An Alternative Test to Evaluate the Assistance Provided by Back-Support Exoskeletons: A Pilot Study

OCCUPATIONAL APPLICATIONS The test proposed in this study, assimilable to a maximal isometric mid-thigh pull, appears to detect the acute assistance provided by the exoskeleton, expressed as the difference in performance with and without donning the device; the test can be performed with minimal equipment and setup, ensuring that the test can be replicated widely by practitioners and employed in an occupational setting. These preliminary results point to the test being feasible; additionally, the exoskeleton was perceived as giving support and being moderately comfortable. Future research could further validate the test by comparing different models of back-support exoskeletons, passive or active, different populations, and repeating the test over multiple sessions. TECHNICAL ABSTRACT Background: Despite the surge in researchers’ interest in exoskeletons and their diffusion in workplace settings, many tests either lack direct applicability or rely on purely subjective measures. This paper aims to propose an evaluation methodology to combine the reliability of objective measures with the interpretability of subjective ones, testing exoskeletons at high muscular force outputs. Methods: A simple test consisting of an isometric maximum mid-thigh pull was investigated. Thirty-four healthy adult males participated in the study. The participants performed three trials, both with and without an exoskeleton, in a random order, then filled in the NASA-TLX and an exploratory questionnaire on wearability and perception of the device. Results: Within-session reliability was good to excellent (ICC > 0.75), with ICC = 0.87 (95% CI = 0.74, 0.93) for the three lifts without exoskeleton, ICC = 0.91 (95% CI = 0.85, 0.95) for the three lifts with exoskeleton. Using the exoskeleton resulted in a significant performance increase (mean ± SD: 9.4 ± 16.5 kg, p = 0.0022), compared to the trials without the device, with an effect size of g: 0.56 (95% CI: 0.20, 0.92). Participants perceived the exoskeleton as safe and giving good support, although comfort was average and autonomous wearability was an issue. Two components were identified through exploratory factor analysis, dubbed performance support and usability. Overall, Cronbach’s alpha of the perception questionnaire (Likert items only) was atot = 0.80 (95% CI = 0.68, 0.87), which dropped to a1 = 0.78 for the first factor, and a2 = 0.69 for the second. Conclusions: The proposed test was feasible and reliable, and the exoskeleton model tested was associated with higher isometric strength production. Additionally, the device was perceived as safe and useful, although wearability remains a barrier.