Does a higher training age attenuate injury risk in junior elite rugby league players?

Introduction

Rugby league (RL) is a popular contact sport played at representative levels in Australia from the age of 14 years onwards. ¹ Potentially high injury incidences are associated with RL, given participation rates and game exposure time when compared to other sporting contexts.^2,3 The injuries most commonly observed include injuries to the lower limb (e.g. ankle, knee, hamstring) ⁴ in the form of muscular strain, ligament sprain and contusions. ⁵ Injury incidence typically increases with higher playing levels and intensities. ⁶ However, previous studies have demonstrated that higher playing levels, ⁷ higher pre-season training load (TL) ⁸ and increased aerobic capacity (VO2 max) ⁹ have been associated with reduced injury risk at various ages and playing levels. ¹⁰ These findings potentially suggest that training and physical preparation experience may help reduce the likelihood of lower limb injury risk.

Injury risk is of particular concern during the developmental stages of an athlete’s playing career due to under developed musculoskeletal physiology (bone, connective tissue), and the repeated injury risk following an initial occurrence of many common injuries sustained in contact sports. ¹¹ Recent studies have suggested relationships between relative age, maturation and physiological characteristics. ¹² During these developmental years, players’ also commonly begin to enter structured training programs according to chronological age-related playing groups and more frequent competitive schedules at representative playing levels involving higher physical playing intensities. Further, as prior injury is one of the greatest predictors of subsequent injury, ¹¹ it can be hypothesised that developing athletes may benefit from structured progressive training experience which potentially reduces the likelihood of injury. However, very little research has examined such possible relationships between prior training history and injury risk.

Developing athletes require years of structured, progressive and intense training incorporating a multidisciplinary approach. In particular, increasing physical TLs over extended time periods are required to provide adequate stimulus to attain the desired performance enhancements. ¹ However, high progressive TL can be accompanied by increased exposure to negative outcomes, such as injury and illness. ¹³ At the crucial developmental level, TL must increase to elicit adaptations that match the requirements of higher performance levels. ¹⁴ Therefore, TL must be carefully monitored in order to minimize exposure to negative outcomes.

Training age (TA) is defined as the number of formalized years of strength and conditioning training ¹⁵ and is an often overlooked measure of an athlete’s ability to tolerate prescribed TLs and intensities. Little is known about the potential benefit from TA and the length of time spent in a training modality to injury risk. TA could be a potentially useful measure of training experience in youth athlete assessment when precise measurements of TL are considered. For instance, indirectly related studies have shown higher degrees of performance enhancement in maximal strength at lower resistance TAs in RL players, ¹⁶ while soccer players with a lower playing age have demonstrated more asymmetrical lower limb strength, ¹⁷ a known predictor of injury in sporting populations. At the junior elite level, RL players are exposed to three fundamental training modalities; RL training (skills based training), cardiovascular training (energy systems conditioning) and resistance training (strength based training). In this study, we examined whether these three different TA domains, Rugby League Training Age (RLTA), Cardiovascular Training Age (CVTA) and Resistance Training Age (RTA) were associated with injury risk in a sample of junior elite RL players.

Materials and methods

A multi-season prospective cohort design was conducted at one elite RL club at two representative levels (U16 and U18) over two seasons (2015–2016). Binary logistic regression with RLTA, CVTA and RTA as independent predictors were modelled on injury/no injury outcomes. Relationships between TA and injury characteristics were also examined using linear regression. Previous injury was also modelled in a binary logistic fashion.

Procedures

TA data were collected prior to the commencement of pre-season training via the use of a standardised questionnaire. Each player was provided detailed instructions on how to complete the questionnaire which requested detailed information on the players’ medical history and previous training history, including details and examples of RLTA, CVTA and RTA. Each domain was defined as the period in which the participant had previously engaged in consistent structured training for that training modality and was reported in the number of years and months.

