https://orcid.org/0000-0001-8580-9834
ABSTRACT
Loaded ruck marching is a common training and operational task for many members of the military. It is known to cause fatigue, affect soldier readiness, and can lead to traumatic and overuse injuries. Quantifying the gait changes that occur over the course of a loaded ruck march may provide a better understanding of injury mechanisms and potentially allow for development of individualized injury-prevention training programs. This study examined the change in soldiers’ gait patterns over the course of a loaded ruck march in order to examine the correlation between fatigue and kinematic parameters. Fatigue is a subjective term that may encompass factors such as energy expenditure, muscle exhaustion, and cognitive engagement. Since it can be difficult to quantify, the current study makes the broad assumption that fatigue increases in some (potentially nonlinear) fashion during a loaded ruck march.
Three platoons of soldiers participated in a field training exercise with inertial measurement sensors placed on their chests and ankles to record gait parameters throughout a 7-mile ruck march. The effects of fatigue on stride length, stride width, ankle yaw, and torso lean (anterior-posterior [AP] and side-to-side [SS]) were compared using one-way repeated measure analyses of variance.
In comparing the first and last quarters of the ruck march, stride length decreased, stride width increased, stride width variability increased, AP torso lean variability increased, and SS torso lean variability increased.
Although they do not describe a direct relationship to injury, these results can inform enhanced approaches to quantify and predict soldier fatigue and more reliably prevent future injury.
INTRODUCTION
Soldiers commonly operate in physically demanding circumstances that require traversing long distances while carrying heavy equipment. Loaded marching, or rucking, is a key training technique that is often used as a performance metric for members of the U.S. Army. Numerous schools in the Army require soldiers to perform a loaded march to graduate, including Air Assault School, Mountain Warfare School, and Ranger School. Soldiers regularly participate in loaded marches to maintain the physical fitness required for combat and prepare for the unpredictable terrain that may be traveled on the battlefield. These loaded marches can be quantified by time, distance, or the necessities of the mission or training exercise.1 The load carried, time limit, and distance can vary based on the school, award, or certification the soldier is working toward. Although loaded march training differs based on the preparation that is necessary for the unit and the commander’s intent, the loads carried by soldiers can be as heavy as 60% of the soldier’s body weight, causing an increase in injury risk and fatigue.2
Fatigue is defined by the International Journal of Environmental Research and Public Health as “the diminished individual capability to complete actions at the expected level due to lassitude or exhaustion of mental or physical strength.”3 Physical exertion is the primary condition leading to fatigue,3 leading to the natural hypothesis that fatigue, and likely diminished soldier performance, would develop over the course of ruck marches because of their often long and strenuous nature.4 In addition to the degradation of performance, fatigue can have a deteriorating effect on the health of a soldier, resulting in cognitive and physical missteps, incidents, and injuries in an operational setting.3 Understanding the effect of fatigue on kinematic parameters can play a significant role in detecting and reducing performance degradation as well as health-related consequences including chronic injury, lost duty time, and significant medical costs.3–5
Several studies have assessed the effects of fatigue and load carriage on physical performance. For example, time to complete an obstacle course significantly increases when participants start in a fatigued state.6 Increased body-borne loading can have similar negative effects on gait performance. Previous research has shown significant decreases in the range of motion of knee flexion/extension and pelvic rotation, increases in the adduction/abduction and rotation of the hip and pelvic tilt, decreases in the preferred stride length, and increased forward torso lean.4,5,7 Although these findings are indicative of kinematic changes because of fatigue or load carriage, the combined effect of a fatiguing load carriage event has, to date, not focused on the concept of fatigue but rather total load weight.
