Are Gait Patterns during In-Lab Running Representative of Gait Patterns during Real-World Training? An Experimental...

Area of Science:

• Biomechanics
• Sports Science
• Wearable Technology

Background:

• Biomechanical assessments of running are typically conducted in controlled laboratory settings.
• The representativeness of in-lab gait data for real-world running conditions remains unclear.

Purpose of the Study:

• To evaluate how well in-lab gait data represent real-world running gait patterns.
• To assess the utility of consumer-grade wearable sensors in capturing gait variability.

Main Methods:

• Two cohorts of runners (N=49 and N=19) were equipped with wearable sensors to collect gait data (speed, step length, vertical oscillation, stance time, leg stiffness).
• Data were collected during both in-lab treadmill sessions and multiple real-world runs.
• Univariate and multivariate overlap statistics were used to quantify the agreement between in-lab and real-world gait data.

Main Results:

• Individual in-lab gait data poorly represented real-world gait (32.5% overlap for all metrics).
• Pooling data across multiple subjects significantly improved distributional overlap (89.3-90.3%) compared to individual data.
• Excluding non-flat, non-straight segments of real-world running did not improve data overlap.

Conclusions:

• Individual biomechanical gait patterns observed in laboratory settings are not representative of a runner's real-world gait.
• Utilizing pooled gait data from diverse runners can enhance the prediction of individual gait behavior for clinical and performance applications.