Biomechanics of graded running: Part I - Stride parameters, external forces, muscle activations

Effects of Foot-Strike Pattern on Neuromuscular Function During a Prolonged Graded Run

PURPOSE

To study whether, during typical-level running, non-rear-foot strikers (non-RFS) or rear-foot strikers (RFS) presented a similar or different extent of neuromuscular fatigue after a prolonged graded run.

METHODS

Sixteen experienced male trail runners (8 non-RFS and 8 RFS) performed a 2.5-hour treadmill graded running exercise. Before and after exercise, neuromuscular tests were performed to assess neuromuscular fatigue of the plantar flexors. Biomechanical gait parameters were acquired with an instrumented treadmill, and electromyographic activity of the lower-limb muscles was collected as an index of muscle activation.

RESULTS

There were no significant time × foot strike interactions for neuromuscular (all P ≥ .742), muscle activation (all P ≥ .157), or biomechanical (all P ≥ .096) variables.

CONCLUSIONS

A dominant level running foot-strike pattern did not directly affect the extent of neuromuscular fatigue during a prolonged graded run. This suggests that no ideal running foot-strike pattern exists to minimize neuromuscular fatigue during prolonged-duration races wherein cumulative uphill and downhill segments are high, such as in trail running.

Peak Oxygen Uptake is Slope Dependent: Insights from Ground Reaction Forces and Muscle Oxygenation in Trained Male Runners

BACKGROUND

The aim of this study is to explore the effect of treadmill slope on ground reaction forces and local muscle oxygenation as putative limiting factors of peak oxygen uptake in graded maximal incremental running tests. Thirteen trained male runners completed five maximal incremental running tests on treadmill at - 15%, - 7.5%, 0%, 7.5% and 15% slopes while cardiorespiratory and local muscle oxygenation responses as well as ground reaction forces were continuously recorded. Blood lactate concentration and isometric knee extensor torque were measured before and after each test.

RESULTS

Peak oxygen uptake was lower at - 15% slope compared to all other conditions (from - 10 to - 17% lower, p < 0.001), with no difference between - 7.5 and + 15% slope. Maximal heart rate and ventilation values were reached in all conditions. The negative external mechanical work increased from steep uphill to steep downhill slopes (from 6 to 92% of total external work) but was not correlated with the peak oxygen uptake reduction. Local muscle oxygenation remained higher in - 15% slope compared to level running (p = 0.003).

CONCLUSIONS

Similar peak oxygen uptake can be reached in downhill running up to - 7.5% slope. At more severe downhill slopes (i.e., - 15%), greater negative muscle work and limited local muscle deoxygenation occurred, even in subjects familiarized to downhill running, presumably preventing the achievement of similar to other condition's peak oxygen uptake.

KEY POINTS

Trained male runners can reach like level running V̇O2peak at moderate but not at severe negative slope. Negative external mechanical work increases with increasing negative slope. At maximal intensity Vastus Lateralis muscle oxygenation is higher in steep negative slope. Knee extensor isometric muscle torque is preserved after maximal level and uphill running, but reduced after downhill running, despite lower blood lactate. Progressive reduction of V̇O2 at maximal effort with increasing negative slope might be related to the metabolic consequences of increased lower limb negative external work (i.e., eccentric muscle actions).

Changes in muscle activation with graded surfaces during canter in Thoroughbred horses on a treadmill

