The acute effectiveness of ankle taping in dancers with chronic ankle instability – a randomized crossover study Authors: Jojo Hoi-Ching Lai, Tracy Cheuk-Chi Tang, Lucy Sum-Yu Lo, Cheryl Shu-Ming Chia, Samuel Ka-Kin Ling, Patrick Shu-Hang Yung Published: 2026 ABSTRACT BackgroundLateral ankle sprains and subsequent chronic ankle instability (CAI) are common among dancers, often presenting with impaired static and dynamic balance. Ankle taping is a conservative treatment option, but its effectiveness remains controversial. Dancers, who often exhibit greater generalized ligamentous laxity compared to other athletes, may respond differently to ankle taping, yet research specific to this population is limited. Hence, this study aimed to investigate the acute effects of taping on ankle stability and subjective functional outcomes in dancers with CAI, as well as to compare any outcome differences between CAI and non-CAI dancers.MethodsThis was a randomized crossover study in which eligible dancers were recruited in local dance institutions on a voluntary basis. Dancers underwent 4 ankle functional tests (single-leg stance, Y balance, side hop, figure-of-8 test) with and without ankle taping, clinical screening for ankle instability and generalized ligamentous laxity, as well as a questionnaire on their subjective ankle functional outcome (i.e., Cumberland Ankle Instability Tool/CAIT) and their perception of ankle taping. The primary outcome measure was single-leg stance result parameters, while secondary outcome measures included results from the remaining three ankle functional tests and the CAIT score. CAI in this study was defined as dancers with at least one ankle sprain and CAIT ≤24, the recommended cutoff by the International Ankle Consortium (IAC). Control group was defined as dancers with no ankle sprain or CAIT >24.ResultsEighty-nine eligible dancers from local dance institutions were recruited, with 85 (170 ankles) completing the study. 74.1% reported at least one ankle sprain, and 64.7% met the criteria for CAI. Ankle taping significantly improved static balance in dancers with non-structural instability (i.e., functional CAI or non-hypermobile dancers), but not for those with structural instability (i.e., mechanical CAI or hypermobile dancers). No significant improvements were observed in dynamic balance or other functional tests across all dancers.ConclusionWhile this study showed evidence that ankle taping can improve static balance in CAI dancers with non-structural instability (e.g., functional CAI, non-hypermobile), the effectiveness of reducing reinjury risks in dancers with CAI remained uncertain. These findings provided some support for the use of ankle taping as an adjunct for dancers with CAI. There is a need to explore alternative conservative treatments tailored to this population. Introduction Chronic ankle instability in dancers Chronic ankle instability (CAI) is highly prevalent in dancers. Lateral ankle sprain is the most common acute traumatic injury in dancers,1, 2, 3 and up to 83% of dancers experienced symptoms associated with CAI as a result.4 Most dancers prefer conservative over operative management,5 mainly due to their concerns particularly about time-loss in training, rehearsals, and performances. Undertreated CAI may eventually lead to premature post-traumatic ankle osteoarthritis as an end-stage sequelae. This inevitably takes a great toll on the healthcare system in managing these conditions, not to mention the time loss and psychophysiological impact on dancers, their later disability-adjusted life years, as well as revenue loss to dance companies. Current evidence on ankle taping in dancers with CAI Studies focusing on the conservative management of CAI in dancers are currently limited, particularly regarding the effectiveness of ankle taping. Unlike other ankle protective adjuncts, such as ankle braces and anti-inversion splints, which are commonly used in other sports, ankle taping may be a more feasible option because it is less bulky and easier to be accommodated by dancers due to its aesthetic nature. Ankle taping has been widely used in sports medicine and has been studied extensively over the years. Yet evidence on its effectiveness remains controversial with mixed results.6, 7, 8, 9 In dancers, one limited study10 suggested that ankle taping may reduce inversion-eversion rotation, increase inversion stiffness of the ankle complex, but there was no statistically significant difference in terms of jump landing time for stabilization. The latest systematic review and meta-analysis by Chamorro-Moriana et al.11 revealed that biomechanical or functional bandages applied to subjects with CAI are associated with improved patient perception, dynamic and static balance, kinematics, agility, and motor control. Yet, these pooled effect sizes were very small, and the meta-analysis showed no statistical significance. Dancers differ from other athletes in that they are associated with a higher prevalence of generalized ligamentous laxity (up to 76%),12,13 and are shown to perform better in balance tests.14 One limited study by Botsis et al.3 also demonstrated that dancers with CAI behave differently, including reduced eccentric plantar flexor strength and increased vertical ground reaction forces upon landing. While these dancers may respond differently to ankle