Can movement tests predict injury in elite orienteerers? An 1-year prospective cohort study

ABSTRACT

The purpose of this study was to determine the predictive value of the movement test, the nine test screening battery (9TSB) and an orienteering-modified version of the 9TSB (M9TSB), for lower extremity injury in adolescent elite orienteerers. Design Prospective cohort study. Participants Forty adolescent (15–19 years), male and female orienteerers from two Swedish orienteering high schools performed the 9TSB, M9TSB, and recorded injuries based on a web-based questionnaire for 52 weeks. Results The results showed no difference in composite scores between injured and non-injured orienteerers for either 9TSB (p = 0.75) or M9TSB (p = 0.83). The optimal cut-off score was calculated at 25 for the 9TSB, with sensitivity and specificity of 74% and 41% respectively, and 17 for the M9TSB, with sensitivity and specificity of 47% and 61%, respectively. There was no association between 9TSB or M9TSB and injury (OR1.38, 95% CI: 0.39–4.92). Including athletes with a history of injury did not result in improved prediction of injury for the 9TSB or M9TSB (OR 2.84, 95% CI: 0.50–16.10). Conclusion Low sensitivity and specificity were obtained for both the M9TSB and the 9TSB. Thus, it is not recommended that physiotherapists use the nine test screening battery to predict lower extremity injury in orienteerers.

KEYWORDS:

• Adolescents
• screening
• orienteering

Introduction

Most injuries occur during competitions in orienteering, where the injury incidence is reported to be 7.3–15.4/1,000 competition hours (Ekstrand and Tropp 1990; Linko et al. 1997) compared to 2.2–3.0/1000 training hours (Johansson 1986; Roos et al. 2015). The majority of injuries in orienteering are located in the lower extremities, with a reported injury incidence proportion between 70% and 94% (Johansson 1986; Linko et al. 1997; Roos et al. 2015). Ankle sprains appear to be the most frequent injury (Ekstrand and Tropp 1990; Hintermann and Hintermann 1992; Linde 1986; Linko et al. 1997; Mclean 1990) with an incidence proportion between 8% and 37% (Johansson 1986; Linde 1986; Roos et al. 2015). However, Leumann et al. (2010) reported that 86% of athletes on the Swiss National Orienteering Team had experienced an acute ankle sprain and 72% of those athletes had sustained recurrent ankle sprains. The different injury numbers are likely related to the use of different injury definitions, such as time-loss, injuries reported by medical staff or using the all physical complaints definition. To date, only two studies have been published, describing risk factors for injuries in adolescent elite orienteerers focusing mainly on extrinsic risk factors (Roos et al. 2015; Von Rosen et al. 2016). Von Rosen et al. (2016) indicated high training volume, long competition time and running on asphalt and paths were identified as extrinsic risk factors for injuries, whereas Roos et al. (2015) identified several training variables and sex as risk factors for injury severity. Thus, a limited number of risk factors for injuries have been explored in orienteerers.

The nine test screening battery (9TSB) (Frohm et al. 2012), is a functional screening test, recently developed from the movements tests of the Functional Movement Screen (FMS) (Cook et al. 2006a, 2006b). The 9TSB, which originally consisted of nine complex functional movement tests, assesses quality of movement in terms of stability and mobility and aims to identify faulty movement patterns and side differences (Frohm et al. 2012). Two tests were recently added, such that the 9TSB nowadays consists of eleven different functional movement tests. The screening test is used to identify intrinsic risk factors by modifying these risk factors, such as implementing training protocols, it is suggested that injury occurrence may be prevented. However, in two recent published meta-analyses on the FMS, low predictive validity for injury (Dorrel et al. 2015) and subsequent injury (Moran et al. 2017) was found. In the only published report of predictive validity for the 9TSB, the total score was not associated with lower extremity injury in football players (Bakken et al. 2018). In orienteering, no study has analyzed the predictive validity of a screening battery, although the risk of lower extremity injury is high in this population (Von Rosen et al. 2016). Based on the literature of orienteering, where lower limb injuries are prevalent (Johansson 1986; Linde 1986; Linko et al. 1997; Roos et al. 2015; Von Rosen et al. 2016), an adjusted version of the 9TSB for orienteering may be more relevant for predicting injury in this population. This is in line with recommendations by Flodstrom et al. (2016) and Frohm et al. (2012), who suggest that future studies explore and identify the most relevant tests of the 9TSB in a specific population.

