The Impact of Relative Load on Dynamic Postural Stability in Recreationally Active Adults: Implications for Tactical Readiness

Abstract

Background/Objectives: Musculoskeletal injuries (MSIs) continue to be a significant challenge in military populations. Load carriage is cited as a key contributor to postural stability (PS) impairments and therefore may contribute to injury risk. Therefore, the purpose of the present study was to examine the influence of load per kilogram of body mass (LpBM) on dynamic postural stability index (DPSI) percentage difference between unloaded and loaded conditions, while moderating for biological sex. Methods: Thirty-three recreationally active adults (16 males, 17 females) participated in a cross-sectional study. Each participant performed single-leg landing (SLL) tasks under unloaded and loaded conditions, and DPSI was calculated using ground reaction force data collected over the first three seconds post-landing. The loaded condition (22–23 kg, varies based on helmet and vest size) required individuals to wear a full combat load. A moderated multiple regression with robust standard errors was run to determine whether the relationship between percentage difference in DPSI between unloaded and loaded conditions and LpBM carried is different for female and male participants. Results: There was not a statistically significant moderator effect of the DPSI percentage difference, as evidenced by the addition of the interaction term explaining an additional 0.94% of the total variance, p < 0.643. Follow-up standard multiple regressions revealed that there was a statistically significant positive linear relationship (0.887 ± 0.320) between DPSI percentage difference and LpBM (p = 0.010). It was also observed that females did not have statistically significantly higher DPSI percentage difference than males (1.210 ± 4.392, p = 0.785). Conclusions: The results suggest that as LpBM increases, stability becomes more difficult to maintain. These findings highlight the importance of considering relative load when assessing injury risk and designing load carriage training protocols in tactical populations.

1. Introduction

Musculoskeletal injuries (MSIs) are a significant concern in military populations, accounting for 72% of incident injuries among U.S. active-duty service personnel in 2021 [1]. In 2022 alone, active-duty soldiers experienced more than two million medical encounters for injury, with 75% attributed to cumulative micro-traumatic MSIs [2]. Lower-extremity injuries comprised 40% of all mechanical injuries among all body regions [2]. Slips, trips, and falls are a notable subset, representing 20% of outpatient MSIs [2] and highlighting the importance of postural stability (PS) in MSI prevention [3]. Understanding PS and injury rates in military personnel is crucial for injury prevention and career longevity [4].

PS is defined as the ability to maintain control over one’s stance and balance in varying conditions [5,6]. This ability depends on an integrated feedback system involving vestibular, visual, and somatosensory input that enables neuromuscular responses to maintain upright posture during movement [6]. Military tasks often add complexity to this process due to load carriage, which is required in training and operational settings [4,7,8,9,10,11,12,13]. Typical loads donned include the Modular Lightweight Load-carrying Equipment System (MOLLE), Army Combat Helmet, and body armor [9]. Load carriage may vary considerably between soldiers [14] and is based on the march types (i.e., fighting, approach or emergency march), terrain, climate, and stress that a soldier can face [15]. Carrying these increased loads may elevate the risk of injury by placing greater strain on the lower extremities, altering gait, and affecting balance and stability [9,16,17,18].

To date, the available literature suggests that increasing external loads may diminish both dynamic [7,13] and static [7] PS. Even loads as low as 15% of an individual’s body weight (BW) can significantly compromise PS, especially during uphill walking, which may lead to MSIs [17]. Notably, greater load magnitudes typically result in worsened PS, and individual biomechanical responses can vary slightly [16]. The impact of load on PS is not limited to experienced soldiers; it affects all levels of military personnel. A study involving first-year and fourth-year Reserve Officers’ Training Corps (ROTC) military cadets explored the effects of carrying loads of 16 kg and 32 kg on gait variability [19]. In both ROTC groups within the study, heavier loads and experience levels had a significant impact on gait speed, stride time, and stride length. These findings suggest the importance of load carriage percentage as a critical factor in maintaining PS and injury prevention in military populations [19]. The impact of load per kilogram of body mass (LpBM) carried on PS is a critical area of investigation. A study involving special police officers examined the effect of carrying loads of 5 kg, 25 kg, and 45 kg on PS [18]. The results revealed that as the load increased, there were significant changes in the center of pressure (COP) path length, average velocity, and the lengths of the minor and major axes of the foot, which suggest a reduction in PS [18].

