Journal Logo


Cleveland Clinic Postural Stability Index Norms for the Balance Error Scoring System


Author Information
Medicine & Science in Sports & Exercise: October 2018 - Volume 50 - Issue 10 - p 1998-2006
doi: 10.1249/MSS.0000000000001660


The maintenance of postural stability is achieved through the integration of somatosensory, visual, and vestibular afferent inputs and the resultant coordinated motor response (1). Postural stability is influenced by factors including aging, injury, or disease, and its precise measurement is used to determine the effects of each of these factors. For example, in older adults, clinical balance measures are used to predict fall risk and to drive rehabilitation interventions aimed at reducing the risk of falls. In concussion, the objective assessment of postural stability is also critical, as it contributes to the diagnostic process and is a key element in the management of concussion in terms of determining readiness to return to play and for referral to physical therapy for those with prolonged postural instability (2,3). Because of its importance in the management of concussion, the evaluation of postural stability is recommended in the position statements of numerous panels and medical groups outlining best practice guidelines (2,4–6) The Balance Error Scoring System (BESS) is a clinical balance assessment designed specifically for concussion (3), and its truncated version, the modified BESS, remains the most commonly recommended and used measure for concussion assessment from the sideline to the clinic.

The BESS was designed to be used as a rapid and inexpensive tool conducive to sideline and clinical use to provide an evaluation of postural stability. Briefly, the BESS uses three variations of the original Romberg stances conducted on a firm and foam surface for a total of six individual trials (7). Eyes remain closed for all six stances to negate the use of visual feedback, thus testing mainly somatosensory and vestibular function and the resultant efferent output. Postural stability is evaluated by counting the number of errors a participant commits during each 20-s stance and summing errors for all six stances (7,8). The scoring criteria limits errors to a maximum of 10 per condition resulting in total BESS scores ranging from 0 to 60, with higher error scores suggestive of postural instability (3,7). The utility of the BESS, however, has been scrutinized due to its marginal validity (9,10), reliability (11), and practice (12), learning (13), and fatigue effects (14). Therefore, a fundamental gap in concussion management remains, because medical providers lack an objective measure of postural stability in diagnosing concussion and determining recovery and clearance for return to play.

A key component of the effective management of concussion, or any complex condition, is the engagement of a diverse group of providers. Concussion specifically includes athletic trainers, team physicians, primary care physicians, neurologists, and rehabilitation specialists (4,5). An inherent challenge in the cohesive utilization of a multidisciplinary team is effective handoffs between providers (15), which are further complicated if assessments are subjective or prone to rater bias (9–11). It was reasoned that the handoff of concussion patients across the multidisciplinary team of providers could be facilitated by the development and utilization of tools and objective outcomes. To address the continuity of care issue specifically as it related to the evaluation of postural stability, we recently developed and validated a tool using a biomechanical approach to objectively quantify performance of the BESS (16).

We have demonstrated that data gathered from the accelerometer and gyroscope of the iPad (Apple Inc., Cupertino, CA) could be used to precisely quantify postural stability (16–19). Briefly, results indicated that linear and angular accelerations of the movement of center of mass (COM) across three directions (anterior–posterior (AP), medial–lateral (ML) and trunk rotation (TR) during the six BESS stances were significantly correlated with those obtained by traditional motion capture approaches (16,17). A 95% ellipsoid volume, termed the instrumented Balance Error Scoring System (iBESS), was computed using acceleration of the COM in ML and AP directions from the iPad’s accelerometer, and TR acceleration from the iPad’s gyroscope (16,17). The iBESS was sensitive in detecting differences between the stance conditions (double-leg, single-leg, and tandem) and between support surfaces (firm and foam), whereas the traditional error scoring method did not detect differences between the double-leg stances on firm and foam surfaces (16).

The iBESS provided a three-dimensional (3D) quantification of postural sway incorporating the ML, AP, and TR directions (16). To enhance clinical utility of using objective assessments in evaluating postural stability, a unitless metric using the biomechanical data from the iPad is presented in the current study to characterize postural stability in three directions of movement (ML, AP, and TR) using normalized path length (NPL) of acceleration. A standardized z-score representing NPL as a percentile relative to the normative population was computed for each individual to summarize movement in each direction across BESS stances. This percentile metric is termed the Cleveland Clinic Postural Stability Index (CC-PSI). The aims of this project were to introduce the CC-PSI as a clinically interpretable approach of applying biomechanics to an existing clinical measure of postural stability, determine the effect of age and sex on CC-PSI values and report normative values for healthy active children, adolescents, and young adults. Based on existing literature related to the development of postural stability, it was hypothesized that CC-PSI values would be lower for youth athletes compared to high school and college athletes.



