First iteration of a User-Centered Design process to develop an in-bed leg press

Abstract

BACKGROUND:

Physical inactivity is prevalent among hospitalized patients and the daytime spent lying should be utilised for exercise. However, implementing new interventions in hospitals’ complex daily routines is difficult and requires a participatory approach.

OBJECTIVE:

Exploring clinical settings and clinicians’ perspectives regarding exercise in hospitalized patients to gauge the potential of further development of an in-bed training device and to formulate development goals.

METHODS:

A User-Centered Design approach was employed, consisting of work shadowing and focus groups with physiotherapists and nurses. Content and network analyses of the focus group data were performed. Personas were then developed and used to create clinical scenarios.

RESULTS:

Some clinicians perceived in-bed exercise counterproductive, while others recognized potential for unsupervised training. The most important design characteristics appeared to be a small size, low weight and simplicity of use to facilitate storage, transportation, and administration, respectively. The scenarios revealed that the device’s use could increase the physical activity time by 1.4%, although it would also increase the working time of clinicians.

CONCLUSION:

The study highlighted the difficulties in developing a usable training device but encouraged the authors to pursue their efforts under the strict condition of following the formulated development goals.

Keywords

Sedentary behavior exercise hospitalization user-centered design

1. Introduction

Decreased physical activity levels are frequently found among patients in hospital and rehabilitation settings [1], leading to functional decline and further diseases [2]. The clearest example of deconditioning is found in critically ill patients from the Intensive-Care Unit (ICU) who may develop ‘ICU-Acquired Weakness’ [3]. However, physical inactivity is a phenomenon that also concerns many other patient groups [2], particularly in elderly. Functional decline can be observed on multiple systems of the human body, including physical and cognitive [4, 5]. For example, it has been shown that bed rest during hospitalization can lead to significant loss of strength in the lower extremities [6], representing an important risk factor for falls in elderly [7]. Other studies have shown that even mental health is affected by physical inactivity in hospital [8]. Altogether, the consequences are that after discharge, patients not only have to recover from their condition that caused hospitalization and/or rehabilitation, but also from the hospital-acquired deconditioning (HAD). It is therefore not surprising that HAD represents a major risk factor for readmission to hospital after discharge (e.g. [9, 10]).

Given this serious issue of HAD, it is important to improve the level of physical activity in hospitalised patients. However, implementing interventions for this purpose is a complex task [11] and it is important to provide as many innovative options as possible to facilitate it. For example, to counteract ICU-Acquired Weakness, an in-bed leg press-typed training device has recently been developed [12]. The device consisted of a base containing two cylinders to which two pedals were connected. It could be placed in patients’ bed allowing them to push the pedals against adjustable resistance. The device presented high levels of usability [12] based on indicators, such as the system usability scale [13] or the Kano model [14].

Nevertheless, ICU patients represent a very limited proportion of patients, which limits the impact of the device. Hence, it appeared interesting to pursue the idea of adapting this leg press to suit other patient groups. The starting point was the notion that the current version of the leg press would likely under-challenge many patients outside the ICU. To increase the challenge, it was assumed that adding more functions to the device was necessary. For example, adding multiple degrees of freedom (DOF) to allow diverse movement patterns and utilizing the potential of exergame-technologies appeared to be sensible ideas.

But since there is a need to shift towards User-Centered Design’ (UCD) processes in the development of healthcare devices [15, 16], our development ideas were only considered one way of looking at the problem and had to be opposed to clinical practice routines and end-users’ perspectives to ensure optimal usability. Consequently, this study placed its focus on potential end-users and aimed to objectify their clinical settings and to understand their views regarding exercise in hospitalized patients. The finale objectives were to gauge the potential of such a device outside the ICU and to formulate first evidence-based development goals.

2. Methods

2.1 General methodology

This study represented the first iteration of a UCD process. To achieve the study’s aim, multiple data sources were required, allowing triangulation of objective and subjective data. This method affords more robust conclusions [17].

First, work shadowing sessions were carried out to objectify clinical settings and to observe clinical practices. Then, based on these results, a list of relevant topics for discussion was generated for focus groups. Focus groups were performed, and a network analysis of the data was conducted to identify impactful development domains. The focus group data were also used to define user profiles (personas). Based on these intermediate results, the initially conceived design of the device was revised and clinical scenarios involving the personas were developed. All data were used to formulate development goals. Figure 1 illustrates the overall procedure.

Figure 1.

Overall study procedure. The black rectangles (‘Work Shadowing’ & ‘Focus Groups’) represent data collection phases, whereas the ellipses represent how the data was utilized and/or analyzed. The lowest rectangle (‘Development Goals’) represents the aim of the study.

2.2 Ethics

The study did not fall under the Swiss Human Research Act because it neither investigated human pathology or physiology, nor involved collection of health-related data [18]. Hence, no ethical clearance was required. Ethical principles based on the Declaration of Helsinki [19] were respected, nonetheless. All participating clinicians were informed about the goals of the study, their tasks, the voluntary nature of their participation (including their right to withdraw at any time), and the anonymization of the data. Informed consent was obtained from all participating clinicians.