The influence of sex on load carriage is another factor to consider [4,8,9,10,11,12,20,21,22]. Physiological and anthropometric differences between males and females may affect MSI risk [4,10,12]. However, findings remain mixed: some studies report no sex-based differences in load carriage [9,21,22], while others have suggested that body armor fit and anthropometric mismatches can disproportionately affect female soldiers [8]. LaGoy et al. [11] tested dynamic PS by using a different type of single-leg landing (SLL) and stabilization task under unloaded and loaded conditions. The study found that dynamic postural stability index (DPSI) scores were significantly affected by load but not sex or the sex-by-load interaction, with DPSI increasing between 0%, 20%, and 30% BW conditions [11]. These findings suggest that increased load negatively affects DPSI, with similar performance observed in men and women [11]. Similarly, a study done by Thomas et al. [9] found no significant difference in Landing Error Scoring System (LESS) scores between males and females. Although the LESS metric was used, the findings still reflect comparable results to LaGoy et al. [11], further supporting the finding that there was no significant difference in dynamic PS between males and females [9].

Current literature has primarily examined absolute load carriage without adequately accounting for load relative to body mass, leading to inconsistent findings on PS outcomes and sex differences. Future studies that focus on LpBM can provide a more individualized understanding of how load impacts dynamic PS, clarify sex-related effects, and better inform injury prevention strategies in military populations. By addressing LpBM, research can bridge the gap between laboratory measures and operational relevance for service members. Therefore, the purpose of the present study was to examine the influence of LpBM on DPSI percentage difference between unloaded and loaded conditions, while moderating for biological sex. It was hypothesized that LpBM would be a significant predictor of DPSI percentage difference. It was also hypothesized that biological sex would not be a significant moderator for the relationship between LpBM and DPSI percentage difference.

2. Materials and Methods

This study recruited 34 recreationally active individuals. However, one participant did not complete all the testing procedures; thus, their data was not included in the final analysis. The final sample consisted of 33 participants, which included 16 males (age 21.8 ± 2.8 years, mass 88.1 ± 18.9 kg, height 179.3 ± 8.2 cm) and 17 females (age 22.4 ± 4.0 years, mass 68.4 ± 14.0 kg, height 168.3 ± 9.3 cm). A recreationally active individual was defined as being a person who regularly engaged in moderate activity, such as tennis, biking, jogging, or weightlifting, 2 or 3 times a week for at least 30 min [7,9,23]. The loaded condition required all participants to wear the same standardized combat load. Participants reported having no prior military experience. All participants were aged 18–35 years. Individuals were excluded from the study if they had experienced one or more of the following: (a) suffered a shoulder, back, or lower-extremity injury within the past six months or (b) surgery to the shoulder, back, hip, knee, or ankle within the past two years. Before beginning the study, all participants read and signed an approved university IRB consent document (protocol #19-28).

2.1. Protocol

Participants reported for one testing session wearing athletic gear and sneakers. Height and weight were measured on each participant. Height was measured using a portable stadiometer (Invicta Plastics Ltd., Leichester, UK). Weight was measured with a Tanita TBF 300A scale and body composition analyzer (Tanita Corporation of America, Inc., Arlington Heights, IL, USA). For the unloaded task, participants wore shorts, a t-shirt, and combat boots (Belleville 390 Hot Weather Boots, Belleville Boot Company, Belleville, IL, USA). The same boots were worn for both conditions. The loaded condition (22–23 kg, varies based on helmet and vest size) requires individuals to wear a full combat load of equipment and gear (EQG) [helmet (1.4 kg), rucksack (18.1 kg), and tactical vest (5.4 kg)] and standard-issue combat boots (1.6 kg). The load carried for each participant was calculated by finding the difference between their weight with EQG and without EQG to account for differences in load based on helmet and vest size. For safety reasons, all balance tests were first performed without EQG before any EQG conditions.