Each training session and competitive match was attended by the lead researcher, and all relevant training and injury data were documented at the time of occurrence. Overall TLs were assessed using both external (Global Positioning System technology (GPS)), variables measured included total distance (m), high-speed distance (<18 km/h), very high speed distance (<28 km/h), total number of accelerations and decelerations (>2.78 m/s^–2) and total sprint (>1s above 5 m/s^–1) distance (m), ¹⁸ and internal (session rate of perceived exertion (sRPE)) ¹⁹ measures. Thirty GPS units (SPI HPU 15Hz position sampling, 100 Hz, 16 G tri axial accelerometer) were randomly allocated (15 per playing group, per session) evenly among positional groups (forwards and backs) across both squads (U16 and U18) at each training session by the lead researcher. A total of 86.9 ± 3.7 observations per player were analysed. All players reported an RPE score (0–10)²⁰ within 30 min of training session cessation. sRPE was calculated as RPE×session duration (min) and expressed as arbitrary units (AU). Session and weekly mean ± SD were calculated for all external and internal TL variables. Weekly mean ± SD were also calculated for pre-season and in-season training phases.

Injury was defined as “the loss of bodily function or structure that is the object of observations in clinical examinations or any immediate sensation of pain, discomfort or loss of functioning that is the object of athlete self-evaluations”. ²¹ Injury occurrences were recorded at the time of occurrence (if a player was unable to continue playing or training or voluntarily left the field) directly by the lead researcher or the club physiotherapists. Only injuries sustained during training or matches related to these specific playing squads were included in the analyses. Injuries were categorised as incidence, site, type and mechanism (Table 1) according to categories defined in the standardised injury recording template and previous RL studies. ¹ Overall injury incidence was reported as number of injuries/1000 training hours. Injury characteristics were dichotomised according to whether or not each individual player sustained an injury of that specific type, site or mechanism. Previous relevant injuries were then dichotomised per individual participant in the same manner.

OPEN IN VIEWER

Table 1.

Mean weekly training load variables.

		Internal load	External Load measures using GPS
	Duration (min)	sRPE (AU)	Distance (m)	Distance VHIR (>28 km.hr–1)	Total distance HIR (>18 km.hr–1)	Sprint count	Total Acc/Dec >2.78 m.s–1	Repeat sprints	Total sprint distance (m)
Pre-season	392.4 ± 116.7	2516.3 ± 785.8	18329.3 ± 5480.8	15.1 ± 22.5	1127.1 ± 551.1	209.8 ± 77.6	171.2 ± 61.9	108.0 ± 53.1	2412.4 ± 877.1
In season	287.7 ± 78.2	1691.7 ± 516.5	13813.7 ± 3435.6	11.2 ± 14.1	879.3 ± 444.8	171.7 ± 56.1	146.7 ± 57.4	90.5 ± 44.8	1931.3 ± 630.9

AU: arbitrary units; min: minutes; m: metres; HIR: high-intensity running; sRPE: session rating of perceived exertion; VHIR: very high intensity running; km.hr^–1: kilometres per hour; Acc: acceleration; Dec: deceleration; m.s^–1: metres per second.

Results

TA at the beginning of the study was reported as RLTA (9.1 ± 2.8 years), RTA (1.64 ± 1.38 years) and CVTA (2.4 ± 2.3 years).

Weekly mean TL decreased from the pre-season training phase to the in-season training phase for both internal and external measures (see Table 1). sRPE decreased over the same period by 32.7% from an average of 2516.3 AU to 1691.7 AU per week. Weekly external measures also showed a reduction from pre- to in-season in total distance (–24.6%), total high intensity running >18 km/h distance (–22%) and total very high intensity running >28 km/h (–25.8%).

Total injuries sustained over the two seasons were 300 (Table 2). Injury incidence over the two seasons of tracking was 4.03/1000 h. There were 163 match injuries with an incidence of 35.67/1000 match hours and 137 training injuries with an incidence of 3.82/1000 training hours. Overall match and training injury incidences increased from 2016 to 2017, as match injury incidence increased from 33.7 to 37.64/1000 match hours, while training injury incidence increased from 3.22 to 4.47/1000 training hours.

OPEN IN VIEWER

Table 2.

Injury characteristics.

