Standardized Lab Shoes Do Not Decrease Loading Rate Variability in Recreational Runners

in Journal of Applied Biomechanics
View More View Less
  • 1 University of Maryland
  • 2 Algovation, LLC
  • 3 Kyung Hee University

Studies of running mechanics often use a standardized lab shoe, ostensibly to reduce variance between subjects; however, this may induce unnatural running mechanics. The purpose of this study was to compare the step rate, vertical average loading rate, and ground contact time when running in standardized lab shoes versus participants’ normal running shoes. Ground reaction forces were measured while the participants ran overground in both shoe conditions at a self-selected speed. The Student’s t-test revealed that the vertical average loading rate magnitude was smaller in lab shoes versus normal shoes (42.09 [11.08] vs 47.35 [10.81] body weight/s, P = .013), while the step rate (170.92 [9.43] vs 168.98 [9.63] steps/min, P = .053) and ground contact time were similar (253 [25] vs 251 [20] ms, P = .5227) and the variance of all outcomes was similar in lab shoes versus normal shoes. Our results indicate that using standardized lab shoes during testing may underestimate the loads runners actually experience during their typical mileage.

Studies of running mechanics often have all subjects use standardized “lab shoes” with the same make and model for all subjects.14 The objective of using lab shoes is ostensibly to remove a source of variance between subjects, so that the differences between subjects can be more confidently attributed to the independent variable(s) of the study. However, the responses of running mechanics to a change in footwear can vary widely between individuals, even if there is no main effect statistically.5,6 There is, therefore, a risk of inducing unnatural running mechanics when subjects run in unfamiliar lab shoes rather than their own shoes. This risk is a concern for studies that seek to draw conclusions on running outside the lab (eg, injury risk, race performance) from data collected in the lab. While many studies have compared the mechanics and energetics of running in different types of shoes, the effect on running mechanics of switching from a participant’s own normal running shoes to standardized lab shoes does not appear to be widely known. This gap in knowledge is important because it is possible that lab shoes may not necessarily meet the goal of reducing variance between subjects and that using lab shoes may affect the generalizability of results by inducing mechanics that deviate from a subject’s typical gait.

Running injuries are attributed to forces during the contact phase of running,7 and various running injury studies compare how these forces are applied in a variety of runner groups and/or running conditions.1,4,5,815 High vertical average loading rate (VALR) has been prospectively4 and retrospectively1 associated with running injuries in rearfoot strike runners. More cushioned shoes are often recommended for impact shock attenuation associated with these higher loads; however, the amount of cushioning in running shoes was classically shown to have no effect on the loading rate.8,9 More recent studies have found cushioning to both increase16 and decrease17 loading rate. Recent studies have found similar inconsistent effects of shoe cushioning on spatiotemporal parameters.18 Ground contact time (GCT) was similar between shoes with large differences in cushioning,18 while increased shoe stiffness associated with mileage accumulation increased GCT.19 The effects of shoes with different levels of cushioning and stiffness on VALR and GCT are important because of the force–time interaction during the contact phase of running16,17 and the frequency with which these variables are studied as injury and performances outcomes.

Therefore, the purpose of this study was to compare the common running gait outcomes of step rate (SR), VALR, and GCT when running in standardized lab shoes versus the participants’ normal running shoes. The first hypothesis was that SR, VALR, and GCT would not differ between standardized lab shoes versus the participants’ normal running shoes. Second, because footwear has widely varying subject-specific effects on ground reaction forces in running20 (Figure 1), the second hypothesis was that the between-subjects variance in running mechanics would not decrease when running in standardized lab shoes compared with the participants’ normal running shoes.

Figure 1
Figure 1

Conceptual framework for the second hypothesis based on Bates et al,20 demonstrating hypothetical responses to the vertical ground reaction force loading rate when running in different shoes. The highly variable responses in A are more typical than the consistent responses in B. Both data sets have the same mean and variance.

