Despite strides, gaps between lab and practice remain

It’s an exciting time for clinicians involved in preventing sports injuries. Increasing numbers of studies are identifying risk factors associated with specific injuries and documenting the effectiveness of preventive interventions for reducing injury rates.

But in many ways, it’s also a confusing time, especially when it comes to designing and refining evidence-based interventions. Studies can identify risk factors associated with an injury, but that doesn’t necessarily mean those risk factors actually contribute to the injury or that targeting those risk factors will reduce injury rates. Other studies can demonstrate a preventive benefit of an intervention without offering any insight into why it works, or why it works for some athletes, but not others.

Such gaps between the sports injury prevention literature and clinical practice are not as wide as they once were, but still represent significant challenges for researchers and clinicians alike. During the triennial IOC World Conference on Pre­vention of Injury and Illness in Sport, held in Monaco in March, this theme was re­visited continually—by keynote speakers, researchers presenting their latest findings, and attendees commiserating and collaborating during coffee breaks.

Wrestling with risk factors

As most sports medicine professionals are well aware, the model for sports injury prevention¹ introduced by van Mechelen and colleagues in 1992 proposes that, after identifying the need to reduce the risk of a particular injury (step 1), researchers should identify risk factors and mechanisms that contribute to that injury (step 2), after which they should develop preventive interventions (step 3), and rigorously evaluate the effect of each intervention (step 4).

But Willem van Mechelen, MD, PhD, now a professor of occupational and sports medicine at Vrije Universiteit in Amsterdam, opened the conference with a keynote speech noting many ways in which sports injury prevention has turned out to be more complicated than he and his colleagues envisioned. For one thing, there have been far more papers published on topics related to steps 1 and 2 of the model than those related to steps 3 and 4.

Designing effective interventions based on identified risk factors, it turns out, is not as straightforward as experts had hoped. The reasons for this were explored in another keynote presentation by Roald Bahr, MD, PhD, professor in the Department of Sports Medicine at the Norwegian School of Sport Sciences, chair of the Oslo Sports Trauma Research Center, and head of the Aspetar Sports Injury & Illness Prevention Programme in Doha, Qatar.

The idea that identified risk factors should be used to screen athlete populations for those at the highest risk, and that interventions should be implemented based on risk factor based cutoffs, is particularly problematic, Bahr said—echoing the theme of a 2016 paper he published in the British Journal of Sports Medicine.²

“Identifying risk factors does not mean we can identify players at risk,” Bahr said.

As an example, he cited a 2004 study in which athletes with a history of hamstring injury were 7.42 times more likely to suffer a future hamstring injury than those with no history.³ Despite the impressive odds ratio, Bahr pointed out, only 10 of the 19 hamstring injuries in that study occurred in athletes with a history of injury, so screening based solely on that risk factor would have missed nearly half of the athletes who ended up being injured.

“When screening for injury risk, statistically significant association is not the same as prediction,” Bahr said.

In Monaco, this theme was revisited multiple times during oral research presentations on risk factors, particularly those related to lower extremity injuries.

In a study of 306 youth basketball and floorball (a type of floor hockey) players, researchers from the UKK Institute for Health Promotion in Tampere, Finland, analyzed frontal plane projection angle during single-leg stance at baseline and then assessed new time-loss injuries for 12 months.⁴ They found that a frontal plane knee projection angle greater than one standard deviation above the mean was associated with a 5.57 times higher risk of lower extremity injury and a 2.37 times higher risk of ankle injury.

But first author Anu Raisanen, PT, PhD, project coordinator for the Tampere Research Center of Sports Medicine, cautioned conference attendees in Monaco to consider the findings in light of Bahr’s comments.

“As you know, an association does not equal prediction,” Raisanen said. “So we have to keep that in mind. Most likely, we can use this [information] in combination with other factors.”