Injury site		Injury type		Injury mechanism		Injury severity
Head/face	41	Abrasion	10	Being tackled	93	Transient	135
Neck	8	Laceration	12	Tackling	71	Mild	68
Shoulder	32	Contusion	65	Player contact	41	Moderate	68
Arm	11	Concussion	16	Ground contact	9	Severe	29
Elbow	1	Dislocation	5	Running	69	Total	300
Wrist	6	Fracture/bone	21	Accelerate	17
Hand	24	Joint	7	Total	300
Sternum	10	Cartilage/disc	18
Rib	6	Ligament sprain	81
Lumbar	25	Muscle strain	50
Hip	2	Tendon	0
Groin	8	Nerve	15
Quad	13	Total	300
Hamstring	15
Knee	34
Lower leg	25
Ankle	30
Foot	9
Total	300

RLTA was positively associated with risk of bone-related injuries (Table 3) (e.g. fracture, stress fracture and shin splints). RLTA explained 4.7–9.6% of the variance in 89.3% cases of bone injury (OR 1.34, p = 0.03). This equated to a 33.7% increase in the risk of bone injury per year of RL training previously accumulated.

OPEN IN VIEWER

Table 3.

Relationship between training age and injury characteristics.

	N	Sig.	Exp (B)	95% C.I. for Exp (B)
				Lower	Upper
Rugby league training age – bone injuries	26	0.025	1.337	1.038	1.722
Resistance training age – hamstring injuries	15	0.008	1.055	1.014	1.098
Resistance training age – running mechanisms	86	0.009	1.026	1.007	1.047
Cardiovascular training age – muscle strains	50	0.050	1.156	1.000	1.027
Cardiovascular training age – ankle injuries	30	0.049	1.18	1.000	1.031
Cardiovascular training age – shoulder injuries	32	0.033	1.192	1.001	1.032

C.I.: Confidence interval.

CVTA was positively associated with injury including absolute risk of injury (OR 1.65, p = 0.02), suggesting a 65% increase in absolute injury risk per year of CVTA previously undertaken (Table 4). Between 4.3 and 5.8% of variance in overall injury risk in 56.5% cases was accounted for by player CVTA. CVTA also showed several positive relationships with injury characteristics (Table 3), specifically, muscle strains, ankle and shoulder injuries. Between 7.4 and 11.2% of variance in 74.6% cases of muscular strain was identified (OR 1.16, p = 0.05), suggesting 15.6% increase in injury risk per additional year of CVTA previously conducted. Between 5.2 and 9.1% variance in 85.2% cases of ankle injuries (OR 1.18, p = 0.05) and between 4.3 and 7.8% variance in 85.8% cases of shoulder injuries (OR 1.19, p = 0.03) were also attributed to increased CVTA.

OPEN IN VIEWER

Table 4.

The relationship between training age and injury risk.

	P	Exp (B)	95% C.I. for Exp (B)
			Lower	Upper
Injury risk (n = 169)
Rugby league training age	0.251	0.798	0.543	1.173
Resistance training age	0.885	0.968	0.624	1.501
Cardiovascular training age	0.021	1.650	1.078	2.526

C.I.: Confidence interval.

RTA identified a positive relationship with injuries sustained during high-intensity running mechanisms (sprinting/accelerating) and hamstring injuries (Table 3). RTA accounted for between 4.1 and 5.6% of variance in 65.7% cases of injuries sustained during explosive running activities (OR 1.31, p = 0.01), suggesting a 31.2% increase in the risk of this type of injury per additional year of prior resistance training undertaken. Between 5.6 and 12.9% of variance in 91.7% cases of hamstring injury was also accounted for by RTA. A binary regression identified 66% increase in the risk of hamstring injury per additional year of resistance training previously undertaken (OR 1.66, p = 0.01). No significant relationships were defined between previous injury and injuries sustained during the course of this study.

