Discussion

Our main finding was that adoption and implementation of the NH exercise programme at the highest level of male football in Europe, the UCL, was low and, perhaps surprisingly, that it was equally low in the Norwegian Premier League, Tippeligaen, where the programme was pioneered. Of the 150 club-seasons covered by the study, the NH programme was completed in full in 10.7% and in part in an additional 6% of cases. Consequently, 83.3% were classified as non-compliant. Therefore, as four out of five teams were non-compliant with the protocol, it comes as no surprise that hamstring injury rates remain high in UCL and Norwegian elite football.29 ,30

The compliance cut-off selected, that is, using the initial NH programme for at least 5 weeks and the maintenance programme at least most weeks on >50% of players for the rest of the season, may be questioned. First, we do not know what the minimum effective dose is, only that efficacy has been demonstrated for the full, 10-week programme in the trials by Arnason et al1 and Petersen et al.28 As we do not know if lower doses can prevent injuries as well, we chose this threshold. Consequently, the majority of teams were classified as non-compliant, even if they did report using the NH exercise to some extent. An indication that the full 10-week programme needs to be completed is that most of the injuries observed in the intervention group in the study by Petersen et al28 occurred within the initial, 10-week progression period. This suggests that completing this exercise progression may be critical for programme efficacy. A recent systematic review on the efficacy of the NH programme and other eccentric interventions also documents that programme compliance is essential, showing a 65% reduction in hamstring injury risk among highly compliant participants, with no effect when including less compliant participants.34 In the current study, clubs reported doing no initial NH training or no maintenance training for 95 of the 150 seasons included (see table 2); we therefore feel confident in concluding that the vast majority of teams did not use the NH programme as prescribed. Second, classifying programme use retrospectively over three seasons, based on a simple questionnaire, is difficult. Recall bias may be an issue, for example, medical staff for some teams had changed during the period in question. Designing questions to capture compliance accurately is also difficult, and the questionnaire has not been validated against prospective reporting or direct observation, which would be the preferred method.35

Hamstring injury prevention does seem to be a priority among teams, as the majority of clubs reported using specific exercises/exercise programmes other than the NH programme to prevent injuries. In the same way, a recent study revealed that the NH programme is only ranked fifth among preferred exercises to prevent non-contact injury in a survey of 44 premier league clubs.36 As we anticipated team practices to range widely, we made no attempt at creating pre-determined categories to classify their use of other exercises in the same way as for the NH programme. In response to an open question, clubs reported a plethora of exercises, as expected. These include isokinetic eccentric exercises, single leg balance exercises, lunges, trunk/core stabilisation exercises, the Askling rehabilitation exercise protocol,37 ,38 gymball supine exercises, high-speed running and sprint exercises, slider board exercises, pulley exercises, pilates-based exercises, yoga-based flexibility and strength exercises, eccentric leg curls in a yo-yo device39 and several more. Clubs also reported using preseason isokinetic tests and functional movement screening, as well as fatigue monitoring in various forms. However, all these have one thing in common: there is currently no evidence to document their effectiveness in preventing hamstring injuries in football (or in any other sport, for that matter).

In contrast, the efficacy of the NH programme on hamstring injury risk is well documented,1 ,28 and confirmed in a recent study on male Dutch amateur players documenting a 72% reduction in the risk for hamstring injuries.40 The NH exercise addresses the mechanism of injury, believed to be excessive maximal eccentric loading,23–25 as well as one key risk factor, eccentric strength. A number of candidate risk factors have been proposed for hamstring injuries,41 the most prominent internal factors being: previous injury, age, reduced hip range of motion and poor hamstrings strength.26 A history of previous hamstring injury greatly increases injury risk, as documented in numerous studies.26 ,41–43 Older players are at increased risk for hamstring injury, and although older players will be more likely to have a previous injury because of a longer career, increased age is also an independent risk factor for injury.15 ,41 ,43 However, studies on football players suggest that hamstring flexibility is not a significant risk factor for injuries,15 ,44 and a stretching programme did not reduce hamstring injury risk in footballers.1 In contrast, several studies suggest that players with low hamstring strength or low hamstrings:quadriceps strength ratio (or side-to-side strength imbalances) may be at increased risk of injury.26 Inadequate hamstring strength would mean that the eccentric load necessary to resist knee flexion and start hip extension during maximal sprints could surpass the tolerance of the muscle-tendon unit. In particular, the relationship between the ability of the quadriceps to generate speed and the eccentric capacity of the hamstrings to resist the resulting forces, the hamstrings to quadriceps strength ratio, is thought to be important. The 10-week NH exercise programme has been shown to increase the concentric hamstrings to eccentric quadriceps strength ratio by as much as 11%.27 Notably, Opar et al45 recently used a novel field testing device to record hamstring strength, providing a reliable measure of eccentric knee flexor forces during the NH exercise. They documented that low maximal eccentric strength in the NH exercise at the end of the preseason resulted in a fourfold increased risk of sustaining a hamstring injury during the season.45

The question is, therefore: Why do teams not adopt and implement an exercise programme with a well-documented effect on both injury and re-injury risk, but continue using exercises with no or limited evidence? This is very difficult to explain, based on the data from the current study. Reach was good: 88% of clubs reported being familiar with the NH exercise programme. Also, the vast majority of the teams reported few complaints from coaches and players. Programme satisfaction was good among teams using the programme, and very few teams disagreed with pro-programme statements such as “It really reduces injuries”, “It makes more players available for team selection”, “Players can return to play sooner after injury”, “Players perform better”, etc. However, a significant minority felt that it is not easy to get players to do the programme and that it causes muscle soreness in players, despite the fact that the Mjølsnes protocol has been shown to cause no or minimal delayed onset muscle soreness.27 Nevertheless, it appears that we have to look for reasons for non-compliance among other factors, such as limited influence by the medical team on coaching practices. This needs to be explored in future studies.

We were somewhat surprised to see that compliance was no better in Tippeligaen than in the UCL group. The initial NH studies were carried out in Tippeligaen, as early as 200127 and 2002,1 and since then the NH programme has received ample attention in basic and continued education, with books and Norwegian websites aimed at medical personnel. However, there has not been a similar focus on education programmes targeting coaching staff.

In conclusion, adoption and implementation of the NH exercise programme at the highest levels of male football in Europe is low; too low to expect any overall effect on acute hamstring injury rates.