The Prevention of Sport Injury: An Analysis of 12 000 Published Manuscripts

Abstract

Objective: 

To identify the nature and extent of research in sport injury prevention with respect to 3 main categories: (1) training, (2) equipment, and (3) rules and regulations.

Data Sources: 

We searched PubMed, CINAHL, Web of Science, Embase, and SPORTDiscus to retrieve all sports injury prevention publications. Articles were categorized according to the translating research into injury prevention practice model.

Results: 

We retrieved 11 859 articles published since 1938. Fifty-six percent (n = 6641) of publications were nonresearch (review articles and editorials). Publications documenting incidence (n = 1354) and etiology (n = 2558) were the most common original research articles (33% of total). Articles reporting preventive measures (n = 708) and efficacy (n = 460) were less common (10% of the total), and those investigating implementation (n = 162) and effectiveness (n = 32) were rare (1% of total). Six hundred seventy-seven studies focused on equipment and devices to protect against injury, whereas 551 investigated various forms of physical training related to injury prevention. Surprisingly, publications studying changes in rules and regulations aimed at increasing safety and reducing injuries were rare (<1%; n = 63) with a peak of only 20 articles over the most recent 5-year period and an average of 10 articles over the preceding 5-year blocks of time.

Conclusions: 

Only 492 of 11 859 publications actually assessed the effectiveness of sports injury prevention interventions or their implementation. Research in the area of regulatory change is underrepresented and might represent one of the greatest opportunities to prevent injury.

INTRODUCTION

Physical activity is associated with multiple health benefits including primary and secondary disease prevention and reduced mortality rate.¹ The American College of Sports Medicine and the American Heart Association recommend endurance training for at least 30 minutes on 5 days each week for all healthy adults.² However, participation in sport increases the risk of injury.^3,4 In the United States, an estimated 1.4 million injuries occur annually as a result of participation in high school sports.⁵ Of those receiving medical attention for sports-related and recreation-related injuries in the United States, one-fifth of school children and more than one-quarter of working adults experience 1 or more days of lost time from school or work.⁶ In addition to these short-term effects, injured athletes may have an increased risk of long-term sequelae, such as osteoarthritis, with concomitant reduction of physical capacity.^7-11 Thus, participation in sport and exercise is a double-edged sword: the profound health benefits gained versus the risk of injury and the associated morbidity and costs.

Recently, sport injury prevention has received greater attention. For example, the Fédération Internationale de Football Association (FIFA) has established worldwide FIFA Medical Assessment and Research Centers to initiate injury prevention programs and improve athlete health.¹² The first and second World Congress on Sport Injury Prevention were held in Norway (Oslo 2005 and Tromso 2008) and focused on musculoskeletal injury. The third congress will be cohosted with the International Olympic Committee in Monaco (2011). Recently, several consensus statements have tried to establish definitions and standards for sports injury prevention including systematic methods of gathering information.^13-16

To date, few prevention programs have actually been widely implemented. One reason is that the design of good studies and the implementation of interventions can be very difficult. van Mechelen et al¹⁷ explained that sport injury prevention research requires 4 stages: (1) identifying the magnitude of the problem, (2) identifying risk factors that may contribute to injuries, (3) developing potential injury prevention programs in the laboratory setting (eg, the mechanical effectiveness of a knee brace) or as pilot projects, and (4) testing the injury prevention programs under ideal conditions. Finch¹⁸ added 2 more stages to examine barriers and facilitators to the implementation of proven injury prevention programs and termed the new model the “translating research into injury prevention practice” (TRIPP) framework (Figure 1).

FIGURE 1: Categorization of sports injury prevention articles. This figure shows the categorization of sport injury prevention articles. Adapted from van Mechelen et al¹⁷ and Finch¹⁸.

