National Strategy for Preventing Injuries from Skiing and Snowboarding in Switzerland

Giannina Bianchi¹, Othmar Brügger¹

¹ bfu – Swiss Council for Accident Prevention, Research Department, Berne, Switzerland

Abstract

Skiing and snowboarding are very popular sports in Switzerland. However, every year, around 67,000 skiers and 18,000 snowboarders are injured so seriously that they need medical treatment. Moreover around 8 people die on the slopes in Switzerland. Therefore, there is a need to reduce the burden of injuries in snow sports. The Swiss Council for Accident Prevention (bfu) developed a strategy that utilized scientifically-based injury prevention, using an effect-oriented prevention cycle. The aim of this study is to present this systematic and evidence-based strategy for snow sports. Throughout the prevention cycle, the incidence and severity of injuries are established, risk factors and possible prevention measures are identified and rated, prevention goals are set, prevention programs are developed and implemented, and the success of the measures and processes are monitored. In addition, the effect-oriented prevention cycle involves collaboration with other prevention-minded partners. This strategy allows the bfu to carry out its legal mandate to prevent non-occupational accidents and to coordinate prevention measures throughout Switzerland that are effective, efficient, and practical. Furthermore, the collaboration with other prevention-minded agencies greatly improves the implementation.

Zusammenfassung

Ski- und Snowboardfahren sind beliebte Sportarten in der Schweiz. Dabei verletzen sich jedoch jedes Jahr durchschnittlich 67 000 Skifahrer und 18 000 Snowboarder so schwer, dass sie ärztlich behandelt werden müssen. Zudem ereignen sich jedes Jahr rund 8 tödliche Unfälle auf Schweizer Schneesportpisten. Daher ergibt sich im Schneesport ein Präventionsbedarf. Zur effizienten Verletzungsprophylaxe stützt sich die Strategie der bfu – Beratungsstelle für Unfallverhütung auf einen wirkungsorientierten Präventionskreislauf. Ziel dieses Artikels ist, diese systematische und evidenzbasierte Strategie anhand des Schneesports aufzuzeigen. Im Verlauf des Präventionskreislaufes werden das Ausmass und der Schweregrad der Verletzungen ermittelt, Risikofaktoren und mögliche Präventionsmassnahmen identifiziert und bewertet, Präventionsziele gesetzt, Präventionsprogramme erarbeitet und implementiert sowie der Erfolg der Massnahmen und Programme evaluiert. Dabei werden alle Schritte des Kreislaufes in Zusammenarbeit mit Präventionspartnern umgesetzt. Dieses strategische Vorgehen ermöglicht es der bfu, ihren gesetzlichen Auftrag, Nichtberufsunfälle zu verhüten und wirksame, effiziente und umsetzbare Präventionsmassnahmen in der Schweiz zu koordinieren, wahrzunehmen. Zudem verbessert die Einbindung aller Akteure in den ganzen Prozess die Implementierung von Präventionsmassnahmen.

Introduction

Skiing and snowboarding are very popular sports in Switzerland with more than 1.7 million skiers and 350,000 snowboarders aged 10 to 74 years (Lamprecht et al., 2008a; Lamprecht et al. 2008b). Beside Austria and Andorra, Switzerland has the highest ratios of visiting skiers to inhabitants (1.66) and visiting skiers to foreign visitors (1.60) worldwide (Vanat, 2014). The 240 ski areas in Switzerland registered 24.8 million visiting skiers during the winter 2011/12 season.
Each year, around 400,000 men and women residents in Switzerland are injured in a sports accident either in or outside Switzerland (bfu – Swiss Council for Accident Prevention, 2013), with an average of 135 fatalities (Bianchi and Brügger, 2013). In addition, on average, there are 61 fatalities among foreigners who suffer a fatal sports accident in Switzerland. Accidents largely occur while skiing, snowboarding, biking, bathing, swimming, playing football, and hiking in the mountains (Brügger, 2012).
Accidents are avoidable; therefore, the Swiss Council for Accident Prevention (bfu) and like-minded organizations in Switzerland aim to reduce the frequency and/or severity of accidents and injuries by implementing effective prevention measures.
The bfu is a foundation and has a legal mandate to prevent non-occupational accidents in the sectors of sports, home and leisure, and road traffic, and to coordinate prevention measures throughout Switzerland. Thus, decisions are made without economic and political vested interests. The aims and key aspects of the bfu are based on the need for action and knowledge of effective prevention measures; however, social and political aspects and personal resources are considered.
The aim of this study is to present the systematic and evidence-based strategy of the bfu to prevent injuries in snow sports in Switzerland.

