# Which pronator are you? New perspectives from an unsupervised clustering approach in running

## TL;DR

This study identifies three distinct pronation movement patterns in runners and proposes updated thresholds for excessive pronation to better assess injury risk.

## Contribution

The study introduces a new unsupervised clustering approach to classify pronation patterns and proposes updated excessive pronation thresholds.

## Key findings

- Three distinct functional pronation movement patterns were identified in runners.
- Specific thresholds for excessive pronation were proposed based on kinematic variables from clusters 1 and 2.
- The findings suggest that separating runners by pronation type could improve injury risk assessment.

## Abstract

Introduction: Excessive pronation is still considered as a factor partially involved in the running-related injury risk despite inconsistent evidence. The between-runner variability and the way excessive pronation is quantified are potentially involved issues. The purpose of this study is to highlight the different functional pronation movements in runners to be able to propose suitable and up-to-date excessive pronation thresholds.

Methods: 234 (overground) and 190 (treadmill) recreational runners ran at a self-selected speed while the lower limb dynamic pronation was measured with skin-mounted markers to calculate seven common pronation-related kinematic variables of the rearfoot and the tibia.

Results: These variables were shown to provide different, complementary, information regarding pronation, which influenced the unsupervised hierarchical clustering. Three distinctive functional pronation movements were identified: fast with large excursions (cluster 1), prolonged with high peaks (cluster 2), and overall low (cluster 3). Excessive pronation thresholds were proposed from the typical variables of clusters 1 and 2: −16.9° of maximal rearfoot eversion, 25.8° of rearfoot excursion, −10.0° of maximal tibia internal rotation, 20.2° of tibia internal rotation excursion, −849 °/s of maximal rearfoot eversion velocity, 0.273 s of rearfoot eversion duration, and 1.18 of ratio of excursion of the rearfoot eversion to the tibia internal rotation.

Discussion: To get greater evidence of pronation involvement in injury risk, future studies should refer to these results to separate runners with and without excessive pronation for studying the effect of a given intervention on these groups and/or for assessing them in longitudinal follow-up studies.