Article Text

Determining key clinical predictors for chronic ankle instability and return to sports with cost of illness analysis: protocol of a prospective cohort study
  1. Jente Wagemans1,2,
  2. Jan Taeymans2,
  3. Kevin Kuppens1,
  4. Heiner Baur2,
  5. Chris Bleakley3,
  6. Dirk Vissers1
  1. 1Department of Rehabilitation Sciences and Physiotherapy, University of Antwerp, Antwerpen, Belgium
  2. 2Department of Physiotherapy, Bern University of Applied Sciences, Bern, Switzerland
  3. 3School of Health Sciences, Faculty of Life and Health Science, Ulster University, Newtownabbey, UK
  1. Correspondence to Jente Wagemans; jente.wagemans{at}


Introduction Ankle sprains are common in sports and the general population. Although considered innocuous, a large proportion has residual complaints such as recurrent ankle sprains and develop chronic ankle instability. Although some predicting factors are identified, there is no unequivocality regarding the development of chronic ankle instability, nor about the optimal rehabilitation for an acute ankle sprain. Alongside the biomechanical impairments, ankle sprains are a burden on society due to substantial economic costs. Therefore, we aim to identify key clinical predictors of chronic ankle instability or recovery after acute lateral ankle sprain. Additionally, we aim to determine cost-of-illness of patients who developed chronic ankle instability.

Methods and analysis This prospective cohort study ( NCT05637008 - pre-results) aims to recruit adult (18–55 years) patients with an acute lateral ankle sprain who are active in sports. Clinical assessments and patient-reported outcome measures will be used to collect data at 7–14 days, 6 weeks, 12 weeks and 12 months after enrolment in the study. The primary outcome will be chronic ankle instability at 12-month follow-up. Salient outcomes will be analysed by logistic regression to determine association with the development of chronic ankle instability. Participants will fill in a cost diary containing direct and indirect costs related to their injury.

Ethics and disseminations The ethical committee of the Antwerp University Hospital (B3002022000138) has given approval of the protocol and consent forms on 10 October 2022. We perform this study according to the Helsinki Declaration. We will present results at conferences or webinars and publish in peer-reviewed articles.


  • health economics
  • foot & ankle
  • musculoskeletal disorders
  • rehabilitation medicine

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

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.

Strengths and limitations of this study

  • First known prospective cohort study with long-term follow-up that inquires about clinical predictors for the development of chronic ankle instability.

  • First known prospective cohort study that inquires about clinical predictors for return to sports.

  • First prospective cohort study to evaluate cost of illness of patients who develop chronic ankle instability.

  • We do not provide a rehabilitation programme, which can lead to variability in content of rehabilitation among participants. This may influence various outcome measures. However, content of rehabilitation will be assessed and included as outcome variable.

  • The comprehensive protocol can lead to drop-outs.


Lateral ankle sprains (LAS) remain the most common musculoskeletal injury in athletic population.1 2 Approximately 1 in 10 000 people suffer an ankle sprain everyday,3 with 49.3% occurring during sporting activity.4 Fifty-nine per cent of professional basketball players and football players have reported to have sustained LAS.5 Indoor sports are most prone to incur LAS,6 followed by field sports and outdoor sports.7 Even though incidence and prevalence are high, LAS are still considered innocuous.8 Yet, research in different sports report high reinjury rates and residual symptoms: 19% in football, 28% in basketball, 43% in American football and 46% in volleyball.9 Furthermore, up to 73% develop chronic ankle instability (CAI)10; a heterogeneous condition characterised by reinjury at its epicentre.11 Several aetiological models exist for CAI,12–17 but these are mostly based on cross-sectional and case-controlled studies.18 Only one prospective study has examined predictive factors for developing CAI after an acute LAS: lower postural control and the inability to perform a jump landing and a drop vertical jump.19 Other factors contributing to CAI are delayed peroneal reaction time, decreased eversion muscle strength and impaired proprioception.20