Meeuwisse et al. (2007) have previously described a model of etiology in sports injury, emphasizing intrinsic risk factors that predispose athlete to injury. In orienteering, intrinsic risk factors such as flexibility, stability and strength deficits have been neglected in favor of extrinsic risk factors (Roos et al. 2015; Von Rosen et al. 2016). However, intrinsic risk factors have been found to explain injury risk in runners (Wen 2007), whose injuries may share similar mechanisms as those in orienteering. In addition, overuse injuries in runners have been associated with faulty movement patterns (Wen 2007), which supports the use of exploring injury risk in orienteering based on movement screening that was developed to assess quality of movement patterns and side differences. The purpose of this study was to determine the predictive value of the movement test, the 9TSB and an orienteering-modified version of the 9TSB (M9TSB), for lower extremity injury in adolescent elite orienteerers. It was hypothesized that the score on the M9TSB could predict injury in adolescent elite orienteerers.

Methods

Design

This study is part of the larger KASIP-study (Karolinska Athlete Screening Injury Prevention), a prospective cohort study, aiming to understand injury incidence and associated risk factors in Swedish adolescent elite athletes. The orienteerers performed the 9TSB at their high school during September and October 2013. Injury data were then collected prospectively for 52 weeks.

Participants

The coaches at the two existing national orienteering high schools in Sweden were contacted to determine feasibility. The coaches provided conceptual support for the project. One of the authors visited the two orienteering high schools to inform orienteerers and coaches about the project and to invite the orienteerers to participate. In total, 61 orienteerers, both male and female age 15–19 (median 17), were invited to participate in this project. One orienteerer declined participation. Informed consent was collected from all the included orienteerers. The study was approved by the Regional Ethical Committee in Sweden (2011/749–31/3).

To be included in the data analysis, the orienteerer had to be injury-free at the start of the study, to respond to a minimum of four questionnaires throughout the injury registration period as well as to complete the 9TSB before the injury registration period. The injured orienteerers at start of the study (n = 27) were included in the data analysis after reporting no physical complaint leading to reduced training volume, experience of pain or reduced performance in sporting activity that affected participation in normal training or competition for four consecutive weeks (Clarsen et al. 2013).

Injury registration procedure

Orienteerers received a background questionnaire prior to the start of the study covering personal data and previous injury for the last 12 months. An injury questionnaire was emailed weekly to all orienteerers for 52 weeks, from autumn 2013 to 2014, using Questback online survey software (Questback V. 9.9, Oslo, Norway). One of the two high schools started injury registration the week after they performed the 9TSB and the other school five weeks later due to software issues. If no response was registered, a reminder email was sent four days later. The average response rate over 52 weeks was 62.7%.

The questionnaire used for injury registration was based on the translated Swedish and valid version (Ekman et al. 2015) of the Oslo Sports Trauma Research Centre (OSTRC) Overuse Injury Questionnaire (Clarsen et al. 2013) and a questionnaire regarding new injuries, used by Jacobsson et al. (2013) in an athletic surveillance study. It also included questions about performed training and competition for the last week. The OSTRC Overuse Injury Questionnaire has shown acceptable content and construct validity with a Cronbach’s α of 0.91, as well as reliability (ICC: 0.62) (Clarsen et al. 2013; Jorgensen et al. 2016). All injury data was self-reported and an injury was defined as any new physical complaint in the lower extremity (foot to hip region) that affected participation in normal training or competition, leading to reduced training volume, experience of pain or reduced performance in orienteering (Clarsen et al. 2013). Acute injuries, referring to an injury resulting from a specific identifiable event, were excluded and consequently the injury definition incorporates only overuse injuries in the lower extremity.

9TSB test procedure

The 9TSB aims to analyze the quality of functional movement patterns, originally during nine different tests (i.e. deep squat, one-legged squat, in-line lunge, active hip flexion, straight leg raise, push up, diagonal lift, seated rotation and functional shoulder mobility), where two tests were recently added (i.e. deep one-legged squat, drop jump) (Frohm et al. 2012). The quality of the performance on each test is graded from zero to three. A score of three indicates a test performed correctly without compensatory movements. A score of two indicates a test preformed correctly but with compensatory movements. A score of one indicates a participant’s inability to perform the test correctly without major compensatory movements. A score of 0 is recorded if pain is present during the test. Three attempts at each test were allowed and the score for the most correct performed movement, according to the specific assessment criteria, was recorded. If there was a score difference between the sides, the lower score was recorded as the final score. The total score range is 0–33. The 9TSB has been found to be reliable, both for interrater (ICC: 0.68–0.80) and intrarater (ICC: 0.75) reliability (Frohm et al. 2012).

