Within the athletic population, the ankle joint is one of the most commonly injured joints in the body.1 One traumatic ankle sprain can lead to multiple recurring injuries and episodes of “giving way” resulting in pain, swelling, and loss of function.2 The symptoms of instability and balance impairment are collectively described as chronic ankle instability (CAI).1 Residual symptoms from CAI are reported in up to 72% of those who suffer an ankle sprain.3,4 Strong correlations between CAI and decreased quality of life have been found in both general populations and physically active participants.5,6 Specific health care implications for those who suffer recurrent ankle sprains include significant time loss from physical activity, osteoarthritis, and articular degeneration.7 Of patients with ankle sprains, 13% to 15% are occupationally handicapped for at least 9 months and some for longer than 6 years, with 6% of patients with recurring sprains not returning to work at all.8,9
Because of the negative impact of recurrent ankle sprains on health and physical activity, screening tools for instability of the ankle joint can lead to therapeutic intervention to improve joint stability and prevent future ankle sprains. Several self-report screening instruments have shown good statistical significance in determining the presence of ankle instability in patients.10–12 The Foot & Ankle Disability Index and Foot & Ankle Disability Index Sport Module have been shown to have impaired scores in patients with a history of ankle injury.10,11 The Cumberland Ankle Instability Tool and the Foot and Ankle Ability Measure were both found to have impaired scores in patients with CAI compared to patients with no history of ankle injury.12 Along with self-reported screening instruments, static single leg balance testing is another screening tool often used to assess a patient’s balance and ankle stability, because single leg balance impairments are strongly associated with CAI and can be a predictor of ankle injury.13–17
There are several methods used to assess static postural control, some of which are already included in baseline measurements taken during pre-participation screenings. One such method is the Balance Error Scoring System (BESS), originally developed as a concussion assessment tool that challenges static balance through a variety of stances on stable and unstable surfaces. The BESS has shown moderate to excellent intra-tester reliability with intraclass correlation scores ranging from .6018 to .9219 for overall BESS score and .5020 to .9821 for individual stances. Inter-tester reliability of the BESS was found to be moderate to excellent, with intraclass correlation scores ranging from .5720 to .9622 and four stances having intraclass correlation scores greater than .84.22 The BESS also has high retest reliability on both stable and unstable surfaces (.93 and .92).22 Validity of the BESS as a balance assessment has been established through comparisons of errors scores with results of instrumented balance testing devices. Significant correlations with target sway were observed in five of six stances of the BESS, with the double leg firm resulting in no errors, so no correlation calculation was possible.22 Through the multiple measures mentioned above, the BESS has been shown to be a reliable and valid assessment of static single leg balance.
The positive results of reliability and validity testing of the BESS increase its value as an assessment tool. Researchers using the BESS found a significant difference in number of errors committed during single leg firm, single leg foam, tandem firm, and tandem foam stances between unstable ankle groups and control groups.16 Most clinics and athletic training facilities have access to the minimal equipment necessary for testing and with some baseline testing protocols already in place as part of concussion management guidelines, scores could be easily available for identifying those who may be most in need of therapeutic intervention following residual ankle instability. This allows the BESS to serve a dual purpose as a concussion and an ankle assessment tool. Therefore, the purpose of our study was to determine which stances of the BESS Division I collegiate athletes with CAI perform poorly and to determine cut-off scores. This study could help clinicians streamline identification of individuals who may be most in need of a rehabilitation program aimed at increasing postural control following recurrent ankle sprains.
Fifty-one Division I collegiate athletes volunteered to participate in our case–control study. Thirty-two individuals (15 men and 17 women) with a history of ankle sprains and the sensation of giving way were included in the CAI group (mean age: 20 ± 1 years [range: 18 to 22 years]; mean height: 177 ± 11 cm [range: 157.5 to 205.7 cm] ; mean mass: 71 ± 16 kg [range: 47 to 108 kg]). Nineteen individuals (7 men and 12 women) with no history of ankle injury were included in the control group (mean age: 20 ± 1 years [range: 18 to 22 years]; mean height: 178 ± 8 cm [range: 165.1 to 190.5 cm]; mean mass: 73 ± 16 kg [range: 52 to 100 kg]). All volunteers were Division I mid-major athletes. Inclusion criteria for the CAI group were (1) history of at least one significant ankle sprain, (2) self-reported sensations of giving way during activity, and (3) no signs and symptoms of an acute injury.2 Exclusion criteria for all volunteers were (1) any known knee or hip injuries that limit function, (2) any signs or symptoms of acute injury or injury occurrence within the previous 6 weeks, including concussions, (3) previous history of surgery on involved lower extremity, (4) shoe size greater than 14, (5) any vision deficits (other than myopia, hyperopia, and astigmatism), and (6) any other balance disorder that may affect performance. The study was approved by the university’s institutional review board, and informed written consent was provided by each participant prior to participation in the study.