Citation: Journal of Applied Biomechanics 2020; 10.1123/jab.2019-0337

Methods

Participants

The participants were 19 healthy, habitually shod, recreational runners (10 women and 9 men) recruited from the local community. Participants ran with a rear foot strike pattern in both shoes (foot angle >8° at initial contact21) and with similar self-selected speeds in both shoes (± 5%) and were selected from a larger group of participants based on these criteria.10 This sample size had a minimum detectable difference of Cohen d = 0.68 in a dependent t test, with SE rates (α = .05, β = 0.20). The participants represented a wide range of skill, with an average weekly mileage of 23 miles (range: 9–70 miles). The demographics of the participants are in Table 1.

Table 1

Subject Demographics

VariableAllWomenMen
Age, y283025
 SD10128
 Minimum181918
 Maximum545434
Mass, kg67.159.877.5
 SD15.45.119.5
 Minimum52.352.362.3
 Maximum115.968.2115.9
Height, cm170.2165.3176.4
 SD9.14.010.1
 Minimum157.0157.0165.0
 Maximum196.0173.0196.0

Experimental Setup

The participants brought their own normal running shoes, in which they perform the majority of their weekly mileage to use in the study, and they also used a new pair of “neutral-cushioning” standardized lab shoes provided by the researchers (New Balance 780v5, New Balance, Boston, MA). No participants brought shoes that are contemporarily considered in the “maximalist” category (eg, Hoka One) or the extreme end of the “minimalist” category (eg, Vibram FiveFingers). Other common “minimalist” shoes (eg, Nike Free, New Balance Minimus) were included, but the great majority of the participants’ normal shoes were “traditional” cushioned running shoes similar to the lab shoes. No participants had shoes that were the same make and model as the lab shoes.

The participants wore 33 reflective markers on the pelvis (iliac crests, anterior superior iliac spines, posterior superior iliac spines, and sacrum); lower extremity of the dominant leg, defined as the leg used to kick a soccer ball (greater trochanter, 4-marker thigh cluster, lateral and medial epicondyles, fibula, shank, and lateral and medial malleoli); and both feet (great toe, first and fifth metatarsal, and calcaneus).22 Marker positions were captured using a 13-camera motion capture system (VICON, Centennial, CO), sampling at 200 Hz. Eight embedded force plates (Kistler, Amherst, NY) measured GRF at 1000 Hz. Consecutive placement of the force plates on a 12-m straight stretch of the track defined the motion capture space.

The participants performed a static calibration trial by standing still with their feet shoulder-width apart and shoulders abducted to 90° for 10 seconds. Calibration markers were removed for movement trials, and the participants warmed up at their leisure and then ran around a 50-m indoor track for 3 laps at a self-selected “normal” speed. Specifically, the participants were instructed to run a moderate pace that they might use for the bulk of their training for the “normal” speed, and the participants ran freely, based on these instructions. For all participants, static and running trials were conducted first, while subjects wore their own normal running shoes, followed by performing the same conditions and instructions again in the standardized lab shoes. These running trials were part of a larger investigation that involved discontinuous running at a range of submaximal training speeds for a total of 900 m, substantially less than the participants’ typical training distance.

Data Reduction

Data were processed using Visual 3D software (C-Motion, Inc., Germantown, MD). A forward–reverse fourth-order low-pass Butterworth filter smoothed the data, with cutoff frequencies of 10 Hz for markers and 50 Hz for GRF. A 20-N threshold of the vertical ground reaction forces identified the initial foot contact and toe-off. Speed was determined using the stride length and the stride time between successive heel strikes.

The VALR was calculated as the average slope of the vertical GRF versus time between 20% and 80% of the time from initial contact to impact peak and scaled by body weight (BW).1 When there was no clear impact peak, the “peak” was assumed to occur at 13% of the stance duration.23 An average of 5 total strides were collected for the dominant side of each runner during the 3 laps of each shoe condition. The VALR and stride rate for each stride were averaged for each condition.

