364 ORIGINAL ARTICLE Support for a Reduction in the Number of Trials Needed for the Star Excursion Balance Test Richard H. Robinson, MA, Phillip A. Gribble, PhD ABSTRACT. Robinson RH, Gribble PA. Support for a reduction in the number of trials needed for the Star Excursion Balance Test. Arch Phys Med Rehabil 2008;89:364-70. Objective: To determine the number of trials necessary to achieve stability in excursion distance and stance leg angular displacement for the 8 directions of the Star Excursion Balance Test (SEBT). Design: One-way repeated-measures analysis of variance. Setting: Athletic training laboratory. Participants: Twenty participants (10 men, 10 women) without any known musculoskeletal injuries or neurologic deficits that could have negatively affected their dynamic balance volunteered for the study. Intervention: Participants completed 6 practice and 3 test trials in each of the 8 reach directions of the SEBT. Main Outcome Measures: Excursion distances of the reaching leg normalized to leg length and angular displacement at the hip and knee of the stance leg in all 3 planes of movement were determined. Results: There were significant increases in excursion distance, hip flexion, and knee flexion for 7, 4, and 5 of the 8 reach directions, respectively. Conclusions: For the majority of the reach directions, maximum excursion distances and stance leg angular displacement values achieved stability within the first 4 practice trials, thus justifying a reduction in the recommended number of practice trials from 6 to 4 and supporting the trend toward simplifying SEBT administration. Key Words: Kinematics; Measurement; Musculoskeletal equilibrium; Rehabilitation. 2008 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation DYNAMIC POSTURAL CONTROL IS among the foundational components that underlie the performance of movement skills, and, as a consequence, deficits in dynamic postural control can hinder movement skill performance. 1 Therefore, valid and reliable assessment of dynamic postural control is necessary to determine if it represents a limiting factor in a person s movement skill performance. 1 The Star Excursion Balance Test (SEBT), a multidirectional test of dynamic postural control, involves balancing on 1 leg and reaching maximally with the other leg in 8 different directions including 3 anterior, 2 lateral, and 3 posterior. 2 The From the Department of Kinesiology, University of Toledo, Toledo, OH. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Phillip A. Gribble, PhD, College of Health Science and Human Service, Department of Kinesiology, University of Toledo, Mail Stop: 119, 2801 Bancroft St, Toledo, OH 43606, e-mail: phillip.gribble@utoledo.edu. 0003-9993/08/8902-00315$34.00/0 doi:10.1016/j.apmr.2007.08.139 particular advantage of the SEBT over other dynamic balance tests is that it is more challenging for healthy, athletic populations. 3 The SEBT is considered sensitive to functional deficits that are related to musculoskeletal injuries like chronic ankle instability, 4-7 anterior cruciate ligament reconstruction, 8 and patellofemoral pain, 9 as well as those associated with fatigue. 4,7 Intratester 10,11 and intertester reliability 10 has been established with intraclass correlation coefficients between.67 to.96, depending on the reach direction. The recommended SEBT protocol can be time-consuming for clinicians to administer because it requires 9 trials (6 practice, 3 test trials) for each of the 8 directions. For this reason, research has been undertaken to determine if the SEBT administration can be simplified without diminishing the validity of the test. Factor and correlational analyses of the SEBT have revealed that all 8 directions are associated with a single factor and are significantly correlated. 