A total of 6762 athletes from ages 5 to 23 yr enrolled in Kindergarten to Grade 12 schools and colleges across Northeastern Ohio participated in the study. The sample included 4600 (68%) males and 2162 (32%) females who completed baseline concussion testing in the 2013 to 2014 seasons. Normative values for BESS error scores were recently published using the same sample of athletes (20). As in our previous study (20), data were stratified into six age–sex cohorts: males age 5 to 13 yr (n = 360), 14 to 18 yr (n = 3743), and 19 to 23 yr (n = 497) and females age 5 to 13 yr (n = 246), 14 to 18 yr (n = 1673), and 19 to 23 yr (n = 243) as these age groups roughly corresponded to academic grade level (grade school, high school, and college, respectively). Given the motor developmental differences inherent to the 5 to 13 yr age group (21), the youth cohort was subdivided into males age 5 to 9 yr (n = 122), males age 10 to 13 yr (n = 238), females age 5 to 9 yr (n = 48), and females age 10 to 13 yr (n = 198). Demographic details of participants are provided in Table 1.

Sample demographics.

Balance assessments were completed as part of routine preseason baseline testing at organized youth sport leagues, high schools, and colleges. All participants were neurologically healthy, had not experienced a concussion within at least 6 months and did not present with active musculoskeletal impairments that impacted postural stability or precluded participation in sport. Athletes recovering from musculoskeletal injuries, such as sprains, strains, and fractures, or those recovering from surgery were excluded from the sample. The Cleveland Clinic Institutional Review Board approved this retrospective analysis and allowed for waived consent, as baseline testing was a part of the clinical standard of care.

Data collection

The BESS was administered by athletic trainers (n = 48) and trained clinical personnel within the Cleveland Clinic Concussion Center utilizing standardized testing procedures (22). Before balance testing, an Apple iPad was affixed to each individual onto the sacrum at the level of the iliac crests using a belt with customized housing for the iPad. The sacral level represents the approximate COM during upright stance. Per BESS instructions, each participant completed the six 20-s trials according to BESS protocol: double-leg, single-leg, and tandem stances on firm and foam (Airex Balance Pad, Knoxville, TN) surfaces. Participants placed their hands on their iliac crests, were instructed to stand as still as possible, and upon eye closure, the 20-s trial was initiated. Linear and angular acceleration from the iPad’s native accelerometer (ST Micro LIS331DLH) and gyroscope (ST Micro L3G4200D), respectively, were recorded at 100-Hz sampling frequency with a 3.5-Hz cutoff, fourth-order, low-pass Butterworth filter and stored within the balance module of the Cleveland Clinic Concussion Application (C3 app) (23,24). The C3 app is a custom-built app written in Objective-C, a general purpose programming language used in Apple’s iPhone operating system (iOS). Raw data were stored locally on each device, uploaded to a secure, Health Insurance Portability and Accountability Act-compliant server following testing, and extracted for data analysis.

Data analysis

The CC-PSI metric was computed using NPL of acceleration, which represents the sum of the absolute differences between successive COM accelerations normalized over time as the body oscillates within the base of support. The NPL of acceleration was converted into standardized z-scores representing postural sway for each individual as a percentile relative to the entire normative population. To compute the CC-PSI, the first and last 1% of the filtered linear and angular accelerations were truncated across movement directions. The NPL data were normalized within each movement direction. Specifically, the natural-log of all NPL values was computed to assure the spread of values was normally distributed across balance conditions and cohorts, and to account for the different measurement units. After normalization, standard normal distributions (z-scores) were calculated for each individual using the mean and SD of the natural-log of NPL of the normative population across sway directions (ML, AP, and TR) for each BESS condition. The z-scores were calculated using the equation:

where s is the SD of the natural-log of the NPL of the normative population,

is the mean of the natural-log of the NPL of the normative population, Xsubject is the natural-log of the NPL value within a given sway direction for each individual across age and sex cohorts, and Zsubject is the z-score within a given sway direction to be computed for the individual. A composite z-score was then computed for each balance condition, resulting in one measure summarizing the three sway directions. A positive z-score indicated that balance performance was worse than the mean of the normative population for the specific BESS condition. The composite z-scores were computed to produce a percentile score, termed the CC-PSI, based on the transformed variable, indicating the rank of each individual’s performance in terms of percentages with respect to the normative population.