Statistical Analysis

Statistical analysis was done using a customized script in R (Vienna, Austria).24 SR, VALR, and GCT met statistical assumption of normality (Shapiro-Wilk test) and homogeneity of variance (Levene’s test)25 and were compared between the runners’ normal shoes and the standardized lab shoes using a Student’s paired t-test. Cohen dz effect sizes were calculated, where the numerator was the t statistics from the t-test comparison, and the denominator was the square root of the number of participants.26 This analysis is exploratory; therefore, we determined that effect sizes > 0.3 are worth consideration.14

Results

Running speed in the standardized lab shoes was, on average, 2.4% faster than in the participants’ normal running shoes (3.44 [0.43] vs 3.39 [0.39] m/s), less than the commonly accepted 5% difference threshold for “similar” speeds.4,27 VALR magnitude was smaller in lab shoes versus the participants’ normal running shoes (42.09 [11.08] vs 47.35 [10.81] BW/s, P = .013, dz = .64). SR and GCT were similar in lab shoes versus normal shoes (170.92 [9.43] vs 168.98 [9.63] steps/min, P = .053, dz = .48) and (253 [25] vs 251 [20] ms, P = .523, dz = .15), respectively (Figure 2). Variance in lab shoes compared with normal running shoes was 4.4% lower, 4.8% higher, and 31.2% higher for SR, VALR, and GCT, respectively, but the Levene test for homogeneity of variance found that the between-subjects variance was similar for all variables in both shoe conditions (Figure 3).

Figure 2
Figure 2

(A) SR, (B) VALR, and (C) GCT of participants running at a self-selected “normal” speed in a standardized lab shoe and their own normal running shoes. Connected gray dots show individual subjects, and group means and SDs are represented by white triangles and solid black lines, respectively. SR indicates step rate; VALR, vertical average loading rate; GCT, ground contact time. *Significance of P < .05.

Citation: Journal of Applied Biomechanics 2020; 10.1123/jab.2019-0337

Figure 3
Figure 3

Between-subject variance of participants running in a standardized lab shoe (black) versus their own normal running shoes (gray). Variance in lab shoes compared with normal running shoes was 4.4% lower, 4.8% higher, and 31.2% higher for (A) SR, (B) VALR, and (C) GCT, respectively. SR indicates step rate; VALR, vertical average loading rate; GCT, ground contact time.

Citation: Journal of Applied Biomechanics 2020; 10.1123/jab.2019-0337

Discussion

In this study, we compared SR, VALR, and GCT when running in standardized lab shoes and their own normal running shoes. This comparison was made in runners who heelstruck in both shoes and ran at the same self-selected speed in both shoes, as the footstrike pattern and speed both affect running mechanics independent of footwear.11,12 The first hypothesis, that SR, VALR, and GCT would be similar between the 2 shoe conditions, was not supported. We found that VALR was significantly greater when the participants ran in their normal running shoes. The second hypothesis, that between-subjects variance in SR, VALR, and GCT would be similar between the 2 shoe conditions, was supported.

The VALR result is important because this variable is often suspected to affect injury risk.1,4,13,14 VALR is higher in injured runners prospectively4 and retrospectively.1,13 These studies used a standardized lab shoe1,4 or a combination of standardized shoes and runners’ normal shoes.13 Our results indicate that using standardized lab shoes during testing may underestimate the loads runners actually experience during their typical mileage, while not providing the intended benefit of reducing variance due to differences in footwear. Differences in GRF also suggest that there could also be footwear differences in joint moments. We did not perform these more complicated analyses in this article, but they could be meaningful, as joint moments have been associated with some running injuries.28