5 These findings were interpreted as indicating that there was functional redundancy across the 8 reach directions and led to the recommendation that the number of reach directions administered could be decreased. The recommended number of trials for SEBT protocol is based on the results of Hertel et al 10 who identified learning effects in 4 of the 8 reach directions (lateral, posteromedial, posterolateral, posterior), with the longest excursion distances occurring during trials 7 through 9 for all reach directions. However, Hertel administered 4 blocks of 3 trials on 2 separate days. In addition, subjects in the Hertel study were free to use their arms for balance, whereas the recommended protocol requires that the hands be placed on the hips. Allowing arm movements increases the biomechanic degrees of freedom that have to be coordinated and may increase the time necessary to learn a task. 12 Therefore, the first purpose of the study was to administer the recommended SEBT protocol of 6 practice and 3 test trials and determine when normalized maximum excursion distance for all 8 reach directions no longer showed a significant change (eg, performance stability). If the findings support a reduction in the number of trials, then this, along with a decrease in the number of reach directions tested, could substantially streamline the administration of the SEBT by clinicians. Given the functional redundancy of the human movement system, 12 the same performance (eg, normalized maximum excursion distance) can be achieved with different angular displacement combinations; thus, collecting angular displacement data provides a way of determining if movements have stabilized or if the participants are continuing in a search mode. 13 Variability in movement patterns has been shown to precede changes in performance 14 ; thus, the reliability of measurement is bolstered by evidence that movement patterns have also stabilized. Previous kinematic studies of the SEBT were limited in that they only examined sagittal plane angular displacement of the stance leg at the normalized maximum excursion distance in 3 of the 8 reach directions. 4,7 Because the SEBT requires reaching movements in 8 different directions, it seems reasonable to examine frontal and transverse plane movements of the stance leg. Therefore, the second purpose of
DYNAMIC POSTURAL CONTROL MEASUREMENT, Robinson 365 Fig 1. (A) Anterior reach direction, (B) posteromedial reach direction, and (C) posterolateral reach direction. this study was to measure changes in sagittal plane displacement across trials at the knee and sagittal, frontal, and rotational displacement at the hip of the stance leg during the performance of all 8 reach directions to determine when movements have stabilized. METHODS Participants Twenty participants (10 women; age, 21.5 3.3y; height, 1.65 0.07m; weight, 62.2 16.3kg; 10 men; age, 23.2 3.3y, height, 1.79 0.08m; weight, 80.4 16.5kg) volunteered for the study and signed a university-approved informed consent form. Nineteen participants were undergraduate or graduate students in the department of kinesiology. Four participants were intercollegiate athletes (3 played women s basketball, 1 played men s hockey). Participants did not have any known musculoskeletal injuries or neurologic deficits that could have negatively affected their dynamic balance. Protocol Participants reported to the laboratory for a single testing session. The stance leg was determined to be the opposite of a participant s self-reported kicking leg. The length of the stance leg was measured from the anterosuperior iliac spine to the middle of the medial malleolus by using a standard tape measure while participants lay supine on a plinth. Leg length was used to normalize excursion distances. 3 We collected hip and knee angular displacement data by using an electromagnetic tracking system a and MotionMonitor software. b We attached electromagnetic sensors with self-adhesive (Velcro) straps to the participant s sacrum, the lateral midthigh, and the lateral midshank of their stance leg. Anatomic landmarks on the hip, knee, and ankle were digitized with a stylus attached to a fourth electromagnetic sensor in order to create a digitial representation of the stance leg and pelvis. The task goal and performance constraints of the SEBT were explained to participants. Participants completed 6 practice (P1 P6) and 3 test trials (T1 T3) in each of the 8 reach directions with 2 minutes of recovery between reach directions, practice trials, and test trials. The order of the reach directions was randomized. SEBT Performance The SEBT instrument was created by projecting and securing 8 tape measures at 45 o to each other from the center of a circle. Figures 