2.2. Dynamic Postural Stability

The subjects completed three trials of an SLL task with (loaded) and without EQG (unloaded). For the SLL task, participants dropped from a 30 cm box onto a 40 cm × 60 cm force plate (Bertec, Columbus, OH, USA), landing with their dominant leg without losing balance [7]. The dominant leg was defined as the leg one would kick a soccer ball with [7]. The 30 cm box was positioned at 10% of the subject’s height away from the target, as per previous research [7,9]. After landing, they maintained a quiet stance for 10 s. Participants were allowed to perform 1–2 practice trials for each condition to become familiar with the SLL task. Following the practice trials, the participants completed the test trials. Participants rested for 30 s between trials and 5 min between unloaded and loaded conditions. The participant initially performed three trials without EQG, and then the participant performed three more trials with EQG. Ground reaction forces were recorded at a frequency of 1500 Hz. A low-pass filter with a 20 Hz cutoff frequency was used to remove noise from the signal [11,13]. Initial ground contact was defined as the instant when the vertical ground reaction force exceeded 5% of BW. Average DPSI was calculated using the first three seconds of ground contact forces following initial contact. DPSI was calculated using the equation reported in Sell et al. [13]. A higher DPSI value represents worse dynamic PS.

2.3. Statistical Analysis

The DPSI percentage difference between unloaded and loaded conditions was calculated with the following equation:

$${\displaystyle \frac{\left|V_1-V_2\right|}{\left(\left(V_1+V_2\right)/2\right)}}\times 100,$$

(1)

where (V₁) is the DPSI without EQG and (V₂) is the DPSI with EQG [7]. LpBM was computed by dividing the weight of the EQG by the participant’s BW and multiplying by 100, thereby expressing load as a percentage of body mass. Percentage difference was chosen as the metric for the present study because the researchers were interested in calculating a difference between two numbers and not a change from one number to another (i.e., percentage change) [24,25]. A moderated multiple regression was run to determine whether the relationship between DPSI percentage difference and LpBM carried is moderated by biological sex. Linearity was established by visual inspection of a scatterplot, and there was evidence of multicollinearity, as evidenced by two tolerance values less than 0.1. To address this issue, the continuous independent variable, LpBM, was transformed into a mean-centered value. A re-analysis of the data using the mean-centered value of LpBM resulted in the removal of the multicollinearity, as evidenced by no tolerance values being less than 0.448. There was one unusual point (0.437) identified through visual inspection of calculated leverage values. A leverage value was considered to by high if it exceeded a value of 0.36, which was established using 3p/n (for small sample sizes), where p = number of parameters plus the intercept and n = number of observations [26]. There was homoscedasticity, as assessed by visual inspection of the studentized residuals plotted against the predicted values for female and male individuals. The studentized residuals were normally distributed, as assessed by the Shapiro–Wilk test (p = 0.634). Due to the high leverage value, the moderated multiple regression was conducted with and without the unusual point, and there was an appreciable difference in the results [26]. Based on these observations the decision was made to run a moderated multiple regression with robust standard errors. The alpha level was set a priori at 0.05. All statistical analyses were performed using IBM SPSS Statistics (Version 29.0.1.0). The moderated multiple regression with robust standard errors was conducted using the PROCESS v5.0 and Robust Linear Model (RLM) macros by Andrew F. Hayes within IBM SPSS Statistics (Version 29.0.1.0) [27].

3. Results

Means and standard deviations are summarized in

Table 1. DPSI results and demographics by sex.

	Males (n = 16)			Females (n = 17)
	Mean ± SD	Min	Max	Mean ± SD	Min	Max	p-Value
Age (yrs)	21.8 ± 2.8	18	30	22.4 ± 4.0	19	34	0.958
Mass (kg)	88.1 ± 18.9	57.2	126.8	68.4 ± 14.0	49.8	103.6	0.002
Height (cm)	179.3 ± 8.2	161.3	190.5	168.3 ± 9.3	149.9	181.0	0.001
LpBM (%)	28.6 ± 7.5	19.7	43.9	33.3 ± 7.2	20.8	51.9	0.070
DPSI w/o EQG	0.326 ± 0.047	0.201	0.404	0.313 ± 0.047	0.219	0.389	0.433
DPSI w/ EQG	0.486 ± 0.084	0.356	0.634	0.494 ± 0.079	0.377	0.635	0.790
DPSI PD (%)	38.9 ± 14.6	15.1	56.1	44.3 ± 14.6	24.2	72.6	0.290

LpBM = load per kg of body mass; DPSI = dynamic postural stability index; w/o = without; EQG = equipment and gear; w/ = with; PD = percent difference; Min = minimum; Max = maximum; SD = standard deviation.