taping compared to general athletes, research on the effect of ankle taping in this population is highly lacking. There is also no clear treatment consensus or recommendation, particularly on the use of ankle taping in dancers. Thus, this research study aimed to bridge this knowledge gap in the field of dance medicine and science, ultimately hoping to guide the management of dancers with CAI. Objectives and hypotheses The two objectives of this study were (i) to determine whether ankle taping is effective in dancers with CAI in terms of ankle stability and overall functional outcome; and (ii) to determine whether there are differences in the improvements between dancers with CAI and those without. The hypotheses were that ankle taping is effective in improving ankle mechanical and functional outcomes in dancers with CAI, and any improvement in these outcomes may be greater in dancers with CAI than in those without. Materials and methods This was a randomized crossover study (ankle taping versus no ankle taping in both dancers with and without CAI) (Fig. 1). The study protocol complied with the Declaration of Helsinki, and ethics approval was obtained from the joint Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee (Joint CUHK-NTEC CREC Ref no. 2024.561).Fig. 1Study design. Study design. Subject recruitment Dancers were recruited from local dance institutions on a voluntary basis based on the following inclusion and exclusion criteria (Table 1).Table 1Eligibility criteria of this study.Table 1Inclusion criteriaExclusion criteria18 years old or aboveHistory of lower limb operation(s)Both gendersRecent lower limb injuries in the past 3 months5+ years of dance trainingKnown neuromuscular diseasesWith/without chronic ankle instabilitySkin allergy to Kinesio tape Table 1Eligibility criteria of this study.Table 1Inclusion criteriaExclusion criteria18 years old or aboveHistory of lower limb operation(s)Both gendersRecent lower limb injuries in the past 3 months5+ years of dance trainingKnown neuromuscular diseasesWith/without chronic ankle instabilitySkin allergy to Kinesio tape Eligibility criteria of this study. Flow of study Dancers were subjected to 3 parts of the study: (i) online questionnaire, (ii) clinical screening exam, and (iii) ankle functional tests with and without ankle taping. This was a one-off 1-h visit to the CUHK-ORT Sports Injury Research Laboratory, Prince of Wales Hospital.(i)Online questionnaire (i)Online questionnaire Online questionnaire Data on (1) dancers’ demographics, and (2) their subjective ankle functional score (9-item Cumberland Ankle Instability Tool/CAIT15) were collected via online questionnaire (Qualtrics). This was completed by recruited dancers before their visit. Demographical data of interest included age, gender, type(s) of dance genre, years of dance training, number of training hours per week, leg dominance, and number of ankle sprain(s). Results were blinded from assessors upon subsequent clinical screening and ankle functional assessment.(ii)Clinical screening exam (ii)Clinical screening exam Clinical screening exam The second part of the study was performed by a registered orthopaedic surgeon. This included screening for (1) ankle instability (i.e., positive anterior drawer test) and (2) generalized ligamentous laxity (i.e., Beighton's score ≥5).(iii)Ankle functional tests (iii)Ankle functional tests Ankle functional tests Dancers were instructed to perform 4 ankle functional tests with and without ankle taping, namely (1) single-leg stance test, (2) Y balance test, (3) side hop test, and (4) figure-of-8 test. They were randomly allocated to undergo these tests with or without ankle taping first, as an attempt to address any potential confounding effect of muscle memory and fatigue during the study. A demonstration video was shown to dancers before their assessments. Trials were also given before data collection to further mitigate the muscle memory learning effect. Ankle taping was applied by a single registered physiotherapist, using the same brand of Kinesio tape, and the same taping configuration (i.e., Stirrup + 2 figure of 8, at 80% stretching tension) as shown below (Fig. 2). This method was adopted from a pre-existing taping study for dancers,10 as an attempt to reduce cross-study heterogeneity for future reviews on this similar topic.Fig. 2Ankle taping technique used in this study. Ankle taping technique used in this study. The above 4 functional tests were chosen as they were part of a recently published composite clinical score (Ankle-GO) recommended to predict readiness in return-to-sport after lateral ankle sprain or CAI.16,17 These were shown to be well-established tools with excellent internal consistency, construct validity, and test-retest reliability in the literature to assess static and dynamic balance, agility, and overall ankle function. Detailed standardized instructions for each test and the minimal detectable change of score were also available. Hence, results from this study could be cross-evaluated against existing data in the literature.1.Single-leg stance 1.Single-leg stance Single-leg stance This test assessed dancers’ static postural balance control. Dancers were instructed to stand barefoot on the testing leg on a Tekscan pressure mapping sensor plate with the knee slightly flexed (100), hands on the hips for 