Article Text
PDF
Abstract
Objectives To identify individual characteristics associated with the adoption of injury risk reduction programmes (IRRP) and to investigate the variations in sociocognitive determinants (ie, attitudes, subjective norms, perceived behavioural control and intentions) of IRRP adoption in athletics (track and field) athletes.
Methods We conducted a cross-sectional study using an online survey sent to athletes licensed with the French Federation of Athletics to investigate their habits and sociocognitive determinants of IRRP adoption. Sociodemographic characteristics, sports practice and history of previous injuries were also recorded. Logistic regression analyses and group comparisons were performed.
Results The final sample was composed of 7715 athletes. From the multivariable analysis, competing at the highest level was positively associated with IRRP adoption (adjusted OR (AOR)=1.66; 99.9% CI 1.39 to 1.99 and AOR=1.48; 99.9% CI 1.22 to 1.80) and presenting a low number of past injuries was negatively associated with IRRP adoption (AOR=0.48; 99.9% CI 0.35 to 0.65 and AOR=0.61; 99.9% CI 0.44 to 0.84), both during their lifetime and the current season, respectively. These results were supported by higher scores of sociocognitive determinants among athletes who reported IRRP adoption compared with other athletes.
Conclusion Some characteristics of athletes seem to be associated with IRRP adoption either positively (competing at the highest level) or negatively (presenting a lower number of past injuries), whereas all the sociocognitive determinants tested appear to be linked to IRRP adoption. Since many athlete characteristics are difficult or impossible to change, IRRP promotion may be enhanced by targeting athletes’ beliefs and intentions to adopt an IRRP.
- athletes
- behaviour
- athletic injuries
- preventive medicine
Data availability statement
Data are available upon reasonable request. Please contact the corresponding author. Data availability statement: Data are available upon reasonable request. Please contact the corresponding author.
Statistics from Altmetric.com
Introduction
Like other sports, participation in athletics (track and field) leads to injury risk.1 Although to date there is little scientific evidence specifically on risk reduction strategies in athletics,1–4 it seems important to implement effective injury risk reduction programmes (IRRP), as seen in other sports.5–7 However, in athletics,2 8 running9 and other sports,10 11 studies investigating the effects of IRRP tend to show low compliance with suggested risk reduction interventions, limiting the impact of IRRP to those athletes who choose to adopt an IRRP. Hence, a better understanding of the beliefs and intentions of athletes who adopt or do not adopt an IRRP is likely to improve the implementation of IRRP. Specifically targeting athletes’ behaviours may also improve compliance.12 13
The relevance of sociocognitive theories of behaviour change has been highlighted by the work of Chan and Hagger.14 One of the most studied sociocognitive theories of behaviour change15 is the theory of planned behaviour.16 It posits that behavioural beliefs (attitudes, subjective norms and perceived control) predict the intention to perform certain behaviours, which then predicts a change in these behaviours.16 In the theory of planned behaviour, attitudes are defined as overall evaluations of behaviour by an individual; subjective norms as beliefs about what significant others may think of an individual’s behaviour adoption; perceived behavioural control as the individual’s perception of the extent of control over behaviour adoption, and intentions as conscious plans, decisions or self-instructions to exert effort towards adopting a behaviour. Such behavioural beliefs and intentions from the theory of planned behaviour are identified as sociocognitive determinants of behaviour adoption. Previous studies on IRRP adoption assumed that the theory of planned behaviour could be used as part of a framework to better understand the athlete’s compliance with their IRRP.14 17 18
The individual perceptions of IRRP and injuries using qualitative research design were previously investigated in athletes in a study.19 Although this study included a small number of participants, the results suggested that athletes emphasise beliefs such as attitudes (eg, injury prevention is less important than performance), subjective norms (eg, communication with physiotherapists and coaches is necessary) and perceived control (eg, injury prevention is part of training).19 Additionally, Bolling et al 19 showed that other factors such as previous injuries or years of sport experience might determine IRRP adoption. However, studies based on larger samples and using behaviour change theories as background are very rare in the context of IRRP adoption. To our knowledge, only Chan and Hagger17 were able to investigate the sociocognitive determinants of IRRP adoption in elite athletes. Their results were based on the self-determination theory,20 which is a theory of motivation (ie, the reason why adopting an IRRP), and not on the theory of planned behaviour (ie, the beliefs regarding IRRP adoption, and intentions to adopt an IRRP).17
In this context, the present study aimed to (1) identify individual athletes’ characteristics associated with IRRP adoption and (2) investigate the variations in sociocognitive determinants of IRRP adoption among athletics athletes using the theory of planned behaviour as theoretical background. We hypothesised that some athletes’ characteristics and sociocognitive determinants are associated with IRRP adoption.