In the current study, the SD of VALR was ∼11 BW/s in both shoe conditions, compared with ∼18 to 20 BW/s in uninjured rearfoot strikers in a standardized lab shoe.1,4,13 This is important because of the relationship between variance and SD (variance = SD2) and considering that the goal of using a controlled shoe condition is ostensibly to decrease variance due to differences in subjects’ normal running shoes. These outcomes suggest that differences in shoe construction and between-subject variance may be unrelated. The underestimation of runners’ loads in standardized shoes may affect the ecological validity of results; therefore, further investigations on the effects of using standardized shoes on running mechanics are warranted. The mean VALR in both shoe conditions was lower than previous studies using a standardized lab shoe. This is likely due to a difference in running speed since relatively small changes in speed can lead to substantial differences in VALR.10 Running at 3.0 to 3.5 m/s resulted in VALR magnitudes of ∼45 to 54 BW/s,29 similar to the ∼42 to 47 BW/s, we found at an average speed of 3.4 m/s.

The wide range of shoe models and wear in participants’ normal shoes is a major limitation of this study. It is possible that the age of the shoes worn in each condition led to the difference in VALR between shoe conditions. Accumulating mechanical loading equivalent to 660 km over 2 months led to stiffer soles and decreased energy absorption capacity in running.15 We would anticipate shoe mileage sufficient to affect midsole stiffness to subsequently affect mechanics, and only 200 miles (320 km) of running led to higher GCT but similar loading rates in shoes with different levels of cushioning.19 The lab shoes were new and unused prior to this study, and most lab shoes presumably have little mileage compared with running shoes used outside the lab, while the participants’ running shoes likely varied substantially in the accumulated mileage. Therefore, these results may not generalize to lab shoes that are old and have accumulated substantial mileage. Conversely, while the effects of mileage on shoe properties may contribute to differences in gait mechanics, this factor may not explain the differences between shoe conditions in the current study. The similar variability between conditions indicates that, when running in different types of conventional running shoes, regardless of shoe age, the variance in the data is a result of individual subject differences, speed differences, differences in the fit or comfort of shoes, or some other factor, but not due to differences in shoe characteristics. This similarity in variance also provides further support for the “preferred movement path” paradigm that suggests runners maintain their running kinematics when wearing a range of similar running shoes.30 If the goal of using lab shoes is to reduce variance in outcome variables due to subjects running in different shoes, the present results suggest that lab shoes may not necessarily achieve this. In addition, there is a risk; lab shoes could induce running mechanics that do not reflect how the individual runs outside the lab, further limiting the applicability of lab-based outcomes to realistic training conditions. When running in the standardized lab shoes versus their own normal shoes, 75% of the participants ran at a faster speed, 68% of the participants decreased SR, 79% increased VALR, and 52% decreased GCT. An increase in VALR is consistent with faster speed10 and lower SR31; however, whether SR, VALR, or GCT increased or decreased when running in the standardized lab shoes versus their own shoes was not predictable or consistent in individual participants, regardless of speed differences.

There are 2 additional limitations to this study. First is that all participants ran first in their own and then in the lab shoes. This choice was deliberate in an attempt to imitate the typical use of lab shoes, where subjects come into the lab and change into the lab shoes after being used to running in their own shoes, but with this design, it is possible an order effect is present in the data. However, since speed and footstrike pattern were similar between conditions, and since the total amount of running was relatively short, the influence of order or the related influence of fatigue seem unlikely to have substantially influenced the data. Second, the sample of participants was heterogeneous, consisting of a wide range of age, experience, and mixed gender, which resulted in a large variance in outcomes for both shoe conditions. Gait mechanics differences that allow runners to be classified based on experience and gender32 are especially important when between-runner comparisons are being made. In the current study, we instead chose to include runners based on footstrike pattern since this running characteristic has an important effect on GRF characteristics. Relatedly, we did not control for running speed quantitatively because of the range of experience and performance level of the participants. Rather, we allowed the participants to run at a self-selected speed based on a qualitative description. This may also have an effect on the results, considering that gait mechanics change as runners run faster or slower.10

In summary, in a cohort that runs in a range of conventional running shoes, using another conventional model of standardized lab shoe did not affect SR or GCT, but indeed did decrease VALR, while not reducing variability in these variables between subjects, a major purpose of using standardized shoes. We suggest that the practice of using lab shoes is in need of careful consideration, especially when studying injuries. A larger study with a more thorough examination of the effects of lab shoes on running mechanics is warranted, particularly on injury-related variables.