The International Ankle Consortium developed a rehabilitation-oriented assessment (ROAST) ‘guideline to evaluate common impairments associating with LAS: pain, swelling, ankle range of motion (ROM), arthrokinematics, muscle strength, static and dynamic postural balance, gait, physical activity level and patient-reported outcome measures (PROMS).21 More recently, a consensus statement was published with a return-to-sports decision-making framework including assessments which aligning with ROAST: pain severity, ankle impairments, athlete perception, sensorimotor control, sport/functional performance.22 This expert-recommended framework was compiled after a systematic review concluded that evidence-based criteria for return to sport after an acute ankle sprain are non-existent in the current published literature.23

LAS incur substantial economic costs.2 A study in the Netherlands found that the mean cost per ankle sprain is around €820 (2010), based on healthcare consumption and direct costs.24 Comparable numbers are found in the USA and United Kingdom (UK), $1000 (US dollars, 2019) and £940 (2007), respectively.25 26 A nationwide cross-sectional study in the USA documented median healthcare costs of $1029 per ankle sprain (2010).27 Indirect costs due to production loss are also pertinent28; an injury estimation model that included both direct and indirect costs, calculated mean comprehensive costs associated with LAS around $12.000 (US dollars, 2007).29 Moreover, a comprehensive systematic review reveals an average production time loss of 7–29 days due to LAS, overall treatment cost of $1 809 to $5271 and a direct cost of illness of $292 to $2268 (US dollars, 2018) .30 31 Given the high recurrence rates and subsequent development of CAI into account, there will also be significant long-term costs associated with LAS.32 33

Study objectives

Determining key prognostic markers will inform clinical decision-making and help highlight patients most at risk for recurrent ankle injury. This study aims to identify key clinical predictors of the development of CAI—as the primary outcome—or recovery after acute LAS. Participants presenting with acute LAS will be prospectively followed over a 12-month period. Candidate predictors will include personal, demographic, clinical outcomes and patient-reported outcomes. As a secondary objective, we will undertake a health economical study to evaluate the cost of illness of LAS patients who developed CAI.



We will conduct a prospective cohort study. Figure 1 shows an overview of the study outline. We have obtained approval from the local ethical committee (UZA – B3002022000138) and started recruitment as of 23 March 2023. The protocol of this trial is published at (NCT05637008). We followed the Strengthening the Reporting of Observational Studies in Epidemiology-guidelines for cohort studies (online supplemental file 1).34


We contacted emergency departments (EDs) of the Antwerp University Hospital and four other hospital surrounding Antwerp, participating sport physicians and sport associations to provide study information to athletes who sustained an acute ankle sprain for recruitment. Additionally, newsletters and internet advertisement (social media, posters, website: are be distributed. Athletes who have sustained an acute LAS will be included. A more detailed overview of the eligibility criteria is displayed in table 1.

Table 1

Eligibility criteria

Sample size

It is difficult to calculate a precise sample size due to insufficient information regarding associations between presumed prognostic factors and the outcomes. The primary outcome will be CAI at 12-month follow-up. The definition of CAI will follow the consensus definition from the international ankle consortium.10 35 36 After LAS, we can conservatively estimate that 33% of individuals develop CAI within the first year.19 To identify at least 10 events per variable investigated37 and allowing for a 20% dropout rate, we will require 318 recruited subjects to obtain 265 participants with full data when considering a margin of error of 5%.38 We are not bound to any deadlines since we perform this study without funding. We estimate to recruit participants for a period of 4 years—based on the feasibility study mentioned below—and end the study in 2028.


