Effect of an Unsupervised Exercises-Based Athletics Injury Prevention Programme on Injury Complaints Leading to Participation Restriction in Athletics: A Cluster-Randomised Controlled Trial

ctive: To test the ef cacy of the Athletics Injury Prevention Programme (AIPP) to reduce the percentage of athletes presenting at least one injury complaint leading to participation restriction (ICPR) over an athletics season. Methods: During the 2017–2018 athletics season, we included in this cluster randomised controlled trial (ClinicalTrials.gov Identi er: NCT03307434) 840 athletes randomly assigned (randomisation unit: athletic clubs) to a control group (regular

cacy of the Athletics Injury Prevention Programme (AIPP) to reduce the percentage of athletes presenting at least one injury complaint leading to participation restriction (ICPR) over an athletics season. Methods: During the 2017–2018 athletics season, we included in this cluster randomised controlled trial (ClinicalTrials.gov Identi er: NCT03307434) 840 athletes randomly assigned (randomisation unit: athletic clubs) to a control group (regular training) or to an intervention group (regular training plus the AIPP 2/week). Using a weekly online questionnaire, athletes reported the ICPR, training and competition exposures, and, for the intervention group, the compliance with the AIPP. The primary outcome was the percentage of athletes presenting at least one ICPR over the study follow-up. Results: A total of 449 and 391 athletes were included in the intervention and control groups, respectively. From them, 68 (15.1%) and 100 (25.6%) athletes, respectively, provided 100% of the requested information during the follow-up (39 weeks). A total of 6 (8.8%) performed the AIPP 2/week or more. The proportion of athletes who had at least one ICPR over the follow-up period was similar in the intervention (64.7%) and control groups (65.0%), with adjusted odds ratios: 0.81 (95% CI 0.36 to 1.85). There were no between-group differences when comparing separately the subgroups corresponding with the different compliance levels. Conclusion: This cluster randomised controlled trial reported no ef cacy of the AIPP. However, the overall response proportion and the compliance with the AIPP in the intervention group were low. In individual sports especially, efforts should be rst made to improve the implementation and adoption of interventions. Keywords: sports injury prevention; injury prevention program; athletics; track and field;epidemiology; prospective studies 1. Introduction Athletics (track and eld) is an Olympic sport practiced worldwide. Athletics includes several disciplines: sprints, hurdles, jumps, throws, combined events, middle- and long- distance track running, road running (including among others 5 km, 10 km and marathon), Int. J. Environ. Res. Public Health2021,18, 11334.

Int. J. Environ. Res. Public Health2021,18, 11334 2 of 15 and race walking, as well as cross country and mountain and trail running (https://www. worldathletics.org; accessed on 19 October 2021). Participation in athletics invariably leads to a risk of injuries [1]. During any athletic season with a structured prospective injury monitoring, about two third of athletes sustained at least one injury with few variations according to athletes' age, sex, and disciplines [1–3]. The nature of injuries varies according to disciplines, and most frequent reported diagnosis were: hamstring muscle injuries (especially in sprints, hurdles, and jumps), Achilles tendinopathies (especially in sprints, jumps, middle- and long-distances), knee overuse injuries including patellar tendinopathies (especially in sprints, middle- and long-distances), shin splints and stress fractures (especially in sprints, middle- and long-distances), ankle sprains (especially in jumps and throws), and lower back pain (especially in jumps and throws) [1–6]. In many ways, injuries in athletics, directly or indirectly, affect an athletes' training, performance, career, and/or health negatively [7–9]. The reduction in injury risk is thus fundamental to promote healthy and sustainable athletics practice. The lack of studies analysing injury risk reduction programmes in athletics moti- vated Edouard et al. [10] to develop and analyse an exercises-based injury risk reduction programme. The Athletics Injury Prevention Programme (AIPP) was developed based on the literature on the epidemiology of athletics injuries, injury risks factors, and cur- rent evidence-based exercises-based injury prevention programmes [10], and aimed to target the most common athletics injuries (hamstring muscle injuries, Achilles and patellar tendinopathies, low back pain, ankle sprains) [1–3]. The AIPP included strengthening, stretching and neuromuscular control exercises aiming at targeting the most common ath- letics injuries [10]. The AIPP revealed promising ndings in being signi cantly associated with lower risk of injury complaints related to athletics practice that leads to restrictions in athletics participation (called then “injury complaint leading to participation restriction” (ICPR)) in the short (12 weeks), but not in the long (40 weeks) terms [10]. Although this was the rst prospective cohort study (level of evidence 2) on injury risk reduction programme designed speci cally for athletics, the study