Injury Prevention Strategy

The bfu designed a scientifically-based prevention strategy in Switzerland, using an effect-oriented prevention cycle composed of 5 steps (see Figure 1). First, a critical analysis of the incidence and severity of injuries is carried out, etiology and mechanisms of injury are established, and possible preventive measures are framed and rated. Based on this accident research, realistic prevention goals are defined in consideration of the mission and resources. To outline national prevention programs and implement the measures, prevention-minded partners in Switzerland are involved. Here, bottom-up approaches are also considered. Finally, the success of the prevention process and its interventions are monitored and the findings provide the basis for a new prevention cycle. The different stages are not strictly separate from each other and sometimes a step must be repeated.

The steps of accident research and monitoring of success are carried out by the research department of the bfu. The training, advice, and communication departments of the bfu in collaboration with prevention partners plan and implement efficient programs. Prevention goals are set by the management of the bfu. However, each step involves all experts.
Injury prevention strategies have been described by other scientists. Van Mechelen and colleague’s “sequence of prevention” is a four stage approach that identifies and describes the incidence and severity as well as the factors and mechanisms of sports injuries, while introducing and evaluating measures (van Mechelen et al., 1992). Finch et al. considered that the model insufficiently implemented the measures in the real-world and updated the model with two more steps (Finch, 2006). The TRIPP framework of Finch and colleagues complements the “sequence of prevention” by adding stage 5, which describes the intervention in the context of implementation strategies, and stage 6, which evaluates the effectiveness of preventive measures in the context of implementation. Van Tiggelen et al. extended the model of Van Mechelen by considering not only the efficacy, but also the efficiency, compliance, and risk taking behavior, since these factors influence the outcome of the prevention measure (Van Tiggelen et al., 2008).
These prevention strategies correspond in several aspects to the safety strategy of the bfu. However, none of the strategies quantify the clear goals that should be achieved. Each step of the bfu prevention cycle is discussed separately in the following sections.

Accident Research

Accident research detects problems, analyzes and rates the relating risk factors, and identifies and assesses effective prevention measures. Therefore, the following questions are addressed: “What happens?”, “Why does it happen?”, and “How can it be prevented?”(Brügger et al., 2012).
While a lot of studies report on the frequency and etiology of injuries, few evaluate preventive measures and their efficacy (Klügel et al., 2010). Even fewer studies evaluate the implementation and effectiveness of preventive measures. In addition to scientific, peer-reviewed articles, data are also obtained from surveillance systems in Switzerland and some existing grey literature. However, there was not enough information to quantitatively describe the risk factors and preventive measures. Therefore, the opinions of research and implementation experts and qualitative panel discussions were necessary to rate the risk factors and interventions.

