Secondary Logo

Journal Logo


Do External Supports Improve Dynamic Balance in Patients with Chronic Ankle Instability? A Network Meta-analysis

Tsikopoulos, Konstantinos MD, MSc; Sidiropoulos, Konstantinos MD; Kitridis, Dimitrios MD, MSc; Cain ATC, Spencer M. MSc, PhD; Metaxiotis, Dimitrios MD, PhD; Ali, Ashique MD

Author Information
Clinical Orthopaedics and Related Research: February 2020 - Volume 478 - Issue 2 - p 359-377
doi: 10.1097/CORR.0000000000000946



It is believed that as many as 74% of individuals with an initial ankle sprain will experience either a subsequent ankle sprain injury or residual symptoms such as pain, weakness, and giving-way episodes [5, 21, 27]. Those chronic symptoms are commonly designated as “chronic ankle instability” [13] and are associated with decreased levels of physical activity [37]. Importantly, the incidence of posttraumatic osteoarthritis resulting from instability of this sort can be as high as 78% [34, 56]. Rehabilitation emphasizing balance has been shown to effectively reduce the symptoms of ankle instability [53, 54]. The addition of external supports such as bracing, taping, and orthotic insoles in shoes could both yield better outcomes and reduce the time and effort required. In evaluating treatment outcomes in ankle instability patients, postural control deficit is one of the major modifiable impairments clinicians and researchers should focus on [6, 46].

However, given the discrepant results reported in multiple randomized controlled trials (RCTs) assessing the impact of taping and orthotic devices on dynamic postural control in patients with chronic lateral ankle instability [7, 12, 22, 39, 52, 58], we considered that pooling data to elucidate this research topic would be of the essence. Furthermore, earlier authors have failed to draw safe conclusions on the existence of a placebo effect of ankle taping in patients with ankle instability [12, 51]. Therefore, whether patients’ performance is affected by their belief or expectations that taping prevents injury remains unclear.

A network meta-analysis allows readers to assess the relative efficacy of multiple competing treatments and potentially rank them even if they have not been previously compared in head-to-head RCTs. This approach could help us gain a more complete understanding of the role of taping, bracing, and orthotic devices in the care of patients with chronic ankle instability by critically appraising key comparisons drawn from randomized trials in a robust way [9, 19].

Therefore, in the current network meta-analysis, we sought to assess (1) the impact of taping and orthotic devices on dynamic postural control in individuals with ankle instability and (2) the presence of a placebo effect in participants treated with sham taping and complications resulting from the administered treatments.

Materials and Methods

We registered the present systematic review in advance with PROSPERO (CRD42016037849), and abided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for network meta-analyses [38].

Inclusion and Exclusion Criteria

We considered randomized trials investigating the results of real and/or sham taping, ankle bracing, and foot orthotics, or a combination of the above in individuals with chronic ankle instability. Patients with functional ankle instability and/or mechanical ankle instability were eligible for inclusion [33]. We discarded trials dealing with acute ankle sprains. For qualitative synthesis, we considered parallel-group and crossover trials assessing dynamic balance; whereas for quantitative synthesis, we narrowed our criteria to studies evaluating the aforementioned outcome using the Star Excursion Balance Test (SEBT) only. Also, to achieve homogeneity in the treatment arms, we excluded trials considering adjuvant therapies from quantitative synthesis.

Information Sources and Search

We performed electronic database and manual searching to identify completed published and unpublished trials until February 13, 2019. For the database search, we included the databases of PubMed, Web of Science, Scopus, and Cochrane Central Register of Controlled Trials, whereas for the manual search, we considered the trial registries of International Standard Randomized Controlled Trial Number,, and the Australian New Zealand Clinical Trials Registry. Reference lists of pertinent systematic reviews were also examined.

In particular, we searched for the following terms in PubMed to identify potentially relevant studies: “controlled trial,” “random*,” “comparative study,” “ankle sprain,” “ankle,” “chronic lateral ankle,” “instability,” “mechanically unstable,” “functional ankle instability,” “mechanical ankle instability,” “functionally unstable,” “chronic complaint,” “ankle strain,” “re-injury,” “ankle injury,” “tapping,” “bracing,” “passive restraints,” “tape,” “taping,” and “strap*” (see Table, Supplemental Digital Content 1,

Study Selection

Two review authors (DK, KT) performed the search independently without language restrictions to identify potentially relevant records. Thereafter, duplicates were removed, and the titles and abstracts of the retrieved articles were screened for eligibility. Subsequently, for the remainder of the articles, the full-text articles were assessed for inclusion. Any discrepancies in the aforementioned study selection procedure were resolved through discussion.

