Saskia Baltrusch

57 Chapter 3 General Discomfort: In the Exoskeleton condition, the participant was also asked to indicate the general discomfort of the device (“Indicate the comfort of the device in the task you just performed”). The participant had to place a cross on the VAS scale, that ranged from “very comfortable” to “very uncomfortable”. Local Discomfort: Local Discomfort was only assessed for the static holding tasks 3 and 4. After each task, the participant was asked to indicate the local discomfort experienced in different body areas on a VAS scale, ranging from “no discomfort” to “maximal discomfort”. The chosen body areas were chest, abdomen, upper back, lower back, upper legs ventral and upper legs dorsal side. This choice was made based on body areas that were expected to be loaded or unloaded by the device. User Impression: The overall user impression was assessed with a User’s Impression questionnaire, that the participants had to fill in at the end of the assessment. Adjustability, range of motion and efficacy of the device were addressed in the questionnaire through VAS scales (see Appendix 1 for more detailed information). 2.6 Statistical analysis To test for statistically significant differences in functional performance between Exoskeleton and Control condition, each of the objective outcome measures was analyzed with a paired t-test. For perceived task difficulty and local discomfort, the VAS scale values were compared between Exoskeleton and Control condition using the non-parametric Wilcoxon test, since there is sufficient evidence that VAS scales generate ordinal data [19]. To assess whether the objective outcome measures are related to general discomfort, a Spearman rank order correlation was conducted. Alpha of 0.05 was used as the critical level of significance. Corrections for multiple testing were made by dividing α by the number of tests performed for each outcome within each group of tasks. For the general discomfort and the user’s impression, descriptive data are presented, since these parameters were only assessed in the Exoskeleton condition. All statistical analyses were performed using SPSS for Windows (IBM, SPSS Statistics 23.0, USA). 3