Lower-Limb Exoskeletons Appeal to Both Clinicians and Older Adults, Especially for Fall Prevention and Joint Pain Reduction

Exoskeletons are a burgeoning technology with many possible applications to improve human life; focusing the effort of exoskeleton research and development on the most important features is essential for facilitating adoption and maximizing positive societal impact. To identify important focus areas for exoskeleton research and development, we conducted a survey with 154 potential users (older adults) and another survey with 152 clinicians. The surveys were conducted online and to ensure a consistent concept of an exoskeleton across respondents, an image of a hip exoskeleton was shown during exoskeleton-related prompts. The survey responses indicate that both older adults and clinicians are open to using exoskeletons, fall prevention and joint pain reduction are especially important features, and users are likely to wear an exoskeleton in the scenarios when it has the greatest opportunity to help prevent a fall. These findings can help inform future exoskeleton research and guide the development of devices that are accepted, used, and provide meaningful benefit to users.


I. INTRODUCTION
N EARLY 40% of older adults (≥65 years of age) report having a disability, and two-thirds of them have difficulty walking or climbing stairs [1].Falling is also a major mobility risk for older adults, with nearly one in three older adults falling each year [2] and falls occurring in various locations of the home, care facility, and outdoors [3], [4], [5].Many older adults with mobility challenges use mobility aids; Of the 55.8 million older adults currently living in the U.S. [6], 15% use one mobility device, and 9% use multiple [7].Aging causes a wide range of physical and sensory degradations [8], and older adults may be prescribed or recommended a mobility aid for a variety of reasons, such as a balance impairment, muscle weakness, joint pain, peripheral neuropathy, visual impairment, or vestibular impairment [9], [10], [11], [12].Balance is often an explicit reason to prescribe a mobility aid, and many of the aforementioned symptoms for which a mobility device may be prescribed, such as joint pain [13] and sensory impairment [10], are associated with increased fall risk.Mobility aids serve a wide range of purposes and balance assistance is especially important.
While commonly prescribed, mobility aids are imperfect.The limitations of mobility devices are best shown by the continued prevalence of mobility problems such as falls, fear of falling, and joint pain.Shortcomings of current mobility aids can be broadly categorized into two groups: functional and acceptance limitations.Some examples of functional limitations summarized in previous work [9] include: rollators cannot be used on stairs, occupy both hands, and likely increase the energy cost to walk; walkers have the same limitations and require enough strength to lift or slide; white canes can fail to detect some obstacles, don't work in some terrain (rough ground, snow, etc.), and occupy one hand; load-bearing canes occupy one hand, make the user susceptible to slipping, increase energy cost to walk, and can impede balance in some settings.These functional limitations of mobility aids can be summarized by limiting use of upper extremities, impairing balance in some instances, increasing energy to move, and having strength requirements that make them unsuitable for some patients [9].One significant reason that older adults reject mobility aids is stigma [14].No matter how effective the device is when it is used, if the potential user rejects it, the device will not assist them.Therefore, in addition to addressing the functional limitations, device designers should consider that acceptance is critical for device success.
Exoskeletons are a promising technology that has the potential to address the functional limitations of current mobility devices, do not occupy the hands, and may provide additional benefits.Recent work in lower-limb exoskeletons has shown improvements for energy savings in walking [15], [16], [17], [18], [19], [20], running [21], [22], [23], walking on inclines [24], walking with a heavy load [25], increasing walking speed [26], sit-to-stand transitions [27], and joint offloading [28].One device designed and shown to assist walking also influences movements associated with balance, indicating that a single device may be able to assist in multiple ways [29].Simulation work even shows that exoskeletons could directly offset age-related tissue changes [30].Finally, in addition to assisting an older adult while worn, exoskeletons may be able to provide rehabilitative training that benefits a user even while not wearing the device [31].
Exoskeletons have the potential to provide many functional benefits, but it is unclear which should be prioritized for development.Exoskeletons could improve balance, reduce joint pain, raise overall activity levels, speed recovery from injury/surgery, and improve quality of life through increased independence for a wide range of patient populations.Given all of these potential benefits, it is essential to identify important aspects of mobility to monitor and prioritize during future exoskeleton development.
In addition to understanding the prioritization for the functional benefits of exoskeletons, it is critical to understand how well exoskeletons could overcome the acceptance limitations of current mobility aids.Many factors contribute to device acceptance and some important questions surrounding acceptance include: Will exoskeletons appeal to these potential users more than current mobility aids?Will clinicians be open to trying exoskeletons with their patients?Which device features are most important to patients?Which features are most important to clinicians?Do the two groups agree?During which activities/scenarios would potential users be most open to using an exoskeleton?Given the importance of balance assistance for mobility aids, when and where do falls occur?Would potential users be willing to wear an exoskeleton in a scenario in which they are likely to fall?Acceptance of an exoskeleton will depend on individual needs and preferences [32], [33], so identifying common needs and preferences is important for informing future exoskeleton research.
We aim to answer these questions to help guide exoskeleton development toward effective designs that meaningfully improve the lives of older adults with mobility impairments.The purpose of this study was to gauge the likelihood of acceptance and to identify high-priority functional benefits of lower limb exoskeletons for potential users and healthcare providers.We interviewed and surveyed older adults and clinicians and hope the results of these surveys help guide the next era of exoskeleton research.