20 s with eyes closed. Data to be measured included number of errors, as well as 3 Tekscan outcome measurements (center of gravity perturbation (COP) area, anterior-posterior (AP) sway excursion, medial-lateral (ML) sway excursion). One error was recorded when the dancer did one of the following: open eyes, lift hands away from hips, lift forefoot or heel, move thigh into >300 flexion and abduction, or remain out of testing position for >5 s. Dancers were allowed 2 trials before undergoing the test, once on each leg.2.Y balance test (YBT) 2.Y balance test (YBT) Y balance test (YBT) This test assessed dancers’ dynamic postural balance control. Dancers were instructed to stand barefoot on the testing leg at the center of a “Y” and reach as far as possible with the opposite leg in 3 directions: anterior (ANT), posteromedial (PM), and posterolateral (PL), and return to the original position. The trial was omitted when the dancer did one of the following: lose balance and fall, transfer weight onto the non-testing leg, take hands off the hip, raise the heel of the testing leg, or move the testing leg. Dancers were allowed 1 learning trial in each direction for each leg before the actual test, with 3 trials being recorded for each leg. The average distance in the 3 trials in each direction was obtained and normalized in relation to the dancer's leg length (i.e., distance from anterior superior iliac spine to medial malleolus).3.Side hop test 3.Side hop test Side hop test In this ankle functional test, dancers were instructed to hop laterally and medially 10 times as fast as possible between 2 lines 30 cm apart. One time was defined as a hop to both sides, with the first hop always being towards the lateral side. The side-to-side hop was omitted if the dancer touched the line. The time to complete 10 side-to-side hops for each leg was recorded. 4.Figure-of-8 test (Fo8) 4.Figure-of-8 test (Fo8) Figure-of-8 test (Fo8) In this second ankle functional test, dancers were instructed to skip in a figure of 8 configuration around 2 posts 5m apart as fast as possible. The time to complete 2 consecutive laps (a total distance of 20m) was recorded. By the end of the visit, dancers were instructed to complete 4 questions regarding their perception of ankle taping (rating from 0 – never to 5 – always), namely:(i)Do you experience instability during the tasks WITHOUT taping?(ii)Do you think taping improves your instability and function during the tasks?(iii)Have you ever used ankle taping during dance?(iv)Will you consider ankle taping during dance? (i)Do you experience instability during the tasks WITHOUT taping? Do you experience instability during the tasks WITHOUT taping? (ii)Do you think taping improves your instability and function during the tasks? Do you think taping improves your instability and function during the tasks? (iii)Have you ever used ankle taping during dance? Have you ever used ankle taping during dance? (iv)Will you consider ankle taping during dance? Will you consider ankle taping during dance? This 0-5 perception scale represented frequency for question (i) and (iii), and perceived usefulness for question (ii) and (iv). It was arbitrarily divided into halves (i.e., 0-2 and 3-5) for more structured interpretation in discussion section. While these 4 items were not derived from validated instruments, they were designed to assess the potential for translating current research findings into the daily practice of dancers. Sample size calculation Depending on the type of dance genres, prevalence of CAI in dancers was quoted from 28 to 83% in the literature. Given the results from pilot studies, an effect size of 0.854, power at 0.90 and α = 0.05 under a two-tailed hypothesis, the calculated sample size was 60. Sample size was calculated using G∗Power 3.1.9.7. Primary and secondary outcomes The primary outcome was the result parameters of single-leg stance test with and without ankle taping, while secondary outcomes were those from YBT, side hop, and Fo8 test, as well as their CAIT scores. Subgroup definitions (i)CAI and non-CAI (control group) (i)CAI and non-CAI (control group) CAI and non-CAI (control group) In the study, dancers were categorized into the CAI versus non-CAI (control) group based on their questionnaire results, and the above outcomes were compared between the two groups. CAI in this study was defined as dancers with at least one ankle sprain and CAIT ≤24, aligning with the recommended cutoff value by the International Ankle Consortium (IAC).18 Control group was defined as dancers with no ankle sprain or CAIT >24. “Copers” were defined as dancers with at least 1 ankle sprain but did not experience CAI symptoms (CAIT >24).(ii)Mechanical and functional CAI (ii)Mechanical and functional CAI Mechanical and functional CAI Within the CAI group, dancers were further categorized into mechanical versus functional CAI. Mechanical CAI was defined as a positive anterior drawer test with CAIT ≤24, whereas functional CAI was defined as a negative anterior drawer test with CAIT ≤24.