Acknowledgments

This study was supported by a grant from Maryland Technology Enterprises Institute. The authors have no conflicts of interest to disclose.

References

  • 1.

    Milner CE, Ferber R, Pollard CD, Hamill J, Davis IS. Biomechanical factors associated with tibial stress fracture in female runners. Med Sci Sports Exerc. 2006;38(2):323328. PubMed ID: 16531902 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2.

    Franz JR, Wierzbinski CM, Kram R. Metabolic cost of running barefoot versus shod: is lighter better? Med Sci Sports Exerc. 2012;44(8):15191525. PubMed ID: 22367745 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Gruber AH, Umberger BR, Braun B, Hamill J. Economy and rate of carbohydrate oxidation during running with rearfoot and forefoot strike patterns. J Appl Physiol. 2013;115(2):194201. PubMed ID: 23681915 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Davis IS, Bowser BJ, Mullineaux DR. Greater vertical impact loading in female runners with medically diagnosed injuries: a prospective investigation. Br J Sports Med. 2016;50(14):887892. PubMed ID: 26644428 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.

    Hamill J, Bates BT, Knutzen KM, Sawhill JA. Variations in ground reaction force parameters at different running speeds. Hum Mov Sci. 1983;2:4756. doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6.

    Reinschmidt C, Nigg BM. Current issues in the design of running and court shoes. Sport Sport. 2000;14:7184.

  • 7.

    Hreljac A. Impact and overuse injuries in runners. Med Sci Sports Exerc. 2004;36(5):845849. PubMed ID: 15126720 doi:

  • 8.

    Munro CF, Miller DI, Fuglevand AJ. Ground reaction forces in running: a reexamination. J Biomech. 1987;20(2):147155. PubMed ID: 3571295 doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9.

    Nigg BM, Bahlsen H, Leuthi S, Stokes S. The influence of running velocity and midsole hardness on external impact forces in heel-toe-running. J Biomech. 1987;20(10):951959. PubMed ID: 3693376 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10.

    Hunter JG, Garcia GL, Shim JK, Miller RH. Fast running does not contribute more to cumulative load than slow running. Med Sci Sport Exerc. 2019;51(6):11781185. doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11.

    Mercer JA, Horsch S. Heel-toe running: a new look at the influence of foot strike pattern on impact force. J Exerc Sci Fit. 2015;13(1):2934. PubMed ID: 29541096 doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Orendurff MS, Kobayashi T, Tulchin-Francis K, et al. A little bit faster: lower extremity joint kinematics and kinetics as recreational runners achieve faster speeds. J Biomech. 2018;71:167175. PubMed ID: 29472010 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Futrell EE, Jamison ST, Tenforde AS, Davis IS. Relationships between habitual cadence, footstrike, and vertical load rates in runners. Med Sci Sports Exerc. 2018;50(9):18371841. PubMed ID: 29614001 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Chan ZYS, Zhang JH, Au IPH, et al. Gait retraining for the reduction of injury occurrence in novice distance runners: 1-year follow-up of a randomized controlled trial. Am J Sports Med. 2018;46(2):388395. PubMed ID: 29065279 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15.

    Chambon N, Sevrez V, Ly QH, Guéguen N, Berton E, Rao G. Aging of running shoes and its effect on mechanical and biomechanical variables: implications for runners. J Sports Sci. 2014;32(11):10131022. PubMed ID: 24576090 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16.

    Udofa AB, Clark KP, Ryan LJ, Weyand PG. Running ground reaction forces across footwear conditions are predicted from the motion of two body mass components. J Appl Physiol. 2019;126(5):13151325. PubMed ID: 30763160 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Addison BJ, Lieberman DE. Tradeoffs between impact loading rate, vertical impulse and effective mass for walkers and heel strike runners wearing footwear of varying stiffness. J Biomech. 2015;48(7):13181324. PubMed ID: 25814181 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18.