We completed a feasibility study (ethical committee approval, University Hospital Antwerp (UZA)—BUN B3002021000199) to determine recruitment rate, eligibility criteria, adherence and dropout rates and duration of the assessment procedure; the criteria for success were established a priori and are detailed in table 2. We recruited participants from January 2022 to April 2022; we included eight participants (seven men, one woman; mean age: 20.00 years, range 18–21; mean height: 186.2 cm, range 165–196.5 cm; mean weight: 85.47 kg, range 74.2–99 kg). As we were unable to meet the criterion for eligibility, recruitment rate, recruitment time and equipment with the initial protocol, the following amendments were made: (1) time to first assessment was increased from 7 to 14 days, based on the feedback of collaborating physicians—they were unable to recruit participants within the week after their injury event. Recruitment rate went up from one per month to five per month after this disposition. (2) Some materials were insufficient: vacuum wall fixation broke down. We bought stronger wall fixators; everything worked henceforward. (3) Assessment duration at the 6-week time point was too time-consuming (>90 min). After having a consensus meeting with the entire research group, the following adaptations were made: (a) knee muscle strength will be excluded entirely; (b) hip flexion, internal rotation and external rotation will be excluded; (c) joint position sense will be evaluated as of 7.5° and higher; (c) there will be only one target value for force sense: 50% of 1RM; (e) only ankle inversion and eversion force sense will be assessed. The target criterion of 90 min was fulfilled after these changes in the protocol.

Table 2

Feasibility results

Testing procedure and assessments

Participants will be assessed at the following time points postinjury: <15 days, 6 weeks, 12 weeks and 12 months after the initial ankle sprain event. Multiple randomised controlled trials, investigating the effect of physical therapy in the treatment of acute LAS,39–43 showed significant improvements in various outcome measures 6 weeks postankle sprain. In the study of van Rijn et al.44 Twenty-eight per cent of the study participants sustained a recurrent ankle sprain after 12 months of follow-up, with 83% of whom the resprain occurred prior to the 3-month assessment point. Additionally, the proliferation healing phase of ligament tissue merges into the remodelling phase between 6 and 12 weeks postinjury.45 The 12-month time point is substantiated by a position statement of the International Ankle Consortium, recommending criteria for patients with CAI 12 months after sustaining the initial sprain.10 35 36 Furthermore, studies have shown that risk of reinjury is similar to the risk of a first-time LAS after 12 months.46 47

Additionally, there will be a monthly survey to inquire about presence of absence of reinjury. All participants will be assessed on location and all tests will be performed barefoot.

Online supplemental file 2 includes all patient-reported outcome measures.

Assessment time point 1: <15 days post ankle sprain event

Ligament injury severity grading

Diagnostic ultrasound (Telemed Ultrasound MicroUS PRO) will determine the severity of the ankle ligament injury and the ligaments involved. Diagnostic ultrasound has proven to have a high diagnostic value for ankle ligaments.48 Afterwards ligaments will be clinically evaluated:

Integrity of the anterior talofibular ligament (ATFL) will be assessed by the anterior drawer test (ADT). Measurement properties of the ADT show sensitivity of 84% and specificity of 96% to detect ATFL injury in ankle sprain patients.49 Participants are positioned supine on the table with a roll under their knees to obtain the least packed position of 30° of knee flexion, with the most distal part of the lower legs and the ankles over the edge of the table. The examiner stands at the end of the table, medially to the tested ankle. The homolateral hand of the examiner fixates the tibia and fibula by pressing the malleoli in the table. The contralateral hand holds the posterior side of the calcaneus, with the plantar side of the foot resting on the palmar side of the lower arm of the examiner. The examiner pulls the calcaneus anteriorly, while maintaining the tibia and fibula fixated on the table.49

To assess integrity of the posterior talofibular ligament (PTFL), calcaneofibular ligament (CFL) and deltoid ligament, the talar tilt test (TTT) will be performed. Measurement properties of the TTT show test–retest reliability with Intraclass correlation (ICC) =0.66,50 sensitivity of 17.2%–31.7% and specificity of 89.6%–95.5%.51 Similar to the ADT, the participant is positioned supine on the table with a roll under the knees and the distal part together with the ankles over the edge of the table. Positioned at the end of the table, medially to the tested foot, the examiner fixates both tibia and fibula in the table just proximal of the malleoli, while the contralateral hand holds the calcaneus. The examiner pulls the calcaneus to inversion and eversion while fixating the tibia and fibula.51

Additionally, we will inspect for presence or absence of discolouration due to haematoma and palpate to evaluate presence or absence of pain at the site of the above-mentioned ligaments. Research has displayed that sensitivity of ligament integrity assessment increased to 96% when performing additional inspection for bruising and pain palpation.49 The participant’s position is identical to the manual stress tests. The examiner closely inspects both feet for bruising. Thereafter, the examiner palpates the ATFL, PTFL, CFL and deltoid ligament.49

We will also ask the participant’s perspective on the severity grade of the ankle sprain. This will be evaluated by asking the participants to fill in a visual analogue scale (0–100).