injury complaints related to athletics practice that leads to restrictions in athletics participation (called then “injury complaint leading to participation restriction” (ICPR)) in the short (12 weeks), but not in the long (40 weeks) terms [10]. Although this was the rst prospective cohort study (level of evidence 2) on injury risk reduction programme designed speci cally for athletics, the study included a small sample size (62 athletes) and needed to be replicated through a randomised controlled design (RCT). In this context, the primary aim of this cluster RCT was to test the ef cacy of the Athletics Injury Prevention Programme (AIPP) to reduce the percentage of athletes pre- senting at least one ICPR over an athletics season. The secondary aims of the study were to test the ef cacy of the AIPP to reduce ICPR burden (i.e., number of days lost per 1000 h of exposure [11]) and whether the AIPP increased time (weeks) before athletes become injured for the rst time during an athletics season. 2. Methods 2.1. Study Design and Overall Procedure From October 2017 to July 2018 (40 weeks of athletics season), we conducted a researcher-blinded unsupervised cluster RCT (called “PREVATHLE”) including competi- tive athletes who were randomly assigned (allocation ratio 1:1) to a control group where they continued their regular training or to an intervention group where we added the AIPP to their regular training. The study protocol was reviewed and approved by the Committee for the Protection of Persons (CPP Ouest II—Angers, number: 2017-A01980-53) and by the French Federation of Athletics (FFA,; accessed on 19 October 2021) prior to recruitment, and was registered on ClinicalTrials.gov (Identi er: NCT03307434). To produce the present manuscript, we used the Consolidated Standards of Reporting Trials (CONSORT) [12]. Athletes and the public were not involved in the trial design and conduct of the study or the choice of outcome measures. 2.2. Population Recruitment and Inclusion/Exclusion Criteria At the start of the 2017–2018 athletics season, the FFA sent the 23 October 2017 through email to all athletes licensed at the FFA for competition an invitation to participate in this RCT at

the public were not involved in the trial design and conduct of the study or the choice of outcome measures. 2.2. Population Recruitment and Inclusion/Exclusion Criteria At the start of the 2017–2018 athletics season, the FFA sent the 23 October 2017 through email to all athletes licensed at the FFA for competition an invitation to participate in this RCT at an individual level. Detailed information about the study purpose and procedure,

Int. J. Environ. Res. Public Health2021,18, 11334 3 of 15 participant rights and contact information for further questioning was available in the email. Athletes were invited to register online using a speci c secured website called “Prevathle” (Windows Server 2013 R2 64 bits—SP2; IBM DOMINO 9.01 x pack 8) to proceed with the inclusion. The inclusion criterion of a cluster was a club with at least 15 licensed athletes. As the extra materials are usually present in most athletic clubs and were only required for the higher levels of the AIPP, we did not ask the clubs whether they actually had these materials, and this did not represent a cluster inclusion criterion. The inclusion of athletes in the study was performed electronically, during a two-week period (from 23 October 2017 to 6 November 2017), based on the following criteria: athletes must be licensed at the FFA in a club of at least 15 athletes (i.e., included cluster), without any contraindications for competitive athletics activity attested by the license at the FFA, aged between 15 to 40 years, and having access to the Internet. We excluded athletes if they declined to participate in the study or if they were unable to express agreement or signing the informed consent. We did not exclude athletes based on their baseline injury status or history [10]. If they met the inclusion criteria, athletes had to provide written informed consent for participation, as well as their parents for those under 18 years old. 2.3. Randomisation After the inclusion, we randomised each athlete to the intervention or the control group at a club level to minimise the risk of contamination bias between athletes from the same club. An independent statistician performed this cluster randomisation a priori using SAS 9.4 (SAS Institute, Inc., Cary, NC, USA) in a 1:1 ratio that assigned a unique randomisation number to each included athlete. The randomisation was strati ed according to the number of athletes and the national ranking of each club. 2.4. Interventions Apart from their regular training, we asked athletes in the intervention group to perform the Athletics Injury Prevention