(iii)Hypermobile and non-hypermobile dancers (iii)Hypermobile and non-hypermobile dancers Hypermobile and non-hypermobile dancers Hypermobile dancers were defined as those with Beighton score ≥5 out of 9, while non-hypermobile dancers were defined as those with Beighton score <5. Statistical analysis All data were summarised as mean ± SD or median (for numeric data) or N (%) (for categorical data). The Kolmogorov–Smirnov test or the Shapiro–Wilk test was used to assess normality. A paired t-test was used for parametric data, while the Wilcoxon signed-rank test was used for non-parametric data. Parametric tests were determined using histograms and Q-Q plots. Spearman's rank correlation was used to assess associations in non-parametric data, while the biserial point correlation (Pearson correlation) was used to determine any association between ADT and other parameters (e.g., CAIT, Beighton's score). Both missing data and outliers were imputed using the mean from the grouping (i.e., no ankle taping or taping), with outliers assessed using a Q-Q plot and by identifying values more than ±2 standard deviations from the mean.19 Cohen's d (parametric test) and the effect size (r) for (non-parametric test)20 were calculated to determine the magnitude of the treatment effect, with values of 0.2, 0.5, and 0.8 representing small, medium, and large effects, respectively.21 Statistical analysis was carried out using IBM SPSS 29. Statistical significance was set at p < 0.05. Results Dancer recruitment and demographics A total of 89 eligible dancers were recruited, and 85 dancers (170 ankles) with complete data were evaluated. There were 76 female and 9 male dancers, with an average of 7.62 training hours per week. More than 90% of them were aged 18-35, and more than 80% had 10+ years of dance training. The top 3 dance genres in this study included Contemporary/Modern, Chinese dance, and Ballet, with most dancers undertaking multiple genres. In terms of ankle instability, 74.1% (63/85) of dancers had at least 1 ankle sprain, while 45.9% (39/85) had recurrent ankle sprains. In this study, the CAI group represented 64.7% (55/85) of the population, 63.5% (54/85) demonstrated generalized ligamentous laxity, and the percentage of “copers”, those with mechanical versus functional ankle instability were illustrated in Table 2.Table 2Ankle instability characteristics of recruited dancers.Table 2Dancer categoryCAI group (at least 1 sprain + CAIT ≤24)64.7% (55/85)Control group (no sprain/CAIT >24)35.3% (30/85)Generalized ligamentous laxity (Beighton ≥5)63.5% (54/85)Ankle function and instability“Coper” (at least 1 sprain + CAIT >24)9.4% (8/85)Ankle with mechanical CAI (positive ADT and CAIT ≤24)24.7% (42/170)Ankle with functional CAI (negative ADT and CAIT ≤24)45.9% (78/170) Table 2Ankle instability characteristics of recruited dancers.Table 2Dancer categoryCAI group (at least 1 sprain + CAIT ≤24)64.7% (55/85)Control group (no sprain/CAIT >24)35.3% (30/85)Generalized ligamentous laxity (Beighton ≥5)63.5% (54/85)Ankle function and instability“Coper” (at least 1 sprain + CAIT >24)9.4% (8/85)Ankle with mechanical CAI (positive ADT and CAIT ≤24)24.7% (42/170)Ankle with functional CAI (negative ADT and CAIT ≤24)45.9% (78/170) Ankle instability characteristics of recruited dancers. Effectiveness of ankle taping - overall In this study, ankle taping was associated with significantly improved static balance performance in all dancers (170 ankles), irrespective of the order of ankle taping. This was illustrated by a reduced anterior-posterior (AP) and medial-lateral (ML) excursion sway as seen in those with ankle taping compared to those without (Table 3).Table 3Ankle functional test results with and without ankle taping.Table 3Single-leg stance test (median (IQR))No ankle tapingAnkle tapingP valueArea of COP (cm2)20.17 (15.06, 31.68)19.58 (15.44, 25.07)0.072AP Excursion sway (cm)9.33 (7.24, 11.56)8.33 (6.99, 10.47)0.026ML Excursion sway (cm)5.10 (4.47, 9.01)5.03 (4.43, 7.36)0.023Number of errors0 (0,1)0 (0,1)<0.001Y balance test (mean (SD))No ankle tapingAnkle tapingMean differenceP valueAnterior (%)66.94 (6.42)67.30 (6.78)0.360.114Posterolateral (%)107.01 (8.92)107.18 (9.32)0.170.691Posteromedial (%)108.64 (8.64)109.09 (9.05)0.450.166No ankle tapingAnkle tapingP valueSide hop test (seconds)17.58 (13.62, 25.50)17.48 (13.89, 24.07)0.362Figure of 8 (seconds)11.76 (11.10, 12.63)11.70 (10.83, 12.65)0.310 Table 3Ankle functional test results with and without ankle taping.Table 3Single-leg stance test (median (IQR))No ankle tapingAnkle tapingP valueArea of COP (cm2)20.17 (15.06, 31.68)19.58 (15.44, 25.07)0.072AP Excursion sway (cm)9.33 (7.24, 11.56)8.33 (6.99, 10.47)0.026ML Excursion sway (cm)5.10 (4.47, 9.01)5.03 (4.43, 7.36)0.023Number of errors0 (0,1)0 (0,1)<0.001Y balance test (mean (SD))No ankle tapingAnkle tapingMean differenceP valueAnterior (%)66.94 (6.42)67.30 (6.78)0.360.114Posterolateral (%)107.01 (8.92)107.18 (9.32)0.170.691Posteromedial (%)108.64 (8.64)109.09 (9.05)0.450.166No ankle tapingAnkle tapingP valueSide hop test (seconds)17.58 (13.62, 25.50)17.48 (13.89, 24.07)0.362Figure of 8 (seconds)11.76 (11.10, 12.63)11.70 (10.83, 12.65)0.310 Ankle functional test results with and without ankle taping. However, while there was a trend of improvement in dynamic balance and the other 2 functional tests (Side hop, Fo8) with ankle taping, such effect did not reach statistical significance. Effectiveness of ankle taping - CAI vs. non-CAI/control group In this study, ankle taping was shown to be effective in significantly improving static balance performance in dancers with CAI (i.e., AP excursion sway, number of errors), but not in