Profiling injuries sustained following implementation of a progressive load carriage program in United States marine corps recruit training

2.1 Participants

Retrospective musculoskeletal injury data from the 2nd and 3rd Battalions at Marine Corps Recruit Depot San Diego (MCRDSD) were provided. The data were queried from the sports medicine injury database used by the MCRDSD. Injury data from companies completing training in 2020 (OLC) and 2021 (MLC) were compared. The data for both groups (i.e., OLC and MLC) were captured during the same period thus accounting for seasonal differences and variations. All OLC data represented only male recruits and, as such, were only compared against male recruits in the MLC group. Prior to this current year, no females had gone through training at MCRDSD. This study was approved by the Institutional Review Board at Naval Health Research Center, and adhered to Department of the Navy human research protection policies (Protocol NHRC.2020.0008). All participants that engaged in the MLC provided written informed consent. Data from OLC was retrospective and only demographic and injury data queried from the sports medicine injury database.

2.2 Musculoskeletal injury data

Data were stratified by bodily site (e.g., knee, shoulder, foot, etc.), nature of injury (e.g., stress fracture, fractures, sprains, strains, etc.), events during which injuries occurred (e.g., running, conditioning hike, etc.), general causal type (e.g., traumatic, new overuse, or a pre-existing overuse injury), and injury severity (e.g., mild, moderate, or severe). Definitions for casual types and severity were input from sports medicine clinical providers and based upon ICD-10 codes [18].

2.3 Progressive load carriage training intervention

All load carriage hikes occurred at MCRDSD and were on a mix of asphalt and hard-pack dirt with zero percent grade. Recruits hiked in their combat fatigues and boots. Backpacks were packed by recruits under drill instructor supervision and weighed the night prior. Pack loads consisted of standard-issued items that Marines are required to carry and were packed according to USMC standards. The load carriage hikes began between 0730– 0800 and followed a 3.0 mile per hour (4.83 kilometers per hour) pace. Recruits hiked with their weapon (3.6 kg) which was not included in the weight of the backpack load. The OLC program consisted of four conditioning hikes as outlined in Table 1. The hikes began on training day 37 and concluded on training day 48. The MLC included the hikes from OLC; however, six hikes added to the training schedule leading to a load carriage frequency of approximately 1 load carriage session per week. One general physical training session was substituted for a load carriage hike that progressively increased over a six-week period (Table 1). MLC started on training day 4 and concluded on training day 48. Thus participants in MOD engaged in a total of 10 conditioning hikes as compared to four in the OLC.

A population estimate of the OLC: MLC relative risk ratio (RR) was calculated using the following formula [19]: RR = ( OLC injury incident rate ) / ( MLC injury incident rate )

The ninety-five percent confidence interval (95% CI) around the population estimate of each RR was calculated as follows: 95 % CI = exp ( ln - 1.96 × SE ( ln [ RR ] ) ) to exp ( ln [ RR ] + 1.96 × SE ( ln [ RR ] ) ) where SE ( ln [ RR ] ) = ( 1 / [ incident rate OLC ] + 1 / [incidentrate_MLC] -1/n_OLC - 1/n_MLC)   [19].

Further descriptive analyses were conducted to examine injury patterns evident in the data, based on the frequency of incidents that were associated with body parts, nature of injury, events during which injuries occurred, general causal types, and injury severity. Proportions of injuries and the number of injuries drawn from the total number of reported injuries were written with injury prevalence. The number of injuries for the complete cohort population including both injured and not injured, was included for context.

4.1 Rates of injury

The prevalence rates of injury reported in this study ranged from 39% (MLC cohort) to 58% (OLC Cohort). Interestingly, research by Almeida et al. [20] over 20 years ago, in a sample of 1,143 male USMC trainees, reported that 39% of trainees sustained at least one injury during their 12 weeks of training. These findings were similar to those reported in US Army Special Forces Assessment and Selection at 38.4% [21]. Conversely, Piantanida et al. [22] reported a prevalence of 59.5% of injuries for trainees completing Marine Officer Candidate School. Similarly, Robinson et al. [23], reported that, in their sample of 1,810 British Army Infantry recruits, 58% suffered at least one injury; similar to those reported in US Army recruits [24]. As such, the rates of injury observed in this study are in alignment with previously reported other military cohorts [20–24].