II. METHODS
In this study, we first conducted semi-structured interviews with a small sample of older adults and clinicians.We used these pilot interviews to refine the preliminary designs of two larger surveys, one with older adults and one with clinicians.Finally, we analyzed the survey responses to identify important features, use cases, and appeal for a potential exoskeleton device.

A. Pilot Interviews
We piloted our surveys with semi-structured, one-on-one interviews with older adults who were potential users for an exoskeleton device (n=11) and with physiatristss and physical therapists (n=9).These interviewees were screened to include individuals who were most likely to be prescribed/prescribe an exoskeleton to aid mobility.Using these pilot interviews as guides, we designed and disseminated a potential user survey with older adults and a clinician survey to identify the relative importance of various potential qualities of an exoskeleton product.The pilot interview screening used nearly identical inclusion/exclusion criteria as the surveys with one notable difference being that in both the clinician and older adult interview, participants were chosen to be most knowledgeable and receptive to new technology.We screened for these respondents in the interviews as they could most clearly articulate the reasons a device would or would not be adopted for use.Interviews were solely used inform survey design and to pilot potential survey questions.Thus, we did not perform qualitative analysis on the interview transcripts.We have included PDF files of the interview guides in the Supplementary Material.

B. Potential User Survey 1) Survey Respondent Population:
We had a goal of receiving responses from 150 older adults with quotas on fall history, age, and primary language spoken to ensure a reasonably diverse survey sample.We required at least 75 participants to be between 60 and 74 years of age, and 75 participants to be ≥75 years of age.We also required at least 75 participants to have experienced a fall in the last year and 75 participants to not have experienced a fall in the last year.And finally, of the 150 target respondents, we required that a minimum of 20 primarily speak Spanish, and another 20 primarily speak Chinese (Mandarin).Respondents were admitted to the study if they passed our exclusion criteria: (1)  Older adults who had opted in to receive market research opportunities from SIS, a marketing research firm specializing in health care and consumer research surveys, were invited to participate in the survey.Participants provided informed consent prior to participating in the survey and were compensated for their time according to SIS policies.
In the older adult survey, a total of 154 potential users met our inclusion criteria and completed the questionnaire.Of the respondents, 49 self-identified as female (105 male), 78 participants were 60-74 years of age (76 were ≥75 years of age), 76 reported falling in the last year (78 had not fallen in the last year).Of the 154 respondents, 20 primarily speak Spanish, 20 primarily speak Chinese (Mandarin) and 114 primarily speak English.Respondents lived in all geographic regions of the US and had similar rates of disease as the national medicare average, with the exception of high blood pressure (hypertension), which was lower than the national average [34].Table I contains more information on respondent age, home life, and health conditions.Despite a few notable exceptions of being wealthier, older, larger percent male, and having a higher frequency of falling than the US population of older adults [34], [35], [36], this sample is sufficiently representative to offer important insights for exoskeleton research.
2) Older Adult Survey Prompts: The survey prompts fall into the following categories: demographic, medical history, activity habits, interest in technology, experience with medical devices, desired features for mobility devices, and assessment of a potential exoskeleton provided.A list of all survey prompts is provided in the Supplementary Material and the following paragraphs contain a list of summarized questions for each category.
The demographic prompts included: user language, gender, age, dwelling type, zip code, education level, household income, medical device company affiliation, work status, work location, and number of household residents.
The medical history prompts included: pre-existing medical conditions, frequency of falls in previous year, details of most recent fall (time, location, and severity), types of support needed for daily activities, joint replacement(s), spinal surgery or injections, and diagnosis of obesity/overweight.
The activity habit prompts included: regularly drive a motor vehicle, frequency of trips outside the home, difficulty of climbing stairs unassisted, difficulty of walking for 30 minutes outside, and attitude toward managing health and wellness.
The interest in new technology and experience with medical devices prompts included: enthusiasm for new technology, familiarity with mobility aids (e.g.canes, walkers, etc.), current use of mobility aids, previous use of mobility aids for surgery/accident recovery, and prescriber of mobility device.
The prompts on the desired features of a potential exoskeleton focused on the importance of select benefits and usability requirements.Prompts on device benefits included the importance to: reduce falls, trips, or stumbles; enhance balance on uneven surfaces; increase walking stamina; assist sit-tostand transitions; reduce joint pain; increase exercise stamina; and increase household activity stamina.Prompts on usability requirements included the importance of being: light and comfortable; easy to put on without assistance; as easy to learn as a knee or hip brace; primarily paid for by insurance; usable for someone overweight; easy to clean; and comfortable enough for sleeping.
The exoskeleton assessment prompts focused on the appeal and likelihood of use for the pictured hip exoskeleton.The exoskeleton image showed a commercially produced bilateral hip exoskeleton (Honda Walking Assist Device) worn by a user walking in front of a grass field.A nearly identical image is shown in Fig. 1 and the exact image is shown in [37].For a useful reference point, the prompt on overall appeal asked the respondent to rate the appeal of the pictured exoskeleton relative to a walker.Prompts focusing on potential use cases included the likelihood to use the exoskeleton for: recovery from an injury/surgery, everyday use, outdoor exercise, shopping/errands, a doctor's appointment, travelling on an airplane, climbing stairs, and chores/hobbies around the home.Lastly, respondents were asked to imagine beyond the pictured hip exoskeleton and report their preferred joint to have an exoskeleton assist as well as the rationale for selecting that joint.