Our search yielded 744 potentially relevant studies. We deleted duplicates and screened the remaining 680 records for inclusion. Fifty-two articles met our eligibility criteria, based on information provided in the abstract and title. Of these, 31 studies were excluded following full-text screening. Ultimately, 13 studies were included in the meta-analysis (Fig. 1).

Fig. 1:
A flowchart of the study selection procedure is shown; SEBT = Star Excursion Balance Test.

Data Extraction

Two investigators (DK, KT) independently extracted information from the included full-text articles. More specifically, details about the comparators in the treatment groups, countries in which the studies took place, anthropometrics and activity level of the enrolled patients, outcome and follow-up measurements, and losses to follow-up were gathered. They also extracted information about potential side effects resulting from the administered treatments, inclusion criteria, and diagnosis of chronic lateral ankle instability. When there were insufficient data for quantitative synthesis or missing information, these data were requested by the author(s) of the included source studies.

Study Characteristics

In the current systematic review, we included 21 trials published between 2006 and 2019 with a total of 469 unstable ankles (Table 1). Thirteen of those studies were crossover studies and another three referred to nested trials in prospective cohort studies with ankle instability patients and healthy controls. In terms of geographic locations, five investigations were conducted in North or South America [10, 20, 39, 57, 58], three in Australia [24, 35, 51], seven in Asia [1, 7, 22, 23, 43, 52, 55], and six in Europe [3, 4, 12, 14, 15, 28]. The mean age of participants ranged from 14 to 24 years, and the mean height of the participants ranged from 156 cm to 192 cm. In addition, the mean weight of the patients varied between 47 kg and 80 kg (see Table, Supplemental Digital Content 2, There was only one loss to follow-up [57].

Table 1.:
Treatment-related characteristics of the enrolled studies
Table 1-A.:
Treatment-related characteristics of the enrolled studies
Table 1-B.:
Treatment-related characteristics of the enrolled studies
Table 1-C.:
Treatment-related characteristics of the enrolled studies
Table 1-D.:
Treatment-related characteristics of the enrolled studies
Table 1-E.:
Treatment-related characteristics of the enrolled studies
Table 1-F.:
Treatment-related characteristics of the enrolled studies

Outcome Assessment

For outcome analysis, we chose the Star Excursion Balance Test because it is a cost- and time-effective test that can be completed in a straightforward manner. On top of that, it is evidenced that the SEBT is a valid and reliable dynamic balance quantification tool with satisfactory intra- and intertester reliability [16, 26]. To perform this particular test, the individual stands on the involved leg and reaches as far as possible in the prespecified test direction while maintaining their balance. Of the eight different reach directions (anterior, anteromedial, anterolateral, medial, lateral, posterior, posteromedial, posterolateral) of the SEBT, we only focused on the posteromedial direction because it is the most representative of balance deficits in patients with ankle instability [25]. Possible scores vary between 0 and 100 with higher scores denoting more stability. We gathered and analyzed data pertaining to patient performance with the external supports. In particular, 85 patients did not receive any particular treatment (that is, they followed the wait-and-see protocol), 29 were allocated to the placebo group, 99 were treated with taping, 16 with bracing, 27 were administered insoles, and six participants were offered a combination of insoles with bracing (see Figure, Supplemental Digital Content 3, We also assessed for the presence of a placebo effect by comparing the results of sham taping with no treatment and reported complications.

Statistical Analysis

We conducted not only pairwise analyses but also a network meta-analysis using the effect measure of standardized mean differences, with the Star Excursion Balance Test- posteromedial reach as the dependent outcome variable. In addition, a p value < 0.05 denoted statistical significance.