Biomechanical changes related to fatigue and load carriage may be detrimental to an individual’s physical performance, as well as increase their risk of future injury.6 Some of the variables that have been previously studied are torso lean (anterior-posterior [AP] and side-to-side [SS]), stride length, stride width, and toe in/out angle. After a fatiguing road march, changes in balance and sensorimotor gait parameters have been seen,4 such as a decrease step width variability.8 These findings differ from previous research completed on older adults, which has shown a slower walking speed, wider step width, and decreased step width variability in fatigued older adults, which may indicate an impairment to their balance and sensorimotor control system.9 These findings suggest that older adults widen their step width as a strategy to maintain balance in the presence of challenging walking conditions. This is further supported by the correlation that has been found between stride width variability and fall history.9 For example, extreme step width variability (either too much or too little) is associated with falling and balance issues in older adults.9 By studying the step width and other kinematic parameters that are associated with falling and balance issues, research may find that the physical and cognitive limitations that older adults experience could be analogous to those that healthy adults experience when in a fatigued state, such as that imposed by a long-distance loaded march.9 If true, measurable kinematic parameters such as stride length, stride width, toe in/out angle, and torso lean (both anterior-posterior and side-to-side), and their variabilities, may be useful indicators of physical and cognitive limitations.
Although there is little specific research on the changes in toe in/out angle over the course of a fatiguing event, prior research has studied how the toe in/out angle influences lower extremity dynamics. Toe in/out angle is a component of an individual’s gait pattern. According to a study performed by Khan et al., toe in/out angle affects knee joint kinetics as well as lower limb energetics, but the effect is inconsistent at different speeds.10 This change in the toe in/out angle because of different speeds suggests a question related to the effect of fatigue on the toe in/out angle.
Previous studies suggest that both fatigue and load carriage can cause changes in an individual’s physical performance.5–7 However, many of these studies have small sample sizes and were performed in laboratory environments with instrumented treadmills and motion capture systems over short periods of time.2,4,5,7,11,12 Although laboratory testing is useful for controlling variables and accurately measuring desired metrics, it does not provide an accurate depiction of the military operational setting of a loaded ruck march that may include varying pace because of peer socialization during group movements, necessary tactical security, and other external factors. Studying individual gait variability over the entirety of a loaded field ruck march may help researchers gain insight into certain parameters that are related to fatigue or future injuries. For example, other gait metrics such as a decrease in stride length can contribute to tibial stress fractures.7 Load carriage, in particular, has been linked to overuse injuries to the lower extremity and back, which can lead to pain and decreased performance.2,4,5,7,12
The purpose of this study was to determine the effect of fatigue on both the mean and the variability of these kinematic parameters over the course of a loaded ruck march. We hypothesized that stride length would decrease, stride width would increase, and stride length and width variability would remain constant over the course of the loaded foot march. Toe in/out angle was hypothesized to increase in variability and diverge from the baseline as fatigue increased over the course of the loaded ruck march. Additionally, it was hypothesized that AP and SS torso lean and their variabilities would increase as fatigue increased over the course of the loaded foot march.
METHODS
Participants
Three platoons consisting of 70 active duty U.S. Army infantrymen from the 82nd Airborne Division between the ages of 18 and 23 years were enrolled in the field study through which these data were collected. Written informed consent was obtained, and the U.S. Army Combat Capabilities Development Command Armaments Center Institutional Review Board and the Army Human Research Protections Office approved all procedures. Investigators adhered to the policies for protection of human subjects as prescribed in Army Regulation 70-25. All soldiers in the study were medically screened.
Protocol
A 7+-mile long ruck march on relatively flat terrain was conducted at the conclusion of a 72-h field training exercise. These were conducted in squads of six to nine soldiers. The soldiers were made aware of the approximate length of the march, but not the details of the path taken. All soldiers wore their Army Combat Uniform, Fighting Load Carrier, Soldier Plate Carrier System (without armored plates), rucksack, rifle, and any other individual specialized equipment for the loaded marches (average weight carried 86.9 lbs.). For the ruck march, the soldiers moved at their own pace as a squad and were allowed to take breaks at the discretion of their squad leader.