Although a risk factor that is consistent across dynamic tasks would be ideal for injury prediction, a study from Liverpool John Moores University in the UK found that most anterior cruciate ligament (ACL) risk factors are more strongly correlated with some tasks than others.⁵ In 41 female athletes, knee abduction angle at initial contact had moderate to good correlations across multiple dynamic tasks (bilateral and single-leg drop vertical jumps, single-leg hops, and side stepping), but knee flexion angle at initial contact, peak knee abduction moment (KAM), and peak vertical ground reaction force had low to moderate correlations.

“We all think a high load will be a high load regardless of the task, but we can see this is not the case,” said Raihana Sharir, a doctoral student in the university’s Institute for Sport and Exercise Science, who presented the findings in Monaco. “One variable can’t tell us everything.”

A second presentation from the Finnish group provided a more detailed analysis of a study published earlier this year, in which stiff landings were associated with ACL injury risk in 171 young female basketball and floorball players.⁶ In the Monaco presentation,⁷ peak hip flexion during landing from a vertical drop jump was negatively associated with noncontact ACL injury risk, with a hazard ratio of .6.

However, the area under the receiver operating curve for that risk factor was .54, suggesting that it would be nearly as likely to predict a false positive result as a true positive result, said Mari Leppanen, PhD, a researcher in the Tampere Research Center of Sports Medicine and first author of both studies, who presented the findings in Monaco.

“We are nowhere near saying we can predict ACL injury based on hip flexion,” Leppanen said.

Multimodal mysteries

The challenge of making the jump from identifying risk factors to preventive intervention wasn’t the only one mentioned by van Mechelen in his keynote presentation. The increasing evidence that few injuries can be traced to a single risk factor would seem to support the use of multimodal interventions, such as the 11+ warm-up program (formerly known as the FIFA 11+)—and, in fact, such programs have been associated with reductions in injury rates, most recently a 4.25-fold decrease in ACL injuries in elite male soccer players.⁸

In such situations, however, researchers often have no way of knowing which aspects of such multimodal interventions are actually driving the outcomes. And the picture gets even more complicated when one considers that, even if an intervention is effective across a group of athletes, there will almost always be some athletes within that group who respond positively and some who do not.

This was underscored in Monaco by a presentation from the University of Delaware in Newark, in which collegiate women’s soccer players performed the 11+ warm up at least three times per week for two seasons.⁹ On average, peak KAM did not change over time for the intervention group or a control group. However, 14 of 39 players in the 11+ group did demonstrate a decrease in peak KAM; those athletes had smaller hip flexion moments and hip adduction moments than the nonresponders, said Amelia Arundale, PT, DPT, a physical therapist and research assistant at the university who presented the findings in Monaco.

Chasing adherence

Of all the challenges involved in sports injury prevention, perhaps the most vexing for van Mechelen is that even the best intervention will not be effective if athletes don’t actually complete it. Too few intervention studies assess athlete adherence, he noted—likely because it isn’t easy to document objectively, and self-reported adherence is likely to be inflated. But there’s no question that it makes a difference.

In a 2011 randomized controlled trial from the Netherlands, for example, only 23% of athletes in the intervention group were fully compliant with a neuromuscular training program designed to prevent ankle sprain recurrence.¹⁰ Although the program had no effect on the group as a whole, the relative risk of an ankle sprain was significantly lower in those who were adherent compared with those who were not, as well as compared with those in the control group.

“We can design interventions, and we know they are effective. But typically we have lousy adherence, which affects injury outcome,” van Mechelen said.

The good news is that a growing number of sports injury prevention studies—including several presented in Monaco—are confirming that better adherence is associated with better outcomes.

In a study of more than 2400 youth rugby players,¹¹ researchers from the University of Bath in the UK found that a preventive exercise program was associated with a 15% reduced risk of match-related injuries compared with controls, which was not a statistically significant difference. However, in players who completed the program at least three times per week, the relative risk of match injury was reduced by 72%, according to Michael Hislop, MSc, a doctoral student at the university who presented the findings in Monaco.

