Article Text

Download PDFPDF

Increasing compliance with neuromuscular training to prevent ankle sprain in sport: does the ‘Strengthen your ankle’ mobile App make a difference? A randomised controlled trial
  1. M Van Reijen1,
  2. I Vriend1,2,
  3. V Zuidema2,
  4. W van Mechelen1,3,4,
  5. E A Verhagen1,5,6
  1. 1Department of Public & Occupational Health, Amsterdam Collaboration on Health & Safety in Sports, EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, The Netherlands
  2. 2Consumer Safety Institute VeiligheidNL, Amsterdam, The Netherlands
  3. 3Physiotherapy and Population Sciences, School of Public Health, University College Dublin, Dublin, Ireland
  4. 4Faculty of Health and Behavioural Sciences, School of Human Movement and Nutrition Sciences, University of Queensland, Brisbane, Australia
  5. 5Faculty of Health Sciences, Department of Human Biology, UCT/MRC Research Unit for Exercise Science and Sports Medicine (ESSM), University of Cape Town, South Africa
  6. 6Australian Center for Research into Injury in Sport and its Prevention, Federation University Australia, Ballarat, Victoria, Australia
  1. Correspondence to Dr Evert Verhagen, Department of Public & Occupational Health, VU University Medical Center, Van der Boechorststraat 7, Amsterdam 1081 BT, NH, The Netherlands; e,verhagen@vumc,nl


Background E-health has the potential to facilitate implementation of effective measures to prevent sports injuries.

Aim We evaluated whether an interactive mobile application containing a proven effective exercise programme to prevent recurrent ankle sprains resulted in higher compliance as compared with regular written exercise materials.

Methods 220 athletes participated in this randomised controlled trial with a follow-up of 8 weeks; 110 athletes received a booklet explaining an 8-week neuromuscular training programme; 110 athletes participated in the same programme in an interactive mobile App (Strengthen your ankle). The primary outcome was compliance with the exercise programme. Secondary outcome measure was the incidence density of self-reported recurrent ankle sprains.

Results The mean compliance to the exercise scheme was 73.3% (95% CI 67.7% to 78.1%) in the App group, compared with 76.7% (95% CI 71.9% to 82.3%) in the Booklet group. No significant difference in compliance was found between groups. The incidence densities of self-reported time-loss recurrences were not significantly different between both groups (HR 3.07; 95% CI 0.62 to 15.20).

Summary This study shows that the method of implementing the exercises by using an App or a Booklet does not lead to different compliance rates.

New findings The use of a mobile App or a Booklet lead to similar compliance and injury rates in the short term.

Trial registration number The Netherlands National Trial Register NTR 4027. The NTR is part of the WHO Primary Registries.

  • Ankle
  • Injuries
  • Physical activity
  • Randomised controlled trial
  • Exercise rehabilitation

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Ankle injuries are the second most common sports-related injuries, and ankle sprain is the most common type of ankle injury. Ankle sprain may account for as much as 80% of all ankle injuries. The incidence of ankle injury and ankle sprain is especially high in popular sports as rugby, (indoor) soccer, triathlon, handball, volleyball and basketball.1 Both externally applied supports (ie, taping or bracing of the ankle), as well as neuromuscular training programmes prevent recurrent ankle sprains and are cost-effective.2 ,3 ,5

Despite these cost-effective interventions, large-scale community uptake of these measures, and thus actual prevention of ankle sprains, is lagging.4 ,5 The cost-effective neuromuscular training programme3 ,6 has suffered from poor compliance4 and its preventive effect was achieved solely among a subsample of compliant athletes.4 Although analyses have been performed using an intention-to-treat approach, there is still a lot of effectiveness to gain by increasing compliance with preventive measures.

E-health has potential to bridge this so-called implementation gap.7 The company VeiligheidNL developed an interactive mobile application: ‘Strengthen your ankle’ (‘Versterk je enkel’, free for iOS and Android) that contains the cost-effective neuromuscular training programme. Although mobile apps are plentiful and have the cachet of technological advancement, the value of this approach for injury prevention has not been evaluated formally.8 Consequently, we evaluated whether the ‘Strengthen your ankle’ App resulted in higher compliance with the neuromuscular training programme as compared with the regular written exercise materials.



This study was a randomised controlled trial. A detailed description of the study protocol has been published elsewhere.9 The study design, procedures and informed consent procedure were approved by the Medical Ethics Committee (2013/248) of the VU University Medical Center Amsterdam (VUmc), the Netherlands. The trial is registered in the Netherlands Trial Registry (NTR4027).