1A, B, and C show 3 of the reach directions, respectively: anterior, posteromedial, and posterolateral. Participants placed the foot of their stance leg in the middle of the SEBT instrument and aligned it with the tape measures that indicated the anteroposterior (AP) reach directions. As a guide for maintaining foot position, marks were made behind the heel and in front of the toes on the AP tape measures. Successful trials required that hands had to remain on hips, the foot position of the stance leg had to remain as originally positioned, and the heel of the stance leg had to stay in contact with the floor. Participants were instructed to make a maximum reach with the opposite leg in a specific reach direction, a light touch on the floor with the most distal part of the reaching foot, and a return to a double-leg stance without changing the base of support of the stance leg. If these criteria were not met, the trial was discarded, and an additional trial was performed. Reach distances were recorded by having one of the testers place a mark on the tape measure that corresponded to the touchdown point. During the 6 practice trials, the mark was erased before the next trial was performed, but during the 3 test trials the marks of previous trials remained. Angular Displacement Data Collection We collected angular displacement data for the stance leg at 100Hz with the electromagnetic tracking system. The Motion- Monitor software smoothed the data with a Butterworth filter set at 20Hz and calculated hip flexion, hip rotation, hip abduction, knee flexion, and knee rotation for the stance leg. The point of touchdown (maximum excursion distance) was indicated by having a tester depress an electronic trigger, which placed an event marker in the angular displacement data. Data Reduction The excursion distances were normalized by dividing by a participant s leg length and multiplying by 100 (normalized maximum excursion distance). 3 The hip flexion, hip rotation, hip abduction, knee flexion, and knee rotation data for the stance leg were displayed by using the MotionMonitor soft-
366 DYNAMIC POSTURAL CONTROL MEASUREMENT, Robinson Table 1: Results of Repeated-Measures ANOVA for Normalized Maximum Excursion Distance Reach Direction F Ratio* P Effect Size Anterior 4.69.001.21 Anterolateral 3.45.001.16 Anteromedial 1.77.088.09 Lateral 6.04.001.25 Medial 3.85.001.18 Posterior 3.24.002.16 Posterolateral 5.13.001.22 Posteromedial 3.27.002.15 *Degrees of freedom (df): 8,144. ware, and the values that corresponded with touchdown of the reaching leg, as indicated by the event marker, were recorded. Statistical Analysis We administered separate, 1-way repeated-measures analyses of variance (ANOVAs) c for each reach direction and joint displacement variable. When statistical significance was achieved, pairwise comparisons were made by using the least significant difference method. Significance for statistical tests was set a priori at P less than.05. RESULTS The results for normalized maximum excursion distance and angular displacement data will be presented separately. Human error during data collection and processing resulted in a few trials being lost and caused the discrepancies in degrees of freedom for F ratios. This resulted in 1 (anterolateral and posteromedial angular displacement variables) and 2 participants (posteriorlateral angular displacement variables) being eliminated from the statistical analyses for those reach directions. Normalized Maximum Excursion Distance Table 1 presents the results of repeated-measures ANOVA for normalized maximum excursion distance data. Table 2 presents means, standard errors, and indicates in which significant differences between trials occurred for normalized maximum excursion distance data. Every reach direction except anteromedial showed a significant increase in normalized maximum excursion distance across trials. In 6 of the 8 reach directions, P1 represented the shortest normalized maximum excursion distance, and in the other 2 reach directions (posterior and anteromedial) T1 was the shortest. In the 7 reach directions that showed statistical significance, T1 was shorter than T2 and T3. For the practice trials, the pattern of change in normalized maximum excursion distance