Accident analysis

To conduct target-oriented injury prevention, the need for action in Switzerland was first identified. The risk of injury and fatality rates as well as the severity of injuries for different sporting activities identified the sports where preventive measures should be implemented. Thus, the injury prevention effort in Switzerland focused mainly on areas with high frequency and/or severe injuries and fatalities (Figure 2). Due to the injury frequency, skiing and snowboarding were one focal point of injury prevention in Switzerland (Brügger et al., 2012).
Three main sources provided data on the burden of injuries in snow sports in Switzerland: (1) a national injury surveillance system collected data about non-occupational injuries of employed people; (2) reports from ski patrols; and (3) another database gathered data on fatal sport accidents (bfu – Swiss Council for Accident Prevention, 2013; Bianchi et al., 2013). Finally, the bfu makes an extrapolation to estimate the number of injuries. Thus, this is a comprehensive set of data in contrast to studies based only on data from hospitals or emergency departments, since we also included less serious injuries; consequently, there is less risk of underestimating injury occurrences (Mitchell et al., 2010).
Every year, around 51,000 skiers and 15,000 snowboarders who reside in Switzerland are injured on slopes and require medical treatment (bfu – Swiss Council for Accident Prevention, 2013). An estimated 10% of skiers and 5% of snowboarders suffer their injuries on slopes outside of Switzerland. Moreover, around 40% of injured skiers and 25% of injured snowboarders visiting the rescue center are travelers from other countries (bfu – Swiss Council for Accident Prevention, 2011). Altogether, around 67,000 skiers and 18,000 snowboards are injured on the slopes in Switzerland and require medical treatment. Moreover, approximately eight people die on slopes each year while skiing or snowboarding (Bianchi and Brügger, 2013). Thus, approximately 3.1 skiers or snowboarders (Swiss inhabitants or guests from foreign countries) are injured per 1000 skier-days on slopes in Switzerland (Vanat, 2014; bfu – Swiss Council for Accident Prevention, 2013). The Swiss Federal Office for Sports (FOSPO) and the bfu calculated in 2008 that skiing and snowboarding had incidence rates of 46 and 68 injured Swiss inhabitants, respectively, per 100,000 hours of activity (bfu – Swiss Council for Accident Prevention, 2013).
However, not only the incidence but also the severity of sports injuries influences the need for action. In Switzerland, the severity of snow sport injuries is defined by the length of the hospital stay. However, the national injury surveillance system also bases the severity on the medical diagnoses, and future efforts may consider the nature of the injury, as previously proposed (van Mechelen, 1997; Pless and Hagel, 2005). Every 20th injury in skiing and snowboarding leads to hospitalization for at least one week and can be classified as “severe” (bfu – Swiss Council for Accident Prevention, 2013). However, 86% and 93% of all injuries in skiing and snowboarding, respectively, do not require hospitalization.

Risk analysis

As soon as the extent of injury is established, the question of the etiology arises. Risk factors for skiing and snowboarding injuries in Switzerland have been determined while considering that accidents involve multiple internal and external risk factors (Meeuwisse et al., 2007), as well as the inciting event (Bahr and Krosshaug, 2005). Risk factors can be classified temporally (pre-event, event, and post-event) and epidemiologically (human, vector, and environment) (Table 1) (Haddon, 1980).

Once, potential risk factors were gathered, they were rated by an expert group of the bfu based on available data, literature, and panel discussions and in cooperation with external specialists from snow sports in Switzerland. The relevance of each risk factor was based on the level of danger and prevalence (Figure 2, Table 2).

Intervention analysis

The third step in accident research involves how accidents and injuries can be prevented. Using a top-down approach, detailed analyses of the accident event and injury, as well as the risk factors, indicate where preventive measures should be implemented.
To identify potential measures to reduce the frequency and/or severity of injuries in Switzerland, it was considered that interventions can influence the host (participant), the vector (sport or activity), and the environment (both physical and sociocultural) through three strategies: education, enforcement, and engineering (Gilchrist et al., 2007; Gielen and Sleet, 2006). To predict whether preventive measures may or may not work in a particular context, the effectiveness, efficiency, and feasibility of the measures were monitored in the safety strategy in Switzerland (Table 3) (Brügger et al., 2012). Thus, measures that reduce the frequency or severity of injuries in real world conditions, were considered based on financial, practical, and administrative implications including social acceptability, financial feasibility, technical limits, time resources, legality, and ethical aspects. Measures that were not effective or economically feasible were initially excluded (Figure 2). This assessment is partly consistent with methods described previously by Donaldson (2010), Van Tiggelen et al. (2008), and Runyan (1998).
In regards to risk factors, the main aim of the prevention strategy is to avoid an event. However, not every accident is preventable. Therefore, a lot of prevention strategies reduce the amount of energy transferred from the agent to the host (Pless and Hagel, 2005), Factors concerning care and rehabilitation after the accident were not evaluated by the bfu, since the mission of the bfu is to avoid accidents and injury and not to optimize recovery.
The prevention measures listed in table 4 were (highly) recommended for avoiding injuries on slopes in Switzerland, based on their effectiveness, efficiency, and the feasibility of implementing the measure (Brügger et al., 2012).

Prevention Goals

The main goal in injury prevention is to reduce the frequency and/or severity of injuries. In a scientific-based strategy, prevention goals have to be defined. According to the top-down approach, targets arise from accident research. In Switzerland, most injury prevention measures are aimed at hazardous sports, involving lots of accidents and/or severe injuries. However, a bottom-up approach can also be used and the safety needs of the population must be considered, even if it is not a focal point of accidents. Moreover, accidents caused by outside variables have to be prioritized. In defining the objectives the mission, legal mandate, economic and ethical principles, and available resources (people, finances, time, knowledge, etc.) are taken into account. The prevention goal is to reduce the rate of injuries within a certain period of time. Quantitatively, the defined goal in injury prevention in Switzerland between 2010 and 2015 is to reduce the burden of fatalities and severe injuries by 10%.

