an improvement between 18.5-35.9% reflects a true change in performance when retesting an adolescent individual. These percentages are slightly higher than previous studies evaluating the test-retest of the anterior, triple, cross and timed hop (Dingenen et al., 2019; Munro & Herrington, 2011; Ross et al., 2002). Our study population is however comprised of recreationally active adolescents performing in both a closed and open environment suggesting a higher variability in SEMs and SRDs may be more reflective of this population.

When establishing baseline measures for an uninjured adolescent population, the LSI, typically used as an outcome measure in return-to-activity guidelines following an injury to the lower extremity (Kivlan & Martin, 2012; Schmitt et al., 2015; Toole et al., 2017), must be examined. Most studies calculate LSI in uninjured individuals by dividing the score of the dominant limb by the non-dominant limb’s score and obtain an average of >90% (Munro & Herrington, 2011; van Grinsven et al., 2010). Munro & Herrington (2011) found a mean LSI between 98.4-101.6% for the original set of four single-limb hop tests, which is in line with the results of our study (93.5-102.6% and 94.4-101.7% for females and males, respectively). However, a mean LSI of 100% or greater may mask the asymmetries of adolescents within an uninjured group as the dominant limb used in tasks requiring both limbs such as kicking a ball, may not be the same in unilateral tasks such as jumping and landing (van Melick et al., 2017). Approximately 44% (range of 22-66.7% depending on jumping task) of adolescents scored higher on their self-reported non-dominant limb compared to their dominant limb on any given single-limb hop task

To further establish baseline measures for an adolescent population, it is important to ensure that the number of practice and test trials, and location do not influence the results. Ross et al. (2002) examined the test-retest reliability of the anterior, triple, cross and 6-meter timed hops in eighteen adult cadet males and had three practice trials followed by three test trials with a four-week interval between testing days. Others had 1-3 practice trials and 2-4 test trials with a one-week interval between testing days (Dingenen et al., 2019; Munro & Herrington, 2011). All studies consistently revealed significant improvements in scores between sessions that could be attributed to a learning effect. Our study used two practice trials and three test trials with a two-week interval between sessions and found significant improvements in all hops except for the 6-meter timed hop on both limbs. However, no significant differences in distances or times were found between locations except for the triple hop on the dominant limb. Although significant improvements from the first to the second testing session were found, we believe the lack of significance between locations in our adolescent population suggests that the battery of hop tests can be performed within acceptable limits of error in a gymnasium and laboratory. This learning effect should be taken into consideration within this population and further examined over repeated monthly intervals for 6-12 months or the length of a rehabilitation program for a serious lower extremity injury such as an anterior cruciate ligament injury. The repeated testing of these hopping tasks in adolescents would be beneficial since many rehabilitation programs use the patient’s performance at different phases of rehabilitation to determine readiness to return to full activity.

Finally, the results of our study indicated there were sex differences within the scores of the hop tasks regardless of location. Female participants improved their performance in the laboratory compared to the gymnasium except for the cross-hop, whereas male participants scored higher for all hopping tasks except the timed 6-meter hop in the gymnasium. While it is difficult to speculate on why males and females responded differently, factors that may contribute include the psychosocial and physical nature associated with participation in different sports and activities. For example, females are more likely to engage in individual sports and activities, whereas males typically engage in team sports or group activities (Deaner et al., 2012), and the data collections done in the gymnasium were conducted in a group setting. These results suggest that the functional capacity of males and females should be evaluated separately when applying scores from one environment to another.

5 Additional Information

6 References