Introduction

As the world population is growing older, the number of older persons in need of care is ever increasing, expecting to cross the threshold of 16% of the world population by 2050 (United Nations Department of Economic and Social Affairs 2022). At the same time, professionals who are willing to do caregiving work are becoming increasingly scarce (Santos and Miguel 2020; Seidlein et al. 2020) and informal caregivers become overburdened and distressed (Syse et al. 2022). To remedy the growing imbalance between older persons in need of care and available care, research is turning toward new technologies. These include, for example, fall detection sensors (Bet et al. 2019; Momin et al. 2022), cameras in the home (Pool et al. 2022), GPS tracking for people with dementia (Bayat and Mihailidis 2021), to cognitive assistants (Holthe et al. 2022) and artificial social companions to relieve loneliness (Gasteiger et al. 2021; Koh et al. 2021).

The actual use of new technologies necessitates investigations into factors that foster and hinder acceptance of such technologies in caregiving. Acceptance in this context matters both because developing technologies that are then rejected by end-users can be wasteful, and “forcing” interventions that people do not want is morally problematic. As engineers, scientists and politicians make choices regarding which technologies are developed and for whom and what purpose, it is crucial to ensure that these choices align with what society wants and what is within ethically acceptable boundaries (Legault et al. 2018).

Several models and theories have emerged over time to investigate and understand acceptance of technology (AlQudah et al. 2021). On established theory is the technology acceptance model (TAM), (Davis 1989). It proposes two main variables to predict acceptance: perceived usefulness and perceived ease of use (Davis 1989). Another widely used model is the unified theory of acceptance and use of technology (UTAUT) (Venkatesh et al. 2003). UTAUT consists of a combination and refinement of earlier models and has already produced a vast body of research for healthcare technologies (AlQudah et al. 2021). UTAUT uses four dimensions to predict acceptance of technology: performance expectancy, effort expectancy, social influence and facilitating conditions. Performance expectancy refers to the degree to which the user believes that using the technology can help attain goals of productivity, while effort expectancy expresses the ease (or difficulty) associated with using the technology (Venkatesh et al. 2003, p. 450). Social influence is defined as how much the end-user perceives that important others want him/her to use the technology (Venkatesh et al. 2003, p. 451). Facilitating conditions capture the extent to which they believe that the necessary infrastructure exists to support use of the new technology (Venkatesh et al. 2003, p. 453). Both UTAUT and TAM have undergone refinement over the years, and extensions of the models, such as UTAUT 2 or TAM 3, have seen the addition of new variables, such as anxiety and cost, among others (Patil et al. 2020; Rondan-Cataluña et al. 2015; Tamilmani et al. 2021; Venkatesh et al. 2012).

Researchers using these models have made important contributions to the investigation of the facilitators and barriers to uptake of technologies that can also be used in aged care. Examples of investigated technologies include mHealth services (Alam et al. 2020; Hoque and Sorwar 2017; Palas et al. 2022; Rajak and Shaw 2021; van der Waal et al. 2022), wearables (Wang et al. 2020), exergames (Xu et al. 2023) and smart home healthcare services (Kang et al. 2022). Studied demographics are older persons (Alexandrakis et al. 2020; Palas et al. 2022; Xu et al. 2023; Zeng et al. 2023), healthcare professionals (Ketikidis et al. 2012) and informal caregivers for people with dementia (Wójcik et al. 2021), among others. Many of these studies have confirmed the validity of the two models, yet also discussed factors outside of these models that may influence acceptance (Paccoud et al. 2021). To investigate the relations between the different variables of the theories, researchers mostly use surveys and carry out statistical analysis based on such quantitative data. However, some researchers have used these theories to inform their qualitative work (Drehlich et al. 2020; Ehn et al. 2019; Hanif and Lallie 2021; Vandemeulebroucke et al. 2021). Both TAM and UTAUT have, however, been criticized for their arbitrariness (Bagozzi 2007), lack of certain dimensions (Dwivedi et al. 2019) and complexity (Tamilmani et al. 2021). Regarding acceptance of technology in general, qualitative research has uncovered relevant issues that have not yet been discussed in the more quantitative approaches, such as deception, the fear of replacement of human care (Wangmo et al. 2019) or loneliness (Zsiga et al. 2013), among others. Such findings demonstrate that existing models can benefit from qualitative research.

The goal of this qualitative paper is therefore to inform these theories of technology acceptance. More specifically, we aim to (1) validate these theories beyond their usual quantitative validation, (2) provide context and depth to the dimensions of TAM and UTAUT and (3) allow the emergence of additional factors fostering and hindering acceptance of technologies to potential add new, overlooked dimensions important for technology acceptance. To achieve these goals, we are mapping the acceptance of new technologies in aged care among the three stakeholder groups (older persons, professional caregivers and informal caregivers), as well as for a variety of technologies.

Research methodology

Exploring the conditions that facilitate and hinder the adoption of technologies in aged care by older persons and their caregivers in Switzerland was one of the main goals of the RESOURCE project. This mixed-method project involves systematic reviews of literature (Felber et al. 2023; Tian et al. 2024), qualitative interviews on which this paper is based and a quantitative national survey.