Methods
Study design and procedure
We conducted a cross-sectional study through a one-time online survey. We asked athletics athletes licensed at the French Federation of Athletics (FFA, http://www.athle.fr) on their habits, motives, beliefs and intentions to adopt an IRRP. There was no athlete, patient and public involvement in the development of the study questions or conduction of the survey. The study was reviewed and approved by the Saint-Etienne University Hospital Ethical Committee (Institutional Review Board: IORG0007394, IRBN232020/CHUSTE).
Population
The eligible population comprised athletes licensed at the FFA with the following inclusion criteria: aged 18 years or older, licensed as competing athletes and legally able to provide consent to participate in the present study.
Data collection
The survey was developed by one researcher experienced in sports psychology (AR), two sports medicine physicians (PE and MS), one researcher experienced in sports scientist (EV), one athletics coach (SM) and one psychologist (LJ). After two review rounds, all coauthors approved the survey which was then pilot tested on three competitive athletes in February 2020. All coauthors performed the final validation of the survey.
The online survey was composed of four parts: (1) information on the athletes (age, sex, athletics discipline,21 number of years of athletics practice experience and competition level), (2) adoption of an IRRP during the entire career named ‘lifetime’ and during the current season (with options: ‘yes entirely’, ‘yes partially’, ‘not at all’), (3) information on injuries (lifetime number, time since the most recent injury, location (following the classification used by Edouard et al 22) and cause (traumatic or overuse) of the most recent injury and time loss after the most recent injury21), and (4) sociocognitive determinants of behaviour adoption from the theory of planned behaviour for IRRP adoption (attitudes, subjective norms, perceived behavioural control and intentions16). The survey is presented in online supplemental data.
Supplemental material
An IRRP was defined in the survey as a set of specific exercises related to his/her sport which aims to reduce the risk of injury, including, for example, muscle strengthening, stretching or balance exercises. An injury was defined as pain, discomfort or damage to the musculoskeletal system, occurring during sports practice (training or competition), and having resulted in consequences on sports practice (reduction in practice, adaptation or incomplete practice, or discontinuation of the practice), regardless of consultation by a health professional.2 These definitions were provided to the athletes in the survey (online supplemental data).
Items measuring the sociocognitive determinants of the theory of planned behaviour were created following Ajzen’s guidelines.16 The four sociocognitive determinants (attitudes, subjective norms, perceived behavioural control and intentions) were measured with four items each, rated on 7-point Likert scales.16 Individual scores ranged from 1 (lowest possible score) to 7 (highest possible score). The items are available in online supplemental data. A confirmatory factor analysis was performed using the R package lavaan 23 to check the factor structure of the created material. The comparative fit index, the Tucker-Lewis index and the root mean square error of approximation were used to estimate the goodness-of-fit statistics24 25 and revealed an acceptable fit of the factor structure to the data.
The invitation to the survey was distributed via an email sent by the FFA to the registered email address of licensed competing athletes on 22 April 2020. The survey was closed on 7 May 2020, after 15 days, without any reminder after the initial invitation.
Statistical analyses
Statistical analyses were conducted using R (V.4.0.2, Copyright 2020 The Foundation for Statistical Computing (Comprehensive R Archive Network, http://www.R-project.org)). In order to limit the risk of type I errors and given the large size of our sample, the α (ie, significance level) was set at 0.001 for statistical analyses26; p values were set at 0.001 and CIs were set at 99.9%. We first performed a descriptive analysis of the collected data using frequency with percentages for categorical variables, and median and range for continuous variables describing the sample (ie, age and years of experience in the athlete’s main discipline).