Data for all BESS testing was collected in the Sports Medicine Research Laboratory by a single investigator, a certified athletic trainer (BD). All trials were recorded using a video camera and scored individually at a later time. Each stance of the BESS was explained prior to testing using a script. Participants completed one trial for each stance on each surface. The stable surface was the laboratory floor and the unstable surface was an AIREX Balance Pad (Airex, Sins, Switzerland; dimensions: 50.8 × 41.7 × 6.4 cm).
Participants performed the six stances in the following order: double legged (feet side by side) on a firm surface, single legged on a firm surface, tandem (leg with CAI or matched test leg placed directly behind the heel of the contralateral foot) on a firm surface, double legged on a foam surface, single legged on a foam surface, and tandem on a foam surface. The single-legged stances were performed with the non-test leg in slight flexion at the knee and the hip. As a concussion assessment the participants’ non-dominant limb determines limb for single leg stance and rear foot in tandem stance. For this study, the stance leg for the CAI group was determined using a questionnaire about the participant’s injury history, which identified limb with symptoms associated with CAI. Participants with bilateral CAI were tested on the leg they felt was the more unstable of the two. The leg tested in the single leg stance was also tested as the rear foot in the tandem stance. The stance leg for the control group was determined by questionnaire identifying which limb was their non-dominant limb. The leg tested in single leg stance was the self-reported non-dominant limb, and was also the rear foot in the tandem stance.
Before all testing began, participants were instructed to remove their shoes and socks. Prior to each test, participants were instructed to keep their hands on their hips and their eyes closed, and to remain as motionless as possible for 20 seconds. An error was recorded if they lifted their hands off hips, flexed or abducted their hip greater than 30 degrees, lifted their forefoot or heel, opened their eyes, touched down with the non-test foot, or remained out of testing position for more than 5 seconds. Participants were given a chance to familiarize themselves before each test stance with their eyes open. Once they were comfortable in the test stance they were instructed to close their eyes and the test would begin. The total number of errors in each stance was used for analysis.
SPSS software (version 20.0; SPSS, Inc., Chicago, IL) was used for statistical analyses. Means ± standard deviations were calculated for all dependent measures. Effect size values between groups were calculated with Cohen’s d and values of 0.20, 0.50, and 0.80 were defined as low, medium, and high, respectively.23 Sensitivity and 1-specificity values were calculated for each significant dependent measure across the range of possible scores to compute receiver operating characteristic (ROC) curves. Area under the curve (AUC) and asymptotic significant values were then calculated (α = 0.05). AUC is an indicator of the overall value of the variable for accurate discrimination between all possible cut-points for dichotomous categorizations of cases. Next, cut-off scores were computed with Youden’s Index [((Sensitivity + Specificity) – 1) * 100].24 Positive and negative likelihood ratios were calculated from the sensitivity and specificity values. Then, odds ratios were used to determine if a specific cut-off score could distinguish individuals with and without CAI (+likelihood ratio divided by –likelihood ratio).25 The odds ratio was selected as an outcome variable because it is an indicator of the discriminatory power of the variable being analyzed and provides the magnitude of association with a classification of either having CAI or not.25 An odds ratio will exceed 1 if the variable of interest is worse in those with CAI versus stable ankles.25 Furthermore, the greater the odds ratio is, the greater the association with CAI. Finally, a one-tailed Fisher’s exact test determined the statistical significance of the selected cut-off score for each dependent measure by providing a means to identify a substantial deviation from the expected frequencies of occurrence that would result from chance (α = 0.05).26 The smaller the P value, the stronger the evidence the two proportions are truly different.26
Means and standard deviations for each group are displayed in Table 1. All dependent measures (AUC, P values, cut-off scores, sensitivity, 1-specificity, odds ratios, 95% confidence intervals, Fisher’s exact test, and Youden’s index results), are presented in Table 2. ROC analyses found significant AUC values for four stances of the BESS, single leg stance on firm surface, tandem stance on firm surface, single leg stance on foam surface, and tandem stance on foam surface, all with large odds ratios. Significant cut-off scores were found for the single leg stance on firm surface, single leg stance on foam surface, and tandem stance on foam surface.