Experimental Setup
Inertial measurement units (IMUs) (APDM Opal; Portland, OR) capable of measuring 3D acceleration, angular velocity, and magnetic fields were used for data collection. Two sensors were mounted on each soldier’s equipment: one on the plate carrier in the approximate location of the sternum and the other on the outside of each boot at the level of the lateral ankle. This arrangement provided the desired motion data related to both the torso and lower limbs. Before the ruck march was initiated, the sensors were calibrated using principal component analysis from the following activities: quiet standing, toe touches, and forward walking. The sensors were synchronized before execution and sampled at 256 Hz.
Data Processing
All data processing was conducted in MATLAB (Mathworks; Natick, MA). Three-dimensional kinematic data were then transformed into the body’s coordinate system. A continuous wavelet transform of acceleration signals was used to identify the following significant gait events: heel strike, toe off, and moments where foot has zero velocity.13 A specialized dead-reckoning gait estimation algorithm based on the zero velocity update (ZUPT) technique and related signal processing algorithms were used to transform raw data to metrics of interests.14 The ZUPT algorithm was used to integrate acceleration signals between zero velocity points to obtain 3D foot trajectory for each step. Raw data were collected at 256 Hz, and torso metrics were computed by simply transforming the time series data into a body-centered reference frame. Stride metrics, on the other hand, were obtained by identifying critical gait events in the raw data and integrating it over those periods. This resulted in a series of processed data whose time deltas varied depending on the soldiers’ specific gait pattern.
Data Analysis
Custom MATLAB code was written to identify the 10 parameters of interest from the processed data: stride length, stride length variability, stride width, stride width variability, ankle yaw, ankle yaw variability, AP torso lean, AP torso lean variability, SS torso lean, and SS torso lean variability. Because of the placement of the sensor on the ankle, the metric collected was ankle yaw as a surrogate metric for toe in/out angle. Several processing steps were carried out in order to convert the time series data to comparable discrete metrics that characterize each parameter throughout the ruck march, for each participant. First, baseline averages of the relevant metrics were calculated from the first 5 min of walking and subtracted to isolate deviations, which were inferred to reflect latent fatigue. Next, visual inspection was used to identify and remove any obvious rest periods taken throughout the loaded foot march. Rest periods could be seen by visual inspection of accelerometer data, separately for the torso-mounted and ankle-mounted sensors, as the regular movement of a soldier significantly decreases during breaks (see Fig. 1). The start and end times of these rest periods were also recorded for descriptive analysis in postprocessing. The remaining data were divided into quarters, which represented gross divisions of time over which fatigue was assumed to increase.15 Finally, the means and standard deviations of each quarter for each parameter were taken for each participant.
MATLAB data processing method to remove rest periods by visual inspection.
Statistical Analysis
One-way repeated measure analyses of variance (ANOVAs) were conducted using Minitab (Minitab, LLC; State College, PA) to examine the effect of fatigue status on each parameter. Significant ANOVAs were then followed with planned pairwise comparisons between the first quarter and each of the three other quarters. Significance level was set at |$p \le 0.05$|.
RESULTS
Of the 10 kinematic parameters analyzed, two of the mean metrics and three of the variability metrics showed a significant change from the first quarter to the last quarter: mean stride length, mean stride width, stride width variability, AP torso variability, and SS torso variability (see Fig. 2). Twenty rest periods were taken collectively among the platoons during the loaded foot march.
Plots of kinematic parameters as means and variability for each quarter of the loaded foot march.
DISCUSSION
The purpose of this study was to determine the effect of fatigue on both the mean and the variability of the specified kinematic parameters over the course of a loaded ruck march. We hypothesized that stride length would decrease, stride width would increase, and stride length and width variability would remain constant over the course of the loaded foot march. Ankle yaw was hypothesized to increase in variability and diverge from the baseline value as fatigue increased over the course of the loaded ruck march. Additionally, it was hypothesized that AP and SS torso lean and their variabilities would increase as fatigue increased over the course of the loaded foot march. As shown in Table I, our hypotheses were supported for stride length, stride width, stride length variability, and AP and SS torso lean variability. Contrary to our hypothesis, a significant change in stride width variability was found. Lastly, the angle yaw showed no significant change.