. A moderated multiple regression with robust standard errors was run to assess the statistical significance of the interaction term between LpBM and biological sex. All data are represented as mean ± standard error. It was found that biological sex did not moderate the effect of LpBM on DPSI percentage difference, as evidenced by the addition of the interaction term explaining an additional 0.94% of the total variance, which was not statistically significant (p < 0.643). As such, the interaction term was dropped from the model and a follow-up standard multiple regression with main effects only was conducted. This new model statistically significantly predicted DPSI percentage difference, F(2, 30) = 3.84, p = 0.033, R² = 0.229, adjusted R² = 0.178. There was a statistically significant positive linear relationship (0.887 ± 0.320) between DPSI percentage difference and LpBM (p = 0.010). In addition, females did not have statistically significantly higher DPSI percentage difference than males (1.210 ± 4.392, p = 0.785). The resultant predictive equation was as follows: DPSI percentage difference = 40.89 + 0.8869 × mean-centered LpBM + 1.210 × women.

4. Discussion

The purpose of the present study was to examine the influence of LpBM on dynamic PS as indicated by DPSI percentage difference between unloaded and loaded conditions, while moderating for biological sex. The main findings revealed that biological sex is not a significant moderator between LpBM and DPSI percentage difference. In addition, there is a significant linear relationship between LpBM and DPSI percentage difference, in which for every unit increase in LpBM there is an increase of 0.887 in DPSI percentage difference.

The findings of the study did support the first hypothesis, which was that LpBM would be a significant predictor of DPSI percentage difference. These findings suggest that heavier relative loads negatively affect landing stability. The results of the present study are consistent with previous research ([7,11]), demonstrating a load-dependent reduction in stability in which for every unit increase in LpBM there is an increase of 0.887 in DPSI percentage difference. The results of LaGoy et al. [11] demonstrate a consistent decline in DPSI as load increased, reporting incremental increases in DPSI from 0.359 in the unloaded condition to 0.396 at 20% BW load and to 0.420 at 30% BW load. To put this into context, this would be a percentage difference of approximately 9.80% between the unloaded and 20% BW loaded conditions. Furthermore, this percentage difference is greater between the unloaded and 30% BW loaded conditions at approximately 15.66%. The findings of LaGoy et al. [11], although demonstrating lower mean DPSI percentage differences than the present study, reinforce the notion that relative load negatively influences PS. There are limited studies exploring the impact of LpBM directly on PS. Thus, to be able to make meaningful comparisons, the authors of the present study examined research with similar methods and used published data from selected studies to estimate the LpBM carried. This was done by taking the mass (kg) of the donned load and dividing it by the reported mean mass (kg) of the sample population in the selected studies. Using this approach, it was observed that in Sell et al. [13] a relative load of 15.5% was enough to negatively influence DPSI. The same relationship between LpBM and DPSI has been reported in fire and rescue. In a secondary analysis, Kollock et al. [7] reported that LpBM was significantly associated with DSPI values while donning firefighter-specific EQG. Collectively, this suggests that heavier relative loads play a role in decreasing PS.

The same trend has also been observed with other balance metrics such as COP path length and COP average velocity [13]. In a study by Kasović et al. [18], loads of 5.4 kg, 25.6 kg, and 44.7 kg (corresponding to 6%, 29%, and 50% of the LpBM carried) were found to significantly alter COP average velocity in special police officers [18]. The marked increases in COP velocity and path length with greater LpBM reinforce the key findings of the present study: increasing relative load significantly compromises PS. Strube et al. [28] found that carrying 16 kg and 20.5 kg (about 20.3% and 26% LpBM, respectively) significantly increased sway velocity by up to 52%. This finding demonstrates a load-dependent decline in static balance [28]. These findings reinforce the progressive impairment of PS as relative loads increase, as demonstrated in the present study. Similarly, Thomas et al. [9] reported that higher LpBM was significantly associated with poorer landing mechanics as indicated by LESS scores (p = 0.034). Their study reported average LpBM values of 28% for males and 34% for females, which are comparative to those in the present study [9]. While the Thomas et al. [9] study used LESS scores and the present study used DPSI as outcome metrics, both show evidence that increasing LpBM compromises movement quality.