2.3 Work shadowing

We used the ‘fly on the wall’ method [17], where the first author passively observed physiotherapists during their daily work routine. All clinicians were recruited among the six sites of Switzerland’s largest hospital group (Insel Gruppe), including Bern University Hospital. Field-notes were taken on a custom form and photographs were shot to substantiate the notes. The objective was to perform i) worker (i.e. working preferences), ii) environmental, and iii) task analyses. The data were summarized in a table and formed the basis for the development of an interview guide for the focus groups.

Figure 2.

The six themes are depicted as the first column of ellipses on the left. Each theme is composed of thematically similar sub-themes (second column of ellipses). Sub-themes are composed of ‘positions’ (rectangles), which are statements summarizing raw opinion data.

2.4 Focus groups

An interview guide was drafted by the first author, reviewed by two experienced qualitative researchers, and tested with an expert in musculoskeletal physiotherapy. The final version of the guide covered the five following key questions to structure the focus groups:

How do you experience patient mobilization and exercise in patients?

What would be your expectations regarding a device for unsupervised exercise in patients, in order to integrate it into your daily practice?

What do you think are patients’ expectations regarding such a device?

What patient groups could benefit from that device?

What challenges do you see in using such a device in your daily practice?

Target population for the focus groups consisted of physiotherapists and nurses. Personnel from ICU were excluded because this was the topic of the first leg press [12]. Personnel from orthopaedic departments were also excluded because it was thought that patients recovering from orthopaedic surgery would frequently have non-weightbearing rehabilitation regime, making exercise with a leg press difficult. Inclusion criteria were i) having a minimum of 1 year of clinical experience at full-time equivalent, and ii) working at least two hours per day with in-patients. These criteria allowed a selection of participants who have a clear understanding of the work with the patients, of the daily clinical routines, and of the problems that may occur. Participants’ recruitment was based on voluntary response sampling [20], where flyers were distributed in four sites of the Insel Gruppe. In addition, key persons consisting of head physiotherapists and advanced practice nurses from these sites and other hospitals were contacted for a network sampling strategy [20].

Online focus groups were conducted and recorded using Microsoft Teams. First, demographic data and information on work experience were collected, a short presentation of the first leg press version [12] was provided, and then, the discussion was launched.

After the focus groups, intelligent verbatim transcription of the records was performed using the software f4analyse (v. 3.0.0) and finalized transcripts were sent to interviewees for review as quality check [21]. A content analysis was then performed according to Kuckartz and Rädiker [22] with a combination of a concept- and data-driven coding system. Five concept-driven themes were first generated based on the interview questions. Transcript content was coded into these themes and grouped into subthemes based on their topic in a data-driven process. The contents forming these subthemes are called ‘positions’ in this study (Fig. 2). Also, additional data-driven themes were generated where necessary. The software MaxQDA (v. MaxQDA Plus 2022, release 22.1.1) was used for this step.

2.5 Further data processing and utilization

2.5.1 Network analysis

A network analysis of the data from the focus groups was performed [23], in which subthemes represented the nodes. Edges were defined as cases where positions within subthemes had direct practical implications for other positions in other subthemes. Weights were also attributed to the nodes and edges. Node weights were calculated as the sum of all positions they contained and edge weights by the number of times connected positions were mentioned. The analysis regarded all nodes that related to the design of the device (‘design-related’) as independent variables because the design is typically invariable once the device is implemented. Edge weights from these design-related nodes were examined and ranked to indicate what the design would impact most.

2.5.2 Review of the initial design

The results of the network analysis were then used to discuss and reflect on the initially conceived design of the proposed device within the research team. Adaptations to that design were made based on consensus achieved during these discussions. Characteristics of a first hypothetical prototype were sketched.

2.5.3 Personas and scenarios

Using the focus groups data, personas were generated according to the method described in Cooper [24]. The target population for the proposed device was identified from these profiles. Using the target population personas in combination with the observational data from the work shadowing sessions, two user-scenarios were developed to conceptually test the feasibility of using the proposed device (Supplementary material 1). Both scenarios illustrated the hospital stay of a patient, once without (a) and once with the use of the proposed device (b). The scenarios were drafted by the first author with the focus on the description of a patient case that is frequently encountered in a public hospital, and on the patient’s activities during hospital stay. Clinicians from the authors’ institution and from the Insel Gruppe who did not participate in the work shadowing sessions and focus groups were recruited to review the scenarios. They were asked to rate the scenarios regarding their credibility using a scale ranging from 0 (‘The scenario is not credible. It does not reflect a plausible patient case and hospital stay’) to 10 (‘The scenario is credible. It does reflect a plausible patient case and hospital stay’). In addition, they were allowed to comment the scenarios where required. The scenarios were adapted according to the comments. In a final step, the time during which the fictive patient conceivably performed different activities, as well as their intensity was estimated. The intensity of the activities was classified as ‘lying’, ‘sitting’, or ‘moving’ (e.g. walking) [25]. Using the proposed device was classified as a ‘moving’ activity. The estimated activity-times were calculated as a percentage of the total daytime spent in hospital. Daytime was preferred as a reference over the entire hospitalization time, because the importance of sufficient quality sleep at night was acknowledged for optimal convalescence (e.g. [26]). The staff’s working time spent helping the patient in ‘sitting’ and ‘moving’ activities was also counted. The two scenarios were compared using these time-related results.