Understanding how muscle activity changes with different surface grades during canter is essential for developing training protocols in Thoroughbreds because canter is their primary gait in training and races. We measured the spatiotemporal parameters and the activation of 12 surface muscles in the leading limb side of 7 Thoroughbreds. Horses were equipped with hoof strain gauges and cantered at 10 m/s on a treadmill set to grades of -4%, 0%, 4%, and 8%, randomly, for 30 seconds each without a lead change. Integrated electromyography (iEMG) values during stance and swing phases were calculated and normalized to mean iEMG values during stride duration at 0% grade in each muscle. The iEMG values at each grade were compared using a generalized mixed model. Stride duration significantly decreased due to shorter swing duration on an 8% grade (P < 0.001) compared to all other grades, where no significant changes were observed. Compared to a 0% grade, the normalized iEMG values during the stance phase on an 8% grade in five muscles significantly increased (Musculus infraspinatus; +9%, M. longissimus dorsi (LD); +4%, M. gluteus medius (GM); +29%, M. biceps femoris; +47%, M. flexor digitorum lateralis; +16%). During the swing phase, the normalized iEMG values in six muscles significantly increased on an 8% grade compared to a 0% grade (M. splenius; +21%, M. triceps brachii; +54%, LD; +37%, GM; +24%, M. semitendinosus; +51%, M. extensor digitorum longus; +10%). No significant changes were observed in iEMG values on -4% and 4% grades compared to the 0% grade. Although +/- 4% grades had little effect on neuromuscular responses, 8% uphill canter reduced stride duration due to decreased swing duration and required increase of muscle activation during either stance and swing phase. Canter on an 8% grade might strengthen equine muscles to increase propulsive force and stride frequency.

Downhill running increases markers of muscle damage and impairs the maximal voluntary force production as well as the late phase of the rate of voluntary force development

PURPOSE

To examined the time-course of the early and late phase of the rate of voluntary force development (RVFD) and muscle damage markers after downhill running.

METHODS

Ten recreational runners performed a 30-min downhill run at 10 km h-1 and -20% (-11.3°) on a motorized treadmill. At baseline and each day up to 4 days RVFD, knee extensors maximum voluntary isometric force (MVIC), serum creatine kinase (CK) concentration, quadriceps swelling, and soreness were assessed. The early (0-50 ms) and late (100-200 ms) phase of the RVFD, as well as the force developed at 50 and 200 ms, were also determined.

RESULTS

MVIC showed moderate decrements (p < 0.05) and recovered after 4 days (p > 0.05). Force at 50 ms and the early phase were not impaired (p > 0.05). Conversely, force at 200 ms and the late phase showed moderate decrements (p < 0.05) and recovered after 3 and 4 days, respectively (p > 0.05). CK concentration, quadriceps swelling, and soreness increased (p < 0.05) were overall fully resolved after 4 days (p > 0.05).

CONCLUSION

Downhill running affected the knee extensors RVFD late but not early phase. The RVFD late phase may be used as an additional marker of muscle damage in trail running.

Tibial strains are sensitive to speed perturbations, but not grade perturbations, during running

A fatigue-failure process is hypothesized to govern the development of tibial stress fractures, where bone damage is highly dependent on the peak strain magnitude. To date, much of the work examining tibial strain during running has ignored uphill and downhill running despite the prevalence of this terrain. This study examined the sensitivity of tibial strain to changes in running grade and speed using a combined musculoskeletal-finite element modelling routine. Seventeen participants ran on a treadmill at ±10, ±5 and 0 deg; at each grade, participants ran at 3.33 m s-1 and at a grade-adjusted speed of 2.50 and 4.17 m s-1 for uphill and downhill grades, respectively. Force and motion data were recorded in each grade and speed combination. Muscle and joint contact forces were estimated using inverse-dynamics-based static optimization. These forces were applied to a participant-adjusted finite element model of the tibia. None of the strain variables (50th and 95th percentile strain and strained volume ≥4000 με) differed as a function of running grade; however, all strain variables were sensitive to running speed (F1≥9.59, P≤0.03). In particular, a 1 m s-1 increase in speed resulted in a 9% (∼260 με) and 155% (∼600 mm3) increase in peak strain and strained volume, respectively. Overall, these findings suggest that faster running speeds, but not changes in running grade, may be more deleterious to the tibia.