Patient-reported outcome measures

The participant will be asked about personal information (eg, age and gender), sports-related information, injury-related information and status of rehabilitation (eg, content of rehabilitation, amount of sessions, duration of sessions).

Pain during weight bearing will be assessed through a numeric pain rating scale (NPRS) for pain intensity and digital pain drawing for area of pain. Test–retest reliability and validity for the NPRS as a measurement of pain intensity are r=0.95–0.96 and r=0.86–0.95, respectively.52 We will use a 10 point rating scale, with 0 meaning ‘no pain’ and 10 meaning ‘worst pain imaginable’, which is considered applicable due to its high level of repeatability.53 Also digital pain drawing as a measurement method for area of pain showed to have excellent validity (96.6%) and reliability (R²= 0.87).54

To assess specific symptoms of instability, the Cumberland ankle instability tool (CAIT) will be used. Specifically, the Dutch version of the CAIT will be used, which is proven valid and reliable (test–retest reliability ICC=0.94), with high internal consistency (Cronbach’s α=0.86) to assess ankle instability in the Dutch population. A cut-off value of >12 points (out of possible 30) as an indication for stable ankles is calculated, based on the Youden index (11.5 points).55 SE of measurement is 0.82 (2.7%) and a minimal detectable change of 2.28 on individual level and 0.04 on group level, respectively.55

The short version of the foot and ankle ability measure (Quick-FAAM) will be used to evaluate region-specific functional abilities. This questionnaire contains 12 items scored across a 5-point Likert scale.56 Measurement properties of the Quick-FAAM show excellent internal consistency (Cronbach’s α=0.94), acceptable test–retest reliability (r=0.82), acceptable concurrent validity (r=0.76), excellent discriminative capabilities (sensitivity=0.96, specificity=0.85, area under the curve=0.95) with a cut-off value of 94.79% to make the distinguishment between patients with CAI and ankle sprain copers.57 58

The illness perception questionnaire (IPQ) quantitatively evaluates the five components of illness representation: illness identity, cause, timeline, consequences and management.59 This questionnaire revealed to be reliable (ICC=0.53–0.85) with good internal consistency (Cronbach’s α=0.51–0.87).60 We will use the brief version of the IPQ, which has shown acceptable reliability (ICC=0.72; 95% CI 0.53 to 0.82).60

Fear avoidance beliefs pertinent to physical activities and work will be quantitatively evaluated by the fear avoidance beliefs questionnaire (FABQ), which has excellent test–retest reliability (ICC=0.97).61

The 11-questions version of Tampa Scale for Kinesiophobia (TSK-11) will be used to assess pain-related fear of physical movement and activity. This short version has shown similar measurement properties as the original TSK (test–retest reliability: ICC=0.81; internal consistency: Cronbach’s α=0.79).62

To assesses quality of life by addressing five domains: mobility, self-care, usual activities, pain/discomfort and anxiety/depression, the EQ-5D-5L questionnaire will be employed. Each domain is scored across a 5-point Likert scale: 0=no problems, 1=slight problems, 2=moderate problems, 3=severe problems and 4=extreme problems. This version is an update of the EQ-5D-3L, showing superior measurement properties.63

Assessment time points 2 and 3: 6 weeks and 12 weeks postankle sprain event

Patient-reported outcome measures

Similar to assessment time point 1, we will inquire about rehabilitation status, pain, specific symptoms of ankle instability, ankle-specific functional abilities, illness representation, fear avoidance believes, pain-related fear of physical movement and activity and quality of life by the PROMS mentioned previously. Additionally, participants will be asked whether or not they returned to sports and if they sustained a recurrent ankle sprain or giving away episode.