2.5.4 Formulation of development goals

Based on raw (work shadowing and focus groups) and processed data (network analysis, personas, and scenarios), development goals were formulated, discussed, and finalized.

3. Results

3.1 Work shadowing

Three institutions accepted to participate. Two were hospitals for acute care with integrated rehabilitation centers (geriatric, cardiac, and pulmonary) and one was a rehabilitation center for neurological patients. In one institution, two different physiotherapists were observed, making a total of four participating clinicians. A total of 13 patients were viewed by the four clinicians during the work shadowing periods.

3.1.1 Worker analysis

All participants chose to provide active therapy sessions over passive therapy, except for one session in which the patient complained about pain. Nearly all sessions consisted of practising an activity of daily living (ADL), such as walking, or getting up from bed or from the ground. Only one contained the use of an exercise device (digital balance training device). All patients also received instructions to exercise by themselves as adjunct therapy, but these exercises included no or minimal equipment, such as elastic bands.

3.1.2 Environmental analysis

All beds had the same dimensions (2 m long, 0.95 m wide), and the space beside the bed was consistently relatively narrow, ranging from 0.85 to 2 m. The space at the foot of the bed was even narrower, ranging from 0.75 to 1.8 m. Also, when disregarding regular furniture, up to seven additional large objects, such as wheelchairs, walking frames, or monitoring system, were counted in patient rooms (range: 1–7). The dimensions of the lift ranged from 2 to 2.7 m in depth, and 0.7 to 1.8 m in width. In two institutions, the storage room, in which therapeutic devices are stored, was located in the basement, making unplanned use of a device difficult. Only one institution had an auxiliary storage room level to patients’ rooms. In both hospitals, all therapy sessions started and ended in patients’ rooms, whereas in the rehabilitation center the meeting point was either an exercise room or the patient’s room.

3.1.3 Task analysis

Main diagnoses were most often of cardiac, pulmonary, or neurologic nature and the majority of patients were co-morbid. Some patients also were hospitalized for deterioration of their general condition. Overall, 47% of the patients were autonomous, while the other 53% needed help to walk and/or get out of bed.

On average, 22 minutes (range: 16–32 min.) of contact time between therapists and patients were measured per patient. However, therapists had to add time for documentation.

3.2 Focus groups

In total, 10 clinicians (7 physiotherapists and 3 nurses) participated in three focus groups, which lasted, on average, 41 minutes (range: 32–47). Participants had an average working experience of 11 years (range: 1–27 years), with 7 years (range: 1–25) in their current settings. Five participants worked in a hospital, and five in a rehabilitation center. Six participants worked in several departments, where internal medicine was represented most frequently (5 participants), followed by the geriatric, neurologic, pneumological departments (each 2 participants), and oncologic rehabilitation (1 participant). Out of the 10 participants, seven reviewed the transcripts and returned their feedback. Thematically, nothing had to be added, removed, or changed.

Figure 3.

Representation of the network analysis. The six themes (described in the legend) contain the subthemes, which are represented by circles. The size of the circles is proportional to the node weight. The connecting lines represent the edges. The thicker and lighter Gray they are, the larger the edge weight is.

Six themes were generated (Fig. 3). The themes ‘Experiencing patient mobilization’, ‘Expectations of clinicians’, ‘Conceived patient expectations’, ‘Patient groups’, and ‘Challenges for implementation’ were concept-driven, whereas the theme ‘Potential and scepticism’ was data-driven.

3.2.1 Experiencing patient mobilization

All clinicians agreed that mobilizing patients into an upright posture as early as possible is the most important goal in hospitalized patients (subtheme ‘Therapy-related’, $n=$ 24). Clinicians often educate their patients about the importance of physical activity to regain autonomy and ask them to perform exercises on their own, most often using minimal equipment. However, another subtheme highlighted that patient mobilization often relies on collaboration between physiotherapists and nurses (‘Organization-related’, $n=$ 20) and that it highly depends on staffing. This was perceived as a major issue given the understaffing problems affecting the health care sector. Nurses clearly stated that patient mobilization is not their priority, which leads to cases where patients lie in bed the entire weekend if understaffing is severe or if more than one nurse is required for mobilization (e.g. in complex cases). Patients stay even longer in bed if prescription for physiotherapy is not issued on the day of admission. Daily routines were perceived as hectic, which also impacts device-based therapy, such as continuous passive motion. Typically, devices are installed by physiotherapists and later removed by nurses, but due to understaffing nurses are decreasingly willing to help with these tasks. Physiotherapists expressed different reactions to this trend with some developing a guilty conscience for asking nurses, and others becoming rather frustrated. In any case, physiotherapists are generally becoming reluctant to use devices that require nurses’ help. One physiotherapist represented an exception, though, as she did not report experiencing such understaffing problem in nurses in her institution.