To modify the 9TSB for orienteering, an expert group was established, consisting of four physiotherapists with several years of clinical experience in the use of the 9TSB and consulting with injured orienteerers. The experts were asked to fill out a questionnaire about the tests they considered to be clinically relevant in predicting lower extremity injury in orienteering. The questionnaire was then compiled by the first author of this study. Based on majority votes, tests were included or excluded. If no consensus about inclusion or exclusion of a test could be reached the test was included in the final screening. The final modified version, M9TSB, consisted of eight tests including deep squat, one-legged squat, deep one-legged squat, in-line lunge, active hip flexion, straight leg raise, diagonal lift, and the drop jump test (Table 1). The push up, seated rotation and functional shoulder mobility tests were excluded. Each test of the M9TSB was graded in the same way as the 9TSB, and the maximum composite score was 24 (score range 0–24). Still, all orienteerers performed all movement tests and the data for the 9TSB and M9TSB were collected concurrently. Before performing the 9TSB, all orienteerers were orally and visually informed of the procedure for each test, following the instructions of Frohm et al. (2012). All test instructors were physiotherapists with several years of clinical experience of the 9TSB.

Table 1. The experts’ considerations of the tests in the nine test screening battery to be clinically relevant in predicting lower extremity injury.

	Expert 1		Expert 2		Expert 3		Expert 4		All experts
Tests	Yes	No	Yes	No	Yes	No	Yes	No	Yes	No
Deep squat	X		x		x		x		x
One-legged squat	x		x		x		x		x
Deep one-legged squat	x		x		x		x		x
In-line lunge		x	x		x		x		x
Active hip flexion	x		x		x		x		x
Straight leg raises	x		x		x		x		x
Push up		x		x		x		x		x
Diagonal lift		x	x		x		x		x
Seated rotation		x	x			x		x		x
Functional shoulder mobility		x		x		x		x		x
Drop jump	x		x		x		x		x
Total	6	5	9	2	8	3	8	3	8	3

Yes = keep the test, no = exclude the test.

Statistical analysis

Descriptive statistics was used to present the score on the 9TSB and M9TSB. Due to non-normally distributed data of composite scores, assessed by the Shapiro-Wilk test (p < 0.05), a significance test using the Mann-Whitney U test was conducted to detect differences in the score on the 9TSB and M9TSB between the noninjured and injured orienteerers. In the case of multiple injuries for a subject, only the first reported overuse injury was included in the statistical analysis. Receiver operating characteristic (ROC) was used to illustrate the test’s ability to predict injury and a corresponding area under the curve (AUC) was determined. An ROC-curve illustrates the ability of a diagnostic test to distinguish correctly between injured (sensitivity) and non-injured (1-specifity) individuals. It is optimal for a diagnostic test to differentiate between injured and non-injured individuals and thereby receive a sensitivity and specificity of 100%, which corresponds to the upper left corner on the ROC-curve. AUC is a summarized measure of the ability for classification in a diagnostic test. AUC is expressed as a number from 0 to 1.0, where an AUC of 1.0 means that the test can differentiate between noninjured and injured individuals 100% of the time. The optimal cut-off scores determined to be optimal for injury screening were calculated using the formula: (1-sensitivity)² + (1-specificity)², where the cut-off score with the lowest value was chosen (Greiner et al. 2000). Odds ratio, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), and negative likelihood ratio (LR-) were calculated for the cut-off scores and presented for all orienteerers and by sex. Odds ratio for injury was also calculated for athletes with a history of injury sustained in the last 12 months for the identified cut-off score. To control for orienteering exposure, training, and competition time were summed up, presented as mean values (hours/week) and compared between cut-off scores. A p value of < 0.05 was considered to be statistically significant. All analyses were performed for both the 9TSB and the M9TSB and analyzed using SPSS software for Windows, version 23.0 (SPSS, Evanston, IL).