Activity tracking

Participants will be asked to wear an activity monitoring device (wGT3X-BT Smart, ActiGraph, USA) 1 week after each assessment time point to capture and record continuous, high-quality information regarding sleep and physical activities. This activity monitor will both measure volume and intensity of physical activities. The wGT3X-BT Smart has shown excellent validity (91%) and intrasession reliability (ICC=0.95–0.98).64

Range of motion

We will apply a digital goniometer (K-Force Sens, KINVENT, France) and the weight-bearing lunge test to assess ROM. Digital goniometry has shown excellent intra-rater reliability (ICC=0.96–0.99) to assess ankle ROM.65 Also the weight-bearing lunge test has conveyed excellent intra-rater (99%) as well as intra-rater reliability (99%).66 For the digital goniometry ROM assessment, participants will be positioned in long sit on the table with the ankle over the edge of the table in 10–15° of plantarflexion. The examiner will place the digital goniometer just proximal of the metatarsal heads, in the direction of the respective movement. Test subjects will actively perform ankle dorsiflexion, plantarflexion, inversion and eversion.67 For the functional assessment of dorsiflexion by utilising the weight-bearing lunge test, participants will be standing in split stance facing a wall with the tested foot in front. A measurement tape will be laid out on the floor to measure distance from the wall to the first toe. Participants will be instructed to perform dorsiflexion by pushing the knee forwards over the toes while maintaining their heels flat on the ground.68

Isometric muscle strength

Hip and ankle muscle strength will be evaluated by the use of a fixed handheld dynamometer (HHD) (K-Force Sens, KINVENT, France). Fixed HHD has shown to have good to excellent reliability to assess isometric muscle strength (ankle muscle strength: inter-rater reliability ICC=0.78–0.94; intra-rater reliability ICC=0.77–0.88; hip muscle strength: inter-tester reliability ICC=0.76–0.95).69 70 Test subjects will be able to hold on to the sides of the table for fixation during the isometric muscle strength tests.

For ankle muscle strength,71 the participants will be positioned in long sit on the table with the ankles over the edge of the table in 10–15° of plantarflexion. The lower leg will be fixated to the table with a fixation belt, just proximal of the malleoli, to exclude compensations by adjacent joint movements. The HHD will be placed perpendicular to the foot in each respective direction: (1) plantarflexion: the HHD will be placed at the plantar side of the foot, at the level of the metatarsals with the wall as fixation of the HHD; (2) dorsiflexion: the HHD will be placed at the dorsum of the foot, at the level of the metatarsals with belt fixation of the HHD; (3) inversion: the HHD will be placed at the medial side of the foot, at the level of metatarsal I with belt fixation of the HHD; (4) eversion: the HHD will be placed at the lateral side of the foot, at the level of metatarsal V with belt fixation of the HHD.

Hip muscle strength70 will be evaluated by assessing hip extension, abduction and external rotation. For hip extension and external rotation, participants will be positioned prone on the table with the hip in neutral position and the knee flexed 90°. To evaluate hip extension, the patients will be placed perpendicular to the posterior aspect of the upper leg, as distally as possible to the knee joint with belt fixation attached to the examiner as fixation of the HHD. For external rotation, the HHD will be placed on the medial aspect of the lower leg, 5 cm proximal to the proximal edge of the malleolus medialis with belt fixation as fixation of the HHD. For hip abduction, the participants will also be positioned prone on the table but with the knee extended with the distal part of the lower leg and the ankle placed over the end of the table. The HHD will be placed on the lateral aspect of the lower leg, 5 cm proximal to the proximal edge of the malleolus lateralis with belt fixation of the HDD.

For the execution of the isometric muscle strength tests, participant will first be asked to perform a guided, submaximal contraction during the above-mentioned movements, then the test subjects will perform a 5 s maximal isometric contraction.