3.2.2 Expectations of clinicians

Much of the expectations were ‘design-related’ ( $n=$ 40), with the most recurrent concerns being the simplicity of use, not only for clinical staff but also for patients in cases they would use it alone. Moreover, the device should be lightweight and compact, such that handling is easy and storage close to patient rooms. One participant’s idea was even a design that allows the device to be left attached to the patient’s bed and be deployed whenever needed.

Patient safety was also mentioned. The aspect of technology was a dividing issue, with some participants proposing the use of visualization effects (e.g. biofeedback) and exergaming, while others thought using such technology would rather be a barrier because it would increase complexity of use. Some participants were also concerned about the repercussion a sophisticated device could have on the price. For example, frames used to help mobilizing weak patients into a sitting or standing position are cheap, and if the leg press is comparatively expensive, it would make more sense to invest in several of those mobilization frames to achieve the goal of getting out of bed as soon as possible.

Two participants stated that using the device should show clear strengthening effect (‘Effect-related’). Also, they stated that the device should have a motivating effect because patients often have low functional levels and therefore low self-efficacy, self-esteem, and motivation levels. Participants also wished strong evidence showing the benefit of such a device before using it in their clinical practice (‘Knowledge-related’, $n=$ 5). One participant highlighted the importance of knowing the physical limits of the device as well as the physical constraints imposed by the different training modes to remain within the safety limits. Others highlighted that from an organizational perspective, it would be good if patients could install, use, and remove the devices by themselves (‘Organization-related’, $n=$ 2).

3.2.3 Conceived patient expectations

Most conceived expectations were also related to the device’s design (‘Design-related’, $n=$ 14), where the ease of installation for autonomous use was deemed important. This way, patients could train when they wish, rather than when staff is available. Also, it was stated that patients typically react positively to the feeling of success and to encouraging signs. Hence, the device shall provide scalable training levels as well as encouraging texts, icons, and/or sounds.

Here, the level of technology use and technical complexity divided opinions, too. One participant intentionally formulated an exaggerated description:

“…patients aged 80 or older are all adverse [to new technologies], while all patients below 70 years of age are fascinated by all technological gadgets …”.

Patients would also value comfortable design. Finally, participants thought that patients wish clear benefits from using the device in achieving their therapeutic goals (‘Effect-related’, $n=$ 3).

3.2.4 Patient groups

‘Non-autonomous patients’ were considered the patient group that could benefit the most ( $n=$ 10), followed by neurological (‘Neurol. Diseases’, $n=$ 6), surgical patients and those with vascular diseases (‘Post Surgery’, and ‘Vascular Diseases’, $n=$ 4 each). Some participants thought that only patients without cognitive deficiency could benefit from the device due to the possible complexity of use. In contrast, others thought that explicitly cognitive deficient patients could benefit more from such training, given it is likely better guided than free exercises that use body weight or barbels (‘Cognitive Deficiencies’, $n=$ 4). Accordingly, some participants explicitly enumerated patient groups that could or should not benefit from the device (‘Cannot benefit’, $n=$ 13). These were cognitive deficient patients who can be mobilized out of bed or those who suffer from the consequences of long-term lying (e.g. decubitus ulcers).

3.2.5 Challenges for implementation

Most challenges in using a device like the proposed leg press in daily practice were seen to be ‘organization-related’ ( $n=$ 7), where poor introduction to the device could result in several problems. On the one hand, clinicians who were sceptical from start could show even more resistance because they would not exactly know how to install and/or use it. On the other hand, proponents could tend to administer it to too many patients, even some who would not require it, because they would lack knowledge about clear indications of use. Participants also perceive the time to administer the device as a challenge.

Second-most mentioned positions were ‘Design-related’ ( $n=$ 6), where again a sophisticated design may imply high price and hence small device pool. Moreover, such design can have an intimidating effect on patients and cleaning/disinfection can be difficult. Finally, using technology increases complexity of administration and of use, which was perceived as an important barrier by some participants.

Finally, seeing no positive effects from using the device could be perceived as a challenge (‘Effect-related’, $n=$ 2).

3.2.6 Potential and scepticism

The focus groups showed that the interview guide did not cover all relevant topics as some discussions drifted away from the key questions to reveal other important aspects. These contents were coded as theme ‘Potential and scepticism’. They expressed explicit scepticism towards the device or, in contrary, ideas for potential use or further development. Here, ‘scepticism’ differs from the theme ‘Challenges’ in that it addresses negative opinions on the idea of implementing the device while challenges referred to practical aspects if the device was implemented.

First, some participants realized ‘Technical and Practical Potential’ ( $n=$ 13). The leg press could be a good method for patients to exercise on their own, particularly when there is no physiotherapy (e.g. on weekends) and if the design allowed self-administration. Moreover, participants thought there would be even more potential if the design allowed versatile use, such as in a sitting position, and if the device could be rented after discharge for patients to exercise at home. Some participants indeed anticipated the leg press to have a positive effect on muscle strength in the lower limbs, on coordinative abilities, and on patients’ motivation to pursue physical activity (‘Effect-related Potential’, $n=$ 5).