    Mann R, Malisoux L, Urhausen A, Statham A, Meijer K, Theisen D. The effect of shoe type and fatigue on strike index and spatiotemporal parameters of running. Gait Posture. 2015;42(1):9195. PubMed ID: 25953506 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19.

    Kong PW, Candelaria NG, Smith DR. Running in new and worn shoes: a comparison of three types of cushioning footwear. Br J Sports Med. 2009;43(10):745749. PubMed ID: 18801775 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20.

    Bates BT. An assessment of subject variability, subject-shoe interaction, and the evaluation of running shoes using ground reaction force data. J Biomech. 1983;16(3):181191. PubMed ID: 6863333 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.

    Altman A, & Davis I. A kinematic method for footstrike pattern detection in barefoot and shod runners. Gait Posture. 2012;35(2):298300. PubMed ID: 22075193 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    Krupenevich RL, Pruziner AL, Miller RH. Knee joint loading during single-leg forward hopping. Med Sci Sports Exerc. 2017;49(2):327332. PubMed ID: 27669448 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.

    Blackmore T, Willy RW, Creaby MW. The high frequency component of the vertical ground reaction force is a valid surrogate measure of the impact peak. J Biomech. 2016;49(3):479483. PubMed ID: 26783094 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.

    R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing; Vienna, Austria. 2013. www.R-project.org/.

    • Search Google Scholar
    • Export Citation
  • 25.

    Fox J, Weisberg S. An {R} Companion to Applied Regression. 2nd ed. 2011. http://socserv.socsci.mcmaster.ca/jfox/Books/Companion.

  • 26.

    Lakens D. Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Front Psychol. 2013;4:863. PubMed ID: 24324449 doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Willwacher S, König M, Braunstein B, Goldmann JP, Brüggemann GP. The gearing function of running shoe longitudinal bending stiffness. Gait Posture. 2014;40(3):386390. PubMed ID: 24882222 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Stefanyshyn DJ, Stergiou P, Lun VMY, Meeuwisse WH, Worobets JT. Knee angular impulse as a predictor of patellofemoral pain in runners. Am J Sports Med. 2006;34(11):18441851. PubMed ID: 16735584 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Hobara H, Baum BS, Kwon HJ, et al. Amputee locomotion: lower extremity loading using running-specific prostheses. Gait Posture. 2014;39(1):386390. PubMed ID: 24035367 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30.

    Nigg BM, Vienneau J, Smith AC, Trudeau MB, Mohr M, Nigg SR. The preferred movement path paradigm: Influence of running shoes on joint movement. Med Sci Sports Exerc. 2017;49(8):16411648. PubMed ID: 28277405 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31.

    Heiderscheit BC, Chumanov ES, Michalski MP, Wille CM, Ryan MB. Effects of step rate manipulation on joint mechanics during running. Med Sci Sports Exerc. 2011;43(2):296302. PubMed ID: 20581720 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32.

    Kobsar D, Clermont CA, Ferber R, Osis ST, Benson LC. Running patterns for male and female competitive and recreational runners based on accelerometer data. J Sports Sci. 2018;37(2):204211. doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

Hunter, Smith, Sciarratta, Shim, and Miller are with the Department of Kinesiology, University of Maryland, College Park, MD, USA. Suydam is with Algovation, LLC, Chicago, IL, USA. Shim and Miller are also with the Neuroscience & Cognitive Science Program, University of Maryland, College Park, MD, USA. Shim is also with the Department of Mechanical Engineering, Kyung Hee University, Seoul, South Korea.

Hunter (jghunter@umd.edu) is corresponding author.
  • View in gallery

    Conceptual framework for the second hypothesis based on Bates et al,20 demonstrating hypothetical responses to the vertical ground reaction force loading rate when running in different shoes. The highly variable responses in A are more typical than the consistent responses in B. Both data sets have the same mean and variance.