Participants were recruited from October 2013 to April 2014 through physiotherapy and sport physician practices, national sport federations’ websites, digital newsletters and an open invitation via social media, the internet and written media. Active sports participants (athletes) between 18 and 70 years of age who had sustained an ankle sprain within the past 2 months, and who had access to a mobile phone (either Android or iOS), were eligible for inclusion. Respondents were excluded if they had sustained an injury other than a lateral ankle sprain in the same ankle in the preceding year (eg, fracture of the ankle). Before inclusion, the main author contacted all potential participants by phone to confirm study eligibility. The recruitment of participants is shown in figure 1.

Figure 1

Flow chart of participant recruitment and follow-up.

Sample size

Sample size calculations were based on the primary outcome measure compliance and originated from previously established compliance rates to the same programme when advocated through written materials.4 Full compliance rates in the written materials’ group were expected to be around 25%. A doubling of this rate to at least 50% was considered to be clinically relevant. Based on a β of 0.90 and an α of 0.05, a total of 158 athletes were required divided across both study groups. Assuming a dropout rate of 20%, a sample of 190 participants was calculated.

Randomisation procedure

After participants had finished ankle sprain treatment by means of usual care, and after the baseline questionnaire and the informed consent were received, participants were randomly assigned to one of two study groups. The control group received the neuromuscular training programme on paper (Booklet group) and the intervention group received the neuromuscular training programme through the App (App group). Participants were allocated to the study groups through a random number generator.


All participants received the neuromuscular training balance board (machU/MSG Europe BVBA). Both the Booklet and the App contained the same neuromuscular training programme. The App can be downloaded for free from both the App store (http://apple,co/1EcHyFP) and the Google Play store (http://bit,ly/1AHuZkB). Whereas the App provided the user with instructional videos and verbal instructions, the Booklet included only pictures of the exercises that needed to be performed. The embedded neuromuscular training programme consisted of six different exercises to be performed during three sessions a week, with a maximum duration of 30 min per session, for a period of 8 weeks. Exercises gradually increased in difficulty and load during the course of 8 weeks. This programme has been shown to be cost-effective in reducing recurrent ankle sprain.6 A full description of the programme has been published elsewhere.4 Figure 2 shows the exercises, and table 1 gives the scheme that participants were required to follow.

Table 1

Exercise scheme of the 8-week neuromuscular training programme

Figure 2

The exercises of the neuromuscular training programme.

Baseline measurement

An online baseline questionnaire collected information from each participant on demographic variables, physical characteristics, sports and injury history, use of preventive measures, severity and received treatment and/or rehabilitation of the current ankle sprain.

Outcome measures

The primary outcome measure was compliance to the exercise scheme prescribed in the neuromuscular training programme. Compliance was defined as the participant's following of the prescribed intervention, that is, the scheme of exercises (table 1).10 Compliance measurements started after the start of the allocated intervention and took place weekly for the duration of the programme (8 weeks). Participants received a request by email to complete an online compliance questionnaire. After 3 days, a reminder was sent in case of non-response. These weekly measurements gathered information for each participant on the number of executed exercises per session and the number of executed sessions per week. From this information, weekly compliance rates were calculated as the percentage of prescribed exercises conducted, by multiplying the percentage per exercise with the number of times these exercises were performed per week. Previous research has shown that, in order for the neuromuscular programme to be effective, participants have to be highly compliant with the exercise scheme.10 In this study, a compliance rate of over 75% was considered adequate.

Exposure and injury registration

Secondary outcome measure was the incidence density of self-reported recurrent ankle sprains. Injury incidence density was defined as the number of recurrent ankle sprains per 1000 h of sports exposure. During the 8 weeks follow-up, in addition to the questions regarding compliance, participants were asked weekly about their hours of sports exposure and whether they had sustained an ankle sprain in the previous week. Ankle sprain recurrence was defined as a self-reported inversion moment of the same ankle. Both an episode of giving-way, as well as a grade 1, 2 or 3 ankle sprain were registered. Self-reported recurrent ankle sprains were further categorised to severity by looking at recurrences that had led to time loss, defined as the discontinuity of (sport) activity and/or missing (part of) the next planned (sports) activity due to the recurrence.4 This time-loss categorisation is in line with the accepted definition of an ankle sprain.11

Statistical analyses

Mean baseline differences between the App and the Booklet group were determined using an independent sample t test for continuous data (age, weight, height, exposure) and Fisher's exact test (with Monte Carlo 95% CI simulation due to small number of samples) for categorical data (level of sport, severity of ankle sprain, gender). Baseline measurements were based on the total number of participants (n=220) that were allocated to either one of the interventions.