Statistical analysis

A two-way multivariate analysis of variance (MANOVA) was performed with age, sex, and age–sex as the independent variables and the CC-PSI metric for the six BESS conditions as the dependent variables. Shapiro–Wilk’s and Levene’s tests were performed to ensure assumptions of normality and homogeneity of variance for each condition, respectively. Effect sizes are presented as η2 and interpreted by Cohen as small (0.00 < η2 < 0.01), medium (0.01 < η2 < 0.06), and large (η2 > 0.06) (25). Post hoc ANOVA were performed with three factors: age, sex, and age–sex interaction. Sex consisted of two levels (males and females), and age consisted of three levels (youth, high school, and college). Secondary analyses were completed with the youth cohort subdivided into the 5- to 9-yr-old and 10- to 13-yr-old brackets to evaluate developmental effects; thus, age consisted of four levels (youth 5–9, youth 10–13, high school, and college). Post hoc correction for multiple comparisons was performed with Tukey correction when equal variance assumptions were met. Robust ANOVA with Tamhane correction (26) was used to account for unequal variance and account for multiple comparisons when needed. Statistical analyses were performed using SPSS (version 19; IBM Corporation, Armonk, NY) and R (R Core Team, Vienna, Austria).


Youth exhibit worse postural stability than high school and college athletes

A visual representation of COM acceleration using the ninety-five percent ellipsoid volume metric and accompanying CC-PSI percentile scores comparing age cohorts is shown in Figure 1. Normative values for CC-PSI percentile scores stratified by age and sex are shown in Figure 2. The mean CC-PSI percentile scores across BESS conditions ranged from (33.1–46.5), (48.4–51.9), and (55.8–63.3) for the youth, high school, and college cohorts, respectively. Multivariate comparisons of CC-PSI across BESS conditions indicated a significant age effect across all cohorts: youth (5–13 yr), high school (14–18 yr), and collegiate (19–23 yr) athletes: female (12, 13376) = 51.60, P < 0.0001; Wilk’s Λ = 0.913, partial η2 = 0.044 (medium effect size) (Table 2 displays between-subjects effects).

Sex, age, and sex–age differences across BESS conditions (MANOVA).
95% ellipsoid volumes combining linear and angular acceleration of the COM using the tablet’s built-in accelerometer and gyroscope, respectively, across BESS conditions and age cohorts. The CC-PSI values computed using the sensor data are displayed on the top right corner of subplots for each BESS condition, indicating the rank of each age cohort’s performance in terms of percentiles with respect to the entire population.
Mean and SD of CC-PSI values comparing differences in BESS performance across A) male and female populations and B) youth, high school, and college cohorts. *P < 0.05; **P< 0.01; ***P < 0.001.

Among males, the effect of age was significant across all BESS conditions (P < 0.001). Specifically, BESS performance in collegiate male athletes was significantly superior to high school males followed by youth males. During the three most challenging conditions, single-leg on foam, tandem on foam, and single-leg on firm, mean CC-PSI values for youth males were the 29th, 40th, and 25th percentiles, respectively, compared with high school males in the 49th, 47th, and 50th percentiles and college males in the 59th, 56th, and 60th percentiles. Less variability and slightly different trends were observed among females; specifically, the effect of age was significant (P < 0.05) in four of the six BESS stances. However, all three female age cohorts performed similarly during tandem stance on foam, and the youth and high school cohort performed similarly during the double-leg stance on foam. In the four BESS conditions that were significantly different, youth females exhibited significantly lower percentiles than high school and college females. These results are depicted in Figure 2A.

Males exhibited lower CC-PSI percentiles relative to females during the BESS

In the multivariate analysis, the effect of sex was significant for the BESS conditions, without adjusting for age: female (6, 6688) = 19.86, P < 0.0001; Wilk’s Λ = 0.982, partial η2 = 0.018 (medium effect size) (Table 2 and Fig. 2B). When adjusting for age, within both youth and high school cohorts, males had significantly lower percentiles than females across all BESS conditions (P < 0.01). Within the college cohort, there were no significant sex effects across any of the BESS conditions.