The primary finding of this study was three stances of the BESS that identified individuals with CAI most in need of a rehabilitation program aimed at improving postural stability, with significant error scores. The study also identified AUC values greater than .80 in two stances (single leg stance on foam surface and tandem stance on foam surface), both with large odds ratios, large effect sizes, and small Fisher exact test values.
In a previous study of Division I athletes, error scores in participants with CAI in single leg on firm surface (2.9 ± 2.1 errors), single leg on foam surface (7.0 ± 1.6 errors), and tandem stance on foam surface (4.3 ± 2.4 errors) were found to be significant when compared to error scores of a control group.16 The results of this study are similar to the findings of our study for the same stances (single leg on firm surface: 2.28 ± 1.80 errors; single leg on foam surface: 5.84 ± 1.37 errors; tandem stance on foam surface: 3.81 ± 1.28 errors). The similarities in results between these two studies may be attributed to similar participant populations. Both studies consisted of male and female Division I athletes, and both studies had a similar total number of participants (60 total participants in the previous study, 51 total participants in our study). However, there are large differences when comparing effects sizes of the two studies. The previous study found a much larger effect size for the single leg stance on firm surface (0.74)16 compared to ours (0.28). We found much larger effect sizes for the single leg stance on foam surface (1.73) compared to theirs (0.82)16 and tandem stance on foam surface (1.24) compared to theirs (0.78).16 These differences could be related to differences between sport participation of those who volunteered for each study. It would be beneficial to determine error score differences between Division I athletes participating in different sports.
Double leg stance on firm surface and double leg stance on foam surface had nonsignificant findings based on the low error scores and lack of variability between the two groups. As a result, both stances revealed low AUC values (.50 and .47, respectively). This is consistent with previous studies that showed no significant difference between group scores of double leg stances.16,27 One study looked at the reliability of the modified BESS, which removes the double leg stance on firm surface and the double leg stance on foam surface from the BESS testing protocol. They found that removing the double leg stances and testing only the single leg and tandem stances on firm and foam surfaces increased the reliability of the BESS from .60 to .71.18 The previous findings of the consistent lack of variability of error scores of the double leg stances indicates that these two stances of the BESS may not need to be performed when identifying individuals with CAI who may be most in need of rehabilitation.
Division I vs General Population
In a generally active population, error scores in single leg on firm surface (2.53 ± 2.37 errors), tandem stance on firm surface (1.29 ± 1.53 errors), and tandem stance on foam surface (3.71 ± 1.65 errors) were found to be indicators of balance deficits associated with CAI.27 The error scores found in these stances are similar to the results found in our study (single leg on firm surface: 2.28 ± 1.80 errors; tandem stance on firm surface: 1.22 ±1.24 errors; tandem stance on foam surface: 3.81 ± 1.28 errors). Significant AUC values were found in the generally active population (single leg on firm surface .66; tandem stance on firm surface .55; tandem stance on foam surface .60).27 In our study, AUC values for the same three stances were .55, .64, and .80. Significant odds ratios were found for each of the three stances in both studies. Our study found higher odds ratios in both single leg on firm surface (7.39 vs 5.25) and tandem stance on firm surface (3.65 vs 1.63). Only in tandem stance on foam surface was the odds ratio found to be greater in the generally active population than in our study (6.67 vs 4.82).27 Increased difficulty of the foam surface may be one reason for a greater odds ratio in a generally active population in the tandem stance. The single leg on firm surface and tandem stance on firm surface may be difficult enough for a generally active population, but not increasingly difficult for a Division I athlete with CAI.
