The purpose of this study was to investigate the effect of an exercise-based injury-prevention program on the incidence of musculoskeletal injury, motor performance and psychosocial status. 36 Estonian Military Academy cadets were randomly assigned into either an intervention or control group. The intervention group followed a neuromuscular exercise-based injury-prevention warm-up program, 3 times per week for 6 months. The control group continued with the usual warm-up. The main outcome measure was injury incidence during the study period. Additionally, evaluation of isokinetic lower-extremity strength, postural sway, physical fitness and psychosocial status was included pre- and post-intervention.

Injury incidence

The primary outcome of interest in our study was incidence of musculoskeletal injury during the 6-month study period. Injury tracking was performed through the Estonian Defence Forces medical database. Since participants receive health care through the military health care system, any musculoskeletal condition severe enough to seek treatment from a medical provider is registered. Additionally, participants were asked to recall any musculoskeletal conditions at 3 and 6 months from the beginning of the study in order to catch potential injuries not reported to a military medical provider, or for which they might have obtained treatment from a nonmilitary medical care provider.

Motor performance testing

Prior to motor performance testing, baseline body mass (kg) and height (cm) were measured. Based on these body mass index (BMI) was calculated (BMI = weight (kg) / height (m)²). Before testing motor performance participants performed a 10-min individual warm-up - jogging, and range of motion and bodyweight strength exercises.

Isokinetic muscle strength testing

Strength parameters of the knee flexor and extensor muscles were recorded using a computerized dynamometer (CON-TREX® MJ, Physiomed Elektromedizin AG, Germany). Participants were seated upright with a hip flexion of 85°, the trunk stabilized with two 3-point seat belts and arms folded across the chest. The extremity to be tested was supported at the distal thigh with a Velcro strap. The distal shin adapter was attached 2-3 cm proximal to the lateral malleolus using a Velcro strap. The alignment of the rotational axis of the dynamometer was set so it would pass transversely through the femoral condyles, 1.5 finger width above joint space and vertical above head of fibula. The peak torque of knee flexion and extension was measured in concentric mode at an angular velocity of 60°/sec between a knee flexion of 10° and 90° (0° = full extension). For task-specific warm-up, the participants performed three submaximal knee flexion-extension movements. The test procedure included 3 maximal, consecutive flexion-extension movements. The participants were given verbal encouragement to flex and extend the leg as fast and forcefully as possible. Based on flexion-extension peak torque measurements hamstring/quadriceps ratio was calculated (H:Q = (flexion peak torque / extension peak torque) x 100%). Additionally, H:Q and peak torque differences between left and right lower extremity were calculated.

Postural sway measurement

A force platform (Kistler 9286A, Switzerland, dimensions 40 x 60cm) and biomechanical movement analysis system Elite Clinic with Sway software® (BTS S.p.A., Italy) were used to assess the equivalent area (EqArea) of the center of pressure sway. Participants were asked to maintain balance while standing on a single leg for 30 sec. For familiarization, the participants balanced 30 sec on each leg. Based on the measurements obtained, the difference of the EqArea between the left and right legs was calculated.

Physical fitness test

Participant physical fitness was assessed with a test consisting of a 2-min maximal-effort push-up event, a 2-min maximal-effort sit-up event and a 3.2 km timed run.

Psychosocial status

To assess participant psychosocial status, the RAND 36-item Health Survey 1.0 was used. All items in this survey are scored on a scale from 0 to 100, with a high score defining a more favorable health state. For analysis, the following subscales were used: physical functioning, role limitations due to physical health problems, role limitations due to personal or emotional problems, energy/fatigue, emotional well-being, social functioning, bodily pain and general health perception.

Exercise intervention

Participants in the INT group were asked to follow a prescribed injury-prevention program (PP) during warm-up before physical training for 6 months, 4 times per week. The PP was a structured warm-up routine, taking 20 min to finish. It was divided into 3 sequential parts: (1) whole-body range of motion and bodyweight strength exercises, (2) aerobic load - jogging, and (3) jump, balance and running exercises with a change of direction. A written PP description was provided. INT group participants were instructed on correct PP execution during a workshop led by a physical therapist. Also, detailed written instructions and online video materials were provided to ensure correct program execution. During the study period, no supervision on PP execution technique was provided. Participants in the CON group were asked to continue with their regular training and warm-up without any restrictions. The usual warm-up in the CON group involved a combination of aerobic exercises and range of motion exercises.

Statistical analysis

Descriptive statistics (mean, SD) in both groups were calculated. Differences between INT and CON groups were analyzed using an independent sample Student's t-test, and within-group differences between time points were analyzed using a dependent-sample paired t-test if the assumptions of a normal distribution and homogeneity of variances were satisfied. Normality of distribution was controlled using the Shapiro-Wilk test. Homogeneity of variances was controlled with Levene's test. If the previous assumptions for the t-test were not met, the between-group differences were analyzed with the non-parametric Wilcoxon test, and within-group differences between time points were analyzed using the dependent-sample paired Wilcoxon test. Bonferroni correction was applied to account for possible false positives due to the number of tests. Relative risk (RR) and 95% confidence intervals (95% CI) for between-group injury incidence were calculated. The level of significance was set at p < 0.05. R version 4.1.1 and RStudio version 1.4.1717 were used for statistical analysis.

Inclusion criteria:

• Age 18-30 years
• Healthy volunteer

Exclusion criteria:

• Current injuries causing limited duty days
• Current injuries causing inability to participate in the proposed injury-prevention program.