Furthermore, we used Cohen’s rule of thumb to classify the effect sizes. Accordingly, a standardized mean difference value of 0.2 denoted a small effect, a value of 0.5 indicated a moderate effect, and a value of 0.8 demonstrated a large effect [11]. Additionally, for a more clinically relevant interpretation of the results, we back-transformed standardized mean differences to SEBT-PM actual scores by multiplying standardized mean differences with the percentage baseline SDs reported in an included 3-arm trial [10]. Subsequently, to enable judgements on clinical meaningfulness, we considered the minimal detectable change of 14% [10], which represents the established smallest amount of change in SEBT-PM score that ensures the change is not a result of a measurement error. Of note, the above change of 14% corresponds to a large effect size (that is, standardized mean difference of more than 1.2). Importantly, with statistical power and probability of type I error set at 80% and 0.05, respectively, a minimum of 16 patients per treatment group was needed to provide sufficient statistical power and detect a SEBT-PM percentage difference of 14%.

For pairwise quantitative synthesis, we considered a random-effects model, and we used the Review Manager software, version 5.3 (The Nordic Cochrane Center, Cochrane Collaboration, Copenhagen, Denmark) [47]. To incorporate the results of two-arm crossover trials, we performed approximate analyses [29] and, depending on the available information, we either imputed or assumed that the required correlation coefficients were 0.5. For three-arm crossover trials, we considered data from the final period of assessment in the analyses. We also combined treatment subgroups when the same type of external support was implemented [30].

Before conducting the random-effects network meta-analysis, we tested transitivity assumption by assessing the distribution of effect modifiers across treatment comparisons [8, 50]. Thereafter, graphical tools were used (StataCorp LP, Release 13, College Station, TX, USA) to illustrate the network of treatments by means of a network meta-analysis plot (Fig. 2) [8]. In this plot, the thickness of edges was proportional to the number of studies for each comparison, and the size of nodes was proportional to the number of participants randomly assigned to each intervention. Then, the surface under the cumulative ranking probabilities were calculated to rank the efficacy of the included treatments, and the percentage contributions of each direct evidence to the network estimates were depicted [8, 41, 50]. Inconsistency (that is, differences between direct and indirect effect estimates concerning the same comparison [17]) was also assessed by using a global test [31, 45]. Moreover, we compared direct and indirect estimates by plotting their absolute difference and assessed the presence of small study effects, which served as a proxy for the evaluation of publication bias. It is worthy of mention that no evidence of small study effects was documented based on the results of Egger’s statistical test [18] (p = 0.208) and visual inspection of comparison-adjusted funnel plot for controlled trials [8] (see Figure, Supplemental Digital Content 4, Of note, one small trial with highly imprecise results was excluded from this plot [43] because we considered that it could not reflect either small study effects or publication bias. Finally, we created an interval plot to enable predictions on the efficacy of external supports in future trials [8, 32, 48]. Accounting for the results of this plot and statistical power, we concluded that the major finding of this study, which is presented in detail in the following sections, is unlikely to change even if further studies are conducted (Fig. 3).

Fig. 2:
The network meta-analysis plot of the current study is depicted. Nodes represent the included competing treatments and edges indicate direct comparisons between those treatments. The size of the former is proportional to the number of patients allocated to each treatment and the thickness of the latter is commensurate with the number of studies considered in each comparison.
Fig. 3:
This figure shows an interval plot of standardized mean differences with their 95% CIs (black solid lines) and predictive intervals (red extensions). The blue line represents the line of no effect. Wait-and-see policy is considered as reference treatment in this plot; PrI = predictive interval.

Risk of Bias and Evidence Quality Assessment

Two reviewers (KS, KT) independently assessed the risk of bias within and across trials using the Cochrane risk of bias tool. For the risk of bias assessment within trials, we considered the following elements: sequence generation; allocation concealment; masking of participants, blinding of personnel, and outcome assessors; incomplete outcome data; selective reporting; and “other bias.” Each entry was assessed to be at an unclear, low, or high risk of bias. In addition, to rate the quality of an included trial, we considered the domain of randomization to be vitally important (that is, it was considered the key domain).

For the risk of bias assessment across trials, if more than half of the information was from randomized controlled trials at a low risk of bias, we considered the domain to be at a low risk of bias. If most information was from trials at an unclear or high risk of bias, we judged the domain to be at an unclear or high risk of bias, respectively.