C. Clinician Survey 1) Survey Respondent Population:
We aimed to receive responses from 150 clinicians meeting our inclusion criteria of being a practicing physical therapist with a clinical specialization in general practice, orthopedics, geriatrics, or sports, or a medical doctor specializing in rehabilitation medicine (physiatrist).Additionally, the survey respondents were screened out based on the following exclusion criteria: (1) if the respondent, their spouse, or children work for or consult with a pharmaceutical, medical device, or diagnostics company; (2) if <30% of their time was spent treating patients over the age of 65; (3) if they self-reported as being "not at all familiar" or "slightly familiar" with "medical devices to minimize the risk of falls or support safe ambulation in seniors." Qualified clinicians who had opted in to receive market research opportunities from M3, a consulting group In the clinician survey, 152 clinicians completed the questionnaire, comprised of 114 physical therapists and 38 physiatrists.On average, each physical therapist treated 52 people over 65 years old each month, and each physiatrist treated 118.For both physical therapists and physiatrist respondents, the majority of their patients over 65 had osteoarthritis in the hips, knees, ankles, or spine.For both groups, approximately 40% of patients over 65 had fallen in the past year.On average, the physical therapist's prescribed 15 mobility aids per month, and the physiatrists prescribed 60 each month.More complete information on clinician practice and the types of patients they treat can be found in Table II.Overall, this information confirms that our clinician survey respondents are familiar with the medical needs of older adults, familiar with current mobility aids, and likely representative of the broader population of clinicians prescribing mobility aids for older adults in the U.S.
2) Clinician Survey Prompts: The survey prompts fall into the following categories: clinical practice, patient population, device preferences, and attitude toward new technology.A list of all survey prompts is provided in the Supplementary Material and the following paragraphs contain a list of summarized questions for each category.
The clinical practice information prompts included: medical device company affiliation, country working in, percentage of time treating patients ≥65 years of age, familiarity with medical devices for walking or fall prevention in older adults, number of prescriptions written for these devices each month, medical profession, clinical or medical specialization, medical practice description (e.g.physical therapy clinic, home healthcare clinic, etc.), ownership of primary place of work (e.g.private practice, health system, etc.), and number and type of providers part of primary work location (e.g.registered nurse, physician's assistant, etc.).
The patient population prompts included an estimate of the number of patients over 65 years of age seen per month.Additionally, the survey included detailed prompts on these patients, such as the amount that: have chronic or acute neuromuscular condition; are underweight, overweight, or obese; have osteoarthritis in the hip, knee, ankle, or spine; are rehabilitating from joint replacement, spine surgery, or trauma; experienced a fall in the last year; use a walking aid, orthosis, or other mobility device; and have a cognitive impairment potentially preventing them from understanding instructions to use a medical device.
The device preferences prompts include the importance for functionality, usability, and joint location.The functionality preference prompts included the importance of a range of potential benefits: reduce falls, trips and stumbles; enhance balance on uneven surfaces; increase 6 minute walk distance during use; improves 30 second sit-to-stand test during use; reduces joint pain during use; increases user's physical activity; and does not cause muscle deconditioning.The usability preference prompts included a variety of usability metrics: light and comfortable; easy to put on without assistance; as easy to learn as a knee or hip brace; costs no more than jointoffloading orthoses; cost to the patient is primarily covered by insurance; and sized for use by patients of up to 40 BMI.The joint location preference prompts asked for a ranking of highest priority joints to have exoskeleton support among hip, knee, ankle, and lumbar spine, and the associated functional or clinical use.
The prompts related to new technology include: stated interest in new technology, most recent mobility aid or rehabilitation product adopted into clinical practice, likelihood to prescribe an exoskeleton device for patient use in a rehabilitative clinic setting or everyday use in the next 5 years, and functional or rehabilitative use to indicate prescription of the exoskeleton (including diagnosis/clinical situation).

III. RESULTS AND DISCUSSION
In this section, we highlight some particularly interesting results and the implications for exoskeleton design.We provide the full data set including survey prompts, respondent demographics, and responses in the Supplementary Material.
Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.

