A comparison of treadmill preferred walking speed assessments: Reliability of common protocols and the novel TreadPWS Test

BACKGROUND: Human locomotion studies often require participants to self-select a preferred walking speed (PWS), yet there is no consensus or standard protocol for determining this speed. We developed a new protocol to determine PWS (TreadPWS Test). RESEARCH QUESTION: We compared the TreadPWS Test to three other common protocols in the literature (Dingwell PWS Test, Jordan PWS Test, Dal PWS Test) across a range of reliability and implementation metrics. METHODS: Healthy participants (N = 20) attended two sessions within a seven-day period, during which a PWS was determined using each protocol. One test (Dal PWS Test) was conducted overground, and the resulting PWS was then applied to the treadmill. The other three tests were performed entirely on a treadmill. RESULTS: Of the four tests, the Dal PWS Test had the highest intraclass correlation coefficient (0.86) and produced the highest PWS (mean 1.56 m/s), which was rated the least comfortable (via a visual analogue scale) and had the highest rate of perceived exertion (Borg scale). The three treadmill-based tests had similar intraclass correlation coefficients (0.71-0.73) and speeds (1.25-1.37 m/s). Relative to the other treadmill-based tests, our TreadPWS Test (with a PWS of 1.34 m/s) had the lowest smallest detectable change (0.12 m/s), was administered in the shortest time (∼2 min), had comparable comfort ratings, and was associated with a 'very light' perceived exertion. SIGNIFICANCE: Our results enable researchers to compare common PWS tests across a range of reliability and implementation metrics and indicate that our TreadPWS Test is a reliable and quick-to-administer tool for determining the PWS of healthy young adults.