Article Text

Download PDFPDF

Implementing an exercise-training programme to prevent lower-limb injuries: considerations for the development of a randomised controlled trial intervention delivery plan
  1. Caroline F Finch1,2,
  2. Peta White1,
  3. Dara Twomey2,
  4. Shahid Ullah2
  1. 1Australian Centre for Research into Injury in Sport and its Prevention, Monash Injury Research Institute, Monash University, Melbourne, Australia
  2. 2School of Human Movement and Sport Sciences, University of Ballarat, Ballarat, Australia
  1. Correspondence to Professor Caroline F Finch, Australian Centre for Research into Injury in Sport and its Prevention, Monash Injury Research Institute, Monash University, Clayton Campus, Melbourne, Victoria 3800, Australia; caroline.finch{at}


Objective To identify important considerations for the delivery of an exercise training intervention in a randomised controlled trial to maximise subsequent participation in that randomised controlled trial and intervention uptake.

Design A cross-sectional survey, with a theoretical basis derived from the Health Belief Model (HBM) and the Reach, Efficacy/Effectiveness, Adoption, Implementation and Maintenance (RE-AIM) framework.

Participants 374 male senior Australian Football players, aged 17–38 years.

Main outcome measurements Beliefs about lower-limb injury causation/prevention, and the relative value of exercise training for performance and injury prevention. The data are interpreted within HBM constructs and implications for subsequent intervention implementation considered within the RE-AIM framework. Ordinal logistic regression compared belief scores across player characteristics.

Results 74.4% of players agreed that doing specific exercises during training would reduce their risk of lower-limb injury and would be willing to undertake them. However, 64.1% agreed that training should focus more on improving game performance than injury prevention. Younger players (both in terms of age and playing experience) generally had more positive views. Players were most supportive of kicking (98.9%) and ball-handling (97.0%) skills for performance and warm-up runs and cool-downs (both 91.5%) for injury prevention. Fewer than three-quarters of all players believed that balance (69.2%), landing (71.3%) or cutting/stepping (72.8) training had injury-prevention benefits.

Conclusions Delivery of future exercise training programmes for injury prevention aimed at these players should be implemented as part of routine football activities and integrated with those as standard practice, as a means of associating them with training benefits for this sport.

View Full Text

Statistics from


Before efficacious injury-prevention measures including exercise training programmes, can be successfully incorporated into usual player safety behaviours and practices, it is necessary to know about likely barriers towards, and motivators for, their uptake.1 2 A number of studies have described attitudes and behaviours in relation to protective equipment use in sports such as Australian Football,3 rugby union4 and squash,5 and shown how these can lead to suboptimal uptake of this form of preventive measure. To date, there has been surprisingly little attention given to these factors in relation to the delivery and uptake of exercise training interventions for injury prevention, with only three previous studies reporting beliefs related to specific exercise training programmes for lower-limb injury prevention (LLIP) among basketball players,6 Australian Football coaches7 and netball coaches.8

Many factors interact to produce a player's safety status.2 9 10 Application of behavioural theories to understanding and preventing unintentional injury is common,11 but their application to sports-injury prevention has been minimal to date.12 One health-promotion framework (Reach, Efficacy/Effectiveness, Adoption, Implementation and Maintenance (RE-AIM))13 for understanding the implementation and evaluation contexts for prevention measures has recently been advocated for sports-injury prevention.14 Despite its wider recognition in broader health and physical-activity promotion settings, the RE-AIM framework with its reach, effectiveness, adoption, implementation and maintenance dimensions13 has only started to be applied to sports-injury prevention.8 14

While it is critical that sports-safety interventions have strong efficacy evidence, there also needs to be a strong theoretical basis behind strategies aimed at implementing them in the real-world context of community sport delivery.14 The impact of non-adoption of specific exercise-training programme components on intervention effectiveness has been highlighted in recent randomised controlled trials (RCTs) of LLIP measures.15 16 The results from the first implementation studies of the Federation Internationale de Football Association's The 11, for example, have shown limited success because few of the targeted participants adopted the programme,15 and there was a perception that it was not relevant to the real-world community sport setting in which it was implemented.17

Despite the availability of targeted exercise-training interventions for LLIP and increasing RCT efficacy evidence,15 16 18,,21 it is clear that broader LLIP efforts are hampered because little research has focused on better understanding implementation issues, including barriers and facilitators to sustainable programmes. One recent study concluded that Australian Football coaches do not possess the latest knowledge about LLIP and hence do not give adequate attention to the development of training skills most likely to reduce the risk of lower-limb injuries (LLI) in their players.7 There is also a need to know more about players' attitudes and beliefs in relation to LLI causes, predisposing factors and preventive measures because this plays a determining role in their adoption of exercise training programmes for safety gains.

