Perspectives of End Users on the Potential Use of Trunk Exoskeletons for People With Low-Back Pain: A Focus Group Study

Introduction

Low-back pain, often termed a pandemic of the modern world, represents a large socio-economic burden. In the Global Burden of Disease Study from 2010 (Institute for Health Metrics and Evaluation), low-back pain was ranked highest in terms of years lived with disability. A variety of risk factors are believed to contribute to its incidence. Based on epidemiological research, three groups of risk factors for low-back pain have been identified: biomechanical, psychosocial, and individual (Balagué, Mannion, Pellisé, & Cedraschi, 2012; Griffith et al., 2012).

Many studies have investigated the effects of ergonomic interventions, aiming to reduce biomechanical risk factors (Davis & Marras, 2000; Kingma et al., 2004). However, ergonomic interventions are rather limited in their applicability, and reductions in back loading are attenuated by changes in lifting behavior (Faber, Kingma, & van Dieën, 2007; Hoozemans, Kingma, & de Vries, 2008). With recent advances in robotic technologies, a shift has occurred toward external lifting devices that help the user to handle loads that are too heavy or require awkward lifting postures. Still, for handling loads within human capacity, users often prefer manual lifting, as external lifting devices are often slow to use and not always easily accessible (Kazerooni, 2002). This suggests that a wearable lifting device may be more usable. Several studies have found that wearing devices that passively support the user’s trunk reduces spinal loading during lifting, bending, and static holding tasks (Abdoli-Eramaki, Stevenson, Reid, & Bryant, 2007; Graham, Agnew, & Stevenson, 2009; Ulrey & Fathallah, 2013; Wehner, Rempel, & Kazerooni, 2009). Thus, developing such an exoskeleton might be helpful to reduce the risk of low-back pain.

Besides biomechanical risks, psychosocial risk factors need to be considered in this context. Systematic reviews have revealed that psychological factors, such as stress, anxiety, pain behavior, and somatization, play a significant role in the transition from acute to chronic low-back pain (Linton, 2000; Pincus, Burton, Vogel, & Field, 2002). Exposing individuals to negative psychosocial environments or demanding mental tasks during lifting also leads to increased spinal loading (Davis, Marras, Heaney, Waters, & Gupta, 2002; Marras, Davis, Heaney, Maronitis, & Allread, 2000), indicating an interaction between biomechanical and psychological risk factors. Therefore, end users’ perceptions of their environment, such as social, emotional, and occupational aspects, need to be considered to truly minimize the risk of low-back pain (Marras, 2012).

O’Sullivan (2011) already emphasized the great need to better understand the complex mechanisms underlying low-back pain by using qualitative research methods, such as interviews and focus groups, to listen to the patient’s story and explore individual beliefs. Such a user-centered approach, when used in medical device development, improves patient safety (Leape, 2009) and health outcomes (Gosbee, 2002), increases user satisfaction (Harrison, Dowswell, & Milewa, 2002), and ensures usability and functionality (De Vito Dabbs et al., 2009). Involving end users in the early development of assistive technology can help to provide a way of formulating or verifying design specifications (Clarke et al., 2011). Previous studies on developing exoskeletons have not considered psychological factors through involvement of end users. A literature review by Hill, Holloway, Morgado Ramirez, Smitham, and Pappas (2017) has shown that the scientific community responsible for developing exoskeleton technology has not sought to understand the needs and desires of individuals who may ultimately benefit from using an exoskeleton.

For assistive technology for people with low-back pain, both patients and health care professionals should be included as potential end users. According to Darlow et al. (2012), there is strong evidence that health care professionals’ attitudes and beliefs are associated with the beliefs of their patients, indicating that both perspectives are important to identify criteria for an exoskeleton. Health care professionals may be in a position to facilitate or limit uptake of new technology due to communication strategies toward the patients. Therefore, the aim of this focus group study was to identify criteria to be considered in the early development phase of an exoskeleton by exploring the perceptions and expectations of patients and health care professionals toward the potential use of an exoskeleton.