Relationship between biomechanics and energy cost in graded treadmill running

The objective of this study was to determine whether the relationships between energy cost of running (Cr) and running mechanics during downhill (DR), level (LR) and uphill (UR) running could be related to fitness level. Nineteen athletes performed four experimental tests on an instrumented treadmill: one maximal incremental test in LR, and three randomized running bouts at constant speed (10 km h^-1) in LR, UR and DR (± 10% slope). Gas exchange, heart rate and ground reaction forces were collected during steady-state. Subjects were split into two groups using the median Cr for all participants. Contact time, duty factor, and positive external work correlated with Cr during UR (all, p < 0.05), while none of the mechanical variables correlated with Cr during LR and DR. Mechanical differences between the two groups were observed in UR only: contact time and step length were higher in the economical than in the non-economical group (both p < 0.031). This study shows that longer stance duration during UR contributes to lower energy expenditure and Cr (i.e., running economy improvement), which opens the way to optimize specific running training programs.

Correspondence Between Values of Vertical Loading Rate and Oxygen Consumption During Inclined Running

Purpose

The aim of this study was to provide a theoretical model to predict the vertical loading rate (VLR) at different slopes and speeds during incline running.

Methods

Twenty-nine healthy subjects running at least once a week performed in a randomized order 4-min running trials on an instrumented treadmill at various speeds (8, 10, 12, and 14 km h−1) and slopes (− 20%, − 10%, − 5%, 0%, + 5%, + 10%, + 15%, + 20%). Heart rate, gas exchanges and ground reaction forces were recorded. The VLR was then calculated as the slope of the vertical force between 20 and 80% of the duration from initial foot contact to the impact peak.

Results

There was no difference in VLR between the four different uphill conditions at given running speeds, but it was reduced by 27% at 5% slope and by 54% at 10% slope for the same metabolic demand (similar $${\dot{\text{V}}\text{O}}_{{2}}$$ V ˙ O 2 ), when compared to level running. The average VLR measured at maximal aerobic intensity during level running would be decreased by 52.7% at + 5%, by 63.0% at + 10%, and by 73.3% at + 15% slope. Moreover, VLR was dependent on the slope in downhill conditions.

Conclusion

This study highlights the possibility to use uphill running to minimize rate of mechanical load (i.e., osteoarticular load) from foot impact on the ground and as a time-efficient exercise routine (i.e., same energy expenditure than in level running in less time).

Downhill running affects the late but not the early phase of the rate of force development

Purpose

This study aimed to evaluate the acute changes in the knee extensors maximum voluntary isometric contraction force (MVIC), rate of force development (RFD), and rate of EMG rise (RER) following a bout of downhill running.

Methods

MVIC and RFD at 0–50, 50–100, 100–200, and 0–200 ms were determined in thirteen men (22 ± 2 yr) before and after 30 min of downhill running (speed: 10 km h⁻¹; slope: − 20%). Vastus lateralis maximum EMG (EMG_max) and RER at 0–30, 0–50, and 0–75 ms were also recorded.

Results

MVIC, RFD_0–200, and EMG_max decreased by ~ 25% [Cohen’s d = − 1.09 (95% confidence interval: − 1.88/− 0.24)], ~ 15% [d = − 0.50 (− 1.26/0.30)], and ~ 22% [d = − 0.37 (− 1.13/0.42)] (all P < 0.05), respectively. RFD_100–200 was also reduced [− 25%; d = − 0.70 (− 1.47/0.11); P < 0.001]. No change was observed at 0–50 ms and 50–100 ms (P ≥ 0.05). RER values were similar at each time interval (all P > 0.05).

Conclusion

Downhill running impairs the muscle capacity to produce maximum force and the overall ability to rapidly develop force. No change was observed for the early phase of the RFD and the absolute RER, suggesting no alterations in the neural mechanisms underlying RFD. RFD_100–200 reduction suggests that impairments in the rapid force-generating capacity are located within the skeletal muscle, likely due to a reduction in muscle–tendon stiffness and/or impairments in the muscle contractile apparatus. These findings may help explain evidence of neuromuscular alterations in trail runners and following prolonged duration races wherein cumulative eccentric loading is high.