Based on the descriptive analysis, several categorical variables were merged to improve the power of logistic regression analyses. For IRRP adoption, ‘yes entirely’ and ‘yes partially’ were combined as ‘yes’, and ‘not at all’ was considered ‘no’; this choice was made pragmatically to make ‘IRRP adoption’ become a binary outcome variable for logistic regression analyses. For discipline practice, we categorised as ‘explosive’ the following disciplines: ‘sprints’, ‘jumps’, ‘throws’, ‘hurdles’ and ‘combined events’; and as ‘endurance’: ‘middle and long distances’, ‘marathon’, ‘race walking’, ‘road running’ and ‘trail and mountain running’, as previously performed.27 From the four competition-level categories, we combined ‘international’ and ‘national’ as one category and ‘regional’ and ‘departmental’ as another. For lifetime number of injuries, categories from ‘none’ to ‘3’ were considered as separate categories, categories of ‘4’ and ‘5’ were combined as ‘4 or 5’ and categories of ‘6’ to ‘10 or more’ were combined as ‘more than 5’; for time since the most recent injury, ‘current season’ and ‘6 months to 5 years’ were considered as separate categories, and ‘5 to 10 years’ and ‘more than 10 years’ were combined as ‘more than 5 years’. For time loss after the most recent injury, ‘1 to 7 days’ and ‘8 to 28 days’ were combined as ‘minor to moderate’, and ‘29 days to 6 months’ and ‘more than 6 months’ were combined as ‘severe’, this reduced the usual classification of the severity of injuries21 to two categories.
We used binomial logistic regressions to analyse the potential associations between IRRP adoption (lifetime and current season) and individual characteristics (sex, age, sport practised, years of discipline practice experience, competition level, lifetime number of injuries, time since the most recent injury and time loss after the most recent injury), providing OR and 99.9% CI in univariate and multivariable models. We then used multivariable models to calculate adjusted OR by including all athletes’ characteristics as covariates. OR above 1 indicates a tendency for the reference group to adopt IRRP more than the other groups, and OR below 1 indicates a tendency for the reference group to adopt IRRP less than the other groups.
For the comparison of sociocognitive determinants of IRRP adoption between groups (ie, those who adopted an IRRP vs those who did not), based on Shapiro-Wilk and Bartlett tests of normality and homogeneity in variances, parametric group comparisons were performed using the Student’s t-test, as well as analyses of variance with Tukey post hoc tests.
Results
Population
From a list of 75 575 competitive licensed athletes, a total of 8809 replied to the invitation to participate in this study between 22 April 2020 and 7 May 2020, among which 7715 athletes (10.2%) met inclusion criteria, gave their informed consent to participate in the present study and were included in the analysis. The characteristics of the final sample are displayed in table 1.
Characteristics of the 7715 athletics (track and field) athletes included in the present study
IRRP adoption
A total of 5430 (70.4%) athletes declared they never adopted an IRRP during their entire athletics career, 1705 (22.1%) declared having partially adopted an IRRP and 580 (7.5%) declared having already performed an IRRP during their lifetime. Additionally, 5929 (76.9%) athletes were not adopting any IRRP during the current season, 1282 (16.6%) were partially adopting an IRRP during the current season and 504 (6.5%) declared having performed an IRRP during the current season (table 1).
IRRP adoption as a function of athletes’ characteristics
Univariate logistic regression models showed that athletes practising endurance disciplines were less likely to adopt an IRRP during their lifetime (OR=0.72; 99.9% CI 0.60 to 0.87), and that athletes competing at a higher level were more likely to adopt an IRRP both during their lifetime (OR=1.82; 99.9% CI 1.53 to 2.15) and during the current season (OR=1.55; 99.9% CI 1.29 to 1.86) (table 2).