The single leg stance on foam surface was not found to be statistically significant in a study of the generally active population,27 but was found to be statistically significant in multiple categories in our study. In a generally active population, participants with both documented CAI and no history of ankle injury had an increased number of error scores in the single leg stance on foam surface (5.59 ± 1.33 in control; 6 ± 1 in CAI).27 Scores for both groups were so close that an odds ratio between groups could not be calculated. In our study, the single leg stance on foam surface had a significant AUC value of .88, odds ratio of 16.26, and Fisher exact test value of 0.00003, which are all strong indicators of the single leg stance on foam surface as a good measure of ankle instability in Division I athletes. Because the uninjured group in a population consisting of highly athletic and highly competitive participants is able to complete the task of standing single legged on a foam pad with their hands on their hips and their eyes closed better than those with documented CAI, error scores are able to identify those with CAI who are most in need of a rehabilitation program.
The increased levels of athleticism in Division I athletes who train through a variety of planes and motions may better allow them to maintain postural stability in abnormal body positions than the generally active individual. This could indicate the use of fewer stances of the BESS when identifying Division I athletes with CAI who may be most in need of a rehabilitation program.
A couple of limitations were present in our study. One limitation was the evaluator of BESS errors was not blinded to group membership and may have unintentionally influenced error scores. Another study limitation was participant foot size. In pilot testing for this study, it was found that participants with a shoe size greater than 14 were not able to fit properly on the AIREX Balance Pad. A question regarding shoe size was added to the injury history questionnaire, and participants with a shoe size greater than 14 were excluded from both potential CAI and control groups. Finally, even though participants categorized with CAI self-reported a history of at least one significant ankle sprain with sensations of “giving way,” a general self-reported foot and ankle function questionnaire was not part of the inclusion criteria. The clinical relevance of this study was to streamline the process of identifying individuals most in need of a rehabilitation program due to balance deficits associated with CAI. An entire questionnaire to determine self-reported ankle function could add to an already lengthy pre-participation examination process for both the clinician and the athlete. The use of the BESS as a baseline concussion assessment and the simple question of previous history of at least one ankle injury with any residual symptoms will allow clinicians to identify potential candidates for rehabilitation without unnecessary burden.
Future research should be conducted to examine cut-off scores at different levels of competition, as well as examining cut-off scores in athletes of various sports. Further research is also need for development of cut-off scores in a “coper” group. Copers have been described in the literature as individuals who have a history of one lateral ankle sprain with no residual instability.12,28
Implications for Clinical Practice
The purpose of our study was to determine which stances of BESS, with associated cut-off scores, best identify Division I collegiate athletes with CAI most in need of a rehabilitation program. Clinicians can use the single leg stance on a foam surface (≥ 5 errors) and tandem stance on a foam surface (≥ 3 errors) to identify balance deficits associated with CAI in Division I mid-major athletes. This allows clinicians to know specifically which stances to use and what error scores to look for to quickly and accurately identify individuals who may be most in need of rehabilitation programs addressing postural stability.
Neuromuscular training programs have been shown to improve error scores in several stances of the BESS, as well as overall BESS error score.29 Following a 6-week training program, single leg stance on foam surface, tandem stance on foam surface, and total BESS showed statistically significant improvements from pretest to post-test error scores in female high school basketball players who participated in the program.29 This is an important factor in the use of the BESS to measure postural control ability following the identification of CAI. Once a participant is identified as in need of a preventative rehabilitation program, participants can be retested using the BESS to measure improvement in postural control. A participant who scores under documented cut-off scores for each significant stance of the BESS and total BESS score may also show an improvement in self-reported ankle function.
Multiple studies have shown increases in total BESS score following exertion protocols. Total BESS score has shown to increase following fatigue in a generally active population. Total error scores increased after fatigue regardless of gender.19 Similar results were found in a collegiate population; total BESS scores increased after fatigue.30 These findings are important to consider when conducting BESS testing at baseline and after rehabilitation. Participation in exercise may cause increased error scores in the multiple stances of the BESS used for determining those most in need of a rehabilitation program. Pre-practice or pre-workout testing at baseline and after rehabilitation may result in the most accurate representation of ankle stability.
- Hertel J. Functional anatomy, pathomechanics and pathophysiology of lateral ankle instability. J Athl Train. 2002;37:364–374.