Neuromuscular, biomechanical, and energetic adjustments following repeated bouts of downhill running

PURPOSE

This study used downhill running as a model to investigate the repeated bout effect (RBE) on neuromuscular performance, running biomechanics, and metabolic cost of running.

METHODS

Ten healthy recreational male runners performed two 30-min bouts of downhill running (DR1 and DR2) at a -20% slope and 2.8 m/s 3 weeks apart. Neuromuscular fatigue, level running biomechanics during slow and fast running, and running economy parameters were recorded immediately before and after the downhill bouts, and at 24 h, 48 h, 72 h, 96 h, and 168 h thereafter (i.e., follow-up days).

RESULTS

An RBE was confirmed by attenuated muscle soreness and serum creatine kinase rise after DR2 compared to DR1. An RBE was also observed in maximum voluntary contraction (MVC) force loss and voluntary activation where DR2 resulted in attenuated MVC force loss and voluntary activation immediately after the run and during follow-up days. The downhill running protocol significantly influenced level running biomechanics; an RBE was observed in which center of mass excursion and, therefore, lower-extremity compliance were greater during follow-up days after DR1 compared to DR2. The observed changes in level running biomechanics did not influence the energy cost of running.

CONCLUSION

This study demonstrated evidence of adaptation in neural drive as well as biomechanical changes with the RBE after DR. The higher neural drive resulted in attenuated MVC force loss after the second bout. It can be concluded that the RBE after downhill running manifests as changes to global and central fatigue parameters and running biomechanics without substantially altering the energy cost of running.

Study of the kinetics of the determinants of performance during a mountain ultra marathon: Multidisciplinary protocol of the first Trail Scientifique de Clécy 2021 (Preprint)

Background

The growing interest of the scientific community in trail running has highlighted the acute effects of practice at the time of these races on isolated aspects of physiological and structural systems; biological, physiological, cognitive, and muscular functions; and the psychological state of athletes. However, no integrative study has been conducted under these conditions with so many participants and monitoring of pre-, per-, and postrace variables for up to 10 days over a distance close to 100 miles.

Objective

The aim of this study was to evaluate the kinetics of the performance parameters during a 156 km trail run and 6000 m of elevation gain in pre-, per-, and postrace conditions. The general hypothesis is based on significant alterations in the psychological, physiological, mechanical, biological, and cognitive parameters.

Methods

The Trail Scientifique de Clécy took place on November 11, 2021. This prospective experimental study provides a comprehensive exploration of the constraints and adaptations of psychophysiological and sociological variables assessed in real race conditions during a trail running of 156 km on hilly ground and 6000 m of elevation gain (D+). The study protocol allowed for repeatability of study measurements under the same experimental conditions during the race, with the race being divided into 6 identical loops of 26 km and 1000 m D+. Measurements were conducted the day before and the morning of the race, at the end of each lap, after a pit stop, and up to 10 days after the race. A total of 55 participants were included, 43 (78%) men and 12 (22%) women, who were experienced in ultra–trail-running events and with no contraindications to the practice of this sport.

Results

The launch of the study was authorized on October 26, 2021, under the trial number 21-0166 after a favorable opinion from the Comité de Protection des Personnes Ouest III (21.09.61/SIRIPH 2G 21.01586.000009). Of the 55 runners enrolled, 41 (75%) completed the race and 14 (25%) dropped out for various reasons, including gastric problems, hypothermia, fatigue, and musculoskeletal injuries. All the measurements for each team were completed in full. The race times (ie, excluding the measurements) ranged from 17.8206 hours for the first runner to 35.9225 hours for the last runner. The average time to complete all measurements for each lap was 64 (SD 3) minutes.

Conclusions

The Trail Scientifique de Clécy, by its protocol, allowed for a multidisciplinary approach to the discipline. This approach will allow for the explanation of the studied parameters in relation to each other and observation of the systems of dependence and independence. The initial results are expected in June 2022.