Associations between athletes’ characteristics and injury risk reduction programme (IRRP) adoption during lifetime and during the current season using the binomial logistic regression
However, when adjusting the models for all athletes’ characteristics (ie, multivariable models), the results showed that only the association between IRRP adoption and competition level remains significant (table 2).
Results of the associations between IRRP adoption and history of injuries, both in current season and during their lifetimes, showed that athletes with no history of an injury were less likely to adopt an IRRP than those who sustained three or more injuries (table 2). These associations remained significant for athletes who sustained the most injuries (four or more) in multivariable models adjusting for all athletes’ characteristics (table 2).
The results of univariate logistic regression analyses also showed that individuals who sustained an injury more than 5 years before their participation in this study were less likely to adopt an IRRP during the current season (OR=0.62; 99.9% CI 0.48 to 0.81) than those who sustained an injury during the current season. However, there seems to be no significant association between IRRP adoption and the severity or cause (overuse vs traumatic) of the most recent injury (table 2).
Sociocognitive determinants of IRRP adoption
Regarding the sociocognitive determinants of IRRP adoption, the 7715 athletes participating in the present study showed mean scores above the theoretical median of 4 (possible scores ranged from 1 to 7) for attitudes (5.54±1.21) and perceived behavioural control (5.45±1.30), and near the theoretical median for subjective norms (4.17±1.26) and intentions (4.69±1.69).
The comparisons between athletes who reported having adopted an IRRP during their lifetime or current season and those who did not showed that for all sociocognitive determinants (ie, attitudes, subjective norms, perceived behavioural control and intentions), athletes who declared they adopted an IRRP scored higher than athletes who declared they did not (for both lifetime and current season) (table 3).
Variations in sociocognitive determinants by injury risk reduction programme (IRRP) adoption
Variations in sociocognitive determinants across athletes’ characteristics
Women displayed significantly higher scores for attitudes and significantly lower scores for perceived behavioural control than men. Athletes practising disciplines categorised as ‘explosive’ showed significantly higher scores of attitudes and subjective norms than those practising ‘endurance’. Athletes competing at the highest levels displayed significantly higher scores for subjective norms, perceived behavioural control and intentions of IRRP adoption than those competing at the regional or departmental level. More detailed descriptions of the sociocognitive determinant variables can be found in the online supplemental material 1.
Scores of sociocognitive determinants of IRRP adoption significantly increased with the lifetime number of injuries: the more the injuries, the higher the scores. Additionally, a similar tendency was observed for the time since the most recent injury occurred: the more recent the injury, the higher the scores of sociocognitive determinants of IRRP adoption. Furthermore, athletes with the most recent injury categorised as ‘severe’ showed significantly higher scores of intentions to adopt an IRRP than those with a ‘minor to moderate’ most recent injury. However, no difference was found between overuse and traumatic injuries (see online supplemental data).
Discussion
The main findings of the present study were that competing at the highest level, presenting a larger number of past injuries and sustaining a most recent injury during the last or current season were positively associated with IRRP adoption. Higher scores of sociocognitive determinants supported adopting an IRRP in these categories of athletes. Additionally, athletes who adopted an IRRP during their career or the current season showed higher scores of sociocognitive determinants than those who did not.
Another important finding of the present study is that, in our sample of 7715 adult athletics athletes, more than two-thirds (70%) had never adopted an IRRP and less than a quarter (22%) had only partially performed an IRRP during their lifetime. Hence, only 7.5% of the study athletes had already completely performed an IRRP during their lifetime. These results suggest that there is a need for further work to increase adoption of injury risk reduction strategies in athletics.