Results

Exclusion of participants in data analysis

A total of 20 (33.3%) orienteerers were excluded from the data analysis due to having been injured during the complete registration period (n = 3), reporting less than four weeks of injury registration (n = 7) or failing to perform the 9TSB (n = 10) (Figure 1). Excluded athletes did not differ from the cohort under investigation, in terms of injury at study start (p = 0.19), sex (p = 0.27), or score of the 9TSB (p = 0.99). The final sample, 66.7% (n = 40) consisted of 22 females and 18 males with a mean age of 16 ± 1.0 and BMI of 20.5 ± 1.9.

Figure 1. Flowchart for inclusion of athletes.

Training, injury, and composite score data

The mean exposure to orienteering training and competition, for the 52 weeks was 6.6 hours/week (SD 1.5). Altogether, 53 acute (n = 6) and overuse injuries (n = 47) located in the head (n = 1), hand (n = 1), thoracic spine (n = 1), hip (n = 3), thigh (n = 2), knee (n = 13), lower leg (n = 7), and foot (n = 25) were reported for the one-year follow-up period. Of the 40 included orienteerers, 57.7% (n = 23) reported overuse injury in the lower extremity. No significant difference between injured and noninjured orienteerers was identified in composite scores of the 9TSB (p = 0.75) or the M9TSB (p = 0.83) (Table 2).

Table 2. The frequency score (0, 1, 2, 3) for each test in % and the composite score with median values (25–75th percentiles) of the nine test screening battery, by all athletes and sex.

	All athletes (n = 40)								Female (n = 22)								Male (n = 18)
	Noninjured (n = 17)				Injured (n = 23)				Noninjured (n = 9)				Injured (n = 13)				Noninjured (n = 8)				Injured (n = 10)
Tests	0	1	2	3	0	1	2	3	0	1	2	3	0	1	2	3	0	1	2	3	0	1	2	3
Deep squat	-	25	38	38	4	17	48	30	-	22	33	44	8	23	31	39	-	29	43	29	-	10	70	20
One-legged squat	-	19	63	19	9	13	65	13	-	22	78	-	15	15	62	8	-	14	43	43	-	10	70	20
Deep one-legged squat	-	13	50	38	9	17	39	35	-	11	67	22	8	23	54	15	-	14	29	57	10	10	20	60
In-line lunge	-	-	38	63	4	-	39	57	-	-	44	56	8	-	31	62	-	-	29	71	-	-	50	50
Active hip flexion	-	38	25	38	-	39	39	22	-	22	44	33	-	31	46	23	-	57	-	43	-	50	30	20
Straight leg raises	-	19	50	31	-	13	35	52	-	33	44	22	15	-	39	46	-	-	57	43	-	10	30	60
Push up	-	25	38	38	-	30	39	30	-	44	44	11	-	39	54	8	-	-	29	71	-	20	20	60
Diagonal lift	-	6	81	13	-	9	83	9	-	11	78	11	-	15	77	8	-	-	86	14	-	-	90	10
Seated rotation	-	-	88	13	-	9	65	26	-	-	89	11	-	8	69	23	-	-	86	14	-	10	60	30
Functional shoulder mobility	-	-	6	94	-	-	4	96	-	-	-	100	-	-	8	92	-	-	14	86	-	-	-	100
Drop jump	-	13	25	63	4	9	48	39	-	22	44	33	8	8	54	31	-	-	-	100	-	10	40	50
	Composite score				Composite score				Composite score				Composite score				Composite score				Composite score
9TSB^1	24 (22–27)				24 (22–26)				23 (22–25)				23 (22–25)				26 (23–28.5)				25.5 (24–29)
M9TSB^2	17 (16–19)				17 (15–19)				16 (16–18)				17 (15–17)				18 (16–20.5)				18 (16–20)

9TSB, nine test screening battery; M9TSB, modified nine test screening battery; ¹Score range 0–33; ² Score range 0–24

Prediction of injury

The ability of the 9TSB and M9TSB to predict injury in orienteerers is presented as ROC-curves (Figure 2). The optimal cut-off score was calculated at 25 for the 9TSB with a sensitivity and specificity of 74% and 41%, respectively, and 17 for the M9TSB with a sensitivity and specificity of 47% and 61%, respectively. AUC for the ROC-curve was calculated at 0.53 (95% CI: 0.34–0.72) and 0.48 (95% CI: 0.24–0.67) for the 9TSB and the M9TSB, respectively.