In contrary, some other participants expressed their strong scepticism and warned against ‘tying patients to bed’ with the device. Moreover, they feared that implementing the leg press could tempt physiotherapists to replace therapy with simply administering the device which would contradict their work ethic.

3.3 Network analysis

Figure 3 illustrates all themes (legends) and subthemes (nodes) with their connections (edges). The edges connecting ‘design-related’ nodes, their weight and interpretation are presented in Table 1.

Table 1
Interpretation of the edges found in the network analysis

Node 1	Node 2	Interpretation	Weight
Expectations of clinicians: Design-related	Experiencing patient mobilization: Organization-related	Developing a device that is simple to administer saves time and may even allow self-administration by patients. This helps in situations of understaffing and decrease frustration.	24
	Potential & scepticism: Technical & Practical Potential	Having a device that can be self-administered by patients has great potential because it can increase the chances to improve overall levels of physical activity, especially on weekends. This benefit would be highest in non-autonomous, weak patients.	21
	Conceived patient expectations: Design-related	The use of motivating technology was mentioned in both nodes. Designing a device that has motivating effect satisfies clinicians and probably patients (examples: encouraging signs, simple progression parameters such as scores in an exergame, time exercised, or number of pressing repetitions performed).	12
Conceived patient expectations:	Challenges for implementation: Device-related	Simplicity of use is required, otherwise the implementation in clinical practice is challenging.	9
Deign-related	Challenges for implementation: Effect-related	Patients shall be motivated because they are likely to have self-doubts, low self-esteem or even depressive symptoms. Training with a device that is not adapted would likely further demotivate patients and have negative effect, rather than just no effect.	5
	Expectations of clinicians: Effect-related	The device should have a motivational effect, which was mentioned in both nodes. This is the opposite of the edge described just above.	5
Challenges: Design-related	Challenges: Organization-related	High acquisition costs result in a small device pool per institution which results in coverage problems. A lack of proper instructions can result in inefficient and/or inappropriate use.	6

3.4 Personas

From the content analysis, three archetypes were identified, which are presented in Table 2.

Table 2
Description of the personas, including demographic data and opinions on using a training device in regular clinical practice

		Opinion on implementing a training device
Name (gender)AgeProfession (working experience)	Maria (♀)42Physiotherapist (17)	She sees potential for such a device because of understaffing problems. It could alleviate the problem of inactivity, particularly on weekends. She thinks such a device can improve strength and coordinative capabilities in weak, non-autonomous patients who are at risk of fall. She also sees potential for patients who are bed-ridden following orthopaedic surgery with non-weightbearing rehabilitation scheme as a method to train the non-operated limb if the device can be used with one leg only. Also, some hemiplegic patients and those with vascular diseases could benefit from it. The exercises shall be intuitive and goal-oriented, such that their benefit would help improve real-world activities. The device should include new technologies to allow optimally adapted training and to increase patients’ motivation. To be implemented in daily practice two main criteria that must be met: First, for general acceptance, empirical evidence showing clinical benefits of the device must be present. Second, the device must be easy-to-use and quickly installed to avoid further frustration in the collaboration between physiotherapists and nurses.
Name (gender)AgeProfession (working experience)	David (♂)33Physiotherapist (9)	He is very critical regarding the utility of such a device because he regards such a device as contrary to the core goal of physiotherapy: regaining autonomy. Evidence-based practice is particularly important to him, and he learnt that getting in an upright position was the most efficient method to achieve good clinical outcomes. Therefore, he does not see the reason for a bed-based training. In few cases, there might be some benefit of using such device, but that device shall clearly be used as an adjunct to therapy and not replace it. The only way he could change his opinion is if he was presented strong evidence that hospitalization involving such a device led to superior clinical outcome. On one point, he agrees with Maria: The design shall allow quick installation and removal.
Name (gender)AgeProfession (working experience)	Evelyne (♀)35Nurse (12)	Although she knows that best practice is to mobilize patients out of bed as early as possible, she is not necessarily against the use of such a device. She thinks, this could be a good supplementation to physiotherapy, especially on weekends, when physiotherapists are not present. However, she also knows that this would imply additional tasks for her and her colleagues. A complex design may quickly make her feel overchallenged, particularly if there are too many options and possible settings. Also, such design will likely increase installation and removal time, which is unconceivable in an already hectic daily routine. In her eyes, the device must feature a simple design, even at cost of additional functionality. Second, all personnel must have had the same introduction to the device to improve consistency of use.