  • View in gallery

    (A) SR, (B) VALR, and (C) GCT of participants running at a self-selected “normal” speed in a standardized lab shoe and their own normal running shoes. Connected gray dots show individual subjects, and group means and SDs are represented by white triangles and solid black lines, respectively. SR indicates step rate; VALR, vertical average loading rate; GCT, ground contact time. *Significance of P < .05.

  • View in gallery

    Between-subject variance of participants running in a standardized lab shoe (black) versus their own normal running shoes (gray). Variance in lab shoes compared with normal running shoes was 4.4% lower, 4.8% higher, and 31.2% higher for (A) SR, (B) VALR, and (C) GCT, respectively. SR indicates step rate; VALR, vertical average loading rate; GCT, ground contact time.

  • 1.

    Milner CE, Ferber R, Pollard CD, Hamill J, Davis IS. Biomechanical factors associated with tibial stress fracture in female runners. Med Sci Sports Exerc. 2006;38(2):323328. PubMed ID: 16531902 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2.

    Franz JR, Wierzbinski CM, Kram R. Metabolic cost of running barefoot versus shod: is lighter better? Med Sci Sports Exerc. 2012;44(8):15191525. PubMed ID: 22367745 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3.

    Gruber AH, Umberger BR, Braun B, Hamill J. Economy and rate of carbohydrate oxidation during running with rearfoot and forefoot strike patterns. J Appl Physiol. 2013;115(2):194201. PubMed ID: 23681915 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.

    Davis IS, Bowser BJ, Mullineaux DR. Greater vertical impact loading in female runners with medically diagnosed injuries: a prospective investigation. Br J Sports Med. 2016;50(14):887892. PubMed ID: 26644428 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5.

    Hamill J, Bates BT, Knutzen KM, Sawhill JA. Variations in ground reaction force parameters at different running speeds. Hum Mov Sci. 1983;2:4756. doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6.

    Reinschmidt C, Nigg BM. Current issues in the design of running and court shoes. Sport Sport. 2000;14:7184.

  • 7.

    Hreljac A. Impact and overuse injuries in runners. Med Sci Sports Exerc. 2004;36(5):845849. PubMed ID: 15126720 doi:

  • 8.

    Munro CF, Miller DI, Fuglevand AJ. Ground reaction forces in running: a reexamination. J Biomech. 1987;20(2):147155. PubMed ID: 3571295 doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9.

    Nigg BM, Bahlsen H, Leuthi S, Stokes S. The influence of running velocity and midsole hardness on external impact forces in heel-toe-running. J Biomech. 1987;20(10):951959. PubMed ID: 3693376 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10.

    Hunter JG, Garcia GL, Shim JK, Miller RH. Fast running does not contribute more to cumulative load than slow running. Med Sci Sport Exerc. 2019;51(6):11781185. doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11.

    Mercer JA, Horsch S. Heel-toe running: a new look at the influence of foot strike pattern on impact force. J Exerc Sci Fit. 2015;13(1):2934. PubMed ID: 29541096 doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Orendurff MS, Kobayashi T, Tulchin-Francis K, et al. A little bit faster: lower extremity joint kinematics and kinetics as recreational runners achieve faster speeds. J Biomech. 2018;71:167175. PubMed ID: 29472010 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13.

    Futrell EE, Jamison ST, Tenforde AS, Davis IS. Relationships between habitual cadence, footstrike, and vertical load rates in runners. Med Sci Sports Exerc. 2018;50(9):18371841. PubMed ID: 29614001 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14.

    Chan ZYS, Zhang JH, Au IPH, et al. Gait retraining for the reduction of injury occurrence in novice distance runners: 1-year follow-up of a randomized controlled trial. Am J Sports Med. 2018;46(2):388395. PubMed ID: 29065279 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15.