the control group (Table 4). No statistically significant improvement was observed for dynamic balance and the other 2 functional tests (Side hop, Fo8) for both CAI and control group.Table 4Ankle functional test results in CAI vs. control group.Table 4Single-leg stance test (median (IQR))No ankle tapingAnkle tapingP valueCAI groupArea of COP (cm2)22.24 (15.25, 37.85)19.36 (14.23, 28.85)0.056AP Excursion sway (cm)9.70 (7.38, 12.32)8.06 (6.79, 10.70)0.011ML Excursion sway (cm)5.29 (4.53, 11.03)5.03 (4.39, 8.55)0.051Number of errors1 (0,2)0 (0,1)<0.001Control groupArea of COP (cm2)18.85 (14.75, 28.04)20.13 (15.95, 25.12)0.941AP Excursion sway (cm)8.58 (6.73, 10.70)9.03 (7.19, 10.47)0.745ML Excursion sway (cm)5.08 (4.37, 7.07)4.94 (4.45, 6.13)0.192Number of errors0 (0,1)0 (0,1)0.048YBT test (mean (SD))No ankle tapingAnkle tapingMean differenceP valueCAI groupAnterior (%)67.43 (6.05)67.55 (6.27)0.120.729Posterolateral (%)108.16 (8.44)108.58 (8.55)0.430.522Posteromedial (%)110.03 (8.58)110.35 (8.41)0.330.488Control groupAnterior (%)66.52 (6.72)67.09 (7.20)0.560.073Posterolateral (%)106.06 (9.24)106.01 (9.80)−0.0500.187Posteromedial (%)107.49 (8.56)108.04 (9.47)0.550.930No ankle tapingAnkle tapingP valueSide hop test (seconds)CAI group18.33 (14.28, 26.44)18.18 (13.99, 24.41)0.530Control group16.93 (12.99, 23.44)16.15 (13.32, 23.63)0.452Fo8 test (seconds)CAI group11.98 (11.03, 12.56)11.78 (10.93, 12.90)0.731Control group11.56 (11.10, 12.77)11.54 (10.78, 12.45)0.052 Table 4Ankle functional test results in CAI vs. control group.Table 4Single-leg stance test (median (IQR))No ankle tapingAnkle tapingP valueCAI groupArea of COP (cm2)22.24 (15.25, 37.85)19.36 (14.23, 28.85)0.056AP Excursion sway (cm)9.70 (7.38, 12.32)8.06 (6.79, 10.70)0.011ML Excursion sway (cm)5.29 (4.53, 11.03)5.03 (4.39, 8.55)0.051Number of errors1 (0,2)0 (0,1)<0.001Control groupArea of COP (cm2)18.85 (14.75, 28.04)20.13 (15.95, 25.12)0.941AP Excursion sway (cm)8.58 (6.73, 10.70)9.03 (7.19, 10.47)0.745ML Excursion sway (cm)5.08 (4.37, 7.07)4.94 (4.45, 6.13)0.192Number of errors0 (0,1)0 (0,1)0.048YBT test (mean (SD))No ankle tapingAnkle tapingMean differenceP valueCAI groupAnterior (%)67.43 (6.05)67.55 (6.27)0.120.729Posterolateral (%)108.16 (8.44)108.58 (8.55)0.430.522Posteromedial (%)110.03 (8.58)110.35 (8.41)0.330.488Control groupAnterior (%)66.52 (6.72)67.09 (7.20)0.560.073Posterolateral (%)106.06 (9.24)106.01 (9.80)−0.0500.187Posteromedial (%)107.49 (8.56)108.04 (9.47)0.550.930No ankle tapingAnkle tapingP valueSide hop test (seconds)CAI group18.33 (14.28, 26.44)18.18 (13.99, 24.41)0.530Control group16.93 (12.99, 23.44)16.15 (13.32, 23.63)0.452Fo8 test (seconds)CAI group11.98 (11.03, 12.56)11.78 (10.93, 12.90)0.731Control group11.56 (11.10, 12.77)11.54 (10.78, 12.45)0.052 Ankle functional test results in CAI vs. control group. Effectiveness of ankle taping – mechanical vs. functional CAI In this study, ankle taping was associated with significantly better static balance performance in dancers with functional CAI, but not in those with mechanical CAI. This was reflected in reduced area of COP and AP sway excursion in the functional CAI group (Table 5). However, the results must be interpreted with caution in view of the small effect size. Furthermore, while mixed results with small effect size from ankle taping were yielded in the anterior direction of YBT between mechanical and functional instability group, there was no statistically significant improvement in overall dynamic balance or side hop tests with ankle taping.Table 5Ankle functional test results in mechanical vs. functional CAI.Table 5Single-leg stance test (median (IQR))No ankle tapingAnkle tapingEffect size (r)P valueMechanical CAI (n = 42)Area of COP (cm2)20.10 (13.39, 32.74)18.89 (15.15, 29.32)0.000.995AP Excursion sway (cm)9.68 (6.78, 11.46)8.37 (6.79, 12.07)0.030.866ML Excursion sway (cm)4.82 (4.28, 6.19)4.77 (4.19, 5.58)0.200.187Number of errors0 (0.00,1.00)0 (0.00,0.00)0.420.006Functional CAI (n = 78)Area of COP (cm2)22.34 (16.56, 33.04)19.49 (14.65, 26.15)0.230.042AP Excursion sway (cm)9.27 (7.38, 12.14)8.09 (6.77, 10.08)0.300.008ML Excursion sway (cm)5.49 (4.63, 11.30)5.09 (4.47, 8.48)0.210.070Number of errors0.00 (0.00,1.00)0 (0.00, 1.00)0.38<0.001YBT test (mean (SD))No ankle tapingAnkle tapingMean differenceCohen d (effect size)P valueMechanical CAI (n = 42)Anterior (%)66.50 (5.83)67.73 (6.41)1.230.390.015Posterolateral (%)106.71 (9.29)106.40 (8.98)−0.31−0.120.438Posteromedial (%)105.46 (9.11)105.32 (9.18)−0.14−0.030.854Functional CAI (n = 78)Anterior (%)66.56 (6.48)66.36 (6.91)−0.35−0.800.025Posterolateral (%)106.19 (8.62)106.58 (9.95)0.450.070.474Posteromedial (%)107.62 (8.37)108.03 (9.69)0.410.090.554Side hop test (seconds)No ankle tapingAnkle tapingEffect size (r)P valueMechanical CAI (n = 42)19.95 (5.58, 26.27)19.36 (14.70, 26.06)0.080.600Functional CAI (n = 78)17.12 (14.23, 22.96)16.50 (13.83, 22.46)0.030.786 Table 5Ankle functional test results in mechanical vs. functional CAI.Table 5Single-leg stance test (median (IQR))No ankle tapingAnkle tapingEffect size (r)P valueMechanical CAI (n = 42)Area of COP (cm2)20.10 (13.39, 32.74)18.89 (15.15, 29.32)0.000.995AP Excursion sway (cm)9.68 (6.78, 11.46)8.37 (6.79, 12.07)0.030.866ML Excursion sway (cm)4.82 (4.28, 6.19)4.77 (4.19, 5.58)0.200.187Number of errors0 (0.00,1.00)0 (0.00,0.00)0.420.006Functional CAI (n = 78)Area of COP (cm2)22.34 (16.56, 33.04)19.49 (14.65, 26.15)0.230.042AP Excursion sway (cm)9.27 (7.38, 12.14)8.09 (6.77, 10.08)0.300.008ML Excursion sway (cm)5.49 (4.63, 11.30)5.09 (4.47, 8.48)0.210.070Number of errors0.00 (0.00,1.00)0 (0.00, 1.00)0.38<0.001YBT test (mean (SD))No ankle tapingAnkle tapingMean differenceCohen d (effect size)P valueMechanical CAI (n = 42)Anterior (%)66.50 (5.83)67.73 (6.41)1.230.390.015Posterolateral (%)106.71 (9.29)106.40 (8.98)−0.31−0.120.438Posteromedial (%)105.46 (9.11)105.32 (9.18)−0.14−0.030.854Functional CAI (n = 78)Anterior (%)66.56 (6.48)66.36 (6.91)−0.35−0.800.025Posterolateral (%)106.19 (8.62)106.58 (9.95)0.450.070.474Posteromedial (%)107.62 (8.37)108.03 (9.69)0.410.090.554Side hop test (seconds)No ankle tapingAnkle tapingEffect size (r)P valueMechanical CAI (n = 42)19.95 (5.58, 26.27)19.36 (14.70, 26.06)0.080.600Functional CAI (n = 78)17.12 (14.23, 22.96)16.50 (13.83, 22.46)0.030.786 Ankle functional test results in mechanical vs. functional CAI. Effectiveness of ankle taping – hypermobile vs. non-hypermobile dancers Ankle taping was associated with significantly better static balance in non-hypermobile/control dancers, yet the effect size was small, ranging from 0.30 to 0.38. In contrast, this effect was not observed in hypermobile dancers (Table 6). Similarly, no statistically significant differences were found in dynamic balance or in the two functional tests between hypermobile and non-hypermobile dancers.Table 6Ankle functional test results in hypermobile vs. control group.Table 6Single-leg stance test (median (IQR))No ankle tapingAnkle tapingEffect size (r)P valueHypermobile (Beighton score ≥ 5) (n = 108)Area of COP (cm2)19.95 (14.81, 30.64)19.72 (15.98, 26.47)0.050.938AP Excursion sway (cm)9.26 (7.15, 11.38)8.58 (6.95, 10.76)0.080.371ML Excursion sway (cm)5.06 (4.47, 8.80)4.94 (4.50, 6.20)0.110.269Number of errors0 (0,1)0 (0,1)0.33<0.001Non-hypermobile/control (Beighton score <5) (n = 62)Area of COP (cm2)20.70 (15.58, 32.52)19.58 (14.99, 25.71)0.380.003AP Excursion sway (cm)9.46 (7.29, 12.37)8.12 (6.92, 10.29)0.320.011ML Excursion sway (cm)5.62 (4.46, 12.14)5.10 (4.37, 8.31)0.300.030Number of errors1 (0,1)0 (0,1)0.320.008YBT test (mean (SD))No ankle tapingAnkle tapingMean differenceCohen d (effect size)P valueHypermobile (Beighton score ≥ 5) (n = 108)Anterior (%)67.72 (6.13)68.20 (6.17)0.480.010.097Posterolateral (%)108.04 (8.70)108.34 (9.16)0.300.010.593Posteromedial (%)109.34 (8.14)109.72 (8.66)0.370.040.380Non-hypermobile/control (Beighton score <5) (n = 62)Anterior (%)65.68 (6.75)65.85 (7.53)0.170.170.097Posterolateral (%)105.12 (9.09)105.05 (9.32)−0.08−0.170.593Posteromedial (%)107.35 (9.41)108.00 (9.74)0.660.070.380No ankle tapingAnkle tapingEffect size (r)P valueSide hop test (seconds)Hypermobile (n = 108)17.83 (13.64, 22.56)17.81 (13.68, 25.06)0.010.966Non-hypermobile (n = 62)17.52 (13.51, 27.60)16.38 (14.30, 24.74)0.170.172Fo8 test (seconds)Hypermobile (n = 54)11.62 (11.13, 12.42)11.54 (10.90, 12.21)0.020.214Non-hypermobile (n = 31)12.03 (10.91,12.78)12.15 (10.81, 12.98)0.170.918 Table 6Ankle functional test results in hypermobile vs. control group.Table 6Single-leg stance test (median (IQR))No ankle tapingAnkle tapingEffect size (r)P valueHypermobile (Beighton score ≥ 5) (n = 108)Area of COP (cm2)19.95 (14.81, 30.64)19.72 (15.98, 26.47)0.050.938AP Excursion sway (cm)9.26 (7.15, 11.38)8.58 (6.95, 10.76)0.080.371ML Excursion sway (cm)5.06 (4.47, 8.80)4.94 (4.50, 6.20)0.110.269Number of errors0 (0,1)0 (0,1)0.33<0.001Non-hypermobile/control (Beighton score <5) (n = 62)Area of COP (cm2)20.70 (15.58, 32.52)19.58 (14.99, 25.71)0.380.003AP Excursion sway (cm)9.46 (7.29, 12.37)8.12 (6.92, 10.29)0.320.011ML Excursion sway (cm)5.62 (4.46, 12.14)5.10 (4.37, 8.31)0.300.030Number of errors1 (0,1)0 (0,1)0.320.008YBT test (mean (SD))No ankle tapingAnkle tapingMean differenceCohen d (effect size)P valueHypermobile (Beighton score ≥ 5) (n = 108)Anterior (%)67.72 (6.13)68.20 (6.17)0.480.010.097Posterolateral (%)108.04 (8.70)108.34 (9.16)0.300.010.593Posteromedial (%)109.34 (8.14)109.72 (8.66)0.370.040.380Non-hypermobile/control (Beighton score <5) (n = 62)Anterior (%)65.68 (6.75)65.85 (7.53)0.170.170.097Posterolateral (%)105.12 (9.09)105.05 (9.32)−0.08−0.170.593Posteromedial (%)107.35 (9.41)108.00 (9.74)0.660.070.380No ankle tapingAnkle tapingEffect size (r)P valueSide hop test (seconds)Hypermobile (n = 108)17.83 (13.64, 22.56)17.81 (13.68, 25.06)0.010.966Non-hypermobile (n = 62)17.52 (13.51, 27.60)16.38 (14.30, 24.74)0.170.172Fo8 test (seconds)Hypermobile (n = 54)11.62 (11.13, 12.42)11.54 (10.90, 12.21)0.020.214Non-hypermobile (n = 31)12.03 (10.91,12.78)12.15 (10.81, 12.98)0.170.918 Ankle functional test results in hypermobile vs. control group. Correlations among number of sprains, CAIT, ADT, Beighton score In this study, correlations among multiple factors were also evaluated. While there was weak correlation between number of sprains and ADT result (0.323∗∗), there was minimal correlation between (1) number of sprains and CAIT score (−0.241∗), (2) number of sprains with Beighton score (0.255∗), (3) CAIT score with ADT result (−0.230∗∗), and (4) CAIT score with Beighton score (−0.200∗∗). ∗∗. Correlation is significant at the 0.01 level (2-tailed) ∗. Correlation is significant at the 0.05 level (2-tailed) Dancers' perception of ankle taping While 63.5% (54/85) dancers experienced moderate to severe (rated 3-5/5) ankle instability during the tasks without taping, 84.7% (72/85) found that taping subjectively improved their ankle instability and function during the tasks. 