A. Selected Older Adult Survey Results
The older adults surveyed were experienced with mobility aids, with 78% reporting using a mobility device in daily use, rehabilitation, or both (as shown in Fig. 2).The large number of experienced users surveyed indicates the stated preferences are well informed and valuable for device designers.
Older adults overwhelmingly found the exoskeleton to be more appealing than a walker.Of the survey respondents, 90% found the exoskeleton to be "more appealing" or "much more appealing" than a walker, as shown in Fig. 2.This may mean that there is less stigma associated with the exoskeleton and that potential users could be more open to using it.
Older adult respondents were interested in using an exoskeleton in a large range of scenarios.The majority of older adult respondents said they were "extremely" or "moderately" likely to wear an exoskeleton in a wide variety of activities, such as household chores, outdoor exercise, and travel.Fig. 3 shows a graphical representation of these data.The wide range of scenarios in which older adults were open to using an exoskeleton indicates that if an effective, well-designed exoskeleton were available, older adults may be open to trying it in a range of scenarios.
Respondents indicated that all of the proposed features were important, and reducing joint pain was especially important.Older adult respondents rated "reduc[ing] joint paint" as "extremely important" more frequently than any other proposed feature, and more frequently than a potential exoskeleton device being "primarily covered by insurance," as shown in Fig. 4.This result indicates that potential users want an exoskeleton that reduces joint pain and even if this device was not primarily covered by insurance, may still be commercially successful.
While creating a device that provides all of the features listed would be ideal, prioritization is critical to product development.We asked respondents to rank features by importance to identify top priorities for device designers.Older adults ranked "reduces joint pain;" "reduces falls, trips, or stumbles;" "enhances balance on uneven surfaces;" and "assists sit-tostand transitions" as the most important feature, listed in decreasing priority.All four of these features were ranked as top priority more frequently than the exoskeleton "cost primarily covered by insurance" as shown in Fig. 5.This result again emphasizes the users' priority to reduce joint pain.It also highlights the importance of balance and assisting with sit-to-stand transitions, an activity essential to maintain-ing independence and mobility.The top rankings given to these features relative to other device features and insurance coverage further indicates the likely commercial success of a device that includes them.
Many falls occur during the daytime in various locations around the home as well as outside [3], [4], [5], [38].The falls reported by our respondents broadly match these trends-79% reported falls occurring during the daytime and 51% required medical attention (Fig. 6).These results aligning with previous studies increases our confidence that our sample is usefully representative of the general older adult population.The scenarios in which older adults said they were likely to use the exoskeleton (Fig. 3) overlap with the scenarios in which falls occur, suggesting that if a balance-assisting exoskeleton were commercially available and effective in preventing falls, it is feasible that older adults would wear it when a fall may occur and could therefore help prevent the fall.

B. Selected Clinician Survey Results
Surveyed clinicians showed interest in exoskeletons, with 60% reporting they were "moderately" to "extremely" likely to use an exoskeleton in rehabilitation clinic setting in the next 5 years, and 3% already using an exoskeleton in a rehabilitation clinic (shown in Fig. 7).Clinicians indicating they are open to using exoskeletons with their patients suggests that if an effective exoskeleton existed, clinicians would recommend it to their patients, increasing the likelihood it reaches patients.
In addition to interest in using exoskeletons inside a clinic, 55% of clinicians said they would prescribe or recommend an exoskeleton for use outside of a rehabilitation clinic setting in the next five years as shown in Fig. 7.The interest in using an exoskeleton with patients inside and outside of a clinic setting further demonstrates a likely acceptance of exoskeletons as medical devices by medical care providers.
Clinician respondents reported that reducing falls, trips, or stumbles and enhancing balance on uneven surfaces was most important, followed by not causing muscle atrophy, increasing user's physical activity, and reducing joint pain, listed in descending order of importance, as shown in Fig. 8.This list of important features highlights the importance of balance and fall prevention as a top priority for medical care providers, and further emphasizes the importance of joint pain reduction.

