CHAPTER 3

Methodology

3.1 Participants

Twelve volunteers (6 male and 6 female) participated in the study (age 20.8 ± 1.1 years, mass 65.3 ± 8.3 kg, height 1.70 ± 0.18m). Informed written consent (appendix B) was obtained from each participant prior to data collection, and a pre-test medical questionnaire (appendix C) was also completed. The participants had previous experience of treadmill running and had no previous or current joint injury. Prior to data collection the study received ethical approval from Sheffield Hallam University Ethics Committee (appendix A).

3.2 Procedure

Prior to testing, superficial joint markers were placed on anatomical landmarks and therefore participants were required to wear minimal clothing. This was done following guidelines from Plagenhoef (1971). The anatomical landmarks defined included the upper spine (C7), the sacrum (S10), the hip, knee and ankle. The markers provided a colour contrast to both the skin of the participant and the background.

Each was participant was provided with sufficient time to complete a warm up. After completion of the warm up, the participants ran at 3.5m.s-1 in each condition:

·  Level running

·  Uphill Incline running at a gradient of 3°

·  Uphill Incline running at a gradient of 6°

The trials were randomised by condition and the time between repetitions (5 min) was sufficient for the participants to recover fully (McArdle et al., 1991). At each condition the participants ran for 1 minute, which allowed time for the participant to adjust to the speed and incline of the treadmill. Video data was captured for the last ten seconds of each minute trial using a Sony DCR-TVR950e video camera, operating at 50Hz with a shutter speed of 1000. The camera was mounted on a stationary rigid tripod, placed at a height of 95cm, ten metres from the plane of filming, with the optical axis perpendicular to the sagittal plane. A calibration frame (1m by 1m) was filmed. The tests took place indoors on a treadmill (H-P-Cosmos-Kistler Gaitway II S). The field of view was prepared so that the background was plain and non-shiny, and sufficient artificial lighting was provided.

3.3 Data Analysis

The calibration and running trials were manually digitised using Simi° Motion (Version 5.0, Unterschleissheim, Germany) to create a model consisting of the upper spine (C7), the sacrum (S10), the hip, knee and ankle. For each trial digitising started from touchdown of the right foot through the stride to when the right foot made contact with the ground again. Touchdown was defined as the instant at which the participant made contact with the ground and take off was defined as the instant that the foot broke contact with the ground. The appropriate frames defining touchdown were identified through visual inspection by the researcher who digitised all the trials. One stride was digitised for each participant at each condition. The data was smoothed using a 2nd order low-pass Butterworth filter, with a cut off frequency of 10Hz.

The digitised image that was produced allowed joint segment angles to be calculated. For the touchdown and take-off frames, the angles of the knee, hip and trunk were calculated. The knee angle was calculated by using a three point angle convention, using the hip, knee and ankle as the anatomical markers. The hip angle was calculated using a 4 point angle convention, using C7, the sacrum, hip and knee as the anatomical markers. The trunk angle was determined relative to the horizontal. The line between the upper spine (C7) and the sacrum (S10) was used relative to a horizontal line.

Figure 1: Location of the body landmarks and visualisation of the angles. In the study, knee (a), hip (b) and trunk (d) angle were measured and these have therefore been highlighted for clarity. (Adapted from Paradisis and Cooke, 2001)

The spatio-temporal variables that were assessed included stride length, stride frequency, time spent in contact phase and time spent in non-contact phase. The average time for each stride was calculated, by recording the time for ten strides from the video, and then divided by ten. This then allowed the stride frequency and stride length to be calculated accurately. The following equation was used to calculate stride length:

Running Velocity (V) = Stride Frequency (Hz) * Stride Length (m)

3.4 Statistical Analysis

All the data was analysed using the statistical computer software package SPSS (SPSS version 12.0, Chicago, Illinois, USA). A one-way analysis of variance (ANOVA) was used to establish if there were any significant differences between the conditions. Each dependent variable was analysed using a separate ANOVA. A post-hoc Tukey test was used to determine the differential variability between conditions.