- Delahunt E, Coughlan GF, Caulfield B, Nightingale EJ, Lin CC, Hiller CE. Inclusion criteria when investigating insufficiencies in chronic ankle instability. Med Sci Sports Exerc. 2010;42:2106–2021. doi:10.1249/MSS.0b013e3181de7a8a [CrossRef]
- Braun BL. Effects of ankle sprain in a general clinic population 6 to 18 months after medical evaluation. Arch Fam Med. 1999;8:143. doi:10.1001/archfami.8.2.143 [CrossRef]
- Gerber JP, Williams GN, Scoville CR, Arciero RA, Taylor DC. Persistent disability associated with ankle sprains: a prospective examination of an athletic population. Foot Ankle Int. 1998;19:653–660. doi:10.1177/107110079801901002 [CrossRef]
- Arnold BL, Wright CJ, Ross SE. Functional ankle instability and health-related quality of life. J Athl Train. 2011;46:634–641.
- Houston MN, Van Lunen BL, Hoch MC. Heath-related quality of life in individuals with chronic ankle instability. J Athl Train. 2014;49:758–763. doi:10.4085/1062-6050-49.3.54 [CrossRef]
- Valderrabanno V, Hintermann B, Horisberger M, Fung TS. Ligamentous post traumatic ankle osteoarthritis. Am J Sports Med. 2006;34:612–620. doi:10.1177/0363546505281813 [CrossRef]
- Schaap GR, de Keizer G, Marti K. Inversion trauma of the ankle. Arch Orthop Trauma Surg. 1989;108:273–275. doi:10.1007/BF00932312 [CrossRef]
- Verhagen RAW, de Keizer G, van Dijk CN. Long-term follow-up of inversion trauma of the ankle. Arch Orthop Trauma Surg. 1995;114:92–96. doi:10.1007/BF00422833 [CrossRef]
- Hale SA, Hertel J, Olmsted-Kramer LC. The effect of a 4-week comprehensive rehabilitation program on postural control and lower extremity function in individuals with chronic ankle instability. J Orthop Sports Phys Ther. 2007;37:303–311. doi:10.2519/jospt.2007.2322 [CrossRef]
- McKeon PO, Ingersoll CD, Kerrigan DC, Saliba E, Bennett BC, Hertel J. Balance training improves function and postural control in those with chronic ankle instability. Med Sci Sports Exerc. 2008;40:1810–1819. doi:10.1249/MSS.0b013e31817e0f92 [CrossRef]
- Wright CJ, Arnold BL, Ross SE, Ketchum J, Ericksen J, Pidcoe P. Clinical examination results in individuals with functional ankle instability and ankle-sprain copers. J Athl Train. 2013;48:581–589. doi:10.4085/1062-6050-48.3.15 [CrossRef]
- Wang HK, Chen CH, Shiang TY, Jan MH, Lin KH. Risk-factor analysis of high school basketball-player ankle injuries: a prospective controlled cohort study evaluating postural sway, ankle strength, and flexibility. Arch Phys Med Rehabil. 2006;87:821–825. doi:10.1016/j.apmr.2006.02.024 [CrossRef]
- McGuine TA, Greene JJ, Best T, Leverson G. Balance as a predictor of ankle injuries in high school basketball players. Clin J Sport Med. 2000;10:239–244. doi:10.1097/00042752-200010000-00003 [CrossRef]
- Tropp H, Ekstrand J, Gilquist J. Stabilometry in functional instability of the ankle and its value in predicting injury. Med Sci Sports Exerc. 1984;16:64–66. doi:10.1249/00005768-198401000-00013 [CrossRef]
- Docherty CL, McLeod TCV, Shultz SJ. Postural control deficits in participants with functional ankle instability as measured by the Balance Error Scoring System. Clin J Sport Med. 2006;16:203–208. doi:10.1097/00042752-200605000-00003 [CrossRef]
- Arnold BL, de la Motte S, Linens SW, Ross SE. Ankle instability is associated with balance impairments: a meta-analysis. Med Sci Sports Exerc. 2009;41:1048–1062. doi:10.1249/MSS.0b013e318192d044 [CrossRef]
- Hunt TN, Ferrar MS, Bornstein RA, Baumgartner TA. The reliability of the modified Balance Error Scoring System. Clin J Sport Med. 2009;19:471–475. doi:10.1097/JSM.0b013e3181c12c7b [CrossRef]
- Erkman N, Taskin H, Kaplan T, Sanioglu A. The effect of fatiguing exercise on balance performance as measured by the Balance Error Scoring System. Isokinet Exerc Sci. 2009;17:121–127.