The quality of evidence of the present systematic review was evaluated in terms of the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) framework [49]. More precisely, judgments of the elements of inconsistency, study limitations, imprecision, indirectness, and publication bias were made. Each of these elements were either maintained at a high-quality level or downgraded by up to three levels.

In the assessment of the individual trials, 18 studies were judged to be at a low risk of bias (Table 2). In the evaluation of the risk of bias across trials, the domains of randomization and incomplete outcome data were deemed to be at a low risk of bias. On the contrary, the domains relating to blinding were considered to be at a high risk of bias, and this was mainly attributed to the nature of treatments. In addition, the domains of allocation concealment, selective outcome reporting, and “other bias” data were considered to be at an unclear risk of bias. The quality of evidence of the network of interventions was judged to be robust enough (that is, level A).

Table 2.:
The risk of bias assessment of the included trials

Sensitivity Analyses

To assess for the presence of the carry-over effect, which occurs when the difference between treatments is affected by the order in which they were administered [29], we conducted a sensitivity analysis. In this analysis, only crossover trials allowing for a time interval between the alternation in treatment protocols were considered. We also prespecified a sensitivity analysis on the quality of the enrolled studies, in which we excluded trials at an unclear or high risk of bias. In addition, given the importance of treatment lumping and node-making in network meta-analyses, we proceeded with a further subanalysis to account for the established orthotic device classification [36, 40]. Finally, to confirm the statistical validity of our findings, we adjusted the assumed correlation coefficients required for the SDdif calculation in approximate analysis and found no difference compared to our original analysis.


Assessment of the Efficacy of External Supports as Measured by the Star Excursion Balance Test

Statistical pooling showed no differences in favor of real taping over the watch-and-wait approach and placebo taping (percentage SEBT-PM difference between taping versus wait-and-see and placebo: -2.4 [95% CI -6 to 1.1]; p = 0.18 and -1.3 [95% CI -6.8 to 4.1]; p = 0.64, respectively) (Table 3). Likewise, orthotic insoles demonstrated no advantages over wait-and-see at the end of the treatment protocols (percentage SEBT-PM difference between orthotic insoles and control: -3.1 [95% CI -10.4 to 4.3]; p = 0.41) (Table 3). Although a combination of orthotic insoles with bracing had the highest probability of being among the best modalities (see Figure, Supplemental Digital Content 5,, this finding was not clinically relevant because the 95% CIs crossed unity (percentage SEBT-PM difference between combined orthotics plus bracing versus the wait-and-see approach: -8.9 [95% CI -20.4 to 2.6]; p = 0.13) (Table 3). Only one study considered this particular combined treatment. Importantly, we found no global inconsistency in the network of interventions (p = 0.9949). This was also the case when the node-splitting approach was used (see Figure, Supplemental Digital Content 6,

Table 3.:
A league table of percentage SEBT-PM differences with their 95% confidence intervals

Head-to-Head Comparisons

We considered 13 trials with 262 individuals to conduct the pairwise meta-analyses (Fig. 4). Of note, one trial was excluded from the aforementioned analysis due to the presence of adjuvant treatments in the study groups [55]. Real taping did not provide an additional clinical advantage over no treatment (n = 9 trials; percentage SEBT-PM difference between taping and control: -2.7 [95% CI -6.4 to 1]; p = 0.15; I2 = 0%). This was also the case for placebo taping, which was no more effective than a wait-and-see policy (n = 3 trials; percentage SEBT-PM difference between placebo and wait-and-see: -1.8 [95% CI -9 to 5.7]; p = 0.64; I2 = 0%). Likewise, no clinical advantage of bracing to wait-and-see was documented (n = 2 trials, percentage SEBT-PM difference between bracing and control: -6 [95% CI -14.9 to 2.8]; p = 0.18). Similarly, we observed no superiority of foot orthotics to no treatment (n = 2 studies; percentage SEBT-PM difference between foot orthotics and control: -2.7 [95% CI -10.2 to 4.7]; p = 0.49).

Fig. 4:
Illustrations depict the forest plots of standardized mean differences for the assessment of dynamic postural control in terms of the star excursion balance test in the posteromedial direction. The vertical line demonstrates no difference between the two treatment groups. No statistical differences between the treatment groups are shown; IV = inverse variance; SEBT= Star Excursion Balance Test.