International Registered Report Identifier (IRRID)

RR1-10.2196/38027

Predicting continuous ground reaction forces from accelerometers during uphill and downhill running: a recurrent neural network solution

BACKGROUND

Ground reaction forces (GRFs) are important for understanding human movement, but their measurement is generally limited to a laboratory environment. Previous studies have used neural networks to predict GRF waveforms during running from wearable device data, but these predictions are limited to the stance phase of level-ground running. A method of predicting the normal (perpendicular to running surface) GRF waveform using wearable devices across a range of running speeds and slopes could allow researchers and clinicians to predict kinetic and kinematic variables outside the laboratory environment.

PURPOSE

We sought to develop a recurrent neural network capable of predicting continuous normal (perpendicular to surface) GRFs across a range of running speeds and slopes from accelerometer data.

METHODS

Nineteen subjects ran on a force-measuring treadmill at five slopes (0°, ±5°, ±10°) and three speeds (2.5, 3.33, 4.17 m/s) per slope with sacral- and shoe-mounted accelerometers. We then trained a recurrent neural network to predict normal GRF waveforms frame-by-frame. The predicted versus measured GRF waveforms had an average ± SD RMSE of 0.16 ± 0.04 BW and relative RMSE of 6.4 ± 1.5% across all conditions and subjects.

RESULTS

The recurrent neural network predicted continuous normal GRF waveforms across a range of running speeds and slopes with greater accuracy than neural networks implemented in previous studies. This approach may facilitate predictions of biomechanical variables outside the laboratory in near real-time and improves the accuracy of quantifying and monitoring external forces experienced by the body when running.

Internal Tibial Forces and Moments During Graded Running

The stress experienced by the tibia has contributions from the forces and moments acting on the tibia. We sought to quantify the influence of running grade on internal tibial forces and moments. Seventeen participants ran at 3.33 m/s on an instrumented treadmill at 0 deg, ±5 deg, and ±10 deg while motion data were captured. Ankle joint contact force was estimated from an anthropometrically-scaled musculoskeletal model using inverse dynamics-based static optimization. Internal tibial forces and moments were quantified at the distal 1/3rd of the tibia, by ensuring static equilibrium with all applied forces and moments. Downhill running conditions resulted in lower peak internal axial force (range of mean differences: -9% to -16%, p < 0.001), lower peak internal anteroposterior force (-14% to -21%, p < 0.001), and lower peak internal mediolateral force (-14% to -15%, p < 0.001), compared to 0 deg and +5 deg. Furthermore, downhill conditions resulted in lower peak internal mediolateral moment (-11%to -21%, p < 0.001), lower peak internal anteroposterior moment (-13% to -14%, p < 0.001), and lower peak internal torsional moment (-9% to -21%, p < 0.001), compared to 0 deg, +5 deg, and +10 deg. The +10 deg condition resulted in lower peak internal axial force (-7% to -9%, p < 0.001) and lower peak internal mediolateral force (-9%, p = 0.004), compared to 0 deg and +5 deg. These findings suggest that downhill running may be associated with lower tibial stresses than either level or uphill running.

Sprinting kinematics and inter-limb coordination patterns at varying slope inclinations

Uphill training is applied to induce specific overload on the musculoskeletal system to improve sprinting mechanics. This study aimed to identify unique kinematic features of uphill sprinting at different slopes and to suggest practical implications based on comparisons we early stance phase. At take-off, steeper slopes induced significantly more extended joint angles and higher ROMs during the late stance phase. Compared with moderate slopes, more anti-phase coordination patterns were detected at steeper slopes. Thus, uphill sprinting at steeper slopes shares essential kinematic features with the early acceleration phase of level sprinting. Moderate inclinations induce biomechanical adaptations similar to those in the late acceleration phase of level sprinting. Hence, the specific transfer of uphill sprinting to acceleration depends on the slope inclinations.

