Distance running is an increasingly globally popular sport as reflected by the increasing number of marathon finishers and major running events around the world.1 In spite of potential cardiovascular and mental health benefits related to distance running,2 the incidence of running-related musculoskeletal injury is extremely high. Up to 79% of regular runners may incur an overuse injury in a given year.3,4 Thus, prevention of running-related musculoskeletal injuries has received a lot of attention over the past decades.

A series of retrospective studies have related high vertical impact loading, which is usually expressed as vertical average loading rate (VALR) and vertical instantaneous loading rate (VILR), with a series of running-related ailments, such as patellofemoral pain, tibial stress fracture, and plantar fasciitis.5–7 Recent prospective studies also suggest that high VALR and VILR may be associated with the development of running injuries.8,9 Therefore, different strategies have been proposed to lower the vertical loading rates, including footstrike pattern modification10 and cadence adjustment.11

Specifically, runners exhibiting non-rearfoot strike, i.e. midfoot or forefoot strike, have been shown to experience lower VALR and VILR than rearfoot strikers.12,13 Such reduction in the impact loading can be explained by a lower effective mass during non-rearfoot strike.14 As for spatial parameters, runners with greater cadence and shorter stride length have been reported to place the heel closer to the center of mass at initial contact, which results in a reduction in the braking impulse15 and vertical loading rates.16 Therefore, shortened stride length accompanied with increased cadence for a given velocity also contribute to the reduction of running injuries.11

Anecdotally, many runners believe that adults should mimic the running pattern of children,17 who are supposed to exhibit the most natural running gait18 without being influenced by any external device e.g., shoes.19 Such belief in running biomechanics modification is mainly based on the assumption that children usually land with more non-rearfoot strikes, run with shorter stride length and higher cadence, when compared with adults.19

Limited knowledge exists, however, with regard to the difference in running characteristics between children and adults. To our best knowledge, the majority of previous studies explored differences in walking biomechanics between the two groups.20–22 However, there is a lack of evidence showing the differences in the running biomechanics between children and adults.

Considering running kinetics,23 spatiotemporal parameters,24 and joint kinematics22 become mature and more adult-like at approximately 7–8 years old, the present study compared the running biomechanics between preschoolers (i.e. age <7) and adults. We hypothesized that preschoolers would present lower body weight normalized VALR and VILR than adults. We also expected preschoolers would exhibit more non-rearfoot strikes, greater cadence, and shorter normalized stride length than adults.

Based on the Shapiro–Wilk test, the adult VILR data distribution was determined to be non-parametric (p=0.02). There was no significant difference in the normalized running speed between the two groups (p=0.85). The VALR, VILR, FSA and footstrike pattern, cadence, and stride length in preschoolers and adults are presented in Table 2 and Fig. 1. There was no significant difference in VALR (p=0.48) and VILR (p=0.48) between preschoolers and adults. Preschoolers and adults also demonstrated no differences in the FSA (p=0.85) and footstrike pattern (p=0.29). In terms of spatiotemporal parameters, preschoolers exhibited greater cadence (p<0.001) and shorter normalized stride length (p=0.01) than adults.

Figure 1.

Violin plots, indicating medians (black dotted lines), quartiles (gray solid lines) and data distribution, to compare running biomechanics between preschoolers and adults.

(0.19MB).

This study examined the difference in running biomechanics between preschoolers and adults. There were no significant differences in the vertical loading rates and footstrike pattern between the two age groups. However, preschoolers demonstrated a statistically greater cadence and shorter stride length than adults.

In the present study, preschoolers presented similar VALR and VILR with adults. Originally, we expected that preschoolers might experience lower vertical loading rate than adults, as preschoolers may have less influence from footwear habituation.18,19 A previous study suggested that children who had never worn shoes experienced lower vertical loading rates during running.14 In the present study, however, most of the parents of our preschoolers reported that their children started to wear shoes on a daily basis since the acquisition of walking skill. In view of the normal developmental milestone of independent walking at approximately 12 months,33 and the mean age of our preschoolers i.e., 4.2 years, footwear habit for 3 years may be sufficient to alter the natural running pattern.34 Because of the association between vertical loading rate and running injury,8,9 we originally hypothesized that a lower vertical loading rate would be present in preschoolers, when compared with adults. A prospective study found that 38.5% of adolescent runners sustained at least one injury over a running season.35 This injury rate is actually similar to the risk in adult runners reported by previous studies (i.e. 19–79%).4,36 The comparable vertical loading rates between preschoolers and adults may reflect similar injury risk between groups.