SAS 9.4 (SAS Institute, Inc., Cary, NC, USA) in a 1:1 ratio that assigned a unique randomisation number to each included athlete. The randomisation was strati ed according to the number of athletes and the national ranking of each club. 2.4. Interventions Apart from their regular training, we asked athletes in the intervention group to perform the Athletics Injury Prevention Programme (AIPP) at least twice a week [10]. The AIPP included 8 exercises with levels of progression (from 2 to 5 depending on the exercise): core stability (plank and side plank), postural control (one-leg balance), pelvic strengthening (lunges and hip abductor strengthening), hamstring exercises (stretching and isometric, concentric and eccentric strengthening), and lower leg exercises (stretching and eccentric strengthening) [10]. The speci c exercises, number of repetitions, and levels of progression are presented in Table, and detailed information on the AIPP is provided in (supplementary le Figure S1). Only for athletes in the intervention group, we sent the AIPP via e-mail in both paper (Figure S1) and video versions. The AIPP was also available on the website used for data collection. No further guidance was given. Once the athletes were familiar with the exercises, the AIPP took about 15 min to complete. We asked athletes in the control group to follow their regular training plan. 2.5. Data Collection At the start of the season, we collected baseline information on each included athlete using a survey developed in Google Forms (Google ® ): sex, age, height, body mass, dis- cipline, usual week hours of athletics practice, and history of ICPR during the preceding season (yes or no). During the season, we automatically sent an e-mail every Monday with a secured link to the weekly online questionnaire to all included athletes. It aimed to collect information on the preceding week: number of hours of training and competition (i.e., athletics exposure), number of complete AIPP sessions performed, and possible injury complaints. We sent two automatic reminders 3 and 5 days after the rst e-mail to non-responders. The data were collected using a secured website called “Prevathle” (Windows Server 2013 R2 64 bits—SP2; IBM

aimed to collect information on the preceding week: number of hours of training and competition (i.e., athletics exposure), number of complete AIPP sessions performed, and possible injury complaints. We sent two automatic reminders 3 and 5 days after the rst e-mail to non-responders. The data were collected using a secured website called “Prevathle” (Windows Server 2013 R2 64 bits—SP2; IBM DOMINO 9.01 x pack 8). We calculated the response proportions by dividing the

Int. J. Environ. Res. Public Health2021,18, 11334 4 of 15 number of completed weekly questionnaires by the maximum number of questionnaires expected to be completed. Table 1.Content and structure of the Athletics Injury Prevention Programme. Exercises Repetitions Bonus (If It Is Too Easy, Athlete Can Increase Dif culty) Core stability—the plank (4 sides: prone position, lateral, supine position, lateral): Level 1: both legs 15 sec. per side for 3 min 4 30 sec. for 6 min, then for 12 minLevel 2: alternate legs Level 3: unstable support Single leg balance: Level 1: static 3 15 sec. (each side) 3 30 sec. each sideLevel 2: unstable support Level 3: throwing ball with partner, then on unstable support Pelvis strengthening: Lunges 3 10 rep. 6 10 rep., then 6 10 rep. with medicine ball Hip abductor strengthening—Level 1: leg empty 3 10 rep. (each side) 6 10 rep. (each side), then unstable support Hip abductor strengthening—Level 2: with elastic 3 10 rep. (each side) 6 10 rep. (each side), then unstable support Hamstring exercises: Hamstring stretching (different positions) 3 15 sec. Hamstring strengthening Level 1: isometric contraction on two legs 6 6 sec. 10 10 sec. Level 1: heel to buttock with elastic 3 6 sec. (each side) 5 10 sec. (each side) Level 2: isometric contraction on one leg 6 6 sec. (each side) 10 10 sec. (each side) Level 2: heel to buttock with elastic 3 6 sec. (each side) 5 10 sec. (each side) Level 3: Nordic hamstring with help of upper arms 1 5 rep., then 3 5 rep., then 6 6 rep. Level 4: Nordic hamstring 1 5 rep., then 3 5 rep., then 6 6 rep. Level 5: Pliometric 1 5 rep., then 3 5 rep., then 6 6 rep. Lower leg exercises: Lower leg stretching 3 15 sec. (each side) Lower leg strengthening—Level 1: down to the ground 3 8 rep. (each side) 3 10 rep., then 5 10 rep. each side and then increase load Lower leg strengthening—Level 2: down to the void 3 8 rep. (each side) 3 10 rep., then 5 10