Method

We used focus group discussions to explore and clarify the perceptions of patients and health care professionals as potential end users regarding the usability of exoskeletons. A focus group discussion is a qualitative research method that explicitly exploits group interaction to help participants to identify their own views by discussing, explaining, or disagreeing with others and offer insights into shared experiences and voiced opinions (Kitzinger, 1995; Stewart, Shamdasani, & Rook, 2015).

Procedure and Data Collection

Participants provided written informed consent and completed a short questionnaire to obtain information about demographic details. Patients were asked for their history of low-back pain and health care professionals were asked for their occupation. The focus group discussions were conducted by a moderator (J.K.) and an observer (main investigator S.B.) using a discussion guide (Table 1). During the focus group, participants could try out an existing exoskeleton (for device description, see the appendix) and were provided with a short explanation of the same device. Providing participants with visual material information facilitates inspiration (Bruseberg & McDonagh-Philp, 2002) and helps to give them a clear picture of the current state of the science. Therefore, the tryout was meant to give the participants an idea and a feeling of how an exoskeleton could look and work. Participants tried on the exoskeleton and performed some simple tasks, such as walking, sitting, and bending forward. The participants were clearly informed that the device used in the tryout was only an example and that questions in the focus group were still designed to elicit opinions about the potential use of exoskeletons in general. We did not aim for an evaluation of the tried-out exoskeleton. We concluded both focus groups with a summary of the findings. Any additional remarks were noted. Both sessions were conducted in Dutch language and lasted between 90 and 120 min.

OPEN IN VIEWER

Table 1

Discussion Guide for Patients (Left) and Health Care Professionals (Right)

Patients	Health Care Professionals
1. Round of introductions Introducing yourself with the help of your personal key ring Tell us something about yourself by explaining the keys on your key ring.
2. What was a key moment with regard to your low-back pain?	2. What was a key moment with regard to your work with low-back pain patients?
3. Tell us about your personal situation and how you deal with low-back pain. Exercises, medical devices, pain killer, physiotherapy What do you like/dislike?	3. Which situation sticks most in your memory? What are the problems? What are your treatment methods?
4. Try to think back to a situation in the past in which you had problems due to your low-back pain.	4. Try to think back to a situation in the past in which you had problems with regard to patients with low-back pain.
5. Could patients come up with an idea of how to solve that problem? Collect on a whiteboard
6. Trying out and explaining an existing exoskeleton Video How does this exoskeleton work?
7. What does an exoskeleton need to provide to make you use it? (green card) What could be a problem that would limit your use of an exoskeleton? (red card)	7. What does the exoskeleton need to provide to the patient? (green card) What could be a problem that would limit their use of an exoskeleton? (red card)
8. Do you think an exoskeleton could help you? (in the situation of Point 2)
9. Closing Do you think you would use an exoskeleton?

Results

The participant characteristics of the two focus groups are presented in Table 2. The average age of the focus group with patients and health care professionals were 54.8 and 47.4 years, respectively. Each of the participants was given a pseudonym and a designation to distinguish between focus groups (P = patient, HCP = health care professional).

OPEN IN VIEWER

Table 2

Participants’ Characteristics

Pseudonym	Focus Group	Age	Occupation	Work Experience With LBP Patients	Years With LBP	Impact of LBP on Activities and Participation
Silke	P	53	Housewife	NA	>1 year	Struggles with certain tasks
Paula	P	55	Currently not working	NA	>1 year	Wheelchair bounded
Kai	P	58	Construction foreman	NA	>1 year	Still restricted in his work
Mette	P	53	Nurse	NA	>1 year	Able to work independently
Adriaan	HCP	52	Medical doctor	>6 years	NA
Pieter	HCP	50	Consultant for work-related musculoskeletal disorders	>6 years	NA
Remy	HCP	38	Movement Therapist	>6 years	NA
Robin	HCP	56	Physiotherapist	>6 years	NA
Bram	HCP	30	Movement Therapist	>6 years	NA
Katy	HCP	47	Reintegration Coach	>6 years	NA
Myrthe	HCP	44	Psychologist	>6 years	NA
Carmen	HCP	62	Reintegration Coach	>6 years	NA

Note. LBP = low-back pain; P = patient; HCP = health care professional.