Young males exhibit lower CC-PSI percentiles relative to older cohorts

When further stratifying youth athletes into smaller age brackets, a secondary analysis showed that across five of the six BESS conditions, the 5- to 9 yr-old cohort had significantly greater postural sway than youths 10 to 13 yr of age (BESS conditions, 1–5; P < 0.0001). A performance gap was particularly evident between 5- and 9-yr-old boys and all other athletes, as shown in Figure 3. In fact, 60% and 41% of youth males between ages 5 and 9 yr performed worse than 1.96 SD from the normative population mean value during the single-leg and tandem on firm, respectively (see Table, Supplemental Digital Content 1, Percent of individuals in each condition in which postural stability was in the bottom 5% of the normative population, Comparatively, only 14% and 10% of females in the 5 to 9 yr age cohort had comparably low CC-PSI values during corresponding firm surface conditions. In general, the 5- to 9-yr-old girls performed similar to their older female counterparts.

Mean CC-PSI values (solid lines) and SD (shaded area) depicting differences in performance across the six BESS stances as a function of age (values on the x-axis) and sex (red vs blue). Red solid and shaded lines represent CC-PSI values for females, whereas blue represents males.


The aims of this project were threefold: 1) to introduce the CC-PSI metric as an objective, biomechanical measure of postural stability, 2) to determine the effect of age and sex on CC-PSI values, and 3) to report normative values for healthy active children, adolescents, and young adults. This is the first large-scale project to combine 3D movement of COM into a unitless percentile to provide a biomechanical quantification of balance performance during a clinical balance assessment. In general, our results indicated that applying biomechanics to the BESS resulted in a more psychometrically sound measure, and that the effects of age and sex were significant using the CC-PSI metric.

Development of the CC-PSI

Previous studies demonstrate value in quantifying the BESS using inertial sensor-based methods; however, translating the substantial amount of biomechanical data obtained into a clinically interpretable metric has remained a challenge (16,18,19,27–29). King and colleagues (27) eloquently described these technical challenges, acknowledging that “the metrics generated by an instrumented approach are numerous and complex” and that “although an instrumented modified BESS holds promise, technical questions remain.” While technically sound and valid, the typical metrics derived using biomechanical methods result in values with complicated units that are difficult to interpret across multiple clinical providers with varying exposure to biomechanics (16,19,27,29). These technical challenges led to the development of the CC-PSI because it provides a composite score quantifying COM sway in three directions of movement, producing a clinically interpretable metric comparing an individual’s performance relative to a large sample of normative values. The CC-PSI, which represents NPL as a percentile relative to healthy active controls, provides an immediate interpretation and reference state of the biomechanical data.

Enhancing clinical balance measures by applying biomechanics to quantify postural stability

An advantage of the CC-PSI is the multidirectional quantification of COM movements in the ML, AP, and TR directions. Increasing granularity when quantifying movement strategies to stabilize the COM may aid in elucidating pathophysiological impairments that vary among populations and individuals, in terms of deficits and severity. In concussion, precise measures of postural sway using biomechanical approaches have been used to detect residual deficits not identified using clinical means. Instability in the ML direction has been commonly identified in athletes postconcussion, (30,31) whereas instability in the AP direction only, particularly during the tandem stance conditions where less AP sway occurs naturally, may not reflect postural instability after concussion (32). Enhancing the precision of balance testing may prevent athletes who have not fully recovered from returning to contact sport. In stroke, decreased weight shifting in the ML direction toward the hemiplegic side is a commonly seen deficit that alters balance responses and contributes to asymmetrical gait (33). Obtaining a precise measure of postural stability to identify deficits in a given direction of movement can allow physical therapists to more precisely direct rehabilitation interventions and monitor outcomes of those interventions. Applying biomechanics to measure postural sway can also improve the psychometric properties of clinical balance measures by eliminating floor and ceiling effects (20). Clinical balance measures typically produce a score which quantifies one or more aspect of balance maintenance. For example, although the BESS quantified balance by measuring discrete losses of balance or balance perturbations, tasks within the Berg Balance Scale (34) and the Clinical Test of Sensory Organization and Balance (35) quantify some elements of performance by the length of time an individual is able to maintain a given stance. It is difficult for the scales used in these clinical balance measures to account for subtle differences in postural stability or to sufficiently measure high and low performers. A limitation of the error scoring approach of the BESS is its ability to detect differences in postural stability in the two double-leg stances in which participants often do not commit an error. This limitation is increasingly present when only the modified BESS (firm stances) is used (2), rendering one third of the clinical data unusable, because errors are infrequent during the double-leg stance. This ceiling effect was evident in our recent work using the same sample of participants as the current study in which approximately 97% and 90% of all subjects did not commit an error in the double-leg stances on firm or foam surfaces, respectively (20). Reliance on error counts also limits the utility of the single-leg and tandem stances on the foam surface. Recently, we have demonstrated, regardless of age or sex, that greater than one half of healthy student athletes committed 8 to 10 errors on the single-leg stance on foam, with 20% achieving the worst possible score of 10 errors (20). Considering the 10-error limit per stance, the ability to detect changes in balance as a function of injury or monitor the recovery process is severely compromised when healthy individuals routinely perform at the edges of the scoring continuum. The elimination of floor and ceiling effects with the CC-PSI metric has the potential to improve the clinical utility and interpretability of the BESS. With respect to the less demanding double-leg stances, results from the current project indicate that the CC-PSI can effectively characterize and discriminate the subtle nuances of postural stability that the typical error scoring method often misses, as it discriminated among the age and sex cohorts.