Second, it was hypothesized that biological sex would not be a significant moderator of the relationship between LpBM and DPSI percentage change. This hypothesis was supported as biological sex did not moderate the relationship between LpBM and DPSI percentage difference (p < 0.643). This finding is in line with LaGoy et al. [11], who observed no sex effect on DPSI across varying load conditions. Taken together with the findings exploring the first and second hypotheses, it appears that the percentage of load carriage relative to BW may have a greater influence on injury than sex differences. According to Brager and Starratt [29], Army females are required to weigh no more than 65.8–68.5 kg depending on age, while males are required to weigh no more than 79.4–84.4 kg depending on age. This report highlights that female Army personnel typically have lower BW [29]. Consequently, even when load restrictions are set according to march type, uniform load assignments can be problematic. Lighter individuals may end up carrying loads that approach the maximum allowable percentage of their BW (i.e., 30%, 45%, or 46–70%, depending on the march type) [14]. This higher LpBM may be a potential cause of the higher injury incident rate in female military personnel. Schram et al. [4] reported that females in military settings are more susceptible to overuse injuries, particularly when exposed to high-load tasks. The findings of the present study provide additional insight into the injury disparity between male and female military personnel.

These results of the present study have practical implications for tactical populations but should be generalized to military personnel with appropriate caution as the participants were recreationally active adults with no prior military experience. Additionally, the model only predicted 17.8% of the variance in DPSI percentage difference between unloaded and loaded conditions; thus, there are other confounding variables that contribute to DPSI. For example, lower-extremity joint range of motion, muscular strength, neuromuscular eccentric control, and muscle activation patterns may interact to dissipate force when landing, therefore influencing DPSI [30,31]. Research has demonstrated notable sex-related differences in such factors, specifically regarding their utilization in landing tasks [32,33,34]. Moreover, because DPSI is sensitive to early landing behavior, acute adaptations in landing strategy due to participant task and load unfamiliarity may influence the magnitude of DPSI percentage difference between unloaded and loaded conditions [30]. However, a study of male firefighter cadets demonstrated comparable results to the present study [25]. It was found that a male firefighter with a body mass between 93 and 95 kg could expect a DPSI percentage difference of 39% between SLL tasks with and without familiar, standard firefighting EQG, representing 25.8% LpBM on average [25]. In the present study, recreationally active male participants had a mean body mass of 88.1 kg, mean DPSI percentage difference of 38.9%, and mean LpBM of 28.6%. Overall, LpBM should not be interpreted as the exclusive determinant of DPSI percentage difference, but one of many contributing factors. Future studies should conduct baseline assessments and collect data on such influencing factors when possible to produce more robust models that account for individual differences.

The present findings should be interpreted in light of several limitations. First, although the sample size (n = 33) is consistent with prior biomechanics research, moderated effects were sensitive to a single high leverage observation with an extreme but physiologically plausible LpBM. Future studies with larger samples and a wider distribution of relative loads are needed to more precisely characterize the nature and boundaries of this effect. Second, the cross-sectional design limits the ability to establish causal relationships between LpBM and DPSI. This limitation was echoed in earlier studies examining load carriage and PS [18,28]. Third, DPSI, while a validated and widely used metric, represents a composite index and may not capture directional instability. Future research should include directional stability metrics for a more comprehensive evaluation. Fourth, lower-extremity muscular strength data was not collected for the participants and therefore was not included in analyses. It has been demonstrated that muscular strength, especially of the lower extremities, is associated with greater PS in recreationally active adults [35]. Variability in lower-extremity muscular strength between participants could affect DPSI percentage difference along with LpBM. Fifth, comparisons with the previous literature required estimating LpBM from published values. This introduces potential inaccuracies in interpreting external findings. Sixth, order of conditions for the SLL task was not randomized, allowing for potential order effects. The SLL was first performed without EQG for safety purposes as the participants were recreationally active adults without combat load carriage experience. Lastly, the use of recreationally active individuals rather than trained military personnel limits the generalizability of results to tactical populations. Differences in conditions, load familiarity, and movement strategies could alter performance under load [9,12,18].

5. Conclusions

The goal of the present study was to examine the influence of LpBM on dynamic PS and to assess sex differences in DPSI percentage difference. These findings suggest that heavier relative loads negatively affect PS during an SLL task. The findings of the present study further suggest that the percentage of load carriage relative to BW may have a greater influence on injury than sex differences.