In the current article, our objective is to review all published injury prevention articles with reference to the number and type of publications to help researchers identify knowledge gaps and to help understand why little has changed in actual injury prevention over the years. We categorize these articles using the TRIPP model and the 3 main areas of injury prevention: (1) equipment, (2) training, and (3) rules and regulations (Table 1).¹⁹

TABLE 1: Three Categories of Sport Injury Prevention

METHODS

We searched the online databases of PubMed, CINAHL, Web of Science, Embase, and SPORTDiscus to retrieve all available English language publications related to sports injury prevention up to March 2010 (Appendix 1). After removing duplicates manually and through the reference manager software's internal algorithms (EndNote; Thomson Reuters, Carlsbad, California), one author searched all remaining titles, keywords, and abstracts (when available) and excluded articles if they were completely irrelevant to sports injury primary prevention (eg, treatment-related articles were excluded). We interpreted sport injury prevention in its broadest context and included public health-related articles concerning injury prevention in common recreational activities promoted to achieve an active and healthy lifestyle (eg, bicycle helmet use).²⁰ Military studies were included if they examined the prevention of physical activity-related injuries.

Initial Classification of Articles

We classified articles as (1) review, (2) editorial or letter or comment, or (3) original research either by using the electronic database classification (when available) or manually based on the titles, keywords, and abstracts.

Further Classification of Original Research Articles

We further classified each original research article based on the 6 stages of the TRIPP framework (Figure 1).¹⁸ Articles were considered as injury surveillance if they were descriptive and only reported the incidence of sports injuries (Stage 1). We classified studies as etiology (Stage 2) if they evaluated injury mechanisms or other factors associated with injury causes and severity. Some biomechanical studies and most case reports were included here. Studies categorized as preventive measures (Stage 3) identified potential solutions to an injury problem and developed appropriate prevention programs through well-controlled studies that did not yet include epidemiological phases or field tests (eg, laboratory-based assessments of joint or muscle during injury simulations and biomechanical analysis of restrictions in range of motion because of bracing). We categorized studies as efficacy (Stage 4) if these prevention strategies were studied under ideal conditions (eg, randomized controlled trial studies in a limited population). The implementation study category (Stage 5) included articles that examined how efficacy research could be translated into the “real-world” context (eg, motivators or barriers affecting uptake of successfully proven prevention programs). Finally, effectiveness studies (Stage 6) examined both the implementation of interventions related to risk or health behavior in a “real-world” context and whether the implementation was successful or not.¹⁸

To gain additional insight, we subcategorized all prevention-related and implementation-related articles (Stages 3-6 of the TRIPP model) according to whether the study examined the efficacy or effectiveness of equipment, training, or changes to rules and regulations.¹⁹ The equipment category grouped modifications to and use of any equipment. The vast majority of these examined taping, bracing, protective equipment, shoes, and different forms of shoe inserts. Training studies discussed warm-up, stretching, appropriate training, and psychological training interventions (eg, education related). We categorized studies examining the rules of the game or regional laws as regulation. The remaining articles were categorized as other. These included interventions such as nutrition, supplements, or articles about the attitudes or cost of implementation programs without actually studying their effectiveness.

Analysis

Our results represent the entire population of English language articles published, and therefore, any differences are true differences; normal comparative statistics (eg, confidence intervals and P values, which are used to determine the probability that differences are only due to sampling error) are not required. To analyze the progression of sports injury prevention research, we describe the trends in the number of publications per year within each type of article. The different publication types were grouped together in logically related categories to represent overall trends more clearly. Nonoriginal research articles (reviews, editorials, letters, and comments) were grouped together. We grouped incidence and etiology studies because they are both primarily observational. Prevention programs and efficacy programs were grouped because they represent the first introduction of preventive measures. Implementation and effectiveness studies were grouped because they usually describe and examine prevention programs in the context of a much larger population.

Because the number of articles published per year was small for the original research articles and their subcategories, we present the results in 5-year groupings to facilitate the observation of trends over time without the distraction of year-to-year swings that likely occurred by chance.

RESULTS

Figure 2 illustrates the results of our search strategy. We retrieved a total of 20 296 titles from PubMed (n = 7365), CINAHL (n = 2221), Web of Science (n = 2256), Embase (n = 3298), and SPORTDiscus (n = 5156). Of the 5274 original research articles, there were 1354 incidence, 2558 etiology, 708 preventive measures, 460 efficacy, 162 implementation, and 32 effectiveness-related studies.