To demonstrate the increased sensitivity of the CC-PSI, the distribution of error scores and CC-PSI values in the two double-leg stances in our sample of healthy collegiate males is shown in Supplemental Figure 1 (see Figure, Supplemental Digital Content 2, BESS error scores [top panels] for college males and their corresponding CC-PSI scores [bottom panels] during the double-leg on firm [A] and foam [B] surfaces, Although ceiling effects are apparent in error scores for the two double-leg stances (ie, a large proportion of the sample performed at the high end of the measure resulting in very few errors), the distribution of CC-PSI percentile scores is much broader due to the normalization method of computation. For example, an individual may score in the 50th percentile (CC-PSI score of 50) during healthy baseline testing and experience 0 errors in the double-limb stance on a firm surface. Postinjury, the error score may remain at a 0, but the CC-PSI score is a 25, indicating that his/her postural sway during the same stance has increased substantially. This broader distribution as a result of more precise measurement provides an appropriate continuum in which postural stability can be characterized and interpreted. Considering the CC-PSI does not have an arbitrary cutoff or maximum score, the magnitude of postural instability relative to a baseline or age–sex norms can be appropriately measured and identified.

Importantly, our goal was not to develop a metric to correlate with the error scoring method used in the clinical BESS. Discreet errors or losses of balance measure different aspect of postural stability compared with the more subtle variable of COM sway. In fact, a step, stumble or fall (all errors in the clinical BESS) disrupts the biomechanical measure of COM sway. Thus, our goal was to improve the objectivity and sensitivity of the BESS by providing a biomechanical quantification of postural sway during the BESS stances in a clinically meaningful and interpretable metric.

Clinical interpretation of the CC-PSI

The use of percentile scoring provides context in allowing clinicians to directly compare an individual’s performance to a normative population. For example, a CC-PSI score of 50 indicates that the individual’s performance was equal to the mean of the population. A score of 75 indicates that the person performed better than 75% of the population. By contrast, a score of 10 implies that 90% of the population outperformed the individual. The normative values presented in Figure 2 as a function of age and sex should be used to more precisely determine an individual’s performance both relative to the entire healthy, active population, and relative to his/her peers. For example, for a 19-yr-old male, a score of 60 for double-limb stance on firm surface implies that he outperformed 60% of the healthy, active population. However, when examining the normative scores for college-age males, he performed slightly worse than the mean for his cohort, which was 64.9. By contrast, a score of 60 for a 10-yr-old boy would not only reflect performance better than 60% of the entire population but also indicate that he outperformed his youth male cohort, whose mean score in that stance was 31.3. An optimal comparison for an individual may be to obtain longitudinal data by which to determine changes in performance as a function of injury, rehabilitation, or growth/maturity. For example, an elite female gymnast is likely to have better postural stability than her nongymnastic peers simply due to training. Should she be injured, a score close to the mean of her age–sex cohort may still be indicative of a deficit. In elite athletes with superior balance, using individual baseline data by which to compare postinjury performance is likely a more sound clinical approach. Thus, the CC-PSI can be used to compare performance of an individual at different time points, objectively quantifying the effects of injury or recovery.