The main categories derived from the discussions contributing to design requirements for an exoskeleton included “problems faced as a low-back pain patient,” “experience with current devices,” “concerns and fears,” “positive aspects and design improvements,” and “field of application.”

Discussion

This focus group study was a first step in a patient-centered design approach for a novel trunk exoskeleton. We explored potential end user’s perceptions of an exoskeleton to identify criteria to be considered at the start of the design process. Specifically, we investigated health care professionals’ and patients’ perspectives on the idea of using an assistive device to deal with low-back pain patients’ main limitations.

Main concerns from the patient’s perspective were loss of muscle strength as a long-term effect of using an assistive device and overloading of the low back when taking it off after a long period of use. Previous research on different devices, supporting the trunk, has shown that back muscle activity decreased between 10% and 40% during lifting and up to 10% to 60% during static forward bending (de Looze, Bosch, Krause, Stadler, & O’Sullivan, 2016; Koopman, Kingma, Faber, de Looze, & van Dieën, 2019). In the long-term, this reduction in muscle recruitment might lead to a reduction in back muscle capacity, due to the decreased demand on the back muscles. A long-term study on the effect of wearing an exoskeleton on muscle strength would be needed to deal with the patients’ concern. A requirement that derives from this argument is that an assistive device should be used temporarily and the level of support should be adjustable, as mentioned by the patients. Providing different levels of support allows the user to decide how much support is needed for certain tasks and how much work can still be performed by their trunk muscles.

From the health care professionals’ perspective, disadvantages of using an assistive device would be confirming the patient’s disability, increasing vulnerability and potential dependency on the device. Given that low-back pain patients appear to often underestimate their own capabilities, an assistive device might only confirm their wrong self-perception. Trippolini, Janssen, Hilfiker, and Oesch (2018) showed that people suffering from low-back pain indeed have reduced self-efficacy. Low levels of perceived self-efficacy influence behavior in a way that people shy away from tasks they perceive as a threat with respect to their musculoskeletal disorder (Rahman, Reed, Underwood, Shipley, & Omar, 2008). The focus group with patients confirmed that a change in movement behavior occurs due to fear of movement and fear of recurrent pain. Health care professionals even admitted partial responsibility as a result of wrong beliefs in the past that were communicated to low-back pain patients. Still, patients also confirmed that they believed to be supported in their daily activity tasks when wearing the device and indicated to be potentially more active with the exoskeleton, rather than feeling confirmed in their disability.

Adapting movement behavior is a common way of low-back pain patients to deal with their pain and may consist of lowering physical activity (Björck-van Dijken, Fjellman-Wiklund, & Hildingsson, 2008), increasing trunk muscle co-contraction, and reducing deep trunk muscle activity (van Dieën, Flor, & Hodges, 2017). This leads to consequences such as decreased movement and motor variability, increased spinal loading (van Dieën et al., 2017) and detrimental effects, such as high cumulative load on intervertebral disks and stiffening of the trunk muscles (Hodges & Tucker, 2011). Especially decreased motor control was mentioned by the health care professionals, who explained that low-back pain patients often show guarded movement. They therefore expressed their concern immobilizing patients with the exoskeleton, rather than supporting them in moving again. Patients perceived this “guarding” as more stability in the back and felt more confident.

Translating the concerns on confirming inability and adapting movement behavior into design requirements, an exoskeleton should increase patients’ confidence rather than confirming inability. If users feel more confident in performing tasks when wearing the device, they may be more likely to return to work (Woby, Urmston, & Watson, 2007). This confidence cannot be achieved by only adapting the design of the device but rather depends on the implementation strategy of the device in the rehabilitation setting and on adequate communication between the patient and health care professional.