4.2 Bodily sites

The lower limbs, specifically– the knee, lower leg, ankle, and foot were the leading bodily sites of reported injuries in both cohorts representing the top four injured body sites overall (OLC = 65%; MLC = 70%). These results are not unsurprising given that the lower limbs are a leading site of injury in military populations [25–30]. Lower extremity injuries (knee, shin, foot, and ankle) are well-known sites of injury during military training [23, 31], and specifically with load carriage [1, 32–34] with a notable concern being stress fractures [33, 34].

In both cohorts, the shoulder presented as being of a higher injury proportion than the lower back (OLC: shoulder n = 99, 7.2%, lower back n = 86, 6.3% : MLC shoulder n = 19, 7.1%, lower back n = 11, 4.1 %). These findings differ from other research in military populations, whereby lower back injuries are proportionally higher than shoulder injuries [35]. Furthermore, even noting that upper limb strength and endurance are typically a greater predictor of load carriage performance than lower limb strength [36–38], the majority of load carriage research suggests that the lower back is injured proportionally more than the upper limbs [39].

Considering this, upper limb neurological injuries, notably brachial plexus palsy, is a known risk associated with the requirement of military personnel to carry heavy loads in backpacks [1, 41]. However, the proportion of neurological injuries reported in this study (which could include meralgia [13] and digitalgia [14]) was relatively low (i.e., OLC n = 5, 0.5% : MLC n = 11, 4%) and would not account for the reported proportion of shoulder injuries. Unlike other tactical populations (i.e., military and firefighters), police officers also present with a higher proportion of upper limb injuries. A potential reason for these injuries is postulated to be their use of force when dealing with non-compliant offenders [42]. Thus, other training factors, like basic warrior training and martial arts (sources of injury in this population) may have accounted for the higher numbers of shoulder injuries, as may other activities like obstacle courses, which are likewise known to increase the proportions of shoulder injuries [24].

The proportion of lower back injuries was lower in the MLC (n = 11, 4.1%) as compared to the OLC (n = 86, 6.3%). While lower back injuries are typical of injuries sustained during basic military training [27], the impact of lower back injuries, when completing load carriage events for example, is of note given that soldiers who suffer a lower back injury during a load carriage event (e.g., loaded hike training, Crucible, etc.) are less likely to complete the event [43].

4.3 Nature of injury

Sprains and strains were the leading nature of injury for both cohorts representing approximately a quarter of all injuries. This finding aligns with findings across both USMC [11] and other military populations where sprains and strains are likewise the leading nature of injuries [20].

With stress fractures being of particular concern during training, the reduction of stress fractures from 8.3% (OLC) to 3.4% (MLC) is of note and explains the increase in stress-related changes, a precursor to stress fractures, reported in the MLC (6.3% versus 0.3%). These findings suggest that stress changes did not progress to fractures. Given the impacts of stress fractures in personnel undergoing training (notably the lengthy rehabilitation periods or loss of the trainees [44], this reduction is noteworthy.

4.4 Event during which injury occurred

The most common events during which injury occur (overuse-nonspecific, SULE or Crucible, running, and conditioning hikes) are supported by previous research in a USMC population [11]. While some variations were present, such as the slightly higher prevalence of SULE/Crucible injuries in MLC (12.6% versus 9.1%), the overall prevalence of injuries across the top four events, common to both cohorts, was lower in the MLC (30%) as compared to the OLC (40%). This finding demonstrates the efficacy of the progressive load carriage training contained herein, with the inclusion of six additional hikes. Furthermore, while proportionately higher for the SULE and ‘Crucible’ the prevalence of injury was markedly lower in the MLC cohort (13% v OLC = 31%). Given that the ‘Crucible’ is a key USMC training outcome, this finding highlights the success of the MLC approach. Thus, the progressive load carriage conditioning implementation could be considered a success whereby, even though the exposure to load carriage hikes increased, injuries were lower, especially during the ‘Crucible’.