Effect of race distance on performance fatigability in male trail and ultra-trail runners

The etiology of changes in lower-limb neuromuscular function, especially to the central nervous system, may be affected by exercise duration. Direct evidence is lacking as few studies have directly compared different race distances. This study aimed to investigate the etiology of deficits in neuromuscular function following short versus long trail-running races. Thirty-two male trail runners completed one of five trail-running races as LONG (>100 km) or SHORT (<60 km). Pre- and post-race, maximal voluntary contraction (MVC) torque and evoked responses to electrical nerve stimulation during MVCs and at rest were used to assess voluntary activation and muscle contractile properties of knee-extensor (KE) and plantar-flexor (PF) muscles. Transcranial magnetic stimulation (TMS) was used to assess evoked responses and corticospinal excitability in maximal and submaximal KE contractions. Race distance correlated with KE MVC (ρ = -0.556) and twitch (ρ = -0.521) torque decreases (p ≤ .003). KE twitch torque decreased more in LONG (-28 ± 14%) than SHORT (-14 ± 10%, p = .005); however, KE MVC time × distance interaction was not significant (p = .073). No differences between LONG and SHORT for PF MVC or twitch torque were observed. Maximal voluntary activation decreased similarly in LONG and SHORT in both muscle groups (p ≥ .637). TMS-elicited silent period decreased in LONG (p = .021) but not SHORT (p = .912). Greater muscle contractile property impairment in longer races, not central perturbations, contributed to the correlation between KE MVC loss and race distance. Conversely, PF fatigability was unaffected by race distance.

Level, Uphill, and Downhill Running Economy Values Are Correlated Except on Steep Slopes

The aim of this study was first to determine if level, uphill, and downhill energy cost of running (ECR) values were correlated at different slopes and for different running speeds, and second, to determine the influence of lower limb strength on ECR. Twenty-nine healthy subjects completed a randomized series of 4-min running bouts on an instrumented treadmill to determine their cardiorespiratory and mechanical (i.e., ground reaction forces) responses at different constant speeds (8, 10, 12, and 14 km·h⁻¹) and different slopes (−20, −10, −5, 0, +5, +10, +15, and +20%). The subjects also performed a knee extensor (KE) strength assessment. Oxygen and energy costs of running values were correlated between all slopes by pooling all running speeds (all r² ≥ 0.27; p ≤ 0.021), except between the steepest uphill vs. level and the steepest downhill slope (i.e., +20% vs. 0% and −20% slopes; both p ≥ 0.214). When pooled across all running speeds, the ECR was inversely correlated with KE isometric maximal torque for the level and downhill running conditions (all r² ≥ 0.24; p ≤ 0.049) except for the steepest downhill slope (−20%), but not for any uphill slopes. The optimal downhill grade (i.e., lowest oxygen cost) varied between running speeds and ranged from −14% and −20% (all p < 0.001). The present results suggest that compared to level and shallow slopes, on steep slopes ~±20%, running energetics are determined by different factors (i.e., reduced bouncing mechanism, greater muscle strength for negative slopes, and cardiopulmonary fitness for positive slopes). On shallow negative slopes and during level running, ECR is related to KE strength.

Runners Adapt Different Lower-Limb Movement Patterns With Respect to Different Speeds and Downhill Slopes