Prevention Programs

Preventive programs consist of different interventions and aim to translate scientific evidence into practical tools and frameworks that can be adopted e.g. by coaches, teachers, governments, industry, tourism, institutions, or participants. The results of the programs can be regulations, rules, guidelines, norms, material developments, training programs, instruction materials, handbooks, movies/radio spots, leaflets, posters, and more.
In Switzerland, measures are worked out in collaboration with all relevant national partners (around 50) in the field of snow sports. Evidence-based interventions are presented to an expert working group of stakeholders to critically review measures, discuss advantages and drawbacks, and reach a consensus (Finch, 2010; Brussoni et al., 2006). However, with the “bottom-up” approach, measures that are identified as important by partners, including innovative ideas and approaches, are observed, reviewed, and supplemented, if necessary. Potential operators do not only provide content-related input (e.g. usability, format, and messages) (Winston and Jacobsohn, 2010), but also share their experiences with possible stumbling blocks and promising implementation strategies (Collard et al., 2010; Runyan and Freire 2007). Therefore, the prevention efforts and needs of all networkers in Switzerland can be coordinated, duplication can be reduced, more effective interventions can be developed, and a united strategy can be developed to convince target group (Christoffel and Galagher, 1999; Finch et al., 2011).
Relevant stakeholders include e.g. the Swiss cable car association (SBS), the Swiss Commission for Prevention of Accidents on the Ski Slopes (SKUS), the association of ski- and snowboard schools, Swiss Snowsports, the Swiss Federal Office of Sport (FOSPO) that includes the program Youth+Sports (Y+S), the ski federation Swiss Ski, the Swiss National Accident Insurance Fund (Suva), the Swiss association of sport article shops (ASMAS), the sports offices of each canton, public schools, coaches, athletes, and more.

Implementation of measures

To successfully implement and adopt measures in the real-world, a practical strategy is required (Finch, 2011). As mentioned above, promising interventions are developed in collaboration with relevant stakeholders to optimize the likelihood of success of the interventions. All stakeholders must move in the same direction to strengthen the credibility and importance of a measure, to convince target audiences, and to pool the efforts by communicating the same message (Christoffel and Galagher, 1999).
A target group in Switzerland can be reached by different channels, e.g. the unique network of “bfu safety delegates” in cities and municipalities, by cable car companies, by rules and regulations, by campaigning on radio/television, print media and events, by product safety laws, by training in snow sport courses/camps, by public schools, sport clubs, and private fitness centers.
Studies about implementation of measures in sports are rare (about 1%) (Klügel et al., 2010); therefore, little information exists on how to effectively disseminate measures in sport communities. Institutions and participants are willing to adopt measures if they bring a benefit to their core business and do not conflict with their philosophy (Finch, 2006). For example, tourism is more likely to support a prevention program and implement measures if it brings in more guests; whereas, athletes are more willing to adopt a program if the measure helps them improve their performance. Interventions are also more likely to be implemented if the target group is fairly prepared, the measures are easy to adopt and accepted by peers and role models, and if consumers know the benefits of implementing the measures (Finch, 2006).

Monitoring Success

The last step in the effect-oriented prevention cycle involves evaluating the program to identify problems that limit the effectiveness of measures and to monitor the success of an intervention strategy. To assess an intervention program within the safety strategy in Switzerland, four different kinds of evaluations are used: process, impact, outcome (Lowe et al., 2006) and concept. This allows the quality of implementing the intervention to be assessed as well as target-group attentiveness. In addition, we can find out how and why interventions took effect or not, measure if the defined long-term aim of the program has been reached, and estimate the effective potential of the intervention.
Although monitoring success is the last step in the prevention cycle, evaluation precedes the prevention process. For effective evaluation, a plan has to be made at the beginning of the program (Lowe et al., 2006).
Evaluations by the bfu are based on different qualitative and quantitative strategies and methods: analysis of project documentations, surveys (interviews and observations) of skiers and snowboarders on slopes as well in target groups, analyzing injury data, group discussions, media analysis, expert panel discussions, etc. For example, a campaign to wear a snow sport helmet has been evaluated extensively by external experts on behalf of the bfu (Furrer and Balthasar, 2011; Bianchi et al., 2011). Analysis showed that the campaign concept had the necessary success potential and was implemented as planned. However, the impact of the campaign on the increase of helmet use is unclear. The results suggest that the campaign has reinforced and boosted the effects of various factors that promote helmet use.

