Backpacks are widely used by children to carry books, notebooks, pencil holders, and pedagogic material. Most of the time, children have to carry the material daily from home to school and back due to the classes and homework activities [1, 2]. The literature on the topic has raised concerns about heavy loads carried in backpacks by children [3, 4, 5], some relating them to health issues [1, 6, 7]. However, most of the concerns are associated with the effect of the backpack load on posture and gait patterns, such as reduced pelvic rotation [8]; increments on the angle of the head (an adopted forward position) [9, 10, 11]; excessive trunk flexion [9, 10, 11]; and an increase in ground reaction forces (GRF) [12, 13, 14, 15]. These heavy backpacks may cause the above mentioned postural changes [4]. Additionally, backpacks influence the kinetics and kinematics of children’s walking as the greater the schoolbag load, the slower will be the locomotion and the greater the mechanical load [16, 17, 18].

The mechanical forces influence vertebral growth [19] and high loading rates may have negative effects on bone health [20, 21]. For instance, growing children who use backpacks daily may suffer high GRF level associated with lower-limb [22, 23] and spinal injuries [24]. This leads to a health related problem as already presented in literature [25]. That may be explained by children’s motor activities, such as running or jumping while carrying school backpacks [26]. Efforts have been made to define a load limit to be carried [2, 27], which is most frequently set at 10% of the child’s body weight (BW) [2, 28, 29]. However, this limit is often exceeded [3, 28, 30]. For example, recent studies found that approximately two-thirds of Portuguese children (10–15 years old) and Polish children (7–9 years old) usually carry loads above 15% of their body weight [4, 5].

In order to minimize the effects of carrying heavy backpacks, several models of bags have been proposed. Some of the proposals promote a more upright posture by carrying the load in big pockets at the hip level [31] or replacing the flexible shoulder straps by rigid ones [31]. Interior pockets have also been developed for school backpacks to distribute the weight. The pocket placed near the chest level positively influenced the postural pattern, decreased head deviation [10], and the abdominals and spinal erectors muscular activity [32]. As mentioned before, the load stress is typically measured by the GRF. In an attempt to reduce the GRF peak values, Ren et al. [33] assessed the influence of decreasing suspension stiffness in military material (bags) on the GRFs peak values. The evaluation was performed using a computer simulation of level walking of a backpack suspension model with linear elastic and linear damping components.

Unfortunately, some of the attempts to create alternative backpack models that offer biomechanical and energetic advantages have not reached the market, due to not being fashionable and enjoyed by children. Adding elastic material to the shoulder straps of a common backpack will not change the backpack design much and has theoretical support [33]. Thus, it is important to explore and consider a traditional backpack design that can bring biomechanical advantages, specifically a positive influence on the GRF. Furthermore, backpack weight is not the same in the Portuguese primary education system [4, 5] and children’s physical characteristics also become different as they grow. Regarding the above mentioned information, it is important to understand how the backpack load affects the GRF of children aged between 10 and 15 years old (5th and 9th grade of the Portuguese educational system). Moreover, children usually play (walking and running) with the backpack on, and the GRF will vary according to the type of locomotion. Upon that, it is important to better understand the effect of walking and running on the GRF with a modified backpack. Therefore, the aim of this study was to test the effect of the vertical GRF in children when walking and running with both a common backpack and one with elastics added to the straps, allowing a decrease in the shoulder straps stiffness. It was hypothesized that the vertical GRF will be lower with the added elastics in the backpack compared to the one without them. Moreover, the walking may present a lower GRF compared to running.

The time and force variables (mean, standard deviations, significance, and effect sizes) for the traditional and modified backpack during the walking condition are presented in Table 1. The application of the elastics on the backpack straps did not change the total stance time and had a small influence on the time parameters for the total sample. However, a significant difference with medium effect was observed in the time variable from Fz2 to Fz3 (s) on the G-9 group ($\mathrm{\Delta }$ = –0.008; p = 0.03; d = 0.29). The variables significantly affected in the force variables in the G-9 group were the First peak-Fz1 (N) ($\mathrm{\Delta }$ = 13.61; p = 0.039; d = 0.27); Fz2 (N) ($\mathrm{\Delta }$ = –13.09; p = 0.039; d = 0.27); Fz2 (N.BW${}^{-1}$) ($\mathrm{\Delta }$ = –0.02; p = 0.028; d = 0.29); Total integral (N.s) ($\mathrm{\Delta }$ = –6.12; p = 0.037; d = 0.27); Diff. max-min ($\mathrm{\Delta }$ = 23.05; p = 0.048; d = 0.26); Integral from Fz2 to Fz3 ($\mathrm{\Delta }$ = –4.69; p = 0.028; d = 0.29); Integral after Fz3 ($\mathrm{\Delta }$ = –2.92; p = 0.009; d = 0.35). Moreover, the total sample presented significant differences for Integral after Fz3 between the traditional and the modified backpack ($\mathrm{\Delta }$ = –2.17; p = 0.005; d = 0.28). Regarding the running condition (Table 2), time variables were not significantly different between the traditional and the modified backpack. Furthermore, force magnitudes did not vary significantly, except for: First peak-Fz1 (kN); First peak-Fz1 (N.BW${}^{-1}$); Integral to first peak (N.s); and Integral to second peak (N.s).