Summary of Planned Pairwise Comparisons between Specified Quarters during Loaded Foot March for Significant Variables
| Quarter comparisons | |||
|---|---|---|---|
| Kinetic parameters | 1-2 | 1-3 | 1-4 |
| Stride length mean | 0.77 | 0.09 | 0.00a |
| Stride width mean | 0.00a | 0.00a | 0.00a |
| Stride width variability | 0.29 | 0.98 | 0.29 |
| Anterior-posterior Torso lean variability | 0.25 | 0.92 | 0.01a |
| Side-to-side Torso lean variability | 0.12 | 0.01a | 0.00a |
| Quarter comparisons | |||
|---|---|---|---|
| Kinetic parameters | 1-2 | 1-3 | 1-4 |
| Stride length mean | 0.77 | 0.09 | 0.00a |
| Stride width mean | 0.00a | 0.00a | 0.00a |
| Stride width variability | 0.29 | 0.98 | 0.29 |
| Anterior-posterior Torso lean variability | 0.25 | 0.92 | 0.01a |
| Side-to-side Torso lean variability | 0.12 | 0.01a | 0.00a |
Significantly different (P = 0.05) compared to the first quarter.
No significance (P > 0.05) compared to the first quarter.
Summary of Planned Pairwise Comparisons between Specified Quarters during Loaded Foot March for Significant Variables
| Quarter comparisons | |||
|---|---|---|---|
| Kinetic parameters | 1-2 | 1-3 | 1-4 |
| Stride length mean | 0.77 | 0.09 | 0.00a |
| Stride width mean | 0.00a | 0.00a | 0.00a |
| Stride width variability | 0.29 | 0.98 | 0.29 |
| Anterior-posterior Torso lean variability | 0.25 | 0.92 | 0.01a |
| Side-to-side Torso lean variability | 0.12 | 0.01a | 0.00a |
| Quarter comparisons | |||
|---|---|---|---|
| Kinetic parameters | 1-2 | 1-3 | 1-4 |
| Stride length mean | 0.77 | 0.09 | 0.00a |
| Stride width mean | 0.00a | 0.00a | 0.00a |
| Stride width variability | 0.29 | 0.98 | 0.29 |
| Anterior-posterior Torso lean variability | 0.25 | 0.92 | 0.01a |
| Side-to-side Torso lean variability | 0.12 | 0.01a | 0.00a |
Significantly different (P = 0.05) compared to the first quarter.
No significance (P > 0.05) compared to the first quarter.
Similar to the current study, previous research on gait changes because of body-borne loading has shown decreased step length over the course of loaded foot marches.4,5 This decrease may also be related to balance during walking and an increased variability of stride length,11 which may be indicative of dynamic balance and fatigued walking. The findings of the current study suggest that, as they fatigue, the soldiers may have been walking in a way to better prevent them from falling or stumbling during the march by decreasing their step length and minimizing their step length variability, while simultaneously increasing their stride width and stride width variability.
The results of the current study suggest that the variability of the torso movement patterns, and not the mean values, changes throughout the ruck march. Although it is surprising that the current study did not identify an increase in the AP or SS torso lean mean value with increased fatigue, this could be due to the torso sensor placement being on the outerwear of the soldier. Motion of the soldiers’ plate carriers relative to their bodies could alias true trends in torso lean and obscure significance. Similarly, ankle yaw was likely affected by nonideal placement of the ankle sensor and subject to a low signal-to-noise ratio. Although the magnitude of stride motion was sufficient to measure step length and width relatively accurately, the more subtle ankle angles may not have been detectable among the signal noise. Future studies should consider placing the sensor further out toward the toe to leverage the motion magnification and increase the signal-to-noise ratio. Such a change would it make it more difficult to estimate other potentially insightful metrics using IMU (e.g., ground reaction forces). An additional placement suggestion would be to place an IMU on the tibia and on the laces of the upper foot to compare movements and calculate the resulting ankle angle while rucking.
