Open Access Highly Accessed Research article

Comparison of vertical ground reaction forces during overground and treadmill running. A validation study

Bas Kluitenberg1*, Steef W Bredeweg1, Sjouke Zijlstra1, Wiebren Zijlstra23 and Ida Buist1

Author Affiliations

1 Center for Sports Medicine, University Medical Center Groningen, Hanzeplein 1, Groningen, GZ, 9713, The Netherlands

2 Center for Human Movement Sciences, University Medical Center Groningen, P.O. Box 196, Groningen, AD, 9700, The Netherlands

3 Institute of Movement and Sport Gerontology, German Sport University Cologne, Cologne, Germany

For all author emails, please log on.

BMC Musculoskeletal Disorders 2012, 13:235  doi:10.1186/1471-2474-13-235

Published: 27 November 2012



One major drawback in measuring ground-reaction forces during running is that it is time consuming to get representative ground-reaction force (GRF) values with a traditional force platform. An instrumented force measuring treadmill can overcome the shortcomings inherent to overground testing. The purpose of the current study was to determine the validity of an instrumented force measuring treadmill for measuring vertical ground-reaction force parameters during running.


Vertical ground-reaction forces of experienced runners (12 male, 12 female) were obtained during overground and treadmill running at slow, preferred and fast self-selected running speeds. For each runner, 7 mean vertical ground-reaction force parameters of the right leg were calculated based on five successful overground steps and 30 seconds of treadmill running data. Intraclass correlations (ICC(3,1)) and ratio limits of agreement (RLOA) were used for further analysis.


Qualitatively, the overground and treadmill ground-reaction force curves for heelstrike runners and non-heelstrike runners were very similar. Quantitatively, the time-related parameters and active peak showed excellent agreement (ICCs between 0.76 and 0.95, RLOA between 5.7% and 15.5%). Impact peak showed modest agreement (ICCs between 0.71 and 0.76, RLOA between 19.9% and 28.8%). The maximal and average loading-rate showed modest to excellent ICCs (between 0.70 and 0.89), but RLOA were higher (between 34.3% and 45.4%).


The results of this study demonstrated that the treadmill is a moderate to highly valid tool for the assessment of vertical ground-reaction forces during running for runners who showed a consistent landing strategy during overground and treadmill running. The high stride-to-stride variance during both overground and treadmill running demonstrates the importance of measuring sufficient steps for representative ground-reaction force values. Therefore, an instrumented treadmill seems to be suitable for measuring representative vertical ground-reaction forces during running.

Running; Kinetics; Biomechanics; Validity; Overuse injuries