then 6 6 rep. Lower leg exercises: Lower leg stretching 3 15 sec. (each side) Lower leg strengthening—Level 1: down to the ground 3 8 rep. (each side) 3 10 rep., then 5 10 rep. each side and then increase load Lower leg strengthening—Level 2: down to the void 3 8 rep. (each side) 3 10 rep., then 5 10 rep. each side, and then increase loads sec., seconds; rep., repetitions. 2.6. Injury De nition We chose the term “injury complaint”, as mentioned in the study from Edouard et al. [10], since it refers to self-reported information without medical diagnosis [13], usually de ned as: “a pain, physical complaint or musculoskeletal lesion sustained by an athlete during participation in athletics training or competition, regardless of whether it received medical attention or its consequences with respect to impairments in connection with competition or training” [14]. Athletes reporting an injury complaint were asked to provide the following informa- tion: circumstance of injury occurrence (training (including during the AIPP), competition, outside of athletics), mode of onset (sudden or gradual) [14], injury location [14], and its consequence on athletics participation with four categories [15,16]: full participation with no discomfort, full participation with discomfort, reduced participation due to injury

Int. J. Environ. Res. Public Health2021,18, 11334 5 of 15 complaint, full absence from sport due to injury complaint). In this study, the term “injury complaint leading to participation restriction” (ICPR) corresponds to the last two categories (reduced participation or full absence in athletics). 2.7. Compliance We asked athletes to report an AIPP session only if they had performed all the 8 exercisesof the programme. For each athlete, we measured compliance with the AIPP as the average number of AIPP sessions performed per week. We categorized compli- ance as follows: good (at least 2 weekly AIPP sessions, as requested), moderate (between at least 1 to less than 2 weekly AIPP sessions), and low (less than 1 weekly AIPP ses- sion). We monitored adverse events of the AIPP, and especially ICPR, through the weekly online questionnaire. 2.8. Blinding The independent statistician who performed the cluster randomisation was blinded to the study protocol. The assessment and the delivery of the intervention were blinded as all data were collected via an online questionnaire, and the intervention was delivered online through e-mail and the website, ensuring no in uence or bias of healthcare providers or assessors of outcomes. The principal investigator (PE) and the statistician (DB) were also blinded, and outcome measures were not available to any party until all data had been collected. However, it was not possible to blind athletes and coaches. 2.9. Sample Size We used MedCalc (https://www.medcalc.org; accessed on 23 May 2017) for the sample size calculation. We expected to reduce the average percentage of athletes injured at least once per season from ~70%, as described in the literature [1,2], to 50% using the AIPP. Given a power of 90%, a signi cance level of 5%, an expected cluster size of 15 athletes per club with an in ation factor of 1.7 (calculated as 1 + ICC*(m-1), where ICC = 0.05 and m = 15), and a dropout rate of 40%, we calculated that we needed to recruit a total of 704 athletes (352 per group). 2.10. Study Outcomes The primary outcome was the percentage of athletes who presented at

of 15 athletes per club with an in ation factor of 1.7 (calculated as 1 + ICC*(m-1), where ICC = 0.05 and m = 15), and a dropout rate of 40%, we calculated that we needed to recruit a total of 704 athletes (352 per group). 2.10. Study Outcomes The primary outcome was the percentage of athletes who presented at least one ICPR during the 40 weeks of the athletics season. The secondary outcomes are the ICPR burden over an athletics season, de ned as the number of days lost due to ICPR per1000 h of exposure [11], and the time (in weeks) between the rst week of the study and the week of the rst ICPR. All these variables were calculated using data from the weekly online questionnaire. 2.11. Statistical Methods We carried out statistical analysis using R V.3.6.0 (version 3.6.0,©Copyright 2016 The Foundation for statistical Computing (Comprehensive R Archive Network, R-project.org; accessed on 2 July 2021). Signi cance was accepted at p< 0.05. For the primary outcome, we adjusted a multivariate logistic regression model where the dependent variable was the occurrence (or non-occurrence) of at least one ICPR, and the independent variables were the participant's allocation, compliance with the AIPP, athletics exposure, club, sex, age, and history of ICPR during the previous season. Following the purposeful selection approach described by Hosmer and Lemeshow [17], which includes performing univariate analyses for each of the dependent variables, we included in the nal model those covariates of special clinical relevance (participant's allocation and compliance), as well as the ones that result in a better model [17]. In this analysis, we only included athletes with 100% of weekly response proportion.

This study evaluates an injury prevention program's effectiveness in reducing athlete injury complaints.