was for the first trial to be the shortest and for normalized maximum excursion distance to stabilize after 2 or 3 additional trials. Only the lateral reach direction required more than 4 practice trials before stability was achieved (P5). Angular Displacement Table 3 presents F ratios and effect sizes for the angular displacement variables that showed significant differences. Tables 4 through 7 present means and standard errors and indicate where significant differences occurred between trials for knee flexion, hip rotation, hip abduction, and hip flexion, respectively. In 5 of the 8 reach directions, knee flexion showed a significant increase across trials (lateral direction approached significance, P.06). Hip flexion showed a significant increase in 4 of the 8 reach directions (medial direction approached significance, P.053). In the reach directions in which knee flexion and/or hip flexion showed a significant change, the first practice trial always exhibited significantly less angular displacement than some or all of the other practice or test trials. Only anterolateral (knee flexion) and posterolateral (hip flexion) required more than 4 trials (P5 and P6, respectively) before stability was achieved. Thus, the angular displacement data mirrored the pattern observed for normalized maximum excursion distance in which performance improved for 1 or 2 additional trials before stabilizing. In only 2 reach directions did hip rotation show a significant change (anterior, anterolateral). In both cases, the trend was for hip rotation to first increase across the practice trials but then be less or the least during the test trials. Only the anterior reach direction showed a significant change in hip abduction, with P1 actually showing less hip adduction than all the other trials except one. DISCUSSION For the majority of the reach directions, maximum excursion distances (normalized maximum excursion distance) and stance leg angular displacement values achieved stability within the first 4 practice trials. Every reach direction except anteromedial Table 2: Means Standard Errors (SEs) for Normalized Maximum Excursion Distance (% of leg length 100) (N 20) Anterior 80.1 2.4 81.3 2.3* 82.5 2.4* 83.2 2.4* 82.7 2.3* 83.1 2.4* 82.0 2.4 83.1 2.4* # 83.6 2.4* # Anterolateral 69.2 2.0 70.3 1.9 71.8 1.8* 72.0 2.2* 71.9 1.9* 71.3 1.9* 70.9 1.9 72.4 2.0* # 73.0 2.0* # Anteromedial 86.4 2.5 87.2 2.7 87.8 2.5 87.9 2.4 87.7 2.4 88.4 2.6 85.7 2.0 87.0 2.5 87.5 2.1 Lateral 62.9 4.2 67.3 4.2* 66.3 4.3* 67.2 4.6* 68.1 4.4* 68.9 4.4* 66.1 4.4 69.2 4.4* # 70.9 4.4* # Medial 88.9 2.4 89.1 2.5 90.0 2.5 91.2 2.6* 91.5 2.5* 92.0 2.6* # 89.6 2.5 90.9 2.7 # 91.4 2.8 # Posterolateral 75.7 4.4 78.6 4.6* 79.1 4.5* 80.6 4.8* 81.1 4.7* 81.5 4.8* 79.1 4.1* 81.1 4.3* # 82.4 4.2* # Posteromedial 89.5 3.3 90.6 3.6 92.5 3.3* 92.1 3.4* 92.4 3.3* 93.3 3.4* 89.8 3.3 92.5 3.4 # 92.9 3.5 # Posterior 87.9 4.3 89.8 4.4* 89.3 4.6 91.3 4.4* 91.9 4.2* 92.2 4.4* 87.7 4.1 90.9 4.2 # 92.1 3.9* # Abbreviations: P, practice trial; T, test trial. Significant difference from P2 (P.05). Significant difference from P3 (P.05). Significant difference from P4 (P.05). Significant difference from P5 (P.05). Significant difference from P6 (P.05). # Significant difference from T1 (P.05).
DYNAMIC POSTURAL CONTROL MEASUREMENT, Robinson 367 Table 3: Statistically Significant Angular Displacement Variables Reach Direction Variable F Ratio P Effect Size Anterior Anterolateral Lateral Medial Posterolateral Posteromedial *df 8,144. df 8,136. df 8,128. Knee flexion 2.07*.042.10 Hip rotation 2.69*.009.13 Hip abduction 2.48*.015.12 Knee flexion 3.23.002.16 Hip rotation 3.21.002.16 Hip flexion 2.43.017.13 Hip flexion 2.46*.017.14 Knee flexion 2.23*.029.11 Knee flexion 3.28.002.17 Hip flexion 4.63.001.22 Knee flexion 3.07.003.15 Hip flexion 2.92.005.15 showed a significant increase in normalized maximum excursion distance across trials. For the test trials, T1 was significantly shorter than T2 and T3 in 7 of the 8 reach directions. Knee flexion and hip flexion exhibited significant increases across trials in 5 and 4 of the reach directions, respectively. Only 3 reach directions total showed significant changes in hip rotation (anterior and anterolateral) and hip abduction (anterior). The