The aim of this study was to investigate the influence of slope and speed on lower-limb kinematics and energy cost of running. Six well-trained runners (VO_2max 72 ± 6 mL·kg⁻¹·min⁻¹) were recruited for the study and performed (1) VO_2max and energy cost tests and (2) an experimental running protocol at two speeds, 12 km·h⁻¹ and a speed corresponding to 80% of VO_2max (V80, 15.8 ± 1.3 km·h⁻¹) on three different slopes (0°, −5°, and −10°), totaling six 5-min workload conditions. The workload conditions were randomly ordered and performed continuously. The tests lasted 30 min in total. All testing was performed on a large treadmill (3 × 5 m) that offered control over both speed and slope. Three-dimensional kinematic data of the right lower limb were captured during the experimental running protocol using eight infrared cameras with a sampling frequency of 150 Hz. Running kinematics were calculated using a lower body model and inverse kinematics approach. The generic model contained three, one, and two degrees of freedom at the hip, knee, and ankle joints, respectively. Oxygen uptake was measured throughout the experimental protocol. Maximum hip extension and flexion during the stance phase increased due to higher speed (p < 0.01 and p < 0.01, respectively). Knee extension at the touchdown and maximal knee flexion in the stance phase both increased on steeper downhill slopes (both p < 0.05). Ground contact time (GCT) decreased as the speed increased (p < 0.01) but was unaffected by slope (p = 0.73). Runners modified their hip movement pattern in the sagittal plane in response to changes in speed, whereas they altered their knee movement pattern during the touchdown and stance phases in response to changes in slope. While energy cost of running was unaffected by speed alone (p = 0.379), a shift in energy cost was observed for different speeds as the downhill gradient increased (p < 0.001). Energy cost was lower at V80 than 12 km·h⁻¹ on a −5° slope but worse on a −10° slope. This indicates that higher speeds are more efficient on moderate downhill slopes (−5°), while lower speeds are more efficient on steeper downhill slopes (−10°).

Sacral acceleration can predict whole-body kinetics and stride kinematics across running speeds

Background

Stress fractures are injuries caused by repetitive loading during activities such as running. The application of advanced analytical methods such as machine learning to data from multiple wearable sensors has allowed for predictions of biomechanical variables associated with running-related injuries like stress fractures. However, it is unclear if data from a single wearable sensor can accurately estimate variables that characterize external loading during running such as peak vertical ground reaction force (vGRF), vertical impulse, and ground contact time. Predicting these biomechanical variables with a single wearable sensor could allow researchers, clinicians, and coaches to longitudinally monitor biomechanical running-related injury risk factors without expensive force-measuring equipment.

Purpose

We quantified the accuracy of applying quantile regression forest (QRF) and linear regression (LR) models to sacral-mounted accelerometer data to predict peak vGRF, vertical impulse, and ground contact time across a range of running speeds.

Methods

Thirty-seven collegiate cross country runners (24 females, 13 males) ran on a force-measuring treadmill at 3.8-5.4 m/s while wearing an accelerometer clipped posteriorly to the waistband of their running shorts. We cross-validated QRF and LR models by training them on acceleration data, running speed, step frequency, and body mass as predictor variables. Trained models were then used to predict peak vGRF, vertical impulse, and contact time. We compared predicted values to those calculated from a force-measuring treadmill on a subset of data (n = 9) withheld during model training. We quantified prediction accuracy by calculating the root mean square error (RMSE) and mean absolute percentage error (MAPE).

Results

The QRF model predicted peak vGRF with a RMSE of 0.150 body weights (BW) and MAPE of 4.27  ±  2.85%, predicted vertical impulse with a RMSE of 0.004 BW*s and MAPE of 0.80  ±  0.91%, and predicted contact time with a RMSE of 0.011 s and MAPE of 4.68  ±  3.00%. The LR model predicted peak vGRF with a RMSE of 0.139 BW and MAPE of 4.04  ±  2.57%, predicted vertical impulse with a RMSE of 0.002 BW*s and MAPE of 0.50  ±  0.42%, and predicted contact time with a RMSE of 0.008 s and MAPE of 3.50  ±  2.27%. There were no statistically significant differences between QRF and LR model prediction MAPE for peak vGRF (p = 0.549) or vertical impulse (p = 0.073), but the LR model's MAPE for contact time was significantly lower than the QRF model's MAPE (p = 0.0497).

Conclusions

Our findings indicate that the QRF and LR models can accurately predict peak vGRF, vertical impulse, and contact time (MAPE < 5%) from a single sacral-mounted accelerometer across a range of running speeds. These findings may be beneficial for researchers, clinicians, or coaches seeking to monitor running-related injury risk factors without force-measuring equipment.