Potential reasons for these findings, as will be discussed below, include familiarization with equipment, progressive conditioning, and education on proper packing of packs and biomechanics. Given the additional sessions added to the program, the recruits in MLC had considerably more time to familiarize themselves with wearing back packs and hiking underload. Previous work by Rudzki [10] supports this supposition, where a platoon who substituted weight load walking (pack marching) for running activities was said to perform military tasks (i.e., ‘The Challenge’ event consisting of a 20 km pack march, obstacle course, and test of shooting skills) better than their corresponding cohort platoon, even though their maximal aerobic capacity was slightly lower. Further, the progressive nature of the hike program allowed for conditioning of the musculature for hiking to meet recommended conditioning guidelines of a load carriage session every 7– 14 days [8, 9]. Finally, the constraints of the training schedule in OLC did not allow for proper packing of back packs or adequate time for recruits to familiarize themselves with proper back-pack fit and hike biomechanics. As such, where the OLC traditionally had the load carriage events in a compressed time frame and in a field environment, the additional load carriage sessions enabled instruction from the drill instructors on how to properly pack a back-pack and on hike biomechanics.

4.5 Causal type of injury

Including a ‘chronic’ field in the MLC data made direct comparisons between the two cohorts difficult. In both cohorts new overuse injuries were the most common causal type, followed by acute/traumatic injuries, findings supported by previous research [11]. Furthermore, when the findings of this study for the OLC cohort (new overuse = 61%; acute/traumatic=27%; pre-existing overuse = 12%) are compared to those of Jensen et al. [11], the proportional presentation is similar (new overuse = 61%; acute/traumatic=28%; pre-existing overuse = 11%). Considering this, the decrease in the proportion of new overuse injuries suffered by the MLC may explain the drop in event presentations for overuse-nonspecific injuries. This suggests that even with an increase in the number of conditioning hikes, new overuse injury presentations were lower.

4.6 Severity

Noting that the definitions for injury severity were based upon ICD-10 codes [18], the MLC (60%) presented with proportionately more moderate injuries than the OLC (42%), yet the prevalence of moderate injury severity for both groups was similar (MLC = 23%; OLC = 24%). Furthermore, the MLC presented both a higher proportion and prevalence of severe injuries (MLC = 15% and 6%; OLC = 5% and 3%, respectively). However, 7% of OLC data was unusable due to missing information. Considering this data yield, if most instances were recategorized as severe, then the prevalence of moderate and severe injuries would be similar between the two cohorts. This is unlikely, however, given that reporting typically becomes more stringent with higher severity injuries [17]. This assumption is based on the fact that incidents at a severe level would generally have demanded both command and medical attention, making their accurate recording more likely [17].

4.7 Limitations

This study did present several limitations. Firstly, the data comprised of only male recruits, thus limiting the translation of this work to female recruits who may suffer from different injury presentations, notably due to load carriage [32]. Secondly, the changes in potential fitness and load carriage performance needed to be more detailed. While not the focus of this study, understanding potential changes in fitness due to the increase in the number of load carriage sessions would be of value. Considering this, completion without injury could be considered an outcome measure of performance, whereby fewer injuries to USMC recruits led to greater ability to achieve training outcomes. Further, changes in data categories (e.g., ‘chronic’ in causal type) and missing information (e.g., in injury severity presentations), limit some comparisons between the two cohorts. Thirdly, we recognize that a multi-variable logistic regression would assess the relative contributions of other risk factors would provide more insight as to the effectiveness of the progressive load carriage program; however, no data was collected on known contributors to injury such as history of injuries, history of training, movement skills / movement literacy, nutritional status (e.g., vitamin D and iron) and/or sleep. Thus, the analysis was limited to what is presented herein. Finally, in the OLC group demographic information such as height, weight, and age were not available to run analysis to determine the relationship to MSKI. Future work is needed to determine if the relationship between body size and composition to MSKI is valid in military recruits.