relatively rapid achievement of performance stability seems to conflict with the recommendation of Hertel et al 10 that 6 practice trials be performed to reduce the learning effect. The differences observed may be because the protocols administered in the 2 studies were not identical. As discussed previously, Hertel administered 4 blocks of 3 trials on 2 separate days and allowed subjects to use their arms for balance, which increases the biomechanic degrees of freedom that have to be coordinated and may increase learning time. 12 Constraints theory proposes that movement performance emerges from the interaction of task (rules, goals, equipment), environmental (ambient conditions external to the subject), and individual constraints (structural and functional). 15 The relative lack of improvement in normalized maximum excursion distance and angular displacement measures beyond the initial trials may be because the task constraints, specifically, the rules regarding movement performance (eg, hands on hips, maintain heel contact with the floor), limit movement options for increasing reach distance, as the angular displacement results indicate, primarily, to greater knee and hip flexion. Also, the stabilization of SEBT performance may be indicative of an energy minimization bias in which skill acquisition proceeds in the direction of greater movement economy. 16 If such a bias exists, it would conflict with the task goal (eg, make a maximum reach) and may make it necessary during the administration of the SEBT for participants to be consistently reminded to overcome this tendency (eg, make a maximum reach ). The test trials did not occur directly after the practice trials for that particular reach direction. Instead, all practice trials for each reach direction were completed before any of the test trials were administered. Such a protocol probably produced the warm-up decrement or a decrease in performance after a rest period 17 observed between T1 and the other 2 test trials. If clinicians administer the same or similar SEBT protocol in which there is a substantial rest period between practice and test trials for the same reach direction, then they should probably disregard the first test trial when determining a patient s score. The results of the current study along with the findings of Hertel et al 5 on the relationship between reach directions provides clinicians with further evidence that the administration of the SEBT can be streamlined and remain a valid test of dynamic postural control. Hertel reported that all reach directions loaded on a single factor and that reach directions were significantly correlated, which led to a conclusion of functional redundancy across the reach directions and a recommendation that the number of reach directions administered could be decreased without compromising the validity of the test. Specifically, Hertel recommended testing the anteromedial, medial, and posteromedial reach directions because they showed greater sensitivity to the functional deficiencies of those participants with chronic ankle instability. Participants in the current study did not complete warm-up exercises before performing the SEBT. Given the apparent importance of hip and knee flexion range of motion to reach distance, perhaps the inclusion of warm-up exercises could reduce the number of practice trials ( 4 trials) needed to achieve performance stability. A reduction in practice trials and reach directions tested will make the Table 4: Means SEs (deg) for Knee Flexion (N 20) Anterior 45.21 4.86 48.76 4.55 50.95 4.55* 48.43 4.75 49.15 4.38 49.28 4.99 51.25 5.06* 51.65 5.07* 52.02 5.02* Anterolateral 32.14 5.29 33.51 5.30 35.79 5.37* 36.22 5.36* 38.17 5.32* 33.44 5.45 38.02 5.50* 37.74 5.78* 39.86 5.58* Anteromedial 45.77 5.13 47.48 5.81 48.88 5.54 50.23 5.81 47.84 5.19 50.29 5.61 50.75 4.52 52.60 5.06 50.52 5.31 Lateral 30.72 3.61 34.70 4.68 31.93 4.41 34.40 4.76 34.52 4.56 33.81 4.25 33.47 4.85 36.40 4.94 35.02 5.02 Medial 50.74 4.93 53.54 5.16 53.56 5.09 55.70 5.40* 56.61 5.21* 55.72 5.21* 54.90 5.05 55.90 4.92* 56.80 5.29* Posterolateral 30.23 5.87 32.92 6.03 32.72 5.83 36.26 5.55* 35.14 5.13* 35.46 6.22* 37.39 5.48* 39.84 5.77* 38.33 5.75* Posteromedial 39.19 5.45 46.43 6.07* 47.73 5.42* 48.02 5.92* 47.53 5.65* 49.77 6.01* 49.53 5.14* 48.15 5.51* 51.40 5.46* Posterior 36.17 6.11 38.41 6.34 39.20 6.27 38.21 6.51 40.87 6.73 38.90 6.58 39.51 6.03 42.19 6.63 41.63 6.79 Significant difference from P2 (P.05). Significant difference from P3 (P.05). Significant difference from P5 (P.05). Significant difference from P6 (P.05). Significant difference from T1 (P.05).