According to the health-belief model (HBM),22 four constructs combine to explain the adoption of safety behaviours: perceived susceptibility (ie, beliefs about the risk of being injured), perceived severity (ie, beliefs about the seriousness of sustaining an injury in terms of both health and sporting consequences), perceived benefits (ie, beliefs about of the effectiveness of actions available to reduce injury risk) and perceived barriers (ie, beliefs about the potential negative aspects of adopting a prevention measure). While these four constructs represent an individual's ‘readiness to act,’ the model also includes two other dimensions required to produce the desired outcome: cues to action (ie, factors that would motivate a player to actually do something to prevent their injury) and self-efficacy (ie, a player's belief in their ability to undertake exercise training).22

This paper reports a study conducted to identify the best way to deliver an intervention in a large RCT of an exercise-based neuromuscular training programme to prevent LLI in Australian Football. The methodology protocol for the full RCT has been published elsewhere.23 The specific aim was to determine likely facilitators and barriers towards the delivery and uptake of exercise training programmes in senior Australian Football players before significant efforts and resources were invested in the conduct of the RCT. This paper describes the survey approach used for identifying the barriers and facilitators towards exercise training for LLIP and presents the key findings and their implications for the exercise training programme delivery plan development. This paper also compares players' beliefs about the likely injury prevention and performance value of various training programme components, so as to give additional guidance for the future marketing of the benefits of participating in an exercise intervention of this nature.



All nine senior clubs in the Premier Division (of the Sydney Australian Football League) were selected to participate, and all agreed to do so. The research team attended each club on two consecutive training nights towards the start of the 2005 playing season. All players aged 17+ years in attendance at one or more of these training sessions were invited to participate. Informed written consent was obtained from all consenting players, who then completed a self-report questionnaire during a training session. Approval for the study was obtained from the University of New South Wales Ethics Committee.

Questionnaire design and construct

The questionnaire was modelled on previous studies of risk and safety attitudes in Australian Football players and coaches,3 7 24 adjusted as necessary to focus on LLIs after ensuring face and content validity. While not designed to adhere directly to specific HBM constructs, the questionnaire broadly drew on HBM components as recommended in a ecological model for protective equipment use.2 The RE-AIM dimensions were used as the basis of determining question categories. Table 1 lists the 24 specific belief questions asked and shows their alignment to both the HBM constructs and RE-AIM dimensions. The HBM provided guidance in how to interpret the players' beliefs while RE-AIM assisted with directing how these should inform the intervention delivery plan.

Table 1

Categories of questions used on the survey, their alignment to the health belief model (HBM) and the Reach, Efficacy/Effectiveness, Adoption, Implementation and Maintenance (RE-AIM) framework, and players' summary responses (percentage of responding players)

The self-report survey also collected information about player characteristics (age, number of seasons played, level of competition, LLI in previous season history) and their beliefs about which training-programme components are most beneficial for improving football performance and LLIP. It consisted of closed-option questions with specified tick-box responses, and all belief statements were given as a five-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree.’

Data analysis

All data were double-entered into an SPSS database and analysed using R version All statistical analyses were adjusted for potential clustering effects because players were sampled from different clubs. All 5-point Likert scale variables were collapsed to three levels (strongly agree/agree; uncertain; strongly disagree/disagree) for analysis. Ordinal logistic regression analysis,26 with club as a random effect, was used to identify significant relationships between behaviour belief responses and player characteristics after controlling for all other responses and player characteristics. Descriptive statistics (%) for the listed training programme components were used to identify differences in players' ratings of their importance for both performance and LLIP. A Wilcoxon signed-rank test27 was used to compare belief scores across performance and LLIP situations, taking into account the paired nature of the data.


Three hundred and seventy-four male players (median age 23 years (range: 17–38 years)) completed the survey (95% response rate). Most were experienced players with 47.1% having played for >10 seasons, and 50% were currently playing in the most senior competition. Almost half of the players (49.2%) reported they had sustained an LLI during the previous playing season.

Table 1 shows the players' responses to the general opinion statements. The players' strongest agreement was with the statement that they would play with an LLI if it meant their team reached the finals. They least agreed that LLIs are caused by foul play.

Table 2 provides the odds ratio (OR) comparisons across player characteristics and each of the 24 belief statements. For nine statements, there was a significant relationship with age; for seven statements, there was a significant relationship with competition level; three statements were significantly related to self-reported previous LLI and years of playing experience.