Conclusions

The strategy described in this paper involves a systematic approach to increase the level of safety, using the example of snow sports, in Switzerland. This effect-oriented prevention cycle involves steps from accident research to prevention implementation strategies, in collaboration with other prevention-minded partners in Switzerland. The integration of scientific knowledge and practical experience is often challenging and compromises may have to be made for the interventions to be successful. This effect-oriented prevention cycle provides a good systematic approach for preventing and reducing injuries.

Contributorship

Both authors fulfill the criteria of authorship which are based on substantial contributions to conception and design, drafting the article or revising it critically for important intellectual content and final approval of the version to be published.

Corresponding author

Giannina Bianchi
Research Department
bfu – Swiss Council for Accident Prevention
Hodlerstrasse 5a
3011 Berne, Switzerland
E-Mail: g.bianchi@bfu.ch
Phone +41 31 390 21 59
Fax +41 31 390 22 30

References

1. Bahr R., Krosshaug T. (2005): Understanding injury mechanisms: a key component of preventing injuries in sport. Br J Sports Med. 39(6):324-329.
2. bfu – Swiss Council for Accident Prevention. (2011): STATUS 2011: Statistics on non-occupational accidents and the level of safety in Switzerland, Road traffic, sports, home and leisure. bfu, Berne. http://www.bfu.ch/en/order/everything?k=2.080 (accessed 5 July 2014).
3. bfu – Swiss Council for Accident Prevention. (2013): STATUS 2013: Statistics on non-occupational accidents and the level of safety in Switzerland, Road traffic, sports, home and leisure. bfu, Berne. http://www.bfu.ch/en/order/everything?k=2.118 (accessed 5 July 2014).
4. Bianchi G., Brügger O., Niemann S., Furrer C. (2011): Evaluation of a national campaign in snow sports. Br J Sports Med. 45(4):331-331.
5. Bianchi G., Brügger O. (2013): [bfu statistics on fatal sports accidents in Switzerland, 2000-2012] (in German). bfu – Swiss Council for Accident Prevention, Berne http://www.bfu.ch/en/order/everything?k=2.117 (accessed 5 July 2014).
6. Brügger O., Bianchi G., Hofer F., Walter M., Michel F.I., Müller C. (2012): [Accident Research – Sport: analysis of accidents, risks and interventions] (in German, with summary in English and French). bfu – Swiss Council for Accident Prevention, Berne. http://www.bfu.ch/en/order/everything?k=2.106 (accessed 5 July 2014).
7. Brussoni M., Towner E., Hayes M. (2006): Evidence into practice: combining the art and science of injury prevention. Inj Prev, 12(6):373-377.
8. Christoffel T., Galagher S.S. (1999): Injury prevention and public health: Practical knowledge, skills, and strategies. Aspen Publisher, Gaithersburg.
9. Collard D., Singh A., Verhagen E. (2010): The behavioural approach. In: Sports Injury Research, Verhagen E., van Mechelen W. (eds), Oxford University Press Inc., New York, 2010, S. 157-166.
10. Donaldson A., (2010): The pragmatic approach. In: Sports Injury Research, Verhagen E., van Mechelen W. (eds), Oxford University Press Inc., New York, 2010, S. 139-156.
11. Finch C. (2006): A new framework for research leading to sports injury prevention. J Sci Med Sport. 9(12):3-9.
12. Finch C.F. (2010): Implementing studies into real life. In: Verhagen E, van Mechelen W, eds. Sports Injury Research. Oxford University Press Inc., New York, 2010, S. 213-235.
13. Finch C.F. (2011a): No longer lost in translation: the art and science of sports injury prevention implementation research. Br J Sports Med. 45(16):1253–1257.