368 DYNAMIC POSTURAL CONTROL MEASUREMENT, Robinson Table 5: Means SEs (deg) for Hip Rotation (N 20) Anterior 7.68 2.19 8.15 2.31 8.22 2.17 7.35 2.08 9.62 2.17 9.36 2.13 8.12 2.26 6.88 2.37 5.64 2.44 Anterolateral 11.75 2.50 10.23 2.48 11.84 2.51 11.21 2.52 10.94 2.36 9.79 2.62* 8.86 2.34* 7.25 2.54* 8.67 2.86 Anteromedial 8.36 2.82 9.19 2.96 10.23 3.21 9.95 3.20 10.50 3.21 9.65 2.85 8.49 2.34 7.64 2.68 7.98 2.42 Lateral 1.73 2.52 0.01 2.43 1.39 2.35 1.88 2.46 0.99 2.26 0.80 2.31 1.56 2.78 2.35 3.02 0.81 3.06 Medial 11.34 2.31 12.19 2.72 12.04 2.90 14.60 2.79 14.07 2.80 13.72 2.47 11.41 2.20 13.00 2.51 10.83 2.37 Posterolateral 6.62 3.18 6.50 2.68 8.70 2.58 7.79 3.00 8.14 2.82 8.49 2.88 6.87 2.40 7.12 2.56 4.25 2.79 Posteromedial 11.01 3.30 11.84 3.97 11.73 3.77 11.54 3.61 12.74 3.62 15.14 2.62 8.67 3.11 10.50 2.99 13.70 2.78 Posterior 8.92 3.01 9.31 2.53 11.04 2.81 10.47 2.20 11.77 1.93 10.23 2.09 10.31 2.00 11.46 1.79 11.41 2.14 *Significant difference from P3 (P.05). Significant difference from P5 (P.05). Significant difference from T2 (P.05). Significant difference from T3 (P.05). SEBT less time-consuming for clincians to administer without a loss of information about a patient s dynamic postural control. The knee flexion values for the current study were less in all 8 directions than those reported by Earl and Hertel. 18 For example, Earl and Hertel reported a knee flexion value for the anterior direction of approximately 70, whereas the maximum mean value for any trial in the current study was 52. However, there was fairly close agreement as to where each reach direction ranked in terms of the magnitude of knee flexion. In both studies, the greatest knee flexion occurred during the performance of the medial and anteromedial reach directions (although the order was reversed), and the least amount of knee flexion occurred during the performance of the lateral directions (posterolateral in the current study and lateral in the study by Earl and Hertel). There are at least 4 possibilities for the discrepancy in knee flexion values. The differences in knee flexion values may be a consequence of measuring angular displacement with different devices (electrogoniometer vs electromagnetic tracking). Although it has been assumed that maximum knee flexion occurs at the point of touchdown, perhaps the knee is maximally flexed before or after maximum reach occurs. Furthermore, assuming that maximum knee flexion occurs at touchdown, in the current study this point was indicated by the manual depression of an electronic trigger, thus involving the reaction time of the tester and making it possible that the event markers placed in the data corresponded to less than the maximum knee flexion value. Last, allowing compensatory and counterbalancing arm movements as in the Earl and Hertel 18 study should have increased participants dynamic postural control and may have enabled them to achieve greater knee flexion values. Study Limitations and Future Investigations Methodologically, there are several limitations that could be avenues for improving future investigations. A means for automatically recording foot touchdown (eg, foot switch) at maximum excursion distance as opposed to the manual depression of an electronic trigger (which involves the reaction time of the tester) would ensure exact correspondence with angular displacement data. Similarly, automatic measurement would eliminate errors made by the tester in visually tracking, marking, and reporting maximum excursion distance scores. Perhaps this could be accomplished by placing an electromagnetic sensor on the distal part of the foot and configuring the MotionMonitor software to measure the horizontal distance from the center of the star excursion grid to touchdown. Previously, biomechanic analyses of the SEBT have been limited to angular displacement of the stance leg at specific joints and