FIGURE 2: Inclusion criteria and flow of retrieved articles. This figure shows the flow of articles from initial identification through the various steps of inclusion and exclusion. Nonoriginal research review and editorials/comments/letters represent articles that contain no original data. The descriptors reflect only the format of the article (eg, review, editorial). Original research studies contain collected data that can be classified according to one of the categories listed in Figure 1: incidence, etiology, preventive measures, efficacy, implementation, and effectiveness studies.

Figure 3 shows the number of articles within each publication type for each year from 1938 (the first injury prevention publication identified through our search) to 2009 (last complete year of published data). In general, there is an increasing number of every publication type including the categories of reviews, editorials, letters, and comments. However, most original research articles continue to evaluate incidence and etiology, with many fewer studies investigating preventive measures and efficacy, and even fewer articles investigating implementation and effectiveness. In the most recent 3 complete years (2007-2009), the proportion of articles that are preventive measures and efficacy compared with the total is small (9% for 2007; 17% for 2008; 11% for 2009) and very small for implementation and effectiveness (2% for 2007; 2% for 2008; 1% for 2009).

FIGURE 3: Total numbers of studies by publication type. The total number of studies is shown per year, stratified by publication type. For clarity, publications types are shown in logical groups. We grouped nonoriginal research together (reviews, editorials, letters, and commentaries). Original research was grouped according to the categorization illustrated in Figure 1. We grouped articles related to injury surveillance and etiology (Stages 1 and 2) because they do not directly evaluate prevention programs. Studies that identified potential solutions and developed appropriate programs through well-controlled studies (Stage 3) were grouped with studies examining the efficacy (treatment effect under ideal conditions in a limited population) of prevention programs (Stage 4). Finally, we grouped articles related to implementation. This included research on how efficacy programs could be translated into the “real-world” context (implementation, Stage 5) and effectiveness studies (Stage 6) that actually evaluated risk or health behavior in a “real-world” context.

Figure 4 illustrates the publication trends in articles broken down into the categories of preventive measures, efficacy, implementation, and effectiveness in relation to the main types of prevention: equipment, training, and regulation. Although equipment studies (Figure 4A) show a steady increase across all publication types (total n = 677), the absolute number of effectiveness studies remains very low (n = 8). Implementation studies begin to increase about 15 years after publication of the first studies using preventive measures. Training-related articles (Figure 4B) (total n = 551) increased dramatically in the late 1990s and early 2000s for both prevention programs (n = 321) and efficacy studies (n = 211), whereas implementation (n = 16) and effectiveness (n = 3) studies were rare. The total number of training-related articles from 2005 to 2009 exceeded the number of equipment articles for the first time. The number of studies assessing possible regulatory change (Figure 4C) remained extremely low throughout the years for all publication types (n = 63).

FIGURE 4: Number of articles per subcategory and intervention type. The total number of articles related to each of the original research categories described in Figure 1 are grouped according to whether they investigated interventions related to equipment (A), training (B), or regulation (C). Equipment studies discussed the role of equipment in injury prevention (eg, headgear, taping, bracing). Training articles discussed the use of physical training specifically designed to reduce injuries (eg, appropriate warm-up) but excluded training-related articles for other reasons (eg, effect of stretching on range of motion and gains in strength with muscle resistance exercise programs). Regulation studies referred to the assessment or implementation of rules and regulations designed to increase the safety in sport. Articles that could not be easily categorized (n = 77) are omitted.

DISCUSSION

The increase in the number of sport injury prevention articles published over the years across all publication types (n = 11 859) reflects not only the importance of prevention but also the number of open questions and controversies in this field. However, original research on sport injury prevention represents only half of the total publications (n = 5274). Of those, 74% are descriptive articles (incidence and etiology) (n = 3912) and 26% are intervention articles (preventive measures, efficacy, implementation, and effectiveness) (n = 1362). Of the intervention research articles, most were related to training (n = 587) and equipment (n = 709), and less than 5% (n = 66) were related to rules and regulations.