Force sense

A HHD will be used to assess ankle inversion and eversion force strength as a measure of ankle proprioception.72 Participants will be positioned in long sit on the table with the ankles over the edge of the table in 10–15° of plantarflexion. The lower leg will be fixated to the table with a fixation belt, just proximal of the malleoli, to exclude compensations by adjacent joint movements. HHD positioning is similar to ankle isometric muscle strength testing: (1) inversion: the HHD will be placed at the medial side of the foot, at the level of metatarsal I with belt fixation of the HHD; (2) eversion: the HHD will be placed at the lateral side of the foot, at the level of metatarsal V with belt fixation of the HHD. First, participants will perform two maximal voluntary contractions. The mean will be used as 1 repetition maximum (1RM). Participants will then be asked to build up force until target value is reached (50% of 1RM). The achieved target value will be maintained and mentally visualised for 5 s to determine criterion force for the participants. After 10 s of rest, participants will try to reproduce the criterion force and hold for 5 s when they confirm that they have achieved the confirmed force.

Joint position sense

To evaluate joint position sense as a measure proprioception, the slope-box test will be used.73 This test has an excellent test–retest reliability (ICC=0.92–0.93).74 75 The slope box is a 30 cm by 30 cm wooden box with an adjustable slip proof cover. The cover can be inclined at angles varying by 2.5° between 0° and 25° in the four applicable directions (anterior, posterior, lateral and medial). Because mean estimate errors are larger in the 7.5–25° range,75 we will perform the test from 7.5° to 25° in the four directions, yielding a total of 32 test positions. Patients will be blinded and positioned facing a wall on a normal box of equal height of the slop box, which will be positioned adjacently. They can use the wall for stability when necessary. Test subjects will be asked to step on the slope box with the tested foot and to place their entire weight on that foot. Patients will be asked to estimate the direction and degree of incline onto which they stand. Degree of incline will be represented by a scale from 0 to 15 (0° to 37.5°, table 3), which enables possible overestimation by 12.5°. Order of direction and degree of incline will be determined randomly and all 32 positions will be performed. Participants have the opportunity to test trial the platforms at 0°, 12.5° and 25° of the four directions as a reference.

Table 3

Degrees of incline scale

Sensorimotor control

The foot-lift test will be used to assess static balance. This test has good test–retest reliability (r=0.78; ICC: 0.73; 95% CI 0.40 to 0.89).76 Test subjects will be instructed to initiate the test in single stance position with both hands placed on the iliac cristae. Participants will close both eyes when the examiner gives the signal to do so, after which participants have to remain stable for 30 s. The amount of foot lifts, foot shifts and touches on the ground with the contralateral foot will be counted as errors. Each error counts as 1. When the contralateral side remains on the ground, the amount of seconds will be counted as additional errors to the final score.76 77

To evaluate functional balance, the Y-balance will be employed. This test has shown good to excellent reliability (inter-rater reliability: ICC=0.85–0.93; Standard Error of Measurement (SEM): 2.0–3.5 cm) to assess functional stability in ankle sprain patients.78 The starting position for this test is a single-leg stance on the firm centre footplate of the Y-balance testing apparatus, with both hands placed on the iliac cristae. Prior to testing, lower limb length will be measured from the distal aspect of the malleolus lateralis to the proximal aspect of the trochanter major. The participant will be instructed to reach in the thee directions (anterior, posterolateral and posteromedial) and to try to push the indicator as far as possible with the free leg while maintaining balance. The reach distance evaluated by the indicator will be analysed in proportion to the lower limb length of each test subject.

The side-hop test will be used to assess dynamic balance. This test has a significant correlation with the presence of functional ankle instability (r=0.35; p<0.01).76 Similar to both sensory motor function tests above, participants start in single-leg stance with the hands on the iliac cristae. There will be a 30 cm width embarkment with white elastic tape. Participants will take the starting position lateral to the embarkment. Test subjects will hop laterally 30 cm and back 30 cm over the embarked zone for 10 repetitions as fast as possible and perform a stable landing after the tenth hop, which needs to be maintained for 5 s.76 77


The performance tasks included in this study will be evaluated on completion. Additionally, standing long jump performances will be evaluated on jump distance and drop vertical jump performance will be evaluated by measuring reactive strength index (RSI), contact time and jump height as measures of lower limb power.79 This will be recorded and evaluated by the My Jump 2 app, which has acceptable to excellent reliability (67%–98%) and validity (r=0.66–0.98) for the assessment of RSI, jump height and contact time.80