Figure 2. Receiver operating characteristic (ROC) curves for the composite score of the nine test screening battery, 9TSB (A), and the modified nine test screening battery, M9TSB (B), and injury status. In each figure a diagonal line is showed and a circle symbol for the optimal cut-off score.

There was no association between the 9TSB (OR 1.98, 95% CI: 0.52–7.58) or the M9TSB (OR 1.38, 95% CI: 0.39–4.92) and injury for all orienteerers or by sex (Tables 3 and 4). Including athletes with a history of injury for the last 12 months did not improve injury prediction by the 9TSB (OR 2.84, 95% CI: 0.50–16.10) or the M9TSB (OR 2.84, 95% CI: 0.50–16.10). No significant difference in training and competition exposure to orienteering was identified for the cut-off scores of the 9TSB (6.5 vs. 7.0 hours/week; p = 0.35) or the M9TSB (6.3 vs. 7.1 hours/week; p = 0.12).

Table 3. The predictive validity of 9TSB and M9TSB, with the odds ratio for injury and associated p value.

Test score	Sensitivity	Specificity	PPV	NPV	LR+	LR-	AUC (95% CI)	Odds ratio (95% CI)	p-value
9TSB score ≤ 25
All orienteerers	74%	41%	63%	54%	1.25	0.63	0.53 (0.34–0.72)	1.98 (0.52–7.58)	0.32
Orienteerers with previous injury^a	30%	87%	78%	45%	2.28	0.80	0.41 (0.23–0.60)	2.84 (0.50–16.10)	0.38
Female	92%	22%	63%	67%	1.19	0.35	0.44 (0.16–0.71)	3.43 (0.26–45.03)	0.66
Male	50%	63%	63%	50%	1.33	0.80	0.43 (0.18–0.68)	1.67 (0.25–11.07)	0.57
M9TSB score ≤ 17
All orienteerers	61%	47%	61%	47%	1.21	0.87	0.48 (0.24–0.67)	1.38 (0.39–4.92)	0.62
Orienteerers with previous injury^a	30%	87%	78%	45%	2.28	0.80	0.41 (0.23–0.60)	2.84 (0.50–16.10)	0.38
Female	77%	33%	63%	50%	1.15	0.69	0.45 (0.20–0.70)	1.67 (0.25–11.07)	0.69
Male	40%	63%	57%	45%	1.07	0.96	0.49 (0.21–0.76)	1.11 (0.16–7.51)	0.93

9TSB, nine test screening battery; M9TSB, modified nine test screening battery; PPV, positive predictive value; NPV, negative predictive value; LR+, Positive Predictive Value; LR-, Negative Predictive Value; AUC, area under the curve; ^aPrevious injury sustained for the last 12 months.

Table 4. 2 × 2 contingency table for 9TSB and M9TSB scores by injured athletes.

	9TSB			M9TSB
	≤25	>25	Total	≤17	>17	Total
Number injured athletes (%)	17 (74)	6 (26)	23	14 (61)	9 (39)	23
Number noninjured athletes (%)	10 (59)	7 (41)	17	9 (53)	8 (47)	17
Total	27	13	40	23	17	40

Discussion

In the present study, no difference in the composite scores of the 9TSB or the M9TSB, between injured and non-injured adolescent elite orienteerers was identified. The optimal cut-off score was calculated at 25 for the 9TSB with a sensitivity and specificity of 74%, and 41%, respectively, and 17 for the M9TSB with a sensitivity and specificity of 47% and 61%, respectively. However, no association between 9TSB or M9TSB and injury was found. Including athletes with a history of injury did not result in improved prediction of injury for the 9TSB and M9TSB.

Similar score distribution

Clinical use of the 9TSB or the M9TSB to predict lower extremity injury in adolescent elite orienteerers is of questionable value. Only 74% and 61% of the orienteerers were predicted correctly as injured by 9TSB and M9TSB, respectively. The only test where the uninjured athletes had a higher (better) median score than the injured group was the drop jump test (a score of 3 vs. 2). The excluded tests (i.e. push up, seated rotation, and functional shoulder mobility) showed a similar score distribution for the injured and uninjured athletes, possibly explaining the lack of difference in predictive validity for the M9TSB and the 9TSB.