Notably, simply by how percentiles are computed, CC-PSI scores are distributed in a bell-shaped curve. As shown in Supplemental Figure 1, a score between 34 and 68 indicates performance values within one SD of the population mean [see Figure, Supplemental Digital Content 2, BESS error scores (top panels) for college males and their corresponding CC-PSI scores (bottom panels) during the double-leg on firm (A) and foam (B) surfaces,]. A score between 2.3 and 34 or 68 and 97.7 is reflective of performance between one and two SD from the mean, and so on. It is important to understand that CC-PSI scores on either end of the spectrum, particularly beyond 2 SD, are indicative of either very poor (worse than ~98% of the entire population) or outstanding (better than ~98% of the population) performance.

CC-PSI values differ as a function of age and sex

As expected, postural sway improved as a function of age and was impacted by sex (Table 2 and Fig. 3). The physiological maturation of postural control evolves through childhood and into adolescence, reaching full maturity in the latter teenage years (1,36,37). Postural control is a function of the integration of the somatosensory, visual, and vestibular systems, and the resultant motor response (20). It has been estimated that while somatosensory function reaches maturity in children at 3 to 4 yr of age, vision is not fully developed until 14 to 15 yr of age, and the vestibular system at 15 to 16 yr of age (1,37,38). Furthermore, it has been shown that the integration of vestibular responses in postural control develops earlier in females compared with males, with evidence of separation between sexes as young as 7 to 8 yr of age (1,36,37). The effect of sex is not only evident due to changes in the rate of physiological development but also due to differences in anthropometrics (38,39). Body weight and COM contribute to balance maintenance (38,39). In fact, postural stability is inversely proportional to COM, which is typically lower in females due to a smaller waist-to-hip ration, greater thigh girth, and narrower width of the shoulders (40,41). Thus, although significant variability exists in the rate of development and factors contributing to balance maintenance, in general, young females demonstrate superior balance compared with their male peers until the middle to late teenage years (36). As hypothesized, the CC-PSI metric was sensitive in detecting significant differences in postural control across all six stances as a function of age and sex. A visual representation of these differences in postural sway among stances is categorized according to age and depicted in Figure 1. These results, demonstrating the effects of age and sex in healthy children and young adults, underscore the need to use age- and sex-specific normative values when interpreting balance metrics at baseline and as part of clinical management.

Future directions and broader application of the CC-PSI

Although the CC-PSI normative values were obtained during performance of the BESS, the metric itself can theoretically be used to enhance other clinical balance measures by providing a 3D biomechanical quantification of postural sway. The Clinical Test of Sensory Organization and Balance developed by Horak and Shumway-Cook to detect deficits in vestibular function uses several variations of static balance stances optimal for measures of postural sway (35). In fact, Horak and Shumway-Cook (42) even suggested the use of a grid to provide a quantification of postural sway that occurred during each stance, with the test administrator rating sway as “minimal,” “mild,” “moderate,” or a fall. The CC-PSI measured via the iPad has the potential to add precision to this clinical test and enhance the objectivity for health care providers to document deficits in postural stability and improvements over a course of rehabilitation. We recently demonstrated that the CC-PSI metric was sensitive in differentiating healthy older adults from individuals with a chronic neurological disease, Parkinson’s disease (18). Pilot studies are also underway to use the CC-PSI in evaluating the risk of falls in older adults both in an inpatient and primary care setting. Additionally, projects are currently ongoing to determine the utility of the CC-PSI in discriminating balance performance in student-athletes with acute concussion compared to their healthy baseline performance and tracking the recovery of postural stability. Analyses are also underway aimed at determining the minimal clinical important difference in percentile scores when measuring recovery and at evaluating the use of the CC-PSI metric to estimate postural stability during more dynamic-stability tasks (i.e., tandem gait) and dual-task paradigms (i.e., postural stability plus cognitive tasks) in healthy and concussed athletes.


All data from this normative study were collected in healthy, active children and young adults, thus limiting generalizability. Furthermore, because this was a sample of convenience, we experienced large variability in the sample sizes for each age–sex cohort, report relatively low sample sizes for the 5- to 9-yr-old youth cohort, and in particular for 5- to 9-yr-old girls. This resulted in greater variability in CC-PSI values for 5- to 9-yr-old girls. Additionally, we did not account for prior exposure to the BESS. Student athletes undergo a baseline assessment annually, therefore, the possibility of a learning effect influencing the normative values cannot be ruled out.