While compliance was not normally distributed, weekly and overall compliance means and corresponding 95% CI were obtained through bias corrected accelerated bootstrapping with 1000 bootstrap samples. Mean weekly and overall compliance rates were compared between groups by means of a non-parametric Mann-Whitney U test.

Cox regression analyses compared risk of self-reported and time-loss recurrent ankle sprains between the groups. Analyses were checked for confounding by baseline variables, but none were found. Gender was found to be an effect modifier; consequently, analyses were done separately for both males and females.

All analyses were conducted according to the intention-to-treat principle, and differences were considered significant with a significance level of 0.05.



Between October 2013 and April 2014, a total of 220 participants were recruited and randomised to one of the two groups (figure 1). After randomisation, a number of participants indicated that they no longer wished to participate in the study due to time restraints or lack of motivation. For the remainder of the study period, their compliance rates were set to 0%. A small number of participants had to leave the study due to a non-ankle injury (n=4) or due to personal reasons (n=4). The compliance rates of these participants were only included for the weeks they participated in the study. At baseline, both groups were comparable regarding all variables measured (table 2).

Table 2

Baseline characteristics of the study population


Over the complete 8 weeks, the mean compliance to the exercise scheme was 73.3% (95% CI 67.7% to 78.1%) in the App group as compared with 76.7% (95% CI 71.9% to 82.3%) in the Booklet group (table 3). No significant difference in mean overall compliance was found between groups, nor for males or females. In both groups, 82 out of 110 participants (74.5%) complied to more than 75% of the programme.

Table 3

Mean weekly and overall compliance and corresponding 95% CI in both study groups

Compliance gradually declined over the 8 weeks in both groups (figure 3).

Figure 3

Mean weekly compliance rates and corresponding 95% CI of both groups across the 8 weeks of the neuromuscular training programme.

Exposure and recurrent injury

In total, participants took part in 2429 h of sport in the App group and 2547 h of sport in the Booklet group during the 8-week programme. In the 8 weeks of the exercise programme, 93 self-reported recurrent ankle sprains were reported, resulting in injury incidence densities of, respectively, 25.3 self-reported recurrences per 1000 h (95% CI 18.0 to 32.7) in the App group, and 25.6 self-reported recurrences per 1000 h (95% CI 18.3 to 32.9) in the Booklet group (table 4). The injury incidence densities of time-loss ankle sprains were, respectively, 0.82 time-loss recurrences per 1000 h (95% CI −0.3 to 2.0) in the App group, and 2.36 time-loss recurrences per 1000 h (95% CI 0.5 to 4.2) in the Booklet group. No significant differences in injury incidence densities between groups were found for either self-reported or time-loss recurrent injuries.

Table 4

Injury incidence density (IID) of self-reported and time-loss ankle sprain recurrences (95% CI) per 1000 h of sports participation, as well as the HR (95% CI) derived from Cox regression analyses

Gender acted as an effect modifier in the relationship between group allocation and injury recurrence outcome. No significant differences were found between genders.


We found that the method of implementing the ‘Strengthen your ankle’ exercise scheme, by using an App or a Booklet, did not lead to significant different mean overall compliance rates. Also, the percentage of participants that were highly compliant was not different between groups. In both groups, 74.5% of participants complied with the neuromuscular training programme.

Comparison with previous studies

The current study is the first to compare compliance rates to a preventive intervention programme under the influence of different implementation methods. Two previous studies have tested the effectiveness of the same neuromuscular training programme using written materials, and both studies reported compliance rates.4 ,5 The percentage of participants who were fully compliant during the 2 months of the programme was as low as 23%.4 Janssen et al5 described full compliance of 45%. In the current study, the percentage of participants that were highly compliant was 65%; Janssen et al1 ,5 argued that at the time of their study the neuromuscular training programme was more widely accepted in the Netherlands than some years before, explaining the higher compliance rates in their study. It is possible that the neuromuscular training programme at the time of our study has increased acceptance in practice even further. In addition, the previous studies used a printed paper with a simple layout. Both the Booklet and the App used in the current study were updates of the materials that have been used in the previous studies. We speculate that the Booklet and the App employed in our study may have had a more attractive format that resulted in increased compliance rates.