specific times (eg, maximum excursion distance). Some of what remains to be known are (1) the velocity and acceleration patterns at specific joints, (2) intra- and interlimb coordination (eg, relationship of hip and knee joint of the stance leg, interaction of the stance and reaching legs), (3) the kinetics of SEBT performance (eg, center of pressure measurements), and (4) how variability and individuality affect SEBT performance. The dynamic systems approach, because it involves studying the integration and changing relationship of multiple factors 19 as well as the expectation of functional variability 20 and individual specificity 21 in performance, seems best suited for investigating these objectives. The dynamic systems strategy could be used by changing the task constraints of the SEBT (eg, allowing arm Table 6: Means SEs (deg) for Hip Abduction (N 20) Anterior 3.41 1.66 6.02 1.21* 3.85 1.51 5.88 1.15* 6.43 1.26* 5.81 1.62* 6.46 1.41* 6.27 1.40* 5.30 1.50* Anterolateral 7.05 1.34 6.82 1.58 8.36 1.25 8.35 1.43 7.36 1.22 7.54 1.64 7.03 1.72 7.44 1.75 7.57 1.69 Anteromedial 1.31 1.47 1.52 1.52 0.56 1.80 2.06 1.50 1.96 1.61 2.69 1.82 0.62 1.99 1.11 1.79 0.82 1.92 Lateral 6.30 4.10 9.25 5.48 8.92 6.03 10.18 8.33 15.71 10.65 13.58 10.41 8.80 5.51 8.81 6.22 7.90 6.80 Medial 11.53 4.27 11.12 4.29 12.80 4.16 11.39 5.79 11.41 5.40 13.84 5.90 11.46 3.50 15.73 7.85 18.58 7.50 Posterolateral 8.30 5.28 8.74 5.36 9.05 6.26 10.08 6.62 14.29 6.11 7.16 8.02 13.26 8.52 11.98 11.23 12.13 11.02 Posteromedial 0.77 7.07 3.69 7.89 4.54 7.96 3.49 9.03 2.78 7.95 16.42 10.34 1.82 12.36 1.44 12.44 2.45 13.06 Posterior 8.63 12.02 2.63 16.04 1.04 14.90 14.30 13.73 12.19 13.69 2.38 13.93 7.28 10.30 1.31 11.52 0.90 12.51 Significant difference from P3 (P.05). Significant difference from T1 (P.05).
DYNAMIC POSTURAL CONTROL MEASUREMENT, Robinson 369 Table 7: Means SEs (deg) for Hip Flexion (N 20) Anterior 12.14 2.76 12.61 2.64 13.93 2.87 12.34 2.67 14.01 2.80 14.35 2.83 15.33 3.00 14.33 2.85 15.01 2.95 Anterolateral 10.31 2.61 12.25 2.59 12.01 2.81 13.18 2.94 13.05 2.89 12.51 3.01 15.26 3.25* 14.09 2.94* 15.36 3.26* Anteromedial 12.02 3.24 13.18 3.82 14.63 3.89 16.59 4.08 16.04 3.80 16.15 4.49 17.80 3.57 17.45 4.05 16.38 3.48 Lateral 43.51 5.30 46.45 5.41* 45.54 5.58 48.15 5.93 47.50 5.93 47.40 5.76 46.15 5.78 49.11 5.57* 50.78 5.90* Medial 41.00 5.29 43.13 5.71 44.42 5.68 46.72 6.36 45.36 5.95 46.01 5.98 45.07 5.42 45.90 5.49 47.12 5.81 Posterolateral 47.08 5.63* 50.12 5.78* 51.98 5.62* 52.10 5.74* 52.67 5.88* 54.80 6.14* 54.20 5.47* 57.94 5.57* 56.05 5.64* Posteromedial 50.39 6.54 54.04 6.92 56.57 6.43* 56.58 6.52* 56.81 6.73* 58.56 6.44* 58.81 6.36* 59.42 6.65* 61.67 6.39* Posterior 54.06 6.94 56.82 7.00 58.10 7.19 58.45 7.12 60.85 6.87 57.83 6.74 54.85 6.20 56.81 6.41 57.51 6.57 Significant difference from P2 (P.05). Significant difference from P3 (P.05). Significant difference from P4 (P.05). Significant difference from P5 (P.05). Significant difference from T1 (P.05). movements and plantarflexion) and increasing a likely control parameter (eg, maximum excursion distance) and tracking the changes in the order parameters (eg, relative phase) that represent the intra- and interlimb coordinative relationships. 19 If different coordinative relationships are revealed, then this paradigm could be applied in a measurement context to create a more demanding and individually specific test of dynamic postural control. 1 CONCLUSIONS SEBT performance, as represented by normalized maximum excursion distance, showed a significant increase across trials in 7 of the 8 reach directions. Knee flexion and hip flexion exhibited significant increases across trials in 5 and 4 of the reach directions, respectively. Normalized maximum excursion distance and angular displacement measures stabilized after approximately 4 practice trials. The relative lack of change in performance may be caused by restrictive task constraints and possibly an inherent energy minimization bias. The significant differences observed between the first test trial and the other 2 test trials was probably caused by warm-up decrement, which can be managed by disregarding the first test trial and determining a client s score from the last 2 trials. The findings of the current study