Collectively, these results imply that there are at least four kinematic responses to fatigue during a loaded ruck march. There is a decrease in stride length, an increase in stride width, an increase in stride width variability, and an increase in AP and SS torso lean variability. The significant difference between quarters one and four for these parameters suggests that they are affected by fatigue (Table I). Although this study did not seek to detect a direct relationship between kinematic changes and injury, the kinematic adjustments made by the soldiers over the course of the loaded ruck march could lead to an increase in injury susceptibility. As step length decreases over the course of the loaded ruck march, the number of times each foot contacts the ground will increase causing the lower extremities of the soldier to more regularly experience forces that can cause injury. This decrease in step length could cause an overuse injury on the soldier’s lower extremities as the soldier is contacting the ground more often. Previous research has found association between decreased step length and stress fractures of the tibia and metatarsals.7 On the contrary, the decrease in step length may result in a “soften” of the forces felt on the lower limb during the step and instead is being used as a protective mechanism for increasing load over time. These theories could be investigated further with analysis of step time and ground reaction forces.
The advantages of this study were the realistic environment and the relatively large sample size. The outdoor, overground course better represents an operational environment compared to a laboratory environment. However, this does limit the types of data that can be collected and makes controlling variables challenging. In this study, it is likely that the sensor placement on the soldiers’ outer wear, other extensive gear, and disruptions in the march for breaks negatively affected the quality of the data. Data processing provided some measure of offline quality control, even if those steps could only dictate which data to include. Additionally, limitations of this study include the lack of a control group and unintended “weighting” of rest periods. Having no control group makes it impossible to differentiate between effects of fatigue because of the ruck march and the effect of the added load weight. The difference in rest period length may have caused differences in soldier performance therefore effecting the data collected.
CONCLUSIONS
In conclusion, we observed changes in both stride and torso kinematics over the course of a loaded ruck march, which suggests that fatigue causes soldiers to alter aspects of their movement mechanics and may make them more susceptible to injury. This study found fatigue-related decreases in stride length, which could lead to overuse injuries because of an increase in the number of times each foot contacts the ground. Although noncombat injuries in the military have historically been a hidden epidemic, they are now being recognized as a major health problem for the military services. The purpose of this study was to determine the effect of fatigue on both the mean and the variability of the specified kinematic parameters over the course of a load ruck march. This study indicates that there are kinematic changes during loaded ruck marches, which have the potential to cause injuries. Although variability is often a sign of health movement, more research is needed to identify factors that are related to injury risk. This study can be used to better understand the relationship between fatigue and injury and potentially extend beyond the military setting for modeling efforts that seek to predict how the human body will respond to fatigue.
ACKNOWLEDGMENTS
None declared.
FUNDING
The study referred to in this manuscript was financially supported by the U.S. Army Combat Capabilities Development Command (DEVCOM) Soldier Center.
CLINICAL TRIAL REGISTRATION
Not applicable.
INSTITUTIONAL REVIEW BOARD (HUMAN SUBJECTS)
DEVCOM AC Institutional Review Board #18-003.
INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC)
Not applicable.
CONFLICT OF INTEREST STATEMENT
None declared.
INDIVIDUAL AUTHOR CONTRIBUTION STATEMENT
A.E.B. analyzed the data and drafted the original manuscript. J.J.S. and R.A.Z. collected and analyzed the data, assisted in drafting, and reviewed and edited the manuscript. G.M.F. collected data and reviewed and edited the manuscript. V.G.B. and S.E.-F. were principal investigators of the project and reviewed and edited the manuscript. All authors read and approved the final manuscript.
DATA AVAILABILITY
Please contact the authors of this paper with any questions about data availability.
INSTITUTIONAL CLEARANCE
Institutional clearance approved.
REFERENCES
Author notes
Results were presented at the North American Conference of Biomechanics, which was held in August 2022.
The views expressed are solely those of the authors and do not reflect the official policy or position of the U.S. Army, U.S. Navy, U.S. Air Force, the Department of Defense, or the U.S. Government.