In analyzing the effectiveness of the 11+ warm-up program for reducing injury risk in collegiate men’s soccer players, the Delaware research group found four teams had low compliance (1-19 sessions per season), 14 teams had moderate compliance (20-39 sessions), and nine teams had high compliance (40 or more sessions).¹² Overall injury rate was 6.39 per 1000 athlete exposures in the high compliance group, 8.55/1000 in the moderate group, and 10.35/1000 in the low group. Differences between the groups were statistically significant, according to Holly Silvers-Granelli, MPT, a doctoral student at the university who presented the findings in Monaco.

Researchers from Waseda University in Tokyo reported an impressive 89% compliance rate among the 189 collegiate women’s basketball players assigned to an education and hip-focused neuromuscular training program designed to reduce ACL injuries.¹³ The incidence of ACL injury was .08/1000 athlete exposures in the intervention group, compared with .25/1000 in a control group.

The high level of adherence, however, won’t help the Japanese researchers determine whether the intervention’s effectiveness was related to changes in biomechanics, since those were not assessed, noted Yorikatsu Ohmi, RPT, MS, a researcher at the university who presented the findings in Monaco.

Sports injury prevention has evolved considerably since van Mechelen’s model was introduced, but at least one theme has re­­mained the same: Nothing is ever simple.

References:

1. Van Mechelen W, Hlobil H, Kemper HC. Incidence, severity, aetiology, and prevention of sports injuries. A review of concepts. Sports Med 1992;14(2):82-99.
2. Bahr R. Why screening tests to predict injury do not work – and probably never will…: a critical review. Br J Sports Med 2016;50(13):776-780.
3. Arnason A, Sigurdsson SB, Gudmundsson A, et al. Risk factors for injuries in football. Am J Sports Med 2004;32:5S-16S.
4. Raisanen A, Pasanen K, Krosshaug T, et al. Association between frontal plane knee control and acute lower extremity injuries. Br J Sports Med 2017;51(4): 375-376.
5. Sharir R, Vanrenterghem J, Robinson M, et al. What separates an individual at risk of ACL injury? A first step towards an ACL-risk movement passport. Br J Sports Med 2017;51(4):387.
6. Leppanen M, Pasanen K, Kujala UM, et al. Stiff landings are associated with increased ACL injury risk in young female basketball and floorball players. Am J Sports Med 2017;45(2):386-393.
7. Leppanen M, Pasanen K, Krosshaug T, et al. Landing with less hip flexion is associated with increased risk of ACL injuries in young female team sports players. Br J Sports Med 2017;51(4):349.
8. Silvers-Granelli HJ, Bizzini M, Arundale A, et al. Does the FIFA 11+ injury prevention program reduce the incidence of ACL injury in male soccer players? Clin Orthop Relat Res 2017 Apr 7. [Epub ahead of print]
9. Arundale A, Silvers-Granelli H, Marmon A, et al. Biomechanical changes with FIFA 11+ utilization over multiple soccer seasons. Br J Sports Med 2017; 51(4):287.
10. Verhagen EA, Hupperets MD, Finch CF, et al. The impact of adherence on sports injury prevention effect estimates in randomized controlled trials: Looking beyond the CONSORT statement. J Sci Med Sport 2011;14(4):287-292.
11. Hislop M, Stokes K, Williams S, et al. The efficacy of a movement control exercise program to prevent injuries in youth rugby: A cluster-randomised controlled trial. Br J Sports Med 2017;51(4):329-330.
12. Silvers-Granelli H, Bizzini M, Arundale A, et al. Does higher compliance to the FIFA 11+ injury prevention program improve overall injury rate in male soccer (football) players? Br J Sports Med 2017;51(4): 388-389.
13. Ohmi Y, Hirose N. Effects of a prevention program focused on hip joint function on the incidence of anterior cruciate ligament injuries in female basketball players. Br J Sports Med 2017;51(4):366-367.