Effect on injuries

Injury incidence densities of self-reported recurrences between the App group and the Booklet group were not significantly different. In two previous studies that evaluated the same neuromuscular training programme, reductions in injury risk were observed under the influence of the programme.4 ,5 Janssen et al5 reported an overall injury incidence density of 2.51 recurrent ankle sprains per 1000 h of sport (95% CI 1.51 to 3.42) in the group that followed the programme. Hupperets et al4 found an ankle sprain injury incidence density of 1.86 per 1000 h of sport (95% CI 1.28 to 2.75). These injury incidence densities are considerably lower than found in our study. A main difference between the previous studies and the current study is that an inversion moment (giving-way) was also counted as a recurrent event in the current study. Hence, this may explain the disparity in injury incidence densities between our study and previously reported injury incidence densities. Comparing only the injury incidence densities of recurrent injuries that led to time loss, the results of Hupperets et al4 (0.65 per 1000 h of sport; 95% CI 0.38 to 0.92), and Janssen et al5 (0.95 per 1000 h of sport; 95% CI 0.39 to 1.51) are more comparable to our findings.

We note a trend for App users to have reported a lower rate of recurrent ankle sprain with time loss. However, due to the low number of recurrent injuries that were reported and the relative short time span of follow-up, we lacked statistical power to comment on this result. One could speculate that the App may promote better quality execution of the exercises. Where the Booklet contained printed instructions and images on how to properly execute the exercises, the App contained videos and verbal instructions on the correct execution of the exercises. This may have helped participants to correctly execute the exercises. In a recent study among athletes who were ‘compliant’ with an exercise scheme, only 67% performed the exercises as described.12 Incorrect execution of exercises may diminish the preventive effect of an exercise. Our data provide the basis to study this question with appropriate power.

Methodological considerations

One could argue that in practice patients, who are not involved in a study, might have lower compliance rates that those we report here. Participants who volunteered to participate in the study may already be inherently more motivated to perform the exercises. Additionally, the weekly questionnaires used in the study allowed for a weekly reminder and the compliance with the exercises was assessed through self-report. The obligation to report missed exercises may have increased self-reported compliance rates. This should be taken into account when interpreting the reported compliance rates.


The current study showed that the method of implementing the exercises through an App or a Booklet does not lead to different compliance rates. Both methods resulted in around 75% of the participants performing an adequate number of exercises.

What are the new findings?

  • The use of a mobile App or a Booklet lead to equal compliance rates to an injury prevention program.

  • The use of a mobile App or a Booklet lead to similar injury rates in the short term.

  • A injury prevention program embedded in a mobile App is an effective means to reduce the risk on recurrent ankle sprains.

How it might impact on clinical practice in the near future?

  • As both the app and the booklet result in similar compliance and injury rates in the short term both methods can be used to reduce recurrent ankle sprains.

  • Having two difference methods of implementing the Strengthen your ankle programme allows athletes and (para)medics to choose the preferred method.

  • More freedom of choice might further increase the use and compliance of the Strengthen your ankle programme.

  • Future studies will have to determine whether both methods result in similar outcomes in the long term.


The authors would like to thank the following partners for their collaboration and recruitment of study participants: Royal Dutch Society for Physical Therapy (KNGF), Dutch Society for Physical Therapy in Sports (NVFS), Dutch College of General Practitioners (NHG), Dutch Sports Medicine Society (VSG), Dutch Olympic Committee (NOC*NSF), Zilveren Kruis Achmea (ZKA), and Disporta.


View Abstract


  • Twitter Follow Miriam van Reijen at @miriamvanreijen and Evert Verhagen at @Evertverhagen

  • Contributors EAV conceived the research idea; MVR and EAV have written the protocol; MVR has screened and included patients, performed data analyses and is the main author. IV , WvM and VZ contributed to ideas in the protocol. All authors have read and commented on the draft version and approved the final version of the manuscript.

  • Funding This study was funded by the Netherlands Organisation for Health Research and Development (ZonMw) grant number 525001003; Balance boards were provided by Disporta and booklets were provided by VeiligheidNL.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval The study was approved by the medical ethics committee of the VU University Medical Center, Amsterdam, The Netherlands (protocol number 2013/248).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles

  • Warm up
    Adam Weir Johannes L Tol Gustaaf Reurink