The finding that more than 50% of the articles retrieved since 1938 are reviews, commentaries, or letters (Figure 3) indicates a high level of concern about sports injuries and implies difficulty translating these concerns into intervention studies. One reason for the lack of well-designed hypothesis-driven implementation studies is the difficulty of both conception and execution.²¹ Thus, the accumulated literature may simply represent data from particular injury surveillance programs to which the authors have access. In addition, the field of sports medicine tends to lack appropriate injury surveillance for all sports,²² although this is gradually being addressed and remedied.^23,24 Using a classification based on the models by van Mechelen et al¹⁷ and Finch¹⁸ allowed us to categorize the nature and main purpose of each article and to provide an overview of where research has progressed and where gaps are likely still present in sports injury prevention knowledge. It is clear from the paucity of implementation studies that there remains a wide gap between our knowledge of effective prevention programs and our ability to successfully implement them.^25-27

With respect to the specific types of sport injury intervention programs, we grouped them into 3 categories: (1) training, (2) equipment, and (3) regulatory (Table 1). Figure 4 shows a substantial increase in the number of prevention programs studying physical training.^28-30 These studies are usually conducted using resources that may not be available in typical community settings.^31,32 Changing training habits may seem to be an effective prevention strategy, but it requires a willingness on the part of the participant and coach to change behavior. For example, the FIFA Medical Assessment and Research Centers “11+” program, primarily designed to prevent injuries to the lower extremities in football (soccer), consists of 27 different exercises,³³ which, if new to the athlete, require substantial modification of daily habits to be compliant. The relatively few implementation studies may be explained by the fact that it may take as long as 1 to 2 decades for original research to be translated into medical practice.³⁴ The sharp increase in training studies observed in recent years may soon be followed by a concomitant increase in implementation studies.

Equipment studies were more common than training studies 15 years ago (Figure 4). More recently, however, the number of training studies is almost double the number of equipment studies. There are several possible explanations. One explanation is that most equipment research is funded by industry, which has always had a financial incentive to design and test their products. Once equipment usage has enough popular support, industry may simply choose to market updated versions of their products rather than invest more research funds. That said, equipment implementation studies remain more common than training implementation studies. This is likely because these interventions are frequently easier to implement (eg, bicycle helmets)³⁵ compared with training studies, which are time intensive and require changes in coach and athlete behavior. Another possible explanation for this change in research emphasis is that equipment changes by themselves have not had the expected global impact, leading to increased interest in training studies.

One of the most striking findings of this study is the very low number of publications in the third category of research: rules and regulations (Table 1). These represented less than 0.6% of the entire scanned literature (Figure 4). What is equally striking is the fact that studies have shown substantial preventive effects through regulatory change including the dramatic reduction in cervical spine injuries as a result of outlawing spearing in tackle football,³⁶ and the mandatory participation of the New Zealand Rugby Union in a nationwide education program (RugbySmart) on safe techniques for contact that caused a 50% reduction of catastrophic spinal injuries for the country.³⁷ In addition, foot and ankle injuries in baseball were greatly reduced with rule changes requiring breakaway bases,³⁸ and education and rules regulating mouth guard use have led to increased rates of mouth guard usage in contact sports and a decrease in the frequency of dental injuries.³⁹ However, rule changes do not always reduce injuries because they must be enforced to be effective. More importantly, rule changes will likely only be effective in the long term if they change the culture and redefine what is considered acceptable or unacceptable behavior in that sport.

Limitations of Present Study

Our search was limited to the 5 most common electronic databases that included the most common journals. We classified articles based on the TRIPP framework, but misclassification may have occurred because some titles did not include abstracts (more common pre-1990) and data extraction was conducted by a single reviewer. However, the trends observed were so large that any misclassification that occurred is highly unlikely to affect the overall interpretation.

SUMMARY AND RECOMMENDATIONS

Of the 11 859 articles we retrieved on the topic of sport injury prevention since 1938, only 492 were efficacy or effectiveness studies. Of these 492 articles, the majority of interventions were related to training or protective equipment and mechanical devices with <0.6% related to rule changes that govern sport. In addition, less than 2% of the studies over the past 3 years examined the effectiveness of prevention programs in a real-world context. Although this study was not designed to determine why this is so, it is clear that these intervention studies are very difficult to perform. This difficulty, however, should not deter researchers from seeking the evidence to prevent injuries in real-life situations. Research in the area of regulatory change is underrepresented, yet numerous studies have shown that it might represent one of the greatest opportunities to prevent injury.