Single-leg drop landings will be evaluated. Participants will start in a single-leg stance position on a 30 cm box, from which they will have to drop forward and perform a stable landing. Subsequently, subjects need to maintain the stable landing position for 5 s.19

For the standing long jump, test subjects will be instructed to take place in a single-leg stance position at an indicated take-off line. Participants will have to jump as far as possible in a horizontal direction. Swinging of the arms and bending the knees to provide forward drive are permitted.

Starting position for the drop vertical jump is a single-leg standing position on a 30 cm box with the hands on the iliac cristae. Participants will have to drop forward off of the box with the test foot, land with the same foot and immediately perform a maximal jump as high as possible, followed by a landing on the ground in a stable single-leg stance position.19 81

The t-test will be used to assess quickness.82 Participants will have to take place at an indicated starting line. On the signal, they will have to run forward to the centre cone, sidestep 5 m to the right cone, sidestep 10 m to the far left cone and then sidestep back 5 m to the centre cone. Conclusively, participants have to run backward back to the starting line. This test is performed as quickly as possible.82

Assessment time point 4: 12-month post ankle sprain event

Similar to the previous time points, we will ask participants about return to sports, recurrent ankle sprain or the ankle giving away, rehabilitation status, pain, specific symptoms of ankle instability, ankle-specific functional abilities, illness representation, fear avoidance beliefs, pain-related fear of physical movement and activity and quality of life by the previously discussed PROMS.

Monthly assessment

From assessment time point 3 (12 weeks post ankle sprain event) up to assessment time point 4 (12 months postankle sprain event), we will contact participants monthly via e-mail, short message service or telephone call to inquire whether or not—if yes, when—they returned to sports and if—and if yes, when—they sustained a recurrent ankle sprain or giving away episode.

Data processing and analysis

All clinical assessments will be performed two times. The mean of the two measurements will be calculated per outcome measure and used for further statistical analysis. Data from PROMS will be acquired through a secure web application for online surveys and databases (REDCap, Research Electronic Data Capture, USA). Data from ROM and isometric muscle strength will be recorded in the K-Force app (KINVENT, France), from which excel data reports can be extracted for further analysis. The My Jump 2 app will be used to evaluate jump distance of the standing long jump and RSI, contact time and jump height as measures of lower limb power for the drop vertical jump. Excel data reports can be extracted from the K-Force app (K-INVENT, Montpellier, France) and the My Jump 2 app (Carlos Balsalobre, Spain) for further analysis. Participants’ information will be pseudonymised.

Predictor variables

The primary outcome will be CAI at 12-month follow-up. The definition of CAI will follow the consensus definition from the international ankle consortium.10 35 36 After LASs, we can conservatively estimate that 33% of individuals develop CAI within the first year.19 To identify at least 10 events per variable investigated,37 we will require 265 participants with full data. Individuals with missing data will be included within the analysis and assumed to be missing at random, with checks on the plausibility of the assumption.

We will use logistic regression to examine the association between the salient predictors and the primary outcome (CAI). Initially, we will conduct a series of univariate analyses to determine whether any predictor variables are associated with the binary variable of CAI. Correlations among predictor variables will be calculated to screen for any strong colinearity (r>0.8). Predictors demonstrating a p value less than 0.10 on univariate testing will be entered into a multiple logistic regression analysis. The strength of the predictive model will be determined using R2, p values, ORs, with 95% CI.

Secondary outcomes will be time to return to sports and time to reinjury. The relationship between predictors and secondary outcomes will be examined using survival analyses, continuous covariates (predictors) will be categorised using 50th centiles. The time (in days) from the index injury, to a participants’ first event (eg, return to sport/recurrent injury) or the end of their 12-month follow-up period, will be the main end point. Initially, Kaplan-Meier survival curves will be generated and log-rank tests were used to explore survival differences between levels of each covariate. Univariable and multivariable Cox proportional hazard models will be used to evaluate associations between covariates and hazard of injury. Variables with p<0.1 in the univariable model will then be analysed within a multivariate Cox proportional hazard model. HRs will be presented with their 95% CI and we will consider p<0.05 to be statistically significant. Non-proportionality will be checked graphically by assessing Kaplan-Meier survival distribution for each level of the covariate; and stratum-specific log minus-log plots.