Another concern of health care professionals was the fact that low-back pain would mainly be a psychological problem, whereas the assistive device would focus on low-back pain as a biomechanical problem. Previous epidemiological research has shown that both biomechanical and psychological risk factors are associated with low-back pain (Balagué et al., 2012; Griffith et al., 2012). In contrast to our study, a cross-sectional study by Stevens et al. (2016), who investigated patients’ and physiotherapists’ views on triggers for low-back pain, showed that both groups endorse biomechanical risk factors as the most important risk factor category. Asking participants in a short survey to name triggers for a sudden episode of acute low-back pain, Stevens et al. provides only a rough overview of responses. Due to the qualitative approach, our study adds more insight into the spectrum of personal experience in the context of developing an assistive device. This might explain the difference we found in health care professionals’ beliefs compared with the study of Stevens et al. It should still be noted that the use of a spinal assistive device is more likely to show beneficial effects for patients with mechanical low-back pain, as this is the nature of the device.

Important design characteristics that were directly mentioned during the discussions were comfort, adjustability to align the device to the individual user, and independency in taking it on and off. Furthermore, the possibility of gradual adjustment of support was considered as advantageous to provide task-specific support and versatility, hence being able to perform a variety of tasks without hindrance. Previous research has shown that comfort and versatility are indeed limitations of most assistive devices (Baltrusch, van Dieën, van Bennekom, & Houdijk, 2018; Graham et al., 2009), leading to decreased user-friendliness and limited usability in daily life.

When discussing the usability of the exoskeleton and its field of application, health care professionals believed that exoskeletons are mainly useful as preventive measures at work site, supporting tasks like forward bending, lifting, and strenuous physical work. This type of usage is supported by several studies that have shown that forward bending, lifting, and static holding are supported by assistive devices (Abdoli-Eramaki et al., 2007; Graham et al., 2009; Ulrey & Fathallah, 2013; Wehner et al., 2009). Users, however, should be aware of potential negative consequences, such as misuse of the assistive device at work or increased work expectations by the employer. Furthermore, health care professionals considered using the device in the late stage of rehabilitation, namely, vocational reintegration, provided a clear target group and an explicit implementation strategy is defined. Furthermore, long-term supervision would be crucial to ensure the correct use of the assistive device. Patients considered using the device to be of support in their daily activities and to overcome their participation limitations, such as dependency on others and social exclusion, provided the suggested features could be incorporated. The different opinions on application underlines the importance of adequate communication between the health care professional and patient, when aiming to introduce an exoskeleton in rehabilitation. Still, further research is needed to identify the optimal application for an exoskeleton.

Limitations and Futures Research

As the majority of participants in this study were recruited at the same rehabilitation center in the Netherlands, transferability and applicability of the focus group findings are somewhat restricted. Results may be different in other health care centers and other countries. We, however, believe that our findings give a good insight into context-dependent views of potential end users on this important topic in research regarding the use of wearable assistive devices in the prevention and management of low-back pain.

Another limitation of this study is the small sample size of the focus group with patients. As a result, the moderator had to actively facilitate the interaction. However, as an advantage of the small group, patients intensively shared their personal experience with low-back pain.

Conclusion

The findings presented in this study suggest many implications for the design of an exoskeleton, which can be used in future developments. The major issues offering guidance for improvements in future designs are comfort, adjustability to the body, independence in putting it on and taking it off, and gradual adjustment of support. To deal with the main concerns such as loss of muscle strength, dependency, and adapted movement behavior, an adequate implementation strategy is essential, including supervision and behavioral coaching. Furthermore, the optimal field of application is still under debate. Considering that health care professionals bear the responsibility of implementing it when used in rehabilitation, but see greater potential in the vocational reintegration or as preventive measure in the work environment, it is less likely that a spinal assistive device will find its way into the clinical rehabilitation environment.

Key Points