Table 2

Relationship between survey responses and player characteristics

Players were asked if they believed different training-programme components would improve their football performance LLIP (table 3). The four training components that >90% of players believed would improve their performance were drills/set-plays, ball-handling skills, kicking skills and endurance/fatigue training. In contrast, the only three components that >90% of players supported for LLIP were warm-up run, cool-downs and warm-up stretches. There was a statistically significant difference between the rankings of players' opinions about the benefits for performance and LLIP for all components, except for warm-up run, warm-up stretches and jumping/landing training. These ranking differences were greatest for ball-handling skills, drills and set-plays, and kicking skills, all of which favoured performance over LLIP.

Table 3

Players' opinions about the importance of a range of training components for improving football performance or for preventing lower-limb injuries.


This study summarises Australian Football players' attitudes and beliefs around LLIP, particularly the benefits of training. To our knowledge, this is the first sports injury prevention study to apply both HBM constructs associated with a relevant sports injury ecological model2 and the RE-AIM framework to inform the design of a sports injury prevention delivery plan.14 The information is important because it gives clear guidance for the targeting and delivery of exercise training programmes specifically for these players.

Overall, surveyed players had an accurate perception of football-related LLI risk that reflected the reported frequency of LLI in this sport,28,,32 though their relative rankings of knee versus ankle injuries were different from those in the literature.29 31 33 These findings support the need for players to be educated further about the risk of knee injury specifically and also suggest that exercise interventions should be marketed as being of a more general LLIP nature rather than just focusing on knee injuries. This has implications for both the RE-AIM reach and adoption components. As most exercise training programmes that address knee neuromuscular control are also likely to benefit the rest of the lower limb,19 broadening the focus of intervention programmes for targeting purposes would still maintain biomechanical fidelity.

Only about half of all players considered that LLIs could be serious and impact on their ability to play a game, despite published information about this.34 The majority appear to recognise the potential for adverse consequences associated with continuing to play with LLIs in terms of the physical implications this may have later in life. However, there was no strong view about being fully rehabilitated from an LLI before returning to play. This suggests that players need to be better educated about the nature and extent of possible adverse LLI outcomes and that any strategy to recruit them into an RCT and maintain their engagement with exercise training practices should emphasise both LLIP and performance gains. The fact that beliefs about LLI risk and outcomes were related to player age and experience suggests that any overarching approach may need to include different education/promotion strategies for players at different stages of their football career. This has particular implications for the adoption and maintenance RE-AIM components.

Most players understood that specific training and exercises have a role in LLIP. This indicates a likely readiness of players to adopt LLIP programmes when they are embedded into standard football-training sessions. There was a difference in players with/without a previous LLI with regard to views about undertaking additional training at home, indicating that such a delivery strategy would not be adopted by all players. Nonetheless, about a third of players who were uncertain about the LLIP benefits of attending training also indicated a lack of knowledge about specific training components for reducing LLI risk; this is consistent with the level of knowledge exhibited by coaches of the same teams.7 The only other factor that players believed was more important for LLIP than performance was protective equipment use; this was rated even higher than warming up, cooling down and balance training.

Players were in strong agreement that most components of training sessions would improve performance, and the majority agreed that undertaking specific exercises would lead to LLIP. This suggests that extensive efforts aimed at educating players about the benefit of exercises/training programmes are unlikely to be needed before footballers were recruited into an exercise training RCT. However, there is still a need to educate players about the benefits of some aspects of training sessions such as skills training, ball handling and jump/landing, as many rate them as being highly important for performance but not for LLIP.

It is concerning that the vast majority of players would contemplate playing with an LLI if it was an important game, even though they did not feel under pressure to play when injured, consistent with previous findings in elite junior footballers.24 This has important implications in terms of preventing injury recurrence and rehabilitation strategies postinjury. While players recognise the long-term risk of continuing to play with injury, external factors are likely to make it difficult in reality to convince them to take time away from their sport to fully recover, particularly if they perceive their absence as detrimental to the team's performance. Given this, an effective implementation strategy would be delivered to whole teams of players, rather than just individual players, so that the benefits to the team can be stressed, and peer support could influence the adoption.

‘Barriers’ have been identified as the most powerful of the HBM components in terms of predicting a whole variety of behaviours.22 This study highlights important barriers to LLIP that would need to be addressed in the delivery of exercise interventions implemented as part of football training. The majority of players agreed that training should focus more on improving game performance than on LLIP, and while they also have a general belief in the value of exercises for LLIP, they do not want this as the major focus of their training sessions completely. This is further complicated by the clear delineation of players' beliefs as to what components of training contribute most to improved performance and which mostly to LLIP. Players are unlikely to be fooled by LLIP marketed as being performance enhancing, if there are no clear performance outcomes. It is advisable, therefore, that LLIP exercises delivered during training are acknowledged as such and have a shorter duration than training components aimed at performance and game development. Given the strong support for warming-up-type strategies for LLIP, it would seem appropriate also to place other purely LLIP training activities towards the start of training sessions.

