Step Length, But Not Stepping Cadence, Strongly Predicts Physical Activity Intensity During Jogging and Running

Device-based measures often rely on the positive relationship between walking cadence and metabolic equivalents of task (METs) to estimate physical activity. It is unknown whether this relationship remains during jogging/running. The study purpose was to investigate the relationships between METs, cadence, and step length during walking and jogging/running. A treadmill protocol with 5 walking (3.2-6.4 km center dot hr(-1)) and 5 jogging/running stages (8.0-11.3 km center dot hr(-1)) was completed in 43 adults (23 +/- 5 years, 19 female). Predictors of METs during walking and jogging/running were determined by generalized mixed modeling. The strongest prediction models for walking (R-2 = 0.72, P < .001) and jogging/running (R-2 = 0.75, P < .001) included cadence(2), cadence, step length, age, and leg length (all, P < .001). Step length accounted for 49.1% and 78.3% of model variance during walking and jogging/running, respectively. METs are poorly estimated by cadence during jogging/running but step length reduces error. Strategies to measure step length in free-living settings could better predict physical activity intensity.