14. Finch C.F., Gabbe B.J., Lloyd D.G., Cook J., Young W., Nicholson M., Seward H., Donaldson A., Doyle T.L.A. (2011b): Towards a national sports safety strategy: Addressing facilitators and barriers towards safety guideline uptake. Inj Prev. 17(3):1–10.
15. Furrer C., Balthasar A. (2011): [Evaluation of the bfu’s prevention campaign in snow sports] (in German, with summary in English and French). bfu – Swiss Council for Accident Prevention, Berne. http://www.bfu.ch/en/order/everything?k=2.072 (accessed 5 July 2014).
16. Gielen A.C., Sleet D.A. (2006): Injury Prevention and Behavior: An Evolving Field. In: Injury and Violence Prevention: Behavioral Science Theories, Methods, and Applications, Gielen A.C., Sleet D.A., DiClemente R.J. (eds), Jossey-Bass, San Francisco, 2006, S. 1-16.
17. Gilchrist J., Saluja G., Marshall S.W. (2007): Interventions to Prevent Sports and Recreation-Related Injuries. In: Handbook of Injury and Violence Prevention, Doll L.S., Bonzo S.E., Mercy J.A., Sleet D.A. (eds), Springer, Atlanta, 2007, S. 117-134.
18. Haddon W. (1980): Advances in the Epidemiology of Injuries as a Basis for Public Policy. Public Health Rep. 95(5):411-421.
19. Klügl M., Shrier I., McBain K., Shultz R., Meeuwisse W.H., Garza D., Matheson G.O. (2010): The prevention of sport injury: an analysis of 12,000 published manuscripts. Clin J Sport Med. 20(6):407-412.
20. Lamprecht M., Fischer A., Stamm H. (2008a): [Sport in Switzerland 2008: Sport Behaviour in the Swiss Population] (in German and French). Swiss Federal Office for Sport (FOSPO), Magglingen.
21. Lamprecht M., Fischer A., Stamm H. (2008b): [Sport in Switzerland 2008: Report on Children and Adolescents] (in German and French). Swiss Federal Office for Sport (FOSPO), Magglingen.
22. Lowe J.B., Yang J., Heiden E., DiClemente R.J. (2006): Intervention Research and Program Evaluation. In: Injury and Violence Prevention: Behavioral Science Theories, Methods, and Applications, Gielen A.C., Sleet D.A., DiClemente R.J. (eds), Jossey-Bass, San Francisco, 2006, S. 188-209.
23. Meeuwisse W.H., Tyreman H., Hagel B., Emery C. (2007): A Dynamic Model of Etiology in Sport Injury: The Recursive Nature of Risk and Causation. Clin J Sport Med. 17(3):215-219.
24. Mitchell R., Finch C., Boufous S. (2010): Counting organised sport injury cases: evidence of incomplete capture from routine hospital collections. J Sci Med Sport. 13(3):304-308.
25. Pless I.B., Hagel B.E. (2005): Injury prevention: a glossary of terms. J Epidemiol Community Health. 59(3):182-185.
26. Runyan C.W. (1998): Using the Haddon matrix: introducing the third dimension. Inj Prev. 4(4):302–307.
27. Runyan C.W., Freire K.E. (2007): Developing Interventions When There is Little Science. In: Handbook of Injury and Violence Prevention, Doll L., Bonzo S., Sleet D.A., Mercy J.A., Haas E.N. (eds), Springer, Atlanta, 2007, S. 411-431.
28. Vanat L. (2014): 2014 International Report on Snow and Mountain Tourism: Overview of the key industry figures for ski resorts. Vanat L., Geneva. http://www.vanat.ch/RM-world-report-2014.pdf (accessed 14 July 2014).
29. van Mechelen W., Hlobil H., Kemper HC. (1992): Incidence, severity, aetiology and prevention of sports injuries. A review of concepts. Sports Med. 14(2):82-99.
30. van Mechelen W. (1997): The severity of sports injuries. Sports Med. 24(3):176-180.
31. Van Tiggelen D., Wickes S., Stevens V., Roosen P., Witvrouw E. (2008): Effective Prevention of Sports Injuries: a Model integrating Efficacy, Efficiency, Compliance and Risk Taking Behaviour. Br J Sports Med. 42(8):648-652.
32. Winston F.K, Jacobsohn L. (2010): A practical approach for applying best practices in behavioural interventions to injury prevention. Inj Prev. 16(2):107-112.