Occupational exoskeletons are designed to reduce workplace injury risk by decreasing work demands. Due to their relatively recent development, there has been limited research into potential unintended and undesirable consequences of wearing them. The goal of this thesis was to investigate the effects of exoskeleton use on reactive balance in response to simulated slips and trips. Five representative exoskeletons were investigated including leg-, back, and shoulder-support exoskeletons. This thesis consists of two studies: a smaller study investigating one exoskeleton and a larger one investigating multiple exoskeletons. Participants stood on a specialized treadmill, then abruptly and unexpectedly changing treadmill belt speed to simulate trip-like forward losses of balance or slip-like backward losses of balance. The results of the first study showed that a passive leg-support exoskeleton adversely reactive balance for both slips and trips. The results of the second study showed that back-support exoskeletons had a greater adverse effect on reactive balance compared to shoulder-support exoskeletons for both slips and trips. These exoskeletons affected reactive balance due to their interaction with stepping kinematics and movement constraints. This thesis provides important information that can be used to warn users of potential increased fall risks and inform exoskeleton manufacturers who may be able to modify designs to reduce any additional fall risk. / Master of Science / Occupational exoskeletons reduce muscle workload for workers during manual tasks. However, because of their additional weight and how they restrict movement, they can increase the risk of falling after a slip or a trip. The goal of this thesis was to see how exoskeletons affect balance after simulated slips and trips. Five exoskeletons were studied; These exoskeletons supported the legs, back, and shoulders. This thesis includes two studies: a smaller study with one exoskeleton and a larger one with multiple exoskeletons. In order to simulate a slip and trip, participants stood on a treadmill and then the treads would unexpectedly accelerate to a speed to make them lose their balance. The results of the first study showed that an exoskeleton that supported the legs negatively affected balance for both slips and trips. The results of the second study showed that exoskeletons that supported the back negatively affected balance more than those that supported the shoulders for both slips and trips. These exoskeletons affected balance due to them interacting with the legs and affected stepping. This thesis provides important information that can be used to warn workers of potential increased fall risks and inform exoskeleton manufacturers who may be able to help reduce any fall risk.

Problem Statement: Stepping is a common strategy after a perturbation. Stroke survivors display a predilection for stepping with non-paretic leg. Insight into induced stepping between stroke survivors and age-matched control may guide our understanding for reactive postural control training post stroke. Purpose: To investigate the difference in perturbation-induced stepping between chronic stroke survivors and age-matched controls at three phases of the stepping response: preparation, execution, and landing and association with clinical outcome measures. Procedure: Twenty-one community-dwelling chronic stroke survivors (mean age 59y/o ±13yrs) and 17 age- and gender-matched controls (mean age 54.4y/o ±17yrs) completed this study. Clinical measures of gait, balance, range, sensation, and motor control were assessed. A mechanical weight drop of 10% body weight (BW) was used to create the anterior waist pull perturbation during three stance symmetry positions: equal stance (EQ) and two asymmetrical stance (70% BW on dominant leg and 70% BW on nondominant leg). Ten perturbation trials plus two catch trials at 2% BW were given in a standard randomly order at the three stance positions. Kinematic and kinetic data was collected for perturbation steps. Results: The asymmetrical trials resulted in two types of stepping response, steps with the leg bearing 70% BW (loaded steps – LS) and steps with the leg that had 30% BW (unloaded steps – ULS). All subjects initiated steps more often with their unloaded leg (ULS) in the asymmetrical stance trials. In the stroke group the ULS increased paretic leg stepping compared to EQ (p=0.001) and LS (p=0.001). The stroke group had significantly earlier APA onset with both non-paretic leg (p=0.003) and paretic leg (p=0.028), took significantly more steps with paretic (p=0.01) and non-paretic (p=0.07), shorter step length (paretic, p=0.025 and non-paretic p=0.003), and less change in momentum at landing with paretic leg (p=0.01) compared to controls. Conclusion: Reacting to a perturbation is more challenging for chronic stroke survivors than age- and gender-matched control subjects in the preparation, execution, and landing phase of the stepping response regardless of the leg used. Perturbation training should include stepping with both non-paretic and paretic leg.