    Chambon N, Sevrez V, Ly QH, Guéguen N, Berton E, Rao G. Aging of running shoes and its effect on mechanical and biomechanical variables: implications for runners. J Sports Sci. 2014;32(11):10131022. PubMed ID: 24576090 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16.

    Udofa AB, Clark KP, Ryan LJ, Weyand PG. Running ground reaction forces across footwear conditions are predicted from the motion of two body mass components. J Appl Physiol. 2019;126(5):13151325. PubMed ID: 30763160 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.

    Addison BJ, Lieberman DE. Tradeoffs between impact loading rate, vertical impulse and effective mass for walkers and heel strike runners wearing footwear of varying stiffness. J Biomech. 2015;48(7):13181324. PubMed ID: 25814181 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18.

    Mann R, Malisoux L, Urhausen A, Statham A, Meijer K, Theisen D. The effect of shoe type and fatigue on strike index and spatiotemporal parameters of running. Gait Posture. 2015;42(1):9195. PubMed ID: 25953506 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19.

    Kong PW, Candelaria NG, Smith DR. Running in new and worn shoes: a comparison of three types of cushioning footwear. Br J Sports Med. 2009;43(10):745749. PubMed ID: 18801775 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20.

    Bates BT. An assessment of subject variability, subject-shoe interaction, and the evaluation of running shoes using ground reaction force data. J Biomech. 1983;16(3):181191. PubMed ID: 6863333 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.

    Altman A, & Davis I. A kinematic method for footstrike pattern detection in barefoot and shod runners. Gait Posture. 2012;35(2):298300. PubMed ID: 22075193 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22.

    Krupenevich RL, Pruziner AL, Miller RH. Knee joint loading during single-leg forward hopping. Med Sci Sports Exerc. 2017;49(2):327332. PubMed ID: 27669448 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23.

    Blackmore T, Willy RW, Creaby MW. The high frequency component of the vertical ground reaction force is a valid surrogate measure of the impact peak. J Biomech. 2016;49(3):479483. PubMed ID: 26783094 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24.

    R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing; Vienna, Austria. 2013. www.R-project.org/.

    • Search Google Scholar
    • Export Citation
  • 25.

    Fox J, Weisberg S. An {R} Companion to Applied Regression. 2nd ed. 2011. http://socserv.socsci.mcmaster.ca/jfox/Books/Companion.

  • 26.

    Lakens D. Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Front Psychol. 2013;4:863. PubMed ID: 24324449 doi:

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Willwacher S, König M, Braunstein B, Goldmann JP, Brüggemann GP. The gearing function of running shoe longitudinal bending stiffness. Gait Posture. 2014;40(3):386390. PubMed ID: 24882222 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28.

    Stefanyshyn DJ, Stergiou P, Lun VMY, Meeuwisse WH, Worobets JT. Knee angular impulse as a predictor of patellofemoral pain in runners. Am J Sports Med. 2006;34(11):18441851. PubMed ID: 16735584 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29.

    Hobara H, Baum BS, Kwon HJ, et al. Amputee locomotion: lower extremity loading using running-specific prostheses. Gait Posture. 2014;39(1):386390. PubMed ID: 24035367 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30.

    Nigg BM, Vienneau J, Smith AC, Trudeau MB, Mohr M, Nigg SR. The preferred movement path paradigm: Influence of running shoes on joint movement. Med Sci Sports Exerc. 2017;49(8):16411648. PubMed ID: 28277405 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31.

    Heiderscheit BC, Chumanov ES, Michalski MP, Wille CM, Ryan MB. Effects of step rate manipulation on joint mechanics during running. Med Sci Sports Exerc. 2011;43(2):296302. PubMed ID: 20581720 doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32.

    Kobsar D, Clermont CA, Ferber R, Osis ST, Benson LC. Running patterns for male and female competitive and recreational runners based on accelerometer data. J Sports Sci. 2018;37(2):204211. doi:

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 383 383 100
PDF Downloads 81 81 27