Higher level and number of previous injuries associated with higher IRRP adoption
The main characteristics associated with IRRP adoption in multivariable models were a higher level of competition and a higher number of past injuries. These results are in line with previous investigations on the perception of injury prevention in elite athletes. Athletes perceive injury prevention as a learning process that comes with experience, higher training loads and past injuries.19 However, the current challenge in injury prevention may be to generalise efforts among non-elite athletes who have not sustained more than one or two injuries during their careers. Additionally, the current study showed that elite athletes presented higher scores of subjective norms, perceived behavioural control and intention regarding IRRP adoption, as well as the same tendency for athletes who sustained more injuries, hence increasing their likelihood to adopt an IRRP. For these athletes, the promotion and delivery of IRRP may benefit from evidence-based practice in the domain of behaviour change, such as intervention mapping approaches.28
Taking into account the context to enhance IRRP implementation
Bolling et al 29 suggested revising the ‘sequence of prevention’ of sports injuries and highlighted the importance of considering the context to provide better grounds for injury prevention. What athletes (a) think of IRRP (ie, attitudes), what they (b) think their coaches, medical staff, teammates, friends and family think of IRRP (ie, subjective norms), (c) how autonomous they are regarding IRRP adoption (ie, perceived behavioural control), and (d) how much they intend to adopt an IRRP (ie, intentions) are core contextual determinants of their perceptions of injury and injury prevention. Thus, the methods used to promote injury prevention, and improve athletes’ adoption of IRRP, should consider sociocognitive determinants as levers to help athletes change their behaviours.
Kok et al 28 posited that there are three parameters to consider for improving the effectiveness of a method that targets a change in behaviours: ‘(a) it must target a determinant that predicts behaviour; (b) it must be able to change that determinant; (c) it must be translated into a practical application in a way that preserves the parameters for effectiveness and fits with the target population, culture, and context’. Respecting these parameters when promoting IRRP among athletes may increase the chances of achieving higher adherence and compliance rates, and thus the efficacy of IRRP. For example, perceived behavioural control has been shown to be associated with IRRP adoption in our sample of French athletes. Hence, to increase the perceived behavioural control (ie, determinant) of athletes who have never adopted an IRRP because they think they are not in control of such exercises, visualising themselves successfully performing IRRP exercises may be an effective strategy (see Conroy and Hagger30 for an example in health psychology). Most athletes are familiar with using mental imagery (ie, visualisation) as a tool for mental skill training, so this could easily translate to imagining themselves performing exercises from the IRRP.
Limitations
One limitation of the present study is that we focused solely on the sociocognitive determinants of adopting an IRRP and not the perception of the consequences of adopting an IRRP. As highlighted in the reasoned action approach,31 the perceived effects of adopting a behaviour is an important determinant. Previous qualitative research has shown that the perception of an injury may impact athletes’ decisions on their training content.32 Hence, further investigations may be needed and should consider the reasoned action approach (ie, an extended version of the theory of planned behaviour) as a framework for investigating the determinants of IRRP adoption in athletes. Another limitation could be recruitment bias. Indeed, it is possible that athletes who agree with or who have performed IRRP responded preferentially to the survey, and thus could be over-represented compared with the general population of FFA licensed athletes. In addition, although we defined IRRP, no details regarding different IRRP practices and experiences were collected from participating athletes. Among athletes who reported IRRP adoption, understanding and/or experiences of IRRP could differ and thus influence the results.
Additionally, the retrospective design of the present study does not make it possible to conclude that the differences in sociocognitive determinants (measured as beliefs regarding IRRP adoption at the time of the survey) explain the differences in IRRP adoption during previous seasons or throughout athletes’ careers. In order to be able to estimate the associations between sociocognitive determinants and changes in IRRP adoption, a prospective study design is recommended. However, estimating the differences in sociocognitive determinants across groups of individuals who (1) adopt and (2) do not adopt a behaviour is the recommended method for identifying the most relevant determinants to target a behaviour change intervention, as suggested by Crutzen et al.33 Finally, it was not possible to conduct an analysis of non-responders to determine how well the study sample represented the 75 575 eligible athletes.
Perspectives
As described by Bolling et al,29 ‘the study of any health behaviour in isolation from the broader social and environmental context is incomplete’ and will lead to implementation issues. Therefore, further research is encouraged to embrace an ecological perspective of injury prevention by investigating multilevel, contextual and socioecological factors of IRRP adoption. A first step in this direction would be to replicate the present study with multiple stakeholders who are relevant for effective implementation of intervention measures, such as athletics coaches, as stipulated by O’Brien and Finch.34
Given the limited number of athletes who have already performed an IRRP during their lifetime, there is a clear need to improve implementation and adoption of IRRP in order to improve injury risk reduction in athletics. The results of the present study could serve to promote IRRP in athletes. In fact, targeting attitudes, subjective norms and perceived behavioural control in the material of IRRP promotion campaigns could increase IRRP adoption. This could also be achieved via education programmes for athletes and information dissemination.