In contrast to expensive laboratory or subjective clinical approaches, mobile device platforms, such as the iPad, can provide a technically valid, psychometrically sound, inexpensive, and objective assessment of postural stability. Furthermore, its portability and ability to integrate into the electronic medical record make it conducive for field and clinical use across a number of health care providers. Given the significant effects of age and sex, both variables should be considered when clinically interpreting CC-PSI scores. The CC-PSI metric computed from inertial sensor data native to the iPad provides a promising outcome metric to characterize movement of COM in ML, AP, and TR directions as a percentile relative to the normative population to identify individuals with postural instability. Using the CC-PSI metric enables a more reliable interpretation of postural stability assessment for clinical decision making.

This project was supported the Edward F. and Barbara A. Bell Family Endowed Chair to J. L. A. and by the Farmer Foundation. The authors would like to thank the Cleveland Clinic athletic trainers for their assistance with baseline testing.

Conflict of Interest: Intellectual property has been filed by J. L. A., S. M. L., S. J. A., T. D., and M. M. K. protecting the CC-PSI metric as part of the C3 application. The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of this study do not constitute endorsement by ACSM.


1. Forssberg H, Nashner LM. Ontogenetic development of postural control in man: adaptation to altered support and visual conditions during stance. J Neurosci. 1982;2(5):545–52.
2. McCrory P, Meeuwisse W, Dvořák J, et al. Consensus statement on concussion in sport-the 5th international conference on concussion in sport held in Berlin, October 2016. Br J Sports Med. 2017;51:838–47.
3. Guskiewicz KM, Ross SE, Marshall SW. Postural stability and neuropsychological deficits after concussion in collegiate athletes. J Athl Train. 2001;36(3):263–73.
4. Broglio SP, Cantu RC, Gioia GA, et al. National Athletic Trainers’ Association position statement: management of sport concussion. J Athl Train. 2014;49(2):245–65.
5. Harmon KG, Drezner JA, Gammons M, et al. American Medical society for Sports Medicine position statement: concussion in sport. Br J Sports Med. 2013;47(1):15–26.
6. Halstead ME, Walter KD. Council on Sports Medicine and Fitness. American Academy of Pediatrics. Clinical report—sport-related concussion in children and adolescents. Pediatrics. 2010;126(3):597–615.
7. Riemann BL, Guskiewicz KM, Shields EW. Relationship between clinical and forceplate measures of postural stability. J Sport Rehabil. 1999;8:71–82.
8. Riemann BL, Guskiewicz KM. Effects of mild head injury on postural stability as measured through clinical balance testing. J Athl Train. 2000;35(1):19–25.
9. Bell DR, Guskiewicz KM, Clark MA, Padua DA. Systematic review of the balance error scoring system. Sports health. 2011;3(3):287–95.
10. Barlow M, Schlabach D, Peiffer J, Cook C. Differences in change scores and the predictive validity of three commonly used measures following concussion in the middle school and high school aged population. Int J Sports Phys Ther. 2011;6(3):150–7.
11. Finnoff JT, Peterson VJ, Hollman JH, Smith J. Intrarater and interrater reliability of the Balance Error Scoring System (BESS). PM R. 2009;1(1):50–4.
12. Valovich TC, Perrin DH, Gansneder BM. Repeat administration elicits a practice effect with the balance error scoring system but not with the standardized assessment of concussion in high school athletes. J Athl Train. 2003;38(1):51–6.
13. Mulligan IJ, Boland MA, McIlhenny CV. The balance error scoring system learned response among young adults. Sports health. 2013;5(1):22–6.
14. Wilkins JC, Valovich McLeod TC, Perrin DH, Gansneder BM. Performance on the balance error scoring system decreases after fatigue. J Athl Train. 2004;39(2):156–61.
15. Ellis MJ, Leddy J, Willer B. Multi-disciplinary management of athletes with post-concussion syndrome: an evolving pathophysiological approach. Front Neurol. 2016;7:136.
16. Alberts JL, Thota A, Hirsch J, et al. Quantification of the balance error scoring system with mobile technology. Med Sci Sports Exerc. 2015;47(10):2233–40.