Qualitative Synthesis

It should be noted that most of the included trials that did not qualify for meta-analysis did not report any difference in dynamic balance between external supports and control [3, 4, 12, 15, 35, 58].

Assessment of Placebo Effect and Adverse Events

No evidence of a placebo effect was documented when direct and indirect evidence was synthesized at the same time (percentage SEBT-PM difference between sham taping and no treatment in network meta-analysis: -1.1[95% CI -6.9 to 4.7]; p = 0.72). This was also the case when direct evidence was considered only in the analysis (percentage SEBT-PM difference between sham taping and no treatment: -1.8 [95% CI -6.9 to 4.7]; p = 0.55). Of note, there were no reported complications after treatment administration.

Sensitivity Analyses

For our predetermined sensitivity analysis on the washout effect, the network meta-analysis indicated that there were no differences between ankle taping and no treatment (percentage SEBT-PM difference: -2.3 [95% CI -6.6 to 2]; p = 0.3). Likewise, neither bracing nor foot orthotics was more effective than the wait-and-see policy (percentage SEBT-PM difference: -7.5 [95% CI -15.9 to 1] p = 0.08; and -3.1 [95% CI -10.4 to 4.3]; p = 0.41, respectively). We also did not observe any differences after controlling for the quality of the enrolled source studies (see Table, Supplemental Digital Content 7, This was also the case when we adjusted for the material from which the foot and ankle orthotic devices were manufactured (see Table, Supplemental Digital Content 8, as well as correlation coefficients in approximate analyses.


To increase ankle stability in individuals experiencing ongoing subjective symptoms and/or ankle joint laxity after one or more severe ankle sprains, external supports like taping, bracing, and orthotic devices are used sometimes [2], but current evidence on whether external supports are beneficial in improving dynamic postural control in those patients is conflicting. On top of that, the role of the placebo effect of ankle taping in patients with ankle instability has yet to be defined. Because of this, we felt conducting a network meta-analysis to synthesize the available evidence stemming from RCTs would be beneficial because relative effects for all treatment pairings can be considered. Since patients with ankle instability exhibit postural stability deficits [6], we sought to examine the therapeutic effects of external supports on dynamic balance as measured by the Star Excursion Balance Test. Statistical pooling did not reveal any differences in favor of taping, bracing, foot orthotics, and combined bracing and foot orthotics, indicating that external supports of any type appear to offer no additional benefit over wait-and-see treatment protocols. Also, we did not detect any evidence of placebo effect in patients treated with sham taping.


Although the impact of external supports on dynamic postural control can be assessed reliably shortly after their application, consideration of long-term follow-up measurements is essential. In particular, most of the studies we assessed typically evaluated patients before and after treatment application; future studies should focus on conducting longer term follow-up observations to determine whether the effect of external supports remains the same over time. A further study limitation is that blinding the registration protocol details throughout the peer-review process prevented peer reviewers from assessing whether our methodology was defined in advance; it was, and this can be confirmed by reading our prospective registration with PROSPERO (CRD42016037849). Finally, the findings of the current network meta-analysis address only the primary management of chronic lateral ankle instability and not postoperative protocols for surgically reconstructed ankles.

Efficacy of External Supports

Our network meta-analysis found no clinically important benefits to the use of external supports in terms of dynamic postural control between treatment groups and controls. Therefore, external stabilizers cannot be recommended as a standalone treatment in the nonsurgical management of chronic lateral ankle instability. However, we recommend that future research focus on high-quality RCTs assessing external supports in conjunction with rehabilitation. Also, to achieve homogenous and comparable results in this field, we advocate a more standardized manner of quantifying clinician-oriented outcomes. For instance, a considerable amount of data stemming from seven randomized trials [3, 4, 12, 15, 35, 39, 58] was not synthesized in the current systematic review because of the different instruments used to measure dynamic balance.