Conclusions
Athletes’ characteristics seem to be associated with IRRP adoption and sociocognitive determinants of IRRP adoption. As athletes’ characteristics are difficult or even impossible to change, the promotion of IRRP may benefit from targeting athletes’ beliefs and intentions to adopt an IRRP.
Summary box
What are the findings?
Competing at the highest level, presenting a larger number of previous injuries and sustaining the most recent injury in the previous or current season increased the odds of adopting an injury risk reduction programme.
Scores of the sociocognitive determinants (ie, attitudes, subjective norms, perceived behavioural control and intentions) of injury risk reduction programme adoption were higher in athletes who adopted an injury risk reduction programme in the current season or sometime in their lifetime than those who did not.
Athletes who competed at the highest level, sustained a larger number of previous injuries, or the most recent injury in the previous or current season, had higher scores of sociocognitive determinants of injury risk reduction programme adoption.
How might it impact on clinical practice in the future?
While athlete characteristics (eg, athletic discipline, level of competition, history of injuries) are difficult or even impossible to change, it is possible to influence the sociocognitive determinants of their views on injury risk reduction programmes. Thus, a targeted approach on athletes’ beliefs and intentions could increase the adoption of injury risk reduction programmes. Clinicians could increase the adoption of injury risk reduction programmes by educating athletes on the benefits of such programmes (ie, improving attitudes), and how best to implement them into their daily life (ie, increasing perceived behavioural control). The use of social media to visualise elite athletes performing such programmes (ie, increasing subjective norms) could be an effective strategy to increase adoption.
Data availability statement
Data are available upon reasonable request. Please contact the corresponding author. Data availability statement: Data are available upon reasonable request. Please contact the corresponding author.
Ethics statements
Patient consent for publication
Ethics approval
The study was reviewed and approved by the Saint-Étienne University Hospital Ethical Committee (Institutional Review Board: IORG0007394, IRBN232020/CHUSTE).
Acknowledgments
The authors warmly thank the French Federation of Athletics (FFA, https://www.athle.fr) for sharing the questionnaire through their email lists and the athletes who participated in the study. The authors would like to thank Crane Rogers (Chaire ActiFS, Univ Lyon, UJM-Saint-Étienne, Laboratoire Interuniversitaire de Biologie de la Motricité, EA 7424, Saint-Étienne, France) for his contribution to this project.
References
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.
Footnotes
Twitter @Evertverhagen
Contributors PE, MS and AR conceived and designed the study and proposed the study protocol. SM, CH, LJ and EV provided revisions on the study protocol. AR performed data analyses and drafted the manuscript. All coauthors contributed substantially to interpreting the results, provided important revisions and approved the manuscript. PE acts as guarantor of the present study.
Funding The present study was conducted in the context of the FULGUR project (ANR-19-STPH-003) funded by the French Research Agency in the perspective of the Paris 2024 Olympic and Paralympic Games in collaboration with the French Federation of Athletics, Rugby and Ice Sports, Universities of Nantes, Côte d’Azur, Savoie Mont Blanc, Jean Monnet Saint-Étienne, Saclay, the Mines Saint-Étienne, the CEA and the CNRS. The University Jean Monnet Saint-Étienne and the French Institute of Sport (INSEP) are partners of the French-speaking network ReFORM. ReFORM, and the Amsterdam Collaboration on Health and Safety in Sports (ACHSS) are recognised as Research Centres for the Prevention of Injury and Illness and the Protection of Athletes by the IOC and received funding from the IOC to establish a long-term research programme on the prevention of injuries and illnesses in sports to protect athletes' health.
Competing interests None declared.
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.