Low discriminant ability

In the present study, a cut-off score of 17 (maximum score 24) was calculated for the M9TSB. The optimal cut-off score for the FMS varies significantly in previous studies, where: no acceptable cut-off value (Bardenett et al. 2015; O’Connor et al. 2011); a cut-off value of ≤ 14 (Butler et al. 2013; Garrison et al. 2015; Kiesel et al. 2007); ≤ 15 (Bushman et al. 2016); and ≤ 17 (Letafatkar et al. 2014), have been reported. However, a comparison with studies on FMS is challenging since the injury definition, population, and tests and scoring criteria differ. Sensitivity of 61% and specificity of 47% for the M9TSB should be considered low, also confirmed by the AUC value (0.48). In a meta-analysis of the predictive validity of FMS, higher specificity (86%) than sensitivity (25%) was reported (Dorrel et al. 2015). The injury definition and population (number of participants, age, activity level, and sport) differ between these studies, which could be a factor contributing to the reported diverse psychometric properties. In this study, 39% of orienteerers were incorrectly predicted to have an increased risk of injury (PPV 61%), similar to the finding of the FMS (range PPV 42–85%) (Butler et al. 2013). The AUC was calculated at 0.48 for M9TSB, indicating that the test has a low ability to discriminate between injured and noninjured athletes, no better than flipping a coin (Greiner et al. 2000). Female athletes, in both the injured and noninjured group, had a lower median composite score than male injured and noninjured athletes. At the same time, the median value was mostly similar for each test of injured and noninjured athletes and differed only on four tests (i.e. in line-lunge, active hip flexion, straight leg raises, and drop jump), demonstrating limited ability to distinguish between sex and injury.

Movement tests need to be better adapted to orienteering

History of a previous injury, training intensity, training volume, long competition time, and running on asphalt and paths have recently been suggested as risk factors for injury in adolescent elite orienteerers (Jacobsson et al. 2013; Von Rosen et al. 2016). Thus, is questionable whether screening tests aiming to analyze functional movement patterns, with respect to these extrinsic variables alone can predict injury in adolescent elite orienteerers. A further reason for low predictability may be that the included tests are not discriminating enough for elite orienteers. The movement tests may need to be better adapted to orienteering by considering the cause of common orienteering injuries such as ankle sprains, medial tibia stress syndrome, and Achilles tendinopathy (Linde 1986).

Methodological considerations

The strength of this study is the approach of following a homogeneous group of orienteerers over 52 weeks using a valid and reliable web-based injury questionnaire. In addition, two versions of 9TSB were tested using an expert group familiar with it. We also compared training exposure between the identified cut-off scores on the 9TSB and the M9TSB and found no systematic difference, suggesting that training volume did not affect our results. Still, there are some methodological considerations in the present study. The cohort consisted of a limited sample of adolescent elite orienteerers at the only two national sports high schools in Sweden, indicating a possible type II error. However, the limited variance for each test score between injured and noninjured orienteerers suggests a low discriminative ability for the 9TSB. An expert group, with experience of orienteering and the 9TSB, was constructed to identify the most orienteering specific tests of the 9TSB. It is possible that a higher number of experts (> four experts) could have led to inclusion of different movement tests. Nevertheless, due to limited variance in test scores of injured and noninjured orienteerers, selecting different tests would not have changed predictive validity. In one school, the injury registration started five weeks after the performance of the 9TSB, which could have resulted in missing data for new injuries before injury registration started. However, following athletes over 52 weeks can be considered as a satisfactory period of time to identify orienteerers with the highest injury risk. More studies of high quality and with more participants are needed to ensure the ability of the 9TSB to predict injury among young elite athletes in other sports. Furthermore, exploring the discriminative properties of movement tests in relation to different kinds of injuries and severity may still be important.

Conclusions

No difference was found between injured and noninjured orienteerers for the composite score of the 9TSB or the M9TSB. Low sensitivity, specificity, PPV, NPV, and AUC were obtained for the M9TSB and the 9TSB. Physiotherapist should be aware that the 9TSB or the M9TSB, even if injury history is considered, are not sufficient screening tests for predicting lower extremity injury in adolescent elite orienteerers. Since the risk of lower extremity injury is high among orienteerers, other screening tests are warranted.

Declaration of interest

FF and AF have published a book about the nine test screening battery .The other authors report no conflict of interest.

Additional information

Funding

This work was supported by the Swedish National Centre for Research in Sports under Grant [FO2016-0009]