Wrist-worn Accelerometry for Runners: Objective Quantification of Training Load.

Area of Science:

• Sports Science
• Biomechanical Engineering
• Wearable Technology

Background:

• Objective monitoring of physical activity is crucial in sports science.
• Traditional methods like training logs can be subjective and burdensome.
• Wearable accelerometers offer a potential solution for objective data collection.

Purpose of the Study:

• To apply open-source code to wrist-worn accelerometer data for accurate "running" vs. "nonrunning" day classification.
• To develop and compare accelerometer-derived external training load metrics with self-report measures.
• To assess the utility of accelerometry in objective athlete monitoring.

Main Methods:

• Utilized 7-day wrist-worn accelerometer data from 35 experienced runners.
• Date-matched accelerometer data with self-reported training logs.
• Employed receiver operating characteristic analyses and linear regression for metric validation.

Main Results:

• "Most Active-30mins" and "Mins≥400 mg" metrics achieved >94% accuracy in classifying running days.
• Accelerometer metrics explained a high percentage of variance (67%-76%) in self-reported miles, duration, and training load.
• Validation confirmed the robustness of these accelerometer-derived metrics.

Conclusions:

• Wrist-worn accelerometers can objectively and accurately identify running training days in runners.
• This reduces reliance on training logs or user input for prospective research and commercial tracking.
• Accelerometer-derived metrics support the use of accelerometry for prospective, preventative, and prescriptive athlete monitoring.