Reliability and validity of pressure and temporal parameters recorded using a pressure-sensitive insole during running

- The Runalyser is valid and reliable when assessing strike pattern in running.
- Temporal parameters of the Runalyser were validated against the gold standard.
- Strike index provides more meaningful information as a continuous measure.
- The Runalyser is suitable for, and easily applicable in running biomechanics.

Running biomechanics has received increasing interest in recent literature on running-related injuries, calling for new, portable methods for large-scale measurements. Our aims were to define running strike pattern based on output of a new pressure-sensitive measurement device, the Runalyser, and to test its validity regarding temporal parameters describing running gait. Furthermore, reliability of the Runalyser measurements was evaluated, as well as its ability to discriminate different running styles. Thirty-one healthy participants (30.3 ± 7.4 years, 1.78 ± 0.10 m and 74.1 ± 12.1 kg) were involved in the different study parts. Eleven participants were instructed to use a rearfoot (RFS), midfoot (MFS) and forefoot (FFS) strike pattern while running on a treadmill. Strike pattern was subsequently defined using a linear regression (R2 = 0.89) between foot strike angle, as determined by motion analysis (1000 Hz), and strike index (SI, point of contact on the foot sole, as a percentage of foot sole length), as measured by the Runalyser. MFS was defined by the 95% confidence interval of the intercept (SI = 43.9-49.1%). High agreement (overall mean difference 1.2%) was found between stance time, flight time, stride time and duty factor as determined by the Runalyser and a force-measuring treadmill (n = 16 participants). Measurements of the two devices were highly correlated (R ≥ 0.80) and not significantly different. Test-retest intra-class correlation coefficients for all parameters were ≥0.94 (n = 14 participants). Significant differences (p < 0.05) between FFS, RFS and habitual running were detected regarding SI, stance time and stride time (n = 24 participants). The Runalyser is suitable for, and easily applicable in large-scale studies on running biomechanics.