Table 3

Comparison of both clinical scenarios. Top row: credibility evaluation by five clinicians. Middle row: Scenario analysis from a patient’s perspective. Bottom row: Scenario analysis from a clinician’s perspective

		Scenario 1	Scenario 2
Credibility (mean, range)		8.4 (7–10)	7.2 (4–9.5)
Time in activities [hours]	Lying	44.3 h (73.8%)	43.4 h (72.4%)
(% of hospitalized daytime)	Sitting	13.3 h (22.2%)	13.3 h (22.2%)
	Moving	2.4 h (4%)	3.3 (5.4%)
Working time related to activities	Sitting	Evelyne: 130	Evelyne: 130
[minutes]		Maria: 0	Maria: 0
	Moving	Evelyne: 80	Evelyne: 130
		Maria: 91	Maria: 106

3.5 Scenarios

To review the scenarios, one clinician (nurse) from the authors’ institution and four from the Insel Gruppe (two nurses, two physiotherapists) could be recruited. Both scenarios were considered to be credible (Table 3). Some clinicians provided comments, which were used to adapt the scenarios. The comparison of the two scenarios is presented in Table 3. In both scenarios, the patient was thought to be awake between 7 am and 8 pm. Together with the first (4.5 hours) and last day (3.5 hours), the patient spent a total of 60 hours awake (i.e. daytime). Employing the proposed device led to a shift in daytime use from ‘lying’ towards ‘moving’ activities of 1.4% (approximately 50 minutes). With an installation and removal time of 10 minutes each, employing the proposed device increased the working time dedicated to the patient by 50 minutes for the nursing staff and by 15 minutes for the physiotherapist. Detailed results are presented in Supplementary material 2.

3.6 Development goals

Based on the results, the research team agreed on targeting a design that is close to the original leg press because of its simplicity and user-friendliness. The targeted time to install and time to remove the device from the patient’s bed was set to 10 minutes. Additional functionalities were kept to a minimum, including an additional DOF to allow plantar- and dorsiflexion in the ankle joint, the possibility to feature simple exergames and provide bio- and motivational feedback. Versatile use, i.e. in a lying and sitting position, as well as the possibility to keep the device attached to the bed when not in use, were also set as development goals.

A video of a prototype developed based on these development goals with a demonstration of some of its features can be found here: https://www.youtube.com/watch?v=EMQFEoaN-Qs&t=50s.

4. Discussion and conclusion

This study investigated clinical settings and potential end-users’ perspectives regarding exercise in hospitalized patients to define a first evidence-based development framework for subsequent UCD iterations.

In general, all participants recognize the importance of physical activity in patients to achieve targeted clinical outcomes and to prevent deconditioning. They all strive to mobilize patients out of bed and practice ADLs, as became clear from the interviews and from work shadowing. However, they also recognize that mobilizing non-autonomous patients is not always possible due to understaffing and/or complexity of patient cases. For this reason, some participants (personified as ‘Maria’) find that a device, such as the leg press, could form a good alternative for unsupervised exercise. In contrast, others (personified as ‘David’) are strongly opposed to exercising in bed if not absolutely necessary. Despite clear evidence that prolonged lying can have devastating effects on patients [27, 28], it must be considered that hospitalized patients lie in bed for an average of 70–83% of their time, even those with walking abilities [29, 30], and some of this time should be used to exercise. However, patients are often left alone [1], which makes exercising difficult, particularly in the group of non-autonomous patients (53% according to our observations). Hence, the research team believes that unsupervised exercising in a secure position, such as lying, is important.

For successful implementation, the device must be lightweight and compact to allow easy transport and handling, quick installation as well as possible storage in patients’ rooms. These aspects are considered most important due to their large impact on organizational aspects (edge weight $=$ 24), e.g. with respect to interprofessional collaboration, and thus the potential use (edge weight $=$ 21). While the original device was rated as “acceptable and easy to carry over a short distance” in this respect [12], any added functionality will likely increase weight and dimensions. Consequently, the new features must be perceived as significant improvement to be acceptable. Alternatively, the device must be designed such that it remains in the vicinity of the patient to be deployed when needed. For such design, however, the narrow space around the bed and particularly at the foot of the bed (0.75–1.8 m) and the number of other devices found in patients’ rooms will have to be considered. Moreover, the time to install and remove the device must be short. Currently, the targeted time to install or remove the device was set to 10 minutes. Ten minutes were estimated based on the authors’ clinical experience with other devices, such as passive continuous motion. Nevertheless, the clinical scenarios, despite showing an increase of time spent in moving activities of 50 minutes over a hospital stay of 5.5 days (i.e. 1.4% increase), also illustrate that working time rose by a total of 65 minutes for the clinicians. In an already hectic daily routine, this may represent a burden. Possibilities to further reduce the installation and removal time should, therefore, be explored. Another measure that was highlighted by one of the reviewing clinicians would be to target clinical settings with a lower patient-to-nurse ratio, such as in geriatric rehabilitation. Finally, designing enjoyable training modalities that motivate patients to exercise for longer than 20 minutes could represent a way to augment the gain obtained from the effort put in installing and removing the device. These aspects will have to be investigated and tested in future iterations, but the research team is convinced that even small increases in physical activity are likely to contribute to faster recovery and worth the effort [31].