47.1% (40/85) used ankle taping during dance in the past, and 75.3% (64/85) would consider using ankle taping during dance (rated 3-5/5). For those who rated 0-2, they either had no or minimal ankle instability, or they had concerns such as restricted movement, discomfort, troublesome application, and possible reliance on taping. Discussion Definition of CAI In this study, dancers served as the CAI group if they had at least 1 ankle sprain and CAIT score ≤24. Conversely, eight dancers were classified as “copers” (those with at least 1 ankle sprain but CAIT >24) and served as the non-CAI (control) group. All these 8 “copers” had one or two ankle sprains, but they did not experience subsequent instability symptoms. While the IAC consensus typically suggests a threshold of two sprains for "recurrent" instability, a CAIT threshold was adopted in this study to highlight the essential “sensory-perceptual” hallmark component of CAI (i.e., the absence of perceived instability and "giving way” as seen in “copers”). The rationale behind this present classification is to distinguish these "copers" as dancers who have achieved successful rehabilitation following acute trauma without functional sequalae, rather than to perceive them as those suffering from a chronic, persistent condition of CAI. Their ankle functional test performance should therefore lean towards the control group. By incorporating “copers” into control group, this might help reducing type II error. Otherwise, if “copers” were to be reclassified into CAI group as per IAC criteria, they would potentially wash out any significant improvement from ankle taping because “copers” do not have the same baseline deficits as the symptomatic CAI dancers. This approach aims to facilitate internal consistency by grouping dancers into the CAI and control group according to their current condition. However, it is acknowledged that this methodology may affect the generalizability of the findings compared to studies that strictly adhere to the two-sprain IAC criterion. Results should thus be interpreted with caution. Effectiveness of ankle taping in dancers In this study, ankle taping was shown to be effective in improving static balance in dancers, particularly in the CAI group but not in the control group. Current literature suggests that proprioception and cutaneous sensitivity account for 60-70% of standing balance control,22 while ankle taping works by reinforcing ankle proprioception.6, 7, 8 Thus, our result echoed the above as proprioception was shown to be impaired in individuals with CAI10,23 but intact in healthy individuals. No significant benefit from ankle taping would therefore be expected in individuals without CAI in this regard. Furthermore, based on subgroup analyses, ankle taping was shown to be effective in improving static balance in (i) dancers with functional CAI but not mechanical CAI, and (ii) non-hypermobile dancers but not hypermobile counterparts. While mechanical CAI is characterized by pathological laxity which can be detected objectively from physical examination, functional CAI generally lacks frank structural pathologies. Likewise, similar theory applies to hypermobile versus non-hypermobile dancers. Individuals with joint hypermobility and generalized ligamentous laxity were shown to have impaired proprioception, particularly in the lower limbs.24 Compared to non-hypermobile dancers, it would be a double toll for hypermobile dancers with already reduced proprioception from preexisting ligamentous laxity to have an ankle sprain and further impair their proprioception. As a result, this would explain our observation that while ankle taping appeared to be effective in overcoming non-structural instability as seen in functional CAI and non-hypermobile dancers, any significant structural ankle instability with much impaired proprioception in mechanical CAI and hypermobile dancers could not be compensated by ankle taping alone. As for dynamic balance and other functional tests such as side hop and Fo8, there was no statistically significant overall improvement with ankle taping observed in this study. While these functional tests may be more reflective of real-life scenarios in dancing when compared to static balance test, they rely not only on ankle proprioception but also on neuromuscular coordination, proximal joint control (hip and trunk), eccentric muscle strength, and intersegmental movement strategies. In addition, dancers typically demonstrate high baseline balance ability and well-developed compensatory motor control strategies, which may reduce the detectable contribution of ankle taping during dynamic tasks. These potential confounders may hence override any “weak” effect of ankle taping. Strengths of this study To our understanding, this was the first study to investigate the effect of ankle taping in dancers with focus on chronic ankle instability. Although ankle taping has been extensively studied in athletes and results remained controversial, special consideration for dancers is indicated as they are associated with a higher prevalence of generalized ligamentous laxity when compared