Regarding the comparison between males and females, Table 3 presents the differences between the modified backpacks for boys and girls when walking. The Table 4 presents the comparison when running. For the walking condition, only significant differences were observed fot t to Fz1 (s) for boys ($\mathrm{\Delta }$ = –0.008; p = 0.034; d = 0.34) and girls ($\mathrm{\Delta }$ = 0.005; p = 0.041; d = 0.25). The load rate presented significant differences for boys ($\mathrm{\Delta }$ = 0.31; p = 0.014; d = 0.38) as for Integral after Fz3 ($\mathrm{\Delta }$ = –3.21; p = 0.021; d = 0.44), but not for girls.

In Table 4, it is possible to find significant differences for running. The elastics affected the girls’ Time between R1 and PA (s) when running ($\mathrm{\Delta }$ = 0.01; p = 0.037; d = 0.31), but not the boys’. However, no significant differences were observed in force variables for girls. Otherwise, boys presented significant differences, reducing the force levels in most force variables.

The current study aimed to investigate whether the introduction of elastic material on backpack straps can be a solution to decrease the magnitude of vertical GRF peaks and loading rate. It was hypothesized that vertical GRF would be lower with elastics in the schoolbag shoulder straps compared to the traditional one without elastics, and that walking would present a lower GRF compared to running. This hypothesis was assumed based on the concept that the elastics can cause an absorption effect, reducing the GRF. Thus, including elastics in the shoulder straps may lead to lower GRF and allow the design of the backpack to be maintained according to children’s individual preferences. The main results of this study were that the time and force variables were higher in the traditional school backpack compared to the adapted one when running. However, significant differences were only found in the G-9 group when walking. It was observed that the modification on the backpack induced changes in the GRF magnitudes. The most important finding was probably the decrease in the loading rate and force peaks on some tested conditions.

The purpose of a modified backpack came from the need to reduce the GRF caused by the backpack load [4]. By adapting the backpack it should be possible to achieve biomechanical advantages. These changes should be discrete so that children still find the backpack appealing after the applied modifications. It is counterproductive to achieve the most biomechanical beneficial backpack if children won’t make use of it. The introduction of the elastic on the shoulder straps [29] did not significantly change the backpack design or alter the total stance time when compared to the non-modified backpack.

When walking, only G-9 experimented significant differences between the two conditions. When using the modified backpack, G-9 prolonged the phase between Fz2 (vertical force relative minimum) and Fz3 (second vertical force peak). This may be due to the moving up from the center of gravity, originated by the propulsion force produced in this phase, forcing the elastic to elongate. This resulted in the production of more energy in this phase, which is confirmed by the increased integral data. The integral data, impulse, or the “amount of force” produced was higher in the second half of the stance, probably to restore the elastic energy absorbed in the first half. The great advantages presented by using the modified backpack were the decrease in the first vertical force peak and the decrease of the loading rate, which were probably due to the energy absorption induced by the elastics. The increase of the Fz2 (minimum relative) was also a reflex of the elastics action, dissipating energy from the first peak to the following moments. This may be seen as a small phase delay of the load due to the straps, similar to another study with a suspended-load backpack [36]; however, there’s a lack of research in this field. A previous study presented the same results for the walking condition and the integral force was lower compared to running [34]. This may speculate that the straps will allow the decrease of the integral force for each locomotion type. Moreover, the same study [34] presented similar results between loaded and unloaded bags, which probably means the straps will produce a reduction in the load mechanical effects.