To further improve usability of the leg press and facilitate its implementation, it appears important to make the device more versatile (e.g. use in sitting position) and to feature motivational signals, ways to monitor progression (edge weight $=$ 12) and individually adaptable training levels (edge weight $=$ 5). However, care must be taken not to increase difficulty of handling and complexity of use with the incorporation of too many functions and technological features. While ‘Maria’ thinks the device should involve new technologies, such as exergames, another persona named ‘Evelyne’ prefers the simplest design possible. To some extent the difference of views might be due to the difference in professional experience, as it has been shown that clinicians with long-time experience were less intimidated by complex medical devices [32]. To acknowledge the request for simplicity, it was decided to limit additional functionality to only one DOF. The chosen DOF was selected, because training sensorimotor control in plantar- and dorsiflexion in addition to the knee extensors positively impacts the standing balance abilities [33]. The additional exercise feature could thus help achieve important goal-oriented outcomes. With respect to motivational feedback, the literature is not conclusive on the effect of exergaming in elderly [34]. The development of the exergames should therefore be in line with motivational principles in elderly users, such as focussing on providing feedback on health benefits and enjoyability [35].

In conclusion, the results showed that in-patient settings are characterized by hectic daily routines, in which some additional work time can already represent significant barriers to the use of a therapeutic method. However, if following the formulated development goals and if targeting the right patient and end-user groups, the proposed in-bed leg press may have potential in clinical practice.

4.1 Strengths and limitations

First, the main strength of the study is that several data sources building on one another were utilized to collect data. Moreover, the data were processed in different ways, allowing the exploration of diverse aspects. Second, while the study focussed on the development of a particular device, the questions used to frame the focus group discussions were formulated in such a way that the current results can be used for the development of any exercising device for most in-patient settings. It should be noted that highly specialized settings, such as neurological rehabilitation settings might have different requirements, however. The main limitation is that the samples were relatively small (4 for work shadowing, 10 for focus groups).

Funding

The authors report no funding.

Supplementary data

The supplementary files are available to download from https://dx-doi-org.web.bisu.edu.cn/10.3233/THC-240816.

Footnotes

Acknowledgments

The authors thank all participants and their institutions for their participation.

Conflict of interest

The authors declare that they have no conflict of interest.

References

Kuys

Dolecka

Guard

. Activity level of hospital medical inpatients: An observational study. Arch Gerontol Geriatr. 2012; 55: 417-21. doi: 10.1016/j.archger.2012.02.008.

Covinsky

Pierluissi

Johnston

. Hospitalization-associated disability: She was probably able to ambulate, but I’m not sure. JAMA. 2011; 306: 1782-93. doi: 10.1001/jama.2011.1556.

Vanhorebeek

Latronico

van den Berghe

. ICU-acquired weakness. Intensive Care Med. 2020; 46: 637-53. doi: 10.1007/s00134-020-05944-4.

Wilson

Hebert

Scherr

Dong

Leurgens

Evans

. Cognitive decline after hospitalization in a community population of older persons. Neurology. 2012; 78: 950-6. doi: 10.1212/WNL.0b013e31824d5894.

Chen

Almirall-Sánchez

Mockler

Adrion

Domínguez-Vivero

Romero-Ortuño

. Hospital-associated deconditioning: Not only physical, but also cognitive. Int J Geriat Psychiatry. 2022; 37: gps.5687. doi: 10.1002/gps.5687.

Kortebein

Ferrando

Lombeida

Wolfe

Evans

. Effect of 10 days of bed rest on skeletal muscle in healthy older adults. Jama. 2007; 297: 1769-74.

Cho

Bok

Kim

Y-J

Hwang

. Effect of lower limb strength on falls and balance of the elderly. Ann Rehabil Med. 2012; 36: 386. doi: 10.5535/arm.2012.36.3.386.

Guilcher

SJT

Everall

Cadel

Kuluski

. A qualitative study exploring the lived experiences of deconditioning in hospital in Ontario, Canada. BMC Geriatr. 2021; 21: 169. doi: 10.1186/s12877-021-02111-2.

Krumholz

. Post-hospital syndrome – a condition of generalized risk. The New England Journal of Medicine. 2013; 368: 100.

10.

Munir

Goldfarb

. Sedentary time in older adults with acute cardiovascular disease. CJC Open. 2022; 4: 282-8. doi: 10.1016/j.cjco.2021.10.008.

11.

Silva

Godinho

Encantado

Rodrigues

Carraça

Teixeira

, et al. Implementation determinants of physical activity interventions in primary health care settings using the TICD framework: A systematic review. BMC Health Serv Res. 2023; 23: 1082. doi: 10.1186/s12913-023-09881-y.

12.

Rostetter

Jenni

Eggmann

Meyer

Schmitt

K-U

. Implementing an interprofessional user-centered design approach to develop a bedside leg exercise device. THC. 2022; 30: 981-92. doi: 10.3233/THC-213511.

13.

Brooke

. SUS-A quick and dirty usability scale. Usability Evaluation in Industry. 1996; 189: 4-7.

14.

Kano

Seraku

Takahashi

Tsuji

. Attractive quality and must-be quality 1984.

15.

Matinolli

H-M

Mieronkoski

Salanterä

. Health and medical device development for fundamental care: Scoping review. Journal of Clinical Nursing. 2020; 29: 1822-31. doi: 10.1111/jocn.15060.

16.

Shaw

. Innovation and new product development in the UK medical equipment industry. IJTM. 1998; 15: 433. doi: 10.1504/IJTM.1998.002620.

17.