to the general athletic population. They may have a different response to ankle taping as a result. In addition, this study included a crossover design to address potential confounders such as muscle memory from the learning effect, as well as muscle fatigue. It also integrated evaluation from all three perspectives, from dancers’ subjective perception on ankle taping and their ankle instability, their functional test performances, to objective physical examination by clinician. Overall, it shed light on how these functional tests in the lab can correlate with physical examination in actual clinical setting. Limitations There were several limitations in this study. The study population consisted predominantly of female participants. There was also a bias in the types of dances included. However, these reflected the overall profile of Hong Kong's local dance community. Furthermore, smaller subgroup sample sizes may reduce the effect size; hence, the results may lead to a type II error, underestimating the true effect. No subgroup analysis by dance type was performed, as many dancers receive training in multiple dance genres simultaneously. It would be clinically impractical to highlight any differences among genres. In terms of methods, this study involved a one-off testing protocol, such that only the acute effect of ankle taping could be tested. Previous studies suggested that the effectiveness of ankle taping and how long these benefits last may be affected by the duration of application.9 Future studies can introduce serial testing in different time frames after taping (with tape kept intact or removed). This study did not explicitly assess whether taping restricts range of motion. However, there has already been existing evidence which demonstrated that Kinesio taping using the same taping method as our study could restrict ankle motion in the frontal plane (inversion and eversion).10 In addition, possible confounders such as muscle fatigue and muscle memory from learning effect might not be adequately addressed in this study. To address these 2 factors, a crossover design was used in which dancers were randomly allocated to 2 groups: half had their tests performed with ankle taping first, and the other half without. They were given a brief period of rest before they switched to the other group. Trials were also allowed before actual data collection according to the established test protocol15 to further mitigate the possible effect of muscle memory. However, with reference to this protocol, there was no trial given to 2 out of the 4 tests (side hop and Fo8). While there was an attempt in our study to follow this protocol as closely as possible to reduce heterogeneity with existing literature for future comparative studies or systematic reviews, there was an observation that many dancers who had ankle taping first in this study tend to perform better on their second trial in these 2 tests, in which they had their taping removed. Since these tests depend on speed and coordination, they are particularly susceptible to learning effect. This issue is especially relevant given the crossover study design, as dancers who performed one condition first may have improved performance in the second condition due to task familiarization rather than the intervention itself. Such potential influence of learning effect might thereby partially contribute to the non-significant findings in these 2 functional tests as a methodological limitation. Lastly, a few outliers in the functional outcomes were noted in this study, and they were subsequently addressed statistically. Most of these outliers belonged to dancers who had not been actively dancing recently (e.g., dance teachers who mainly demonstrate dance movements, recreational off-season dancers). In retrospect, additional inclusion criteria should be set to address such concerns (e.g., active dance training in the past month). Overall, while this study offered evidence of improving static balance in CAI dancers without structural instability (e.g., non-hypermobile, functional CAI), it is unclear how much of this outcome can be translated into reducing actual reinjury risks in dancers with CAI. Future studies may also further explore the effect of ankle taping on dance performance, which is beyond the scope of the current study. Conclusion This study was the first to investigate the acute effects of ankle taping in dancers with CAI. Our findings suggested that ankle taping may improve static balance in CAI dancers with non-structural instability (e.g., non-hypermobile, functional CAI). However, taping alone was insufficient to address structural ankle instability, particularly in dancers with mechanical CAI and generalized ligamentous laxity. Furthermore, no statistically significant improvement was observed in dynamic balance or other functional ankle tests in all dancers. In conclusion, this study suggested some support for the use of ankle taping for CAI dancers without structural instability as an adjunct. There is still a need to explore alternative conservative treatments tailored to dancers’ unique biomechanical demands. Funding/supporting statement This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors, and no material support of any kind was received. Declaration of interest There are no conflicts of interest related to this research.