Similar to the findings on vertical loading rates, preschoolers demonstrated similar FSA and footstrike pattern when compared with adults. This finding is likely explained by the interaction of footstrike pattern and footwear. Modern footwear usually includes thick and cushioned midsole, rigid heel counter, and protective arch support.37 Although footstrike pattern can be affected by other factors, such as running speed,38 these shoe features have shown to possibly lead to a RFS pattern 14,39. This finding is consistent with a previous study investigating footstrike pattern in a group of preschool children.40 In that particular study, shod running experience encouraged RFS landing in children who were aged 3–6 years and the authors suggested that there may be a footstrike pattern transition from non-RFS to RFS due to the use of shoes.40

Shod running has been reported to result in a reduction of cadence and an increase of stride length in preadolescent children.41 In the present study, preschoolers exhibited greater cadence and shorter stride length than adults, in spite of similar vertical loading rate and footstrike pattern. This finding is consistent with the findings reported in a previous study.42 A plausible explanation is that the spatiotemporal parameters may be less influenced by footwear when compared with footstrike pattern.43

Children, during growth, experience musculoskeletal changes and central nervous maturation.44 It has been suggested that some body growth factors e.g., body mass and body stiffness, affect cadence during running gait development.45 Cadence is almost coincident with the natural frequency of human locomotion during lower-speed running, and the natural frequency is determined by both body mass (m) and body stiffness (k) with relation f=(√(k/m))/2π.18 For children from 2 to 12 years old, the natural frequency of locomotion decreases with age and it is mainly attributed to the reduction in the ratio between body stiffness and body mass.18,46,47 The reduction in the ratio between k/m results in a reduction in both natural frequency and cadence. Cadence appears to be mature at the age of 12 years, as the ratio between k/m becomes constant due to the parallel increase in both k and m with age in the following years.18

The greater cadence may imply shorter stride length in preschoolers given the linear and quadratic stride length–cadence relationship.48 Additionally, lower ankle power generation in preschoolers may also partly explain the shorter stride length compared with adult. Running gait seems to require greater operating effort from the ankle than knee extensors.49 However, preschoolers use their proximal muscles, i.e. hip muscles, more than distal muscles, i.e. ankle plantar flexors, for power generation because of the immaturity of neuromuscular control.44 As ankle power is associated with stride length,50 developing children, especially at a younger age, may therefore present shorter stride length compared with adults.

The two age groups in our study presented comparable loading rates, despite the preschoolers demonstrated higher cadence and shorter stride length, which are commonly considered as low-risk factors during running. Taken together these findings and the potential reasons that children demonstrated different spatiotemporal running gait from adults, simply imitating preschooler-like pattern to run may not be efficacious in decreasing musculoskeletal running injuries.

It is very important to note several limitations in this study. Firstly, the present experiment adopted a cross-sectional design and the causal relationship cannot be determined. Therefore, the footwear effect on the locomotion development remains speculative and future prospective studies are warranted. Secondly, we did not control the testing shoes in the present study as we failed to find a running shoe model for both preschoolers and adults available in the market. Thirdly, we did not collect individual joint kinematics, thus leaving comprehensive analysis of difference in running postures between preschoolers and adults largely unknown. Fourthly, the sample in this study is not enough for subgroup analysis based on the ages ranging from 3 to 6 years to further investigate the impact of time length of shoe wearing and running experience. Finally, the present study was conducted in a laboratory environment, which may not fully reflect the actual gait biomechanics in a natural environment. With the recent advancement of sensor technology, measurement of running biomechanics outdoors using wearable body-worn sensors is now possible.51 Future investigations of gait biomechanics in a natural environment are therefore highly warranted.

Preschoolers in our study demonstrated greater cadence and shorter stride length compared with adults. However, these spatiotemporal performances did not contribute to more frequent non-RFS pattern nor decrease vertical loading rates during initial contact, which were found to be associated with decreased musculoskeletal injuries in running from literature. Therefore, our findings do not support the notion that adults should modify their running biomechanics to resemble preschoolers’ running form for a lower injury risk.

The authors declare no conflicts of interest.