17. Alberts JL, Hirsch JR, Koop MM, et al. Using accelerometer and gyroscopic measures to quantify postural stability. J Athl Train. 2015;50(6):578–88.
18. Ozinga SJ, Koop MM, Linder SM, Machado AG, Dey T, Alberts JL. Three-dimensional evaluation of postural stability in Parkinson’s disease with mobile technology. NeuroRehabilitation. 2017;41(1):211–8.
19. Ozinga SJ, Machado AG, Miller Koop M, Rosenfeldt AB, Alberts JL. Objective assessment of postural stability in Parkinson’s disease using mobile technology. Mov Disord. 2015;30:1214–1221.
20. Ozinga SJ, Linder SM, Dey T, et al. Normative performance for balance error scoring system in youth, high school, and collegiate athletes. J Athl Train. doi: 10.4085/1062-6050-129-17. [Epub ahead of print].
21. Verbecque E, Vereeck L, Hallemans A. Postural sway in children: A literature review. Gait Posture. 2016;49:402–10.
22. McCrory P, Meeuwisse W, Aubry M, et al. Consensus statement on Concussion in Sport—The 4th International Conference on Concussion in Sport held in Zurich, November 2012. Phys Ther Sport. 2013;14(2):e1–13.
23. Bernstein JPK, Calamia M, Pratt J, Mullenix S. Assessing the effects of concussion using the C3Logix Test Battery: An exploratory study. Appl Neuropsychol Adult. 2018:1–8.
24. Simon M, Maerlender A, Metzger K, Decoster L, Hollingworth A, Valovich McLeod T. Reliability and concurrent validity of select C3 logix test components. Dev Neuropsychol. 2017;42:446–459.
25. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Science E, editor. New York, NY: Academic Press; 1988.
26. Tamhane A. A comparison of procedures for multiple comparisons of means with unequal variances. J Am Stat Assoc. 1979;74(366):471–80.
27. King LA, Mancini M, Fino PC, et al. Sensor-based balance measures outperform modified balance error scoring system in identifying acute concussion. Ann Biomed Eng. 2017;45:2135–45.
28. King LA, Horak FB, Mancini M, et al. Instrumenting the balance error scoring system for use with patients reporting persistent balance problems after mild traumatic brain injury. Arch Phys Med Rehabil. 2014;95(2):353–9.
29. Doherty C, Zhao L, Ryan J, Komaba Y, Inomata A, Caulfield B. Quantification of postural control deficits in patients with recent concussion: an inertial-sensor based approach. Clin Biomech (Bristol, Avon). 2017;42:79–84.
30. Parker TM, Osternig LR, van Donkelaar P, Chou LS. Balance control during gait in athletes and non-athletes following concussion. Med Eng Phys. 2008;30(8):959–67.
31. Buckley TA. Acute and Lingering Impairments in Post-concussion Postural Control. In: Slobounov S, Sebastianelli W, editors. Concussions in Athletics. New York, NY: Springer; 2014.
32. Furman GR, Lin CC, Bellanca JL, Marchetti GF, Collins MW, Whitney SL. Comparison of the balance accelerometer measure and balance error scoring system in adolescent concussions in sports. Am J Sports Med. 2013;41(6):1404–10.
33. Balaban B, Tok F. Gait disturbances in patients with stroke. PM R. 2014;6(7):635–42.
34. Berg KO, Maki BE, Williams JI, Holliday PJ, Wood-Dauphinee SL. Clinical and laboratory measures of postural balance in an elderly population. Arch Phys Med Rehabil. 1992;73(11):1073–80.
35. Horak FB. Clinical measurement of postural control in adults. Phys Ther. 1987;67(12):1881–5.
36. Hirabayashi S, Iwasaki Y. Developmental perspective of sensory organization on postural control. Brain Dev. 1995;17(2):111–3.
37. Steindl R, Kunz K, Schrott-Fischer A, Scholtz AW. Effect of age and sex on maturation of sensory systems and balance control. Dev Med Child Neurol. 2006;48(6):477–82.
38. Hayes KC. Biomechanics of postural control. Exerc Sport Sci Rev. 1982;10:363–91.
39. Winter DA, Prince F, Frank JS, Powell C, Zabjek KF. Unified theory regarding A/P and M/L balance in quiet stance. J Neurophysiol. 1996;75(6):2334–43.
40. Kuo AD, Zajac FE. Human standing posture: multi-joint movement strategies based on biomechanical constraints. Prog Brain Res. 1993;97:349–58.
41. Pawlowski B, Grabarczyk M. Center of body mass and the evolution of female body shape. Am J Hum Biol. 2003;15(2):144–50.
42. Shumway-Cook A, Horak FB. Assessing the influence of sensory interaction of balance. Suggestion from the field. Phys Ther. 1986;66(10):1548–50.


Supplemental Digital Content

© 2018 American College of Sports Medicine