Still

Crane

. Fundamentals of User-Centered Design: A Practical Approach. Boca Raton: CRC Press. 2017. doi: 10.4324/9781315200927.

18.

Fedlex – The publication platform for federal law. Federal Act on Research involving Human Beings: Human Research Act (HRA). 2023.

19.

World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. J Am Coll Dent. 2014; 81: 14-8.

20.

Gray

. Doing research in the real world. London; Thousand Oaks, CA: Sage Publications. 2004.

21.

Hagens

Dobrow

Chafe

. Interviewee Transcript Review: Assessing the impact on qualitative research. BMC Med Res Methodol. 2009; 9: 47. doi: 10.1186/1471-2288-9-47.

22.

Kuckartz

Rädiker

. Qualitative Inhaltsanalyse: Methoden, Praxis, Computerunterstützung: Grundlagentexte Methoden. 5. Auflage. Weinheim Basel: Beltz Juventa. 2022.

23.

Pokorny

Norman

Zanesco

Bauer-Wu

Sahdra

Saron

. Network analysis for the visualization and analysis of qualitative data. Psychological Methods. 2018; 23: 169-83. doi: 10.1037/met0000129.

24.

Cooper

. About face: The essentials of interaction design, 4th edition. 4th edition. Indianapolis, IN: John Wiley and Sons. 2014.

25.

Valkenet

Bor

van Delft

Veenhof

. Measuring physical activity levels in hospitalized patients: A comparison between behavioural mapping and data from an accelerometer. Clin Rehabil. 2019; 33: 1233-40. doi: 10.1177/0269215519836454.

26.

Stone

Ensrud

Ancoli-Israel

. Sleep, insomnia and falls in elderly patients. Sleep Medicine. 2008; 9: 18-22. doi: 10.1016/S1389-9457(08)70012-1.

27.

Bergouignan

Rudwill

Simon

Blanc

. Physical inactivity as the culprit of metabolic inflexibility: Evidence from bed-rest studies. J Appl Physiol (1985). 2011; 111: 1201-10. doi: 10.1152/japplphysiol.00698.2011.

28.

Stuempfle

Drury Daniel

. The physiological consequences of bed rest. Journal of Exercise Physiology. 2007; 10.

29.

Pedersen

Bodilsen

Petersen

Beyer

Andersen

Lawson-Smith

, et al. Twenty-four-hour mobility during acute hospitalization in older medical patients. J Gerontol A Biol Sci Med Sci. 2013; 68: 331-7. doi: 10.1093/gerona/gls165.

30.

Brown

Redden

Flood

Allman

. The underrecognized epidemic of low mobility during hospitalization of older adults. J Am Geriatr Soc. 2009; 57: 1660-5. doi: 10.1111/j.1532-5415.2009.02393.x.

31.

Lee

I-M

. Dose-response relation between physical activity and fitness: Even a little is good; more is better. JAMA. 2007; 297: 2137-9. doi: 10.1001/jama.297.19.2137.

32.

Crilly

Dowling

Delaunois

Flavin

Biesty

. Critical care nurses’ experiences of providing care for adults in a highly technological environment: A qualitative evidence synthesis. Journal of Clinical Nursing. 2019; 28: 4250-63. doi: 10.1111/jocn.15043.

33.

Blenkinsop

Pain

MTG

Hiley

. Balance control strategies during perturbed and unperturbed balance in standing and handstand. R Soc Open Sci. 2017; 4: 161018. doi: 10.1098/rsos.161018.

34.

Hwang

Deng

Manninen

Waller

Evans

Schmidt

, et al. Short- and longer-term psychological and behavioral effects of exergaming and traditional aerobic training: A randomized controlled trial. International Journal of Sport and Exercise Psychology. 2023; 21: 120-37. doi: 10.1080/1612197X.2021.2025135.

35.

Subramanian

Dahl

Skjæret Maroni

Vereijken

Svanæs

. Assessing Motivational Differences Between Young and Older Adults When Playing an Exergame. Games Health J. 2020; 9: 24-30. doi: 10.1089/g4h.2019.0082.

First iteration of a User-Centered Design process to develop an in-bed leg press

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Methods

2.1 General methodology

2.3 Work shadowing

2.5 Further data processing and utilization

2.5.1 Network analysis

2.5.2 Review of the initial design

2.5.3 Personas and scenarios

2.5.4 Formulation of development goals

3. Results

3.1 Work shadowing

3.1.1 Worker analysis

3.1.2 Environmental analysis

3.1.3 Task analysis

3.2 Focus groups

3.2.2 Expectations of clinicians

3.2.3 Conceived patient expectations

3.2.4 Patient groups

3.2.5 Challenges for implementation

3.2.6 Potential and scepticism

3.3 Network analysis

Table 1 Interpretation of the edges found in the network analysis

Table 2 Description of the personas, including demographic data and opinions on using a training device in regular clinical practice

3.6 Development goals

4. Discussion and conclusion

4.1 Strengths and limitations

Funding

Supplementary data

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Interpretation of the edges found in the network analysis

Table 2
Description of the personas, including demographic data and opinions on using a training device in regular clinical practice