Thus, for the G-9, when walking, the action of the elastics allowed a decrease in the force peak, “distributing” the force difference over the next stance moments. The G-5 did not present differences between the two backpacks, probably because the carried load was so high that it stretched the elastics completely, leaving no room for them to work. It is possible to compare this effect between loaded and unloaded bags [34]. However, the magnitude of the effects was different in the present study with straps compared with different loads [4]. When running, there were no changes in the time variables, but similarly to walking, the force variables suffered modifications. Again, with the use of the modified backpack, the force peak value decreased as well as the loading rate. Probably, the elastic absorbed a part of the energy generated by the vertical brake of the stance, dissipating it along the time and lowering the maximum level of the vertical force. These results are largely expected based on the literature, in which peak force values are higher in running conditions [34, 37]. As for running, the total generated impulse decreased with the modified backpack. A possible explanation for this could be the reduced rebound of the backpack on the shoulders, promoted by the action of the elastics. With a reduced rebound, the negative vertical acceleration of the backpack would also be reduced, decreasing the need to cause impulse to neutralize it [36].

The G-5 experienced differences between the two backpacks when running, contrary to walking. Although the load carried was the same on both conditions, when running the backpack presented greater rebounds, which could help to explain the change, as they probably promote moments when the elastics could shorten, and then be able to stretch again, absorbing energy [37]. These findings, specially the decrease on the vertical force peaks with the use of the modified backpack, are in agreement with the theoretical model by Ren et al. [33], in which it is suggested that a backpack suspension model with a lower stiffness may offer biomechanical advantages, namely the decrease on the peak values of the GRF. However, it is important to note that the G-5 typically carries higher backpack loads [4].

A comparison between modified and traditional backpacks was carried out for boys and girls. For walking, few differences were observed in the present study. However, for running, the boys presented higher differences regarding force variables. In the present study, boys typically presented higher ground reaction forces. The literature supports these findings, confirming that the locomotion technique between boys and girls explains the differences because boys tend to apply higher force load when running [38]. However, significant differences were only observed in the running condition when comparing traditional to modified backpacks.

The decrease on the vertical GRF and loading rate may represent important advantages in favour of the introduction of elastic material on the shoulder straps of school backpacks. This study can probably encourage a deep understanding of how these benefits can be controlled and maximized. The different modifications observed between G-5 and G-9 could mean that the elastic stiffness needs to be adjusted concerning the carried load, children’s body weight, or the relation between both. After determining this, it could be applied as a combination of elastics that children would apply to their backpack depending on the load they have to carry, their body weight, or a combination of both. If there is the need to combine different elastics with different characteristics, the elastics could be differentiated by the use of different colors. The researchers of the current study believe this is a practical and commercially viable idea.

The differences in the use of the two backpacks in some variables were significant when considering all the elements of the sample; however, this didn’t happen when analyzing the groups separately. That occurred mainly when running and may be evidence that the sample size should be bigger. The backpack was modified introducing the elastics on the shoulder straps but without discontinuing the straps. The effect could be different if the straps were cut and attached exclusively by the elastics, promoting a bigger range of motion for the elastics. This idea is probably worth exploring. In the search for an ideal elastic stiffness, there are many levels of the variables to be tested, such as elastics with different dimensions, characteristics, and resistance, combined with different loads and children’s BW. There is still a very wide field to be analyzed, such as the effects of elastics on the shoulder straps when jumping or rotating. Nevertheless, this should be studied first in a laboratory with a mechanical “subject”, due to the variety of conditions to be tested, and only then, in a second phase, with children, to clarify the results.

The introduction of elastic on the shoulder straps of a school backpack modified the GRF parameters, when walking and when running. The main goal of reducing the vertical peak force levels and loading rate was achieved but not in every condition/age group. More research should be done to define the parameters to be controlled in order to maximize that effect.

GFR, Ground Reaction Force; G-5, 5th grade; G-9, 9th grade.

Conceptualization—MCM, MI, DAM; Data curation—JB, CCS, PF; Formal analysis—JB, PF, HPN; Funding acquisition—DAM; Investigation—JB, CCS, HPN; Methodology—MCM, DAM, MI; Project administration—MCM, MI; Resources—DAM, HPN; Software—JB, CCS; Supervision—MCM, DAM, MI; Validation—PF, HPN; Visualization—JB, CCS, PF, HPN; Roles/Writing - original draft—JB, CCS, PF, HPN; Writing - review & editing—MCM, DAM, MI.

The experimental procedures were conducted according with the International Charter for Ethical Research Involving Children and the Declaration of Helsinki, The study was also approved by the local ethics board (UBI/FCSH/DCD/D974 registration).

Thanks to all the peer reviewers for their opinions and suggestions.

This work is supported by national funding through the Portuguese Foundation for Science and Technology, I.P., under the project UIDB/04045/2020.

The authors declare no conflict of interest. PF, HPN and DAM are serving as one of the Guest editors and DAM is serving as one of the Editorial Board members of this journal. We declare that PF, HPN and DAM had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to AT.