Presence of the Placebo Effect and Assessment of Complications

In the present network meta-analysis, strong evidence revealed no difference in functional performance between sham taping and no-treatment groups, suggesting that a placebo effect was absent. In other words, there was no causal linkage of patients’ belief or expectations for improvement with actual dynamic balance amelioration [42]. In light of this finding, implementation of placebo-controlled study design in future research assessing balance in patients with ankle instability is not recommended; instead, we advocate the use of a no treatment group. Importantly, given the absence of carry-over effect in the results of the current quantitative synthesis, we suggest that crossover study designs be favored over parallel group designs. In terms of complications, we emphasize that the absence of reported side effects in the included studies, which generally enrolled relatively small numbers of patients, does not support the conclusion that the assessed interventions are safe; uncommon complications may not appear in small studies [44].


The major finding of the current network meta-analysis was that external supports of any type did not improve dynamic postural control in patients with ankle instability. We recommend that large-scale trials addressing patient-reported and clinician-assessed outcomes on combinations between rehabilitation and external supports be conducted in the future.


We thank Dimitris Mavridis PhD from the University of Ioannina, Greece for the statistical support he provided in this network meta-analysis.


1. Abbasi F, Bahramizadeh M, Hadadi M. Comparison of the effect of foot orthoses on Star Excursion Balance Test performance in patients with chronic ankle instability. Prosthet Orthot Int. 2019;43:6-11.
2. Ajis A, Maffulli N. Conservative management of chronic ankle instability. Foot Ankle Clin. 2006;11:531-537.
3. Alguacil-Diego IM, de-la-Torre-Domingo C, López-Román A, Miangolarra-Page JC, Molina-Rueda F. Effect of elastic bandage on postural control in subjects with chronic ankle instability: a randomised clinical trial. Disabil Rehabil. 2018;40:806-812.
4. Alves Y, Ribeiro F, Silva AG. Effect of fibular repositioning taping in adult basketball players with chronic ankle instability: a randomized, placebo-controlled, crossover trial. J Sports Med Phys Fitness. 2018;58:1465-1473.
5. Anandacoomarasamy A, Barnsley L. Long term outcomes of inversion ankle injuries. Br J Sports Med. 2005;39:e14; discussion e14.
6. Arnold BL, De La Motte S, Linens S, Ross SE. Ankle instability is associated with balance impairments: a meta-analysis. Med Sci Sports Exerc. 2009;41:1048-1062.
7. Bicici S, Karatas N, Baltaci G. Effect of athletic taping and Kinesiotaping® on measurements of functional performance in basketball players with chronic inversion ankle sprains. Int J Sports Phys Ther. 2012;7:154-66.
8. Chaimani A, Higgins JPT, Mavridis D, Spyridonos P, Salanti G. Graphical tools for network meta-analysis in STATA. PLoS One. 2013;8:e76654.
9. Chaudhry H, Foote CJ, Guyatt G, Thabane L, Furukawa TA, Petrisor B, Bhandari M. Network meta-analysis: users' guide for surgeons: part II - certainty. Clin Orthop Relat Res. 2015;473:2172-2178.
10. Cline J, Fenwick AT, Turner T, Arthur S, Wikstrom EA. Nonelastic and Kinesio Tex Tapes improve perceived stability but not postural control in participants with chronic ankle instability. Int J Athl Ther Train. 2009; 23:195-199.
11. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ, USA: Lawrence Erlbaum Associates, Publishers; 1988.
12. De-La-Torre-Domingo C, Alguacil-Diego IM, Molina-Rueda F, López-Román A, Fernández-Carnero J. Effect of kinesiology tape on measurements of balance in subjects with chronic ankle instability: a randomized controlled trial. Arch Phys Med Rehabil. 2015;96:2169–2175.
13. Delahunt E, Coughlan GF, Caulfield B, Nightingale EJ, Lin CW, Hiller CE. Inclusion criteria when investigating insufficiencies in chronic ankle instability. Med Sci Sports Exerc. 2010;42:2106-2121.
