Comparison of Tibial Shock during Treadmill and Real-World Running

Caleb D. Johnson; Jereme Outerleys; Steve T. Jamison; Adam S. Tenforde; Matthew Ruder; Irene S. Davis


Med Sci Sports Exerc. 2020;52(7):1557-1562. 

In This Article

Abstract and Introduction


The degree to which standard laboratory gait assessments accurately reflect impact loading in an outdoor running environment is currently unknown.

Purpose: To compare tibial shock between treadmill and road marathon conditions.

Methods: One hundred ninety-two runners (men/women, 105/87; age, 44.9 ± 10.8 yr) completed a treadmill gait assessment while wearing a tibial-mounted inertial measurement unit, several days before completing a marathon race. Participants ran at 90% of their projected race speed and 30 s of tibial shock data were collected. Participants then wore the sensors during the race and tibial shock was averaged over the 12th, 23rd, and 40th kilometers. One-way analysis of covariance and correlation coefficients were used to compare vertical/resultant tibial shock between treadmill and marathon conditions. Analyses were adjusted for differences in running speed between conditions.

Results: A significant main effect of condition was found for mean vertical and resultant tibial shock (P < 0.001). Early in the marathon (12-km point), runners demonstrated higher mean tibial shock adjusted for speed compared with the treadmill data (vertical = +24.3% and resultant = +30.3%). Mean differences decreased across the course of the marathon. Vertical tibial shock at the 40th kilometer of the race was similar to treadmill data, and resultant shock remained higher. Vertical and resultant tibial shock were significantly correlated between treadmill and the 12th kilometer of the race (r s = 0.64–0.65, P < 0.001), with only 40% to 42% of the variance in outdoor tibial shock explained by treadmill measures. Correlations for tibial shock showed minimal changes across stages of the marathon.

Conclusions: These results demonstrate that measures of impact loading in an outdoor running environment are not fully captured on a treadmill.


Impact loading has been associated with a number of common overuse injuries in runners.[1–4] These impact loads have primarily been measured by the vertical ground reaction force. Vertical load rates during running are characterized by a sharp increase in vertical ground reaction forces as the foot contacts the ground. Tibial accelerometry has also been used to capture impact characteristics, producing valid surrogates of ground reaction force measures.[5] Peak tibial acceleration, or tibial shock, has been shown to be strongly correlated with vertical loading rates[6,7] and is sometimes used as a surrogate for loading rates. However, the majority of these studies have been conducted in laboratory environments and on a treadmill. Competitive and recreational runners complete most of their training outdoors.[8] Consequently, it is important to establish whether loading profiles measured in laboratory conditions are related to those seen in a runner's natural outdoor environment.

Previous work has provided some indirect evidence that impacts may differ between laboratory and outdoor conditions. Running mechanics are known to be affected by running speed,[9,10] increased perceptual demands,[11,12] fatigue,[13,14] and running surface/slope.[15–17] The majority of evidence suggests that outdoor conditions would result in overall mean increases in impact loading. For example, moving from a rubber to asphalt surface, similar to treadmill to road conditions, has been shown to increase vertical load rates.[15] Small deviations in anterior/posterior surface slope have been shown to increase peak vertical impact forces.[16] Increasing perceptual demands, such as shifting visual gaze, has also been shown to increase tibial shock.[11] Fatigue also manifests itself differently between treadmill and outdoor running. Previous studies have noted increases in impact loading after a fatiguing treadmill protocol.[13,14] However, in a previous report of our marathon data, we found that runners decreased their speed as they progressed through a marathon.[18] This resulted in decreased mean impact loading toward the end of the race, despite the presence of fatigue.[18] It is likely that these collective factors will result in alterations to impact loading during outdoor running compared with treadmill running.

Despite this identified need and rationale, no studies to date have directly compared impact loading between laboratory and outdoor environments. Previously, this was due to the limitations in technology and ease of data collection/analysis. With the recent proliferation of wearable devices, specifically inertial measurement units (IMU), it is now possible to measure tibial shock in an outdoor environment.[19] Measuring tibial shock during outside running may provide more ecologically valid information regarding the relationship between impact loading and running injury. However, it is first important to understand the relationships for tibial shock measures between laboratory and outside conditions.

Therefore, the purpose of the current study was to compare tibial shock between treadmill and outdoor running conditions. Based on the previously described factors that can influence impact loading, we hypothesized that mean tibial shock would be higher during the early stage of the outdoor compared with treadmill condition. Second, we expected that mean differences in tibial shock would decrease as runners progressed through the race, due to fatigue-related changes in running speed.[18] Regarding the correlation in tibial shock measures between conditions, no a priori hypotheses were made as these relationships are relatively unexplored.