Slips, trips, and falls are a serious public health concern, particularly among older adults and within occupational settings, given that falls contribute to a large number of injuries and associate with high medical costs. To reduce the number of falls, there is a need to better understand risk factors contributing to falls, and to develop and evaluate improved balance training interventions to prevent falls. To address these needs, this work has two primary goals: first, to better understand risk factors contributing to falls, including fatigue and balance reactions after a large postural perturbation, and, second, to develop and evaluate improved reactive balance training (RBT) interventions to reduce risk of falls due to slipping and tripping. The first study investigated the effects of performing occupationally-relevant fatigue-inducing physical work on trip and fall risk. Healthy young adults performed a simulated manual material handling (MMH) task, using either heavy or light boxes, for two hours. Gait measures related to risk of tripping and slipping were assessed before and after the task. Reactive balance during one laboratory-induced trip was also assessed after the task. Results showed that performing the heavy MMH task did not affect risk of tripping or slipping, or reactive balance after tripping. These results may have resulted from insufficient fatigue due to the MMH task. The second study investigated the relationship between feet kinematics upon slipping while walking, and the outcome of the slip. Seventy-one laboratory-induced slips were analyzed, which included recoveries, feet-split falls, feet-forward falls, and lateral falls. Feet kinematics differed between these four slip outcomes, and a discriminant model including six measures of feet kinematics correctly predicted 87% of slip outcomes. Two potentially modifiable characteristics of feet kinematics upon slipping that can improve the likelihood of successfully averting a fall were identified: (1) quickly arresting the motion of the slipping foot; and (2) a recovery step that places the trailing toe approximately 0-10% body height anterior to the sacrum. This information may be used to guide the development of improved RBT interventions to reduce risk of slip-induced falls. The third study evaluated the efficacy of two low-cost, low-tech RBT methods for improving reactive balance after slipping. The two methods were: unexpected slip training (UST), which involved repeated unexpected slips while walking and volitional slip-recovery training (VST), which involved practicing balance reactions after volitionally inducing a slip-like perturbation. Young adults completed one session of an assigned intervention (UST, VST, or control), followed by one unexpected, laboratory-induced slip while walking. Compared to controls, UST and VST resulted in a higher proportion of successful balance recoveries from the laboratory-induced slips. UST improved both proactive control and reactive stepping after slipping, while VST primarily improved the ability to arrest slipping foot motion. These results support the use of UST and VST as practical, low-tech methods of slip training. The fourth study evaluated the efficacy of RBT that targets both slipping and tripping. Community-dwelling, healthy older adults (61-75 years) completed four sessions of either RBT (treadmill-based trip-recovery training and VST) or control training (general strength and balance exercises). Reactive balance during unexpected laboratory-induced slips and trips was assessed before and after RBT, and compared between subjects at baseline (before the intervention), after control training, and after RBT. The incidence of slip-induced falls differed between groups in that 80% fell at baseline, 60% fell after control training, and 18% fell after RBT. Post-RBT subjects also exhibited less severe slips, compared to baseline and post-control subjects. The incidence of trip-induced falls did not differ between groups, but margin of stability after tripping was greater for post-RBT subjects, compared to post-control subjects. These results show promise for the use of RBT applied to both slipping and tripping to reduce fall risk among older adults. / Doctor of Philosophy / Slips, trips, and falls are a serious public health concern, given that falls contribute to a large number of injuries and deaths. Falls are particularly concerning among older adults, who are reported to fall more frequently, and within occupational settings, where falls cause a larger number of injuries and a significant economic burden. To reduce the number of falls, there is a need to better understand risk factors contributing to falls, and to develop and evaluate improved balance training interventions to prevent falls. Four studies were conducted to address these needs: two studies aimed to better understand risk factors contributing to falls, including fatigue and balance reactions after slipping, and two studies aimed to develop and evaluate improved balance training interventions to reduce risk of falls due to slipping and tripping. This work focused on slipping and tripping, because slips and trips are reported to cause a large number of injuries and falls among both workers and older adults. The first study investigated the effect of performing occupationally-relevant fatigue-inducing physical work on trip and fall risk among healthy young adults, and results showed that performing a simulated manual material handling task (i.e. moving and stacking boxes using a two-wheeled dolly) did not affect risk of tripping and falling. The second study investigated the relationship between balance reactions after slipping and the outcome of the slip. Results showed that balance reactions of the feet predicted the outcome of the slip (i.e. recovering balance or one of three types of slip-induced falls) with 87% accuracy. We also identified characteristics of balance reactions that can improve the likelihood of successfully averting a fall. The third study evaluated the efficacy of two low-tech reactive balance training (RBT) methods for reducing slip-induced fall risk among young adults. These methods involved practicing balance reactions after slip-like perturbations, induced either unexpectedly or volitionally. Results showed that both RBT methods improved reactive balance after slipping, but through different mechanisms. The fourth study evaluated the efficacy of a RBT intervention targeting both slipping and tripping among older adults. Results showed that RBT improved reactive balance during both slipping and tripping, and reduced the incidence of slip-induced falls. In conclusion, these results help to better understand risk factors contributing to falls, and support the use of practical reactive balance training interventions targeting both slipping and tripping to reduce fall risk.

No description available.