14. Delahunt E, McGrath A, Doran N, Coughlan GF. Effect of taping on actual and perceived dynamic postural stability in persons with chronic ankle instability. Arch Phys Med Rehabil. 2010;91:1383-1389.
15. De Ridder R, Willems TM, Vanrenterghem J, Roosen P. Effect of tape on dynamic postural stability in subjects with chronic ankle instability. Int J Sports Med. 2015;36:321–326.
16. Docherty CL, Valovich McLeod TC, 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.
17. Donegan S, Williamson P, D’Alessandro U, Tudur Smith C. Assessing key assumptions of network meta-analysis: a review of methods. Res Synth Methods. 2013;4:291–323.
18. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629-34.
19. Foote CJ, Chaudhry H, Bhandari M, Thabane L, Furukawa TA, Petrisor B, Guyatt G. Network Meta-analysis: Users' Guide for Surgeons: Part I - Credibility. Clin Orthop Relat Res. 2015;473:2166-71.
20. Gehrke LC, Londero LX, Loureiro-Chaves RF, Souza HH, Freitas GP de, Pacheco AM. Effects of athletic taping on performance of basketball athletes with chronic ankle instability. Revista Brasileira de Medicina do Esporte; 2018;24:477-482.
21. Gribble PA, Delahunt E, Bleakley C, Caulfield B, Docherty C, Fourchet F, Fong DTP, Hertel J, Hiller C, Kaminski T, McKeon P, Refshauge K, Van Der Wees P, Vincenzino B, Wikstrom E. Selection criteria for patients with chronic ankle instability in controlled research: A position statement of the International Ankle Consortium. Br J Sports Med. 2014;48:1014–1018.
22. Hadadi M, Abbasi F. Comparison of the effect of the combined mechanism ankle support on static and dynamic postural control of chronic ankle instability patients. Foot Ankle Int. 2019;40:702-709.
23. Hadadi M, Mousavi ME, Fardipour S, Vameghi R, Mazaheri M. Effect of soft and semirigid ankle orthoses on Star Excursion Balance Test performance in patients with functional ankle instability. J Sci Med Sport. 2014;17:430-433.
24. Halim-Kertanegara S, Raymond J, Hiller CE, Kilbreath SL, Refshauge KM. The effect of ankle taping on functional performance in participants with functional ankle instability. Phys Ther Sport. 2017;23:162-167.
25. Hertel J, Braham RA, Hale SA, Olmsted LC. Simplifying the Star Excursion Balance Test. Med Sci Sport Exerc. 2004;36(Suppl):S187.
26. Hertel J, Miller SJ, Denegar CR. Intratester and Intertester Reliability during the Star Excursion Balance Tests. J Sport Rehabil. 2000;9:104–116.
27. Herzog MM, Kerr ZY, Marshall SW, Wikstrom EA. Epidemiology of ankle sprains and chronic ankle instability. J Athl Train. 2019;54:603-610.
28. Hettle D. The effect of kinesiotaping on functional performance in chronic ankle instability - preliminary study. Clin Res Foot Ankle. 2013;01:1–5.
29. Higgins JPT, Deeks JJ, Altman DG. Chapter 16: Special topics in statistics. In: Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions. Chichester, UK: John Wiley & Sons, 2008: 499, 503-507.
30. Higgins JPT, Deeks JJ. Chapter 7: Selecting studies and collecting data. In: Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions. Chichester, UK: John Wiley & Sons, 2008: 177.
31. Higgins JPT, Jackson D, Barrett JK, Lu G, Ades AE, White IR. Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies. Res Synth Methods. 2012;3:98–110.
32. Higgins JPT, Thompson SG, Spiegelhalter DJ. A re-evaluation of random-effects meta-analysis. J R Stat Soc Ser A Stat Soc. 2009;172:137–159.
33. Hiller CE, Kilbreath SL, Refshauge KM. Chronic ankle instability: Evolution of the model. J Athl Train. 2011;46:133–141.
34. Hirose K, Murakami G, Minowa T, Kura H, Yamashita T. Lateral ligament injury of the ankle and associated articular cartilage degeneration in the talocrural joint: anatomic study using elderly cadavers. J Orthop Sci. 2004;9:37–43.
35. Hopper D, Samsson K, Hulenik T, Ng C, Hall T, Robinson K. The influence of Mulligan ankle taping during balance performance in subjects with unilateral chronic ankle instability. Phys Ther Sport. 2009;10:125–130.