C. Discussion
Combining the results from the clinician and older adult surveys yields important insights for exoskeleton design.Overall, exoskeletons seem to appeal to both groups, both are interested in balance assistance and joint pain reduction, and older adults seem willing to wear an exoskeleton when it will have the greatest opportunity to help prevent a fall.
Both potential users and clinicians are open to using exoskeletons.As shown in Fig. 2, nearly all of the older adults surveyed already have experience with devices intended to aid their mobility such as canes, walkers, crutches, and braces.These experienced users found the presented exoskeleton more appealing than a walker, as shown in Fig. 2. Clinicians who often prescribe canes, walkers, crutches, and braces were   open to prescribing an exoskeleton device (Fig. 7).These results indicating that clinicians are willing to prescribe an exoskeleton and older adults find it more appealing than current mobility aid options suggest that clinicians and older Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.adults would at least be willing to try an exoskeleton as a mobility aid.
Clinicians and older adults both list balance as one of the most important ways in which the device can assist users.Other top attributes for older adults include reducing joint pain as shown in Fig. 4 and Fig. 5, and for clinicians include increasing physical activity, not causing muscle atrophy, and reducing joint pain, as shown in Fig. 8. Interestingly, older adults ranked joint pain reduction as a more important feature than balance, while clinicians placed more emphasis on balance enhancement and fall prevention.While this difference in ordering is notable, both groups emphasized joint pain reduction and balance as important features, indicating that they are the most important features to develop for exoskeletons to be effective mobility aids.
The scenarios in which older adults said they would use the exoskeleton are also when many falls occur, as shown in Fig. 3 and Fig. 6, respectively.These results indicate that the exoskeleton would likely be worn during a possible loss of balance, making it feasible to intervene and prevent a potential fall.
The combined results of these surveys suggest that exoskeletons hold enough potential as mobility aids to be tried by both clinicians and older adults.Clinicians are open to prescribing them, older adults find them more appealing than current mobility aids, older adults and clinicians are aligned in wanting joint pain reduction and balance assistance, older adults are interested in trying them during a wide range of activities, and the times an older adult is most open to wearing the device is when falls are most likely to occur.

D. Limitations
There are several limitations to our study that are worth noting and may affect how this work informs future research.One limitation is inherent to the study type.Responses to survey questions are not guaranteed to accurately predict sustained use of a device.Despite that limitation, our results of older adults indicating positive impressions of the new exoskeleton technology match a trend of older adults using technology more than in the past [39].Additionally, our study design closely resembles methods from previous work studying technology adoption in older adults: interviews followed by questionnaires on acceptance criteria and use cases for the proposed technology [40].While predicting the sustained use of a product is challenging, previous work indicates that questionnaires conducted after a short introduction to a new computer system reasonably predict the real-world use of the computer after several months [41].One important factor in long-term adoption of the technology is the perceived usefulness, and participant ratings of usefulness after several months of use explain even more variance of which users fully adopt the technology.These findings indicate that future work studying exoskeleton adoption should enable study participants to try functional prototypes that demonstrate potential product usefulness to provide even more accurate predictions of the likelihood of exoskeleton adoption.Although the results of this study suggest that exoskeletons appeal to older adults and clinicians, assessing how well the technology would be adopted and used long-term remains an open question for future work.