- Finnoff JT, Peterson VJ, Hollman JH, Smith J. Intrarater and interrater reliability of the Balance Error Scoring System (BESS). Phys Med and Rehabil. 2009;1:50–54.
- Valovich McLeod TC, Perrin DH, Guskiewicz KM, Shultz SJ, Diamond R, Gansneder BM. Serial administration of clinical concussion assessments and learning effects in healthy young athletes. Clin J Sport Med. 2004;14:287–295. doi:10.1097/00042752-200409000-00007 [CrossRef]
- Reiman BL, Guskiewicz KM, Shields EW. Relationship between clinical and forceplate measures of postural stability. J Sport Rehabil. 1999;8:71–82.
- Cohen J. Statistical Power Analysis for the Behavioral Science. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers; 1988.
- Bewick V, Cheek L, Ball J. Statistics review 13: receiver operating characteristic curves. Crit Care. 2004;8:508–512. doi:10.1186/cc3000 [CrossRef]
- Fletcher RH, Fletcher SW. Clinical Epidemiology–The Essentials, 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2005.
- Agresti A, Finlay B. Statistical Methods for the Social Sciences, 3rd ed. Upper Saddle River, NJ: Prentice-Hall International; 1997.
- Linens SW, Ross SE, Arnold BL, Gayle R, Pidcoe P. Postural-stability tests that identify individuals with chronic ankle instability. J Athl Train. 2014;49:15–23. doi:10.4085/1062-6050-48.6.09 [CrossRef]
- Wickstrom EA, Fournier KA, McKeon PO. Postural control differs between those with and without chronic ankle instability. Gait & Posture. 2010;32:82–86. doi:10.1016/j.gaitpost.2010.03.015 [CrossRef]
- McLeod TCV, Armstrong T, Miller M, Sauers JL. Balance improvement in female high school basketball players after a 6-week neuromuscular-training program. J Sport Rehabil. 2009;18:465–481.
- Bell DR, Guskiewicz KM, Clark MA, Padua DA. Systematic review of the Balance Error Scoring System. Sports Health. 2011;3:287–295. doi:10.1177/1941738111403122 [CrossRef]
|STANCE||GROUP (MEAN ± SD)||EFFECT SIZE|
|CHRONIC ANKLE INSTABILITY (n = 32)||CONTROL (n = 19)|
|Double leg firm surface||0 ± 0||0 ± 0||N/A|
|Double leg foam surface||0.06 ± 0.25||0.16 ± 0.50||0.25|
|Single leg firm surface||2.28 ± 1.80||1.84 ± 1.30||0.28|
|Single leg foam surface||5.84 ± 1.37||3.95 ± 0.71||1.73|
|Tandem firm surface||1.22 ± 1.24||0.58 ± 0.77||0.62|
|Tandem foam surface||3.81 ± 1.28||2.47 ± 0.84||1.24|
|DEPENDENT MEASURE||AUC||ASYMPTOTIC SIGNIFICANCE||CUT-OFF SCORE||SEN||SPEC||+LR||−LR||OR||95% CI||P||YI|
|Double leg firm||0.50||1.0||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A||N/A|
|Double leg foam||0.48||.785||1||0.06||0.11||0.55||1.06||0.57||0.07 to 4.39||.47||−5.00|
|Single leg firm||0.55||.527||4||0.28||0.05||5.60||0.76||7.04||0.82 to 60.84||.05a||23.00|
|Single leg foam||0.88||< .001a||5||0.81||0.21||3.85||0.24||16.25||3.94 to 66.95||.00003a||60.20|
|Tandem firm||0.64||.098||2||0.41||0.16||2.56||0.70||3.65||0.88 to 15.11||.06||25.00|
|Tandem foam||0.80||< .001a||3||0.81||0.47||1.72||0.36||4.81||1.36 to 17.05||.01a||33.90|