36. Hsu JD, Michael JW, Fisk JR. Principles and components of lower limb orthoses. In: AAOS atlas of orthoses and assistive devices. Philadelphia: Mosby/Elsevier; 2008.
37. Hubbard-Turner T, Turner MJ. Physical activity levels in college students with chronic ankle instability. J Athl Train. 2015;50:742–747.
38. Hutton B, Salanti G, Caldwell DM, Chaimani A, Schmid CH, Cameron C, Ioannidis JPA, Straus S, Thorlund K, Jansen JP, Mulrow C, Catala-Lopez F, Gotzsche PC, Dickersin K, Boutron I, Altman DG, Moher D. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: Checklist and explanations. Ann Intern Med. 2015;162:777–784.
39. Jackson K, Simon JE, Docherty CL. Extended use of kinesiology tape and balance in participants with chronic ankle instability. J Athl Train. 2016;51:16–21.
40. James A, Yavchitz A, Ravaud P, Boutron I. Node-making process in network meta-analysis of nonpharmacological treatment are poorly reported. J Clin Epidemiol. 2018;97:95–102.
41. König J, Krahn U, Binder H. Visualizing the flow of evidence in network meta-analysis and characterizing mixed treatment comparisons. Stat Med. 2013;32:5414-5429.
42. Kwekkeboom KL. The placebo effect in symptom management. Oncol Nurs Forum. 1997;24:1393-9.
43. Lee BG, Lee JH. Immediate effects of ankle balance taping with kinesiology tape on the dynamic balance of young players with functional ankle instability. Technol Health Care. 2015;23:333-341.
44. Leopold SS. When “safe and effective” becomes dangerous. Clin Orthop Relat Res. 2014; 472: 1999–2001.
45. Lu G, Ades AE. Assessing evidence inconsistency in mixed treatment comparisons. J Am Stat Assoc. 2006;101:447–459.
46. Mckeon PO, Hertel J. Systematic review of postural control and lateral ankle instability, part I : Can deficits be detected with instrumented testing? J Athl Train. 2008;43:293–304.
47. Review Manager (RevMan). 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration 2011. The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011.
48. Riley RD, Higgins JPT, Deeks JJ. Interpretation of random effects meta-analyses. BMJ. 2011;342:d549–d549.
49. Salanti G, Del Giovane C, Chaimani A, Caldwell DM, Higgins JPT. Evaluating the quality of evidence from a network meta-analysis. PLoS One. 2014;9:e99682.
50. Salanti G, Higgins JPT, Ades AE, Ioannidis JPA. Evaluation of networks of randomized trials. Stat Methods Med. Res. 2008;17:279–301.
51. Sawkins K, Refshauge K, Kilbreath S, Raymond J. The placebo effect of ankle taping in ankle instability. Med Sci Sports Exerc. 2007;39:781–787.
52. Someeh M, Norasteh AA, Daneshmandi H, Asadi A. Immediate effects of Mulligan’s fibular repositioning taping on postural control in athletes with and without chronic ankle instability. Phys Ther Sport. 2015;16:135-139.
53. Tsikopoulos K, Mavridis D, Georgiannos D, Cain MS. Efficacy of non-surgical interventions on dynamic balance in patients with ankle instability: A network meta-analysis. J Sci Med Sport. 2018;21:873-879.
54. Tsikopoulos K, Mavridis D, Georgiannos D, Vasiliadis HS. Does multimodal rehabilitation for ankle instability improve patients’ self-assessed functional outcomesa network meta-analysis. Clin Orthop Relat Res. 2018;476:1295–1310.
55. Uddin S, Shafeeq AM, Ahmed F. Effectiveness of ankle taping and semirigid ankle brace on subjects with unilateral chronic ankle instability. Int J Phys Ther Res Pract. 2014;2:11-22.
56. Valderrabano V, Hintermann B, Horisberger M, Fung TS. Ligamentous posttraumatic ankle osteoarthritis. Am J Sports Med. 2006;34:612-620.
57. Wheeler TJ, Basnett CR, Hanish MJ, Miriovsky DJ, Danielson EL, Barr JB, Threlkeld AJ, Grindstaff TL. Fibular taping does not influence ankle dorsiflexion range of motion or balance measures in individuals with chronic ankle instability. J Sci Med Sport. 2013;16:488-492.
58. Wikstrom EA, Arrigenna MA, Tillman MD, Borsa PA. Dynamic postural stability in subjects with braced, functionally unstable ankles. J Athl Train. 2006;41:245–250.

Supplemental Digital Content

© 2019 by the Association of Bone and Joint Surgeons