Although the majority of surveyed players would attend a university to test their skills for LLIP, more would be willing to do so if it helped them to play better. This is important to know because specific recruitment strategies may need to be developed to encourage players to undergo further testing away from their training venue. Similarly, the majority of players would do extra training at home if they thought it would decrease their risk of LLI, but even more would do the extra training if they thought it would help them play better. Taken together, this suggests that LLIP is a strong-enough stimulus to trigger many players to undertake additional skills testing and extra training. This is good news for LLIP measures that require this kind of behaviour uptake. However, the team building and overall performance-goal findings, together with the fact that it is hard to monitor correct techniques outside formal training sessions and to provide the necessary equipment for this, mean that delivery of an intervention through formal training sessions is likely to be most acceptable to players.

This study does have some limitations that need to be acknowledged. The survey was not fully validated, though it was heavily based on similar surveys of safety behaviours in Australian Football players. Although a major strength is the fact that the survey was constructed from a theoretical basis deriving from the HBM and RE-AIM, because the questions were not structured to address a single behaviour, it was not possible to formally test the HBM theory applicability or to explore particular beliefs according to theory predictions in detail. Even though the analyses presented here would suggest that the HBM is relevant to understanding beliefs about LLI and LLIP in football players, there would be further value in formally testing the HBM in this context.

The survey was only conducted within participants of Australian Football, and it is not known to what extent the beliefs and cues to action identified in this study would apply to other sports. Similarly, the surveyed footballers were high-performance players, and it is possible that players at lower levels of competition (eg, in community clubs) could have different views. Both of these aspects of generalisation of results should be assessed in future studies.


This study has examined players' understanding of the importance of training sessions for improving performance and LLIP. This information was then used to develop a specific targeted intervention delivery plan according to RE-AIM principles that would ensure players' maximum participation with all study protocol requirements and maximum adoption of the intervention itself in the subsequent RCT.23 Players generally agreed that doing specific exercises would reduce their LLI risk and would be willing to undertake exercises for LLIP, but not at the cost of reducing training time that is perceived to improve their performance. They appear to be largely unaware of the latest evidence on the role of landing and balance training in LLIP, which is consistent with their coaches.7 It would seem that the optimal approach would be to deliver exercise training LLIP programmes as part of routine football activities, and integrated with those as standard practice, as a means of associating them with training benefits for this sport.

This study has identified important factors that need to be incorporated into the development of delivery plans for implementation in a large RCT of exercise training interventions. While the results from that RCT23 are not yet available, it is certainly expected that they will have more rigour than some of those from other exercise intervention studies, because issues of intervention uptake and sustainability have been addressed from the outset of the RCT intervention delivery planning phase. It is recommended that designers of future sports injury intervention studies, of whatever type of preventive measure, invest significant time and effort prior to design finalisation to also obtain important information about factors that may impact on the uptake and maintenance of those measures.

What is already known on this topic

  • There is considerable efficacy evidence from randomised controlled trials that targeted exercise training programmes can prevent lower-limb injuries.

  • The strength of the results from these studies has been limited by low uptake or compliance with the delivered interventions.

What this study adds

This study provides new information about players' understanding of the importance of training sessions for improving performance and lower-limb injury prevention that can be used to develop specific targeted intervention delivery plans that would ensure players' maximum participation with all study protocol requirements and maximum adoption of the intervention itself.


D Lloyd and B Elliott are thanked for their input in the planning stage of this study. E Roediger and M Romiti conducted the surveys as members of the research team. The data-collection/management phase of this project was conducted from the University of New South Wales (for authors CFF and DT), and the analysis phase was completed at the University of Ballarat (for all authors). The Australian Centre for Research into Injury in Sport and its Prevention (ACRISP) is one of the International Research Centres for Prevention of Injury and Protection of Athlete Health supported by the International Olympic Committee (IOC).


View Abstract


  • Funding The data-collection phase of this study was funded by a University of New South Wales GoldStar Award. The analysis phase was supported by a nationally competitive research grant from the (Australian) National Research Council of Australia (NHMRC) (ID: 400937). CFF was supported by an NHMRC Principal Research Fellowship (ID: 565900), and DT/PW by the NHMRC Project Grant.

  • Competing interests None.

  • Ethics approval Ethics approval was provided by the University of New South Wales Human Ethics Committee.

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

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.