Exploratory analysis

All other clinical and patient-reported outcomes will be recorded at 7–14 days, 6 weeks and 12 weeks postinjury. We will use factor analysis, latent class analysis and item response models to examine the psychometric properties of the various constructs. Latent growth curve models will be used to examine individual trajectories of recovery since the data are nested (level 1) with individuals (level 2). The derived models will then be further tested as a clinically useful predictor of the course of functional recovery.

Cost of illness evaluation alongside prospective study

Aim of the economic evaluation

Athletic patients with an acute ankle sprain receive a prescription for physiotherapeutic rehabilitation after visiting a physician. The maximum amount of sessions prescribed for rehabilitation is by law determined at 18 sessions. This can be a high economic burden for the patient, health insurance agencies and other healthcare policymaking agencies. To the best of our knowledge, it has not yet been researched what the cost of illness is in a Flemish context for patients who sustained an acute ankle sprain. Therefore, this study will be the first to evaluate long-term costs after an acute ankle sprain, with regard of long-term repercussions, such as CAI. In this section, we describe the protocol of a cost-of-illness (COI) analysis alongside the prospective cohort study.

Methods of the health economic evaluation

This health economic evaluation will be conducted for patients with and without a recurrent ankle sprain. All data will be gathered by asking the participants to fill in a PROM containing questions about the economic cost related to the ankle sprain. Patients who reinjured their ankle will be asked to fill in the PROM again. Duration of the trial is 12 months: we will ask participants every month to fill in the questionnaire. The COI analysis will follow an incidence approach, that is, from diagnosis until the disorder has been dissolved. Health services consumed (eg, physician consult, physiotherapy session, ED visit) will be recorded using diaries. Costs will be calculated as units consumed × unit price. Hence, costs will be presented in a non-aggregated form. Direct and indirect costs will be calculated while intangible costs—costs that represent impaired quality of life because of the injury, which exceed the monetary value of tangible effects and services83—will be omitted. Direct costs include costs about diagnosis, treatment, rehabilitation (eg, consultation with physician, physiotherapist session) and non-medical expenditures occasioned by injury (eg, purchase of braces, purchase of wheelchair). Indirect costs include costs regarding travel, productivity loss and other expenses (eg, absenteeism, presenteeism, bus ticket). All costs will be expressed in 2028 Euro’s. This COI study will be conducted from a healthcare and a societal perspective. All results will be presented as means or medians and their corresponding 95% CI or IQR, respectively. A sensitivity analysis will be performed to test robustness of the results against uncertainty on the precision of several input parameters.84

Ethics and dessiminations

The ethical committee of the Antwerp University Hospital (B3002022000138) has given approval of the protocol and consent forms on 10 October 2022. All participants will be informed of the study, after which they will have an opportunity to ask questions. Thereafter, we will ask to fill in an informed consent form. The principal investigator will have access to, and control over the final data set. We perform this study according to the Helsinki Declaration. We will present results at conferences or webinars and publish in peer-reviewed articles.

Ethics statements

Patient consent for publication


We gratefully acknowledge the contribution of Melanie Vandenberghe, MD, orthopaedic surgeon from the AZ Monica Hospital, Antwerp, Belgium. Her contribution during recruitment for the feasibility study was vital for the final shape of the study protocol.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Twitter @jentewagemans

  • Contributors All authors—JW, JT, KK, HB, CB and DV—contributed to the design of this protocol. CB did the sample size calculation and designed the plan for the statistical analysis. JT and DV Implemented the cost of illness evaluation. JW, HB and KK described the assessments. All authors gave critical feedback for revisions of this manuscript and approved the final version of this manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.

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

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.