Discussion

To our knowledge, this is the first study to investigate the effect of TA types on injury characteristics in junior elite RL players. Our findings indicated there was no attenuated (protective) effect from higher TA against injury risk. While a protective effect may indeed exist, it was not apparent based on our sample, methodological approach and analysis. Instead, our findings illustrated a positive association between higher RLTA and bone-related injury; a positive association between CVTA and increased overall risk of injury and increased risk of specific injury characteristics (ankle and shoulder injuries and muscle strains); and, a positive relationship between RTA and risk of specific injury characteristics (hamstring, sprinting/accelerating). To potentially account for these unexpected findings, it is suggested that training intensities and loads experienced at the junior developmental level may have advanced at such a pace (as may have occurred in our sample) to have negated any protective effect of higher TAs (i.e. non-functional overreaching).

Weekly TL (both internal and external) followed a trend consistent with previous studies. ²² While these values do not appear abnormal, when considered according to developmental stage, and compared to elite level RL players, they could still be considered excessive. ⁸ The developmental players in this study undertook an average external TL of 18,329 m per week over an 11-week pre-season. By comparison, chronic weekly TL of >16,095 m has been classified as a high TL in elite level populations. Weekly internal TL has shown elite level RL players engage in between 2665.4 and 2890.4 AU, ²³ sub-elite senior players between 1345.0 and 1891.0 AU²² and junior elite players between 643.0 and 711.8 AU. ²² The developmental players tracked in our study engaged in between 1691.7 and 2516.3 AU per week. Given that TLs should be progressive over time to elicit desired positive adaptations, the sample (and developing athletes generally) may require a slower progression of TL over a greater length of time (or other progressive adaptation strategies) to better help develop tolerances to a given stimulus load. As our study indicates, observed developmental athletes experienced a faster progression of TL and intensities over a shorter pre-season period, potentially counteracting protective benefits of increased TA previously accrued by more experienced junior elite players.

Overall, injury incidence was lower than previously observed in RL populations. ¹ Recent studies have observed similar incidence of injury for developmental RL players to senior professional RL. ²⁴ Positive relationships were identified between CVTA, overall injury risk and muscular strains and ankle injuries. These relationships between CVTA, overall injury risk and risk of muscular strain may be due to players with higher cardiovascular capacities potentially engaging in greater overall TL and intensities through increased ability to maintain distances and speeds at higher thresholds during conditioning drills. Such positive relationships were also apparent for RTA, with increased risk of any injury sustained via high-intensity running mechanisms and hamstring injuries. A greater engagement in structured resistance training sessions by these more well-trained athletes due to selection in school representative teams may lead to higher off-field TL and partially account for present findings.

The unexpected findings from our study point toward the need for accurate monitoring of TL and intensities, particularly at developmental stages of an athlete’s career. A model for systematic TL monitoring and progression in developmental athletes was recently proposed, highlighting how TL could be better matched with player TA. Such considerations may better inform TL prescription and monitoring to avoid non-functional overreaching and injury risk in developmental athletes. ²⁵ When considered within such a model, present findings highlight the possibility of adverse events and increased injuries when TL and intensities are excessive given an individual TA.

Several limitations exist within our study. The study population monitored reflected the training practices specific to one RL club. Training practices more broadly across clubs would need to be assessed to draw more extensive conclusions. Our study examined relationships between TA, injury during a specific developmental window. Further research examining different ages, stages and skill levels of players/athletes is required to enhance present understanding. Developing athletes often participate in multiple competitions (e.g. school RL) and/or playing additional sports (e.g. rugby union, touch rugby), and so data on training practices for additional sporting commitments would help ascertain a better picture of the relationships between TA, TL and injury. Individualised GPS TL data – as opposed to team averages – would also provide a more accurate methodology and data for analysis. Such an approach would provide clarity regarding individual aerobic capacities and high-speed running thresholds, and thus provide valuable insight into acceptable TL for an individual athletes TA in order to minimise injury risk.

Conclusions

Findings related to TA and injury risk contrasted with our hypothesis that higher TA would, in part, attenuate injury risk. Rather, a positive association between CVTA and RTA with overall injury risk and certain injury characteristics (e.g. injuries sustained through running mechanisms, muscle strains, hamstring and ankle injuries) were identified. Present findings may be explained – in part – due to the high accumulations and progression of TL which may have negated any protective effect forms facets of RL specific, cardiovascular and resistance TA experience.