Additional limitations specific to our study includes the respondents opting-in to receive market-research surveys, demographics, and size of our study sample.For both the clinician and older adult survey, respondents opted-in to receive offers to participate in market research surveys.It is possible that the population of respondents that opt-in to participate are not fully representative of the general population.The gender ratio in our survey respondents is 68% male (32% female).That ratio is substantially different from the population of adults over 60 years of age in the U.S., which is 46% male (54% female) [42].This deviation in the sample from the known population demographics may be meaningful for some future research, as falls are affected by exposure.Specifically, there are more opportunities to fall during periods of activity and as men and women have different habits of activity in location and time, this affects when and where falls occur and consequently when it is most beneficial to wear an exoskeleton to prevent a fall [38].Additionally, our sample of older adult respondents does not include individuals living in older adult care facilities, and as activity habits for those individuals likely differ from community-dwelling older adults, our results may not generalize well to that population.In our clinician survey, clinicians report that 26% of their patients over 65 have cognitive issues that limit their ability to understand how to take medication or use medical products.We are likely not fully capturing the preferences of this group in the older adult survey as respondents were screened out based on a diagnosis of Alzheimer's or dementia.Designing a device for users with cognitive impairments could influence device priorities in obvious ways, such as requirements for ease of use, and more subtle ways, such as requirements for compatibility with changes in activity levels associated with cognitive decline [43].It is also worth noting that our sample of older adults uses mobility aids in everyday use more than the U.S. population of older adults, 39% and 24%, respectively [7].This is likely due in part to our sample consisting of 50% adults ≥75 years of age, while in the U.S. population, about 30% of adults in the ≥60 years of age group are ≥75 years age [6].This difference in age ratio is likely a factor, as age has been shown to correlate with device use [7].Additionally, 150 older adults is a relatively small sample size compared to the U.S. population, and some noise is to be expected.This deviation from the U.S. population is not a limitation for our analysis, as we were hoping to get opinions from informed device users, but is worth noting for other researchers who may want to use the provided data for additional analysis.
Due to the small sample size of the older adult respondents relative to the US population, detailed analyses of categorical effects are not feasible.Coarse analyses across gender, age, and fall history match the overall trends reported in this paper for both exoskeleton appeal and priority of exoskeleton functionality.Unpaired t-tests show no statistically significant difference between genders, age, or fall history for exoskeleton appeal (p-value = ≥ 0.10).One interesting finding from these analyses showed that non-fallers ranked joint pain reduction as their top priority (followed by balance), while fallers prioritized balance (followed by joint pain reduction).We include plots of these categorical analyses as well as all survey responses in the Supplementary Material for those who may wish to use the data for additional analysis.It is also worth noting that since our sample was wealthier than the average older adult in the US [35], cost may be a more significant factor for future products than responses from our sample indicate.Assessing the importance of exoskeleton price is also further limited because we were not able to provide a price for the exoskeleton shown to respondents.As the technology required for the hypothetical benefits described in the survey is not yet mature, providing an accurate price estimate is difficult, so we chose to not include a potentially inaccurate price of the exoskeleton in order to avoid biasing the respondents.This lack of price may have influenced enthusiasm for the exoskeleton, especially when comparing the exoskeleton to existing, potentially less expensive, mobility aids (e.g.walker).
Additionally, the lack of price may have biased respondents to value insurance coverage of the exoskeleton less than they would in real life.Although there are limitations to this study, we believe the results are still instructive for exoskeleton designers in identifying important design features for future research and product development.

