Data Processing Techniques May Influence Numerical Results and Interpretation of Single Leg Stance Test

Purposeto evaluate how different data sampling and different analysis methods may effect numerical results and interpretation of single leg stance test using parameters derived from CoP trajectories.MethodsThirty healthy active subjects were recruited for this study on voluntary. Each participant was asked to stand as still as possible for 20 s on the dominant limb, with the supporting foot placed on the force platform. Balancing task in two conditions, with eyes open (EO) and closed (EC). Three trials were collected for each condition.Medial-lateral and anterior-posterior CoP force platform data were obtained and downsampling techniques was applied to get data at original (500 Hz), 100 Hz and 20 Hz of sampling frequencies.Time series data were then analysed to get CoP variables including medial-lateral total path, anterior-posterior total path, total path, maximal excursion for the ML plane and maximal excursion for the AP plane. Sway area was evaluated as 95% confidence ellipse area (CEA) and as 95% prediction ellipse area (PEA)Main findingsSignificant different results were obtained for the same variable evaluated at different sampling frequency. In addition, at all sampling frequencies variables were significantly different (p.<0.05) between EO and EC conditions. High correlation (>0.9) between the same CoP variable calculated at different sampling frequencies was found for all CoP variables. Regarding sway area calculation, both methods were able to distinguish between EO and EC conditions and high correlation was found between CEA and PEA methods.ConclusionOverall results of this study demonstrated the importance of reporting data processing techniques, which includes sampling frequency and variable calculation methods, as they shown to influence one leg stance CoP results, thus data analysed in different manner cannot be directly compared. However, for the variables included in the study, researchers can choose preferred data collection and data analysis methods as they all return same data analysis interpretation as long as they keep consistency in the method.