together with previous research means that clinicians can streamline SEBT administration by reducing the number of reach directions from 8 to 3 (eg, anteromedial, medial, posteromedial) and practice trials from 6 to 4 without affecting the validity of the test. The inclusion of warm-up exercises could further reduce the number of practice trials needed to achieve performance stability. Future research should involve greater automation of data collection and use a dynamic systems approach to investigate the relationship of the multiple factors that underlie SEBT performance. Such an investigation may produce a more demanding and individually specific test of dynamic postural control. Acknowledgment: We thank Katie Rovtar, BS, for her assistance during data collection. References 1. Burton AW, Miller DE. Movement skill assessment. Champaign: Human Kinetics; 1998. 2. Gray G. Lower extremity functional profile. Adrian: Wynn Marketing; 1995. 3. Gribble PA, Hertel J. Considerations for normalizing measures of the Star Excursion Balance Test. Meas Phys Educ Exerc Sci 2003;7:89-100. 4. Gribble PA, Hertel J, Denegar CR, Buckley WE. The effects of fatigue and chronic ankle instability on dynamic postural control. J Athl Train 2004;39:321-9. 5. Hertel J, Braham RA, Hale SA, Olmsted-Kramer LC. Simplifying the Star Excursion Balance Test: analyses of subjects with and without chronic ankle instability. J Orthop Sports Phys Ther 2006;36:131-7. 6. Olmsted LC, Carcia CR, Hertel J, Schultz SJ. Efficacy of the Star Excursion Balance Tests in detecting reaching deficits in subjects with chronic ankle instability. J Athl Train 2002;37: 501-6. 7. Gribble PA, Hertel J, Denegar CR. Chronic ankle instability and fatigue create proximal joint alterations during performance of the Star Excursion Balance Test. Int J Sports Med 2007;28:236-42. 8. Denegar CR, Miller SJ, Challis JH. Factors predictive of anterior reach test balance performance. J Athl Train 2001;37(Suppl 2): 53-S. 9. Earl JE, Hertel J, Denegar CR. Efficacy of a 6-week neuromuscular rehabilitation program on pain, function, muscle activity and joint motion in patients with patellofemoral pain. J Athl Train 2002;38(Suppl 2):83-S. 10. Hertel J, Miller S, Denegar C. Intratester and intertester reliability during the Star Excursion Balance Test. J Sport Rehabil 2000;9: 104-16. 11. Kinzey S, Armstrong C. The reliability of the star-excursion test in assessing dynamic balance. J Orthop Sports Phys Ther 1998; 27:356-60. 12. Bernstein N. Coordination and regulation of movement. New York: Pergamon Pr; 1967. 13. Sparrow WA, Hughes KM, Russell AP, Le Rossignol PF. Effects of practice and preferred rate on perceived exertion, metabolic variables, and movement control. Hum Mov Sci 1999;18:137-53. 14. Haken H, Kelso JA, Bunz H. A theoretical model of phase transitions in human hand movements. Biol Cybern 1985;51:347-56. 15. Newell KM. Constraints on the development of coordination. In: Wade MG, Whiting HT, editors. Motor development in children: aspects of coordination and control. Boston: Martinus-Nijhoff; 1986. p 341-60. 16. Sparrow WA, Newell KM. Metabolic energy expenditure and regulation of movement economy. Psychol Bull Rev 1998;5:173-96.
370 DYNAMIC POSTURAL CONTROL MEASUREMENT, Robinson 17. Newell KM, Mayer-Kress G, Liu YT. Time scales in motor learning and development. Psychol Rev 2001;108:57-82. 18. Earl JE, Hertel J. Lower extremity muscle activation during Star Excursion Balance Tests. J Sport Rehabil 2001;10:93-104. 19. Kurz MJ, Stergiou N. Applied dynamic systems theory for the analysis of movement. In: Stergiou, N, editor. Innovative analyses of human movement. Champaign: Human Kinetics; 2004. p 93-117. 20. James CR. Considerations of movement variability in biomechanics research. In: Stergiou N, editor. Innovative analyses of human movement. Champaign: Human Kinetics; 2004. p 29-62. 21. Bates BT, James CR, Dufek JS. Single-subject analyses. In: Stergiou N, editor. Innovative analyses of human movement. Champaign: Human Kinetics; 2004. p 3-28. Suppliers a. Ascension Technology Corp, PO Box 527, Burlington, VT 05402. b. Innovative Sports Training Inc, 3711 N Ravenswood, Ste 150, Chicago, IL 60613. c. Version 14.0; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.