IV. CONCLUSION
We conducted two surveys, focusing on both clinicians and potential users of an exoskeleton (older adults).The survey responses indicate that exoskeletons appeal to both groups, that fall prevention and joint pain reduction are the highest priority features, and that users are likely to wear an exoskeleton when it will have the greatest opportunity to prevent a fall.These results suggest that an exoskeleton could be a successful mobility aid, and emphasize the importance for exoskeleton designers to focus on balance and joint pain reduction in order to provide the most successful and beneficial product for society.
This study identified broad patient and clinician priorities from remote surveys.While the results from these surveys are informative, they are limited by the remote nature of the study.Future studies focusing on patient and clinician preferences for exoskeletons could conduct in-person device demonstrations or trial periods with interviews and surveys assessing the usertested hardware.This could lead to more nuanced and accurate feedback for exoskeleton designers, and ultimately accelerate the development of exoskeletons into effective mobility aids that greatly benefit society.

Fig. 1 .
Fig. 1.An image of the Honda Walking Assist Device very similar to the one shown to questionnaire respondents.Courtesy of American Honda Motor Co., Inc.

Fig. 2 .
Fig.2.Older adult respondent experience with mobility devices and appeal of exoskeleton relative to walker.

Fig. 3 .
Fig. 3. Older adult likelihood of use in provided scenarios.

Fig. 4 .
Fig. 4. Older adult feature importance.Cost is highlighted in grey as an important reference point.

Fig. 5 .
Fig. 5. Older adult device feature priority ranking.Cost is highlighted in grey as an important reference point.

Fig. 6 .
Fig. 6.Fall time, location, and severity of most recent fall, self-reported in the older adult survey.

Fig. 7 .
Fig. 7. Clinician likelihood to prescribe exoskeleton for use inside or outside a clinic setting.

TABLE I OLDER
ADULT RESPONDENT DEMOGRAPHICS