Interview guide

The purpose of this qualitative part of the project was to explore the opinions of stakeholders living in Switzerland regarding both existing and potential technologies that could be used to care for older persons. These stakeholders were older persons, professional caregivers and informal caregivers, such as family members or friends. To capture as much data as possible while still offering a degree of consistency, the interviews were fashioned in an explorative, semi-structured manner with open-ended questions (Magaldi and Berler 2020). The interview guide was drafted based on the current literature and after consulting experts in assistive technologies for older persons. It was divided into three sections: section one explored the current caregiving-situation and its challenges; section two discussed specific technologies; section three asked general questions regarding conditions and policies facilitating or hindering the adoption of technology in caregiving, for example, “How could the acceptance of smart home technologies be improved in the future?”.

Section two was further divided into two categories: monitoring technologies and devices that were meant for entertainment and companionship. The former category included emergency wrist buttons, wearables and sensors at home (with and without video). The latter category included virtual reality (VR), the robot Pepper (a humanoid robot that is able to interact with users in a variety of ways, such as speech, demonstrating exercises and games, due to a tablet on its chest) (Tanioka et al. 2019) and the robot Paro (a social robot with the appearance of a baby seal that was developed for people with dementia (Shibata 2012)). The technologies were demonstrated through pictures and videos (in German). Participants were asked a variety of questions for each technology, for example, “What is your opinion on wearable sensors that monitor your health-related data, as well as your location (GPS), e.g. heart rate, blood sugar, sleep phases, physical activities?” or “What benefits or problems do you foresee when using wearable sensors?” They were also asked to compare the technologies among each other, with questions such as “which one of these technologies would you most likely use and why?”.

The original English interview guide was translated to German and French by the first author and then underwent back translation by two independent researchers to ensure consistency and coherence between languages. The original interview guide was then slightly adapted for each stakeholder group.

Ethics

The research project was approved by the Ethics Commission of Northwest and Central Switzerland (EKNZ) under ID: AO_2020-00027. A document containing all relevant information regarding the purpose of the study, the content of the interviews, the measures taken to ensure data privacy and a contact person (first author) in case of doubts were distributed before the interview to each participant. Participants then had the opportunity to discuss the document with the interviewer before the interview and thereafter provided written consent. Participants were also informed that they could withdraw at any point during the interview or after the interview prior to anonymization of data. No participant chose to withdraw participation.

Participant recruitment

Participants were recruited through a variety of techniques, including contacting relevant institutions, such as nursing homes, home care organizations and additional services (such as Meals on Wheels), distributing flyers in points of interest, such as organizations centered around age or caregiving, online advertising, for example, in online journals centered around aging, as well as social media and word of mouth. Purposive and snowball sampling were therefore combined. Conditions for inclusion were the following: (1) being 65 years or older OR being a caregiver to a person 65 years or older and (2) living in Switzerland at the time of conducting the interview.

Data collection

Two native female German-speaker interviewers, trained in qualitative methods and interviewing skills by the last author, carried out data collection. The interviews were recorded using a small portable recorder. The first author is one of the interviewers. She was completing her PhD in biomedical ethics at the time. The other interviewer (acknowledged in the paper) was pursuing her master’s degree in medicine. The first interviewer collected 85% of the interview data, while the second collected 15%. The interviews were on average 96 min long (range: 46–189 min). The interviews were usually conducted in one session, except for cases where the participant was either pressed on time (formal caregivers) or unable to continue the interview without a break (older persons). Most of the 60 interviews were conducted individually except in seven cases: three interviews had two older persons present (married couples), one interview had two informal caregivers present (married couple), one interview had two formal caregivers present, one interview had an older person and her formal caregiver present and one had an older person and her informal caregiver present (mother and son). The participants were able to choose the location of the interview and mostly chose their home (in case of older adults and informal caregivers) or their workplace (in case of formal caregivers). Only one interview was carried out over zoom at the request of the participant.

Study participants

A total of 60 interviews took place with 67 participants. Table 1 offers an overview of the demographics of the study participants.

Table 1 Demographics of participants
Full size table

Data analysis

The interviewers transcribed all interviews verbatim into German. The finished transcripts were then analyzed with the support of the co-authors (third and last authors) using thematic analysis (Braun and Clarke 2006, 2019; Guest et al. 2012). The analysis was first conducted in the group using the software MaxQDA. These group sessions served to familiarize the whole team with the data, as well as to ensure consistency in naming codes and agreeing on what each code would encompass. At least 3 transcripts for each stakeholder group were analyzed fully in this group setting. Thereafter, first author analyzed the remaining transcripts for the professional and informal caregivers, while the third author analyzed the interviews of older persons. The initial coding and theme development were conducted in such a way that they left room for reflexive thematic analysis (Braun and Clarke 2019). This method encourages reflection to reach more depth and interconnectedness in the development of themes. Themes developed after the initial sessions were brought back to the group for further sorting and refinement, until all authors agreed on the relevance and interpretation of themes and codes.

Given the qualitative nature of this article, we refrained from presenting the results in a quantitative manner. However, if a certain feature was very prevalent, for example, mentioned by all stakeholder groups, we identify the feature as such. Translated quotes from German into English are provided for illustration purposes.

Results

In this section, we present our study findings related to features that study participants revealed as fostering and hindering acceptance of new technologies. These features are mapped onto the facets of UTAUT and TAM (as well as some facets appearing in the refined models). As the goal of our project was to discover all potential conditions fostering and hindering acceptance of technology, we conducted our analysis inductively at first, then at a later stage compared them to the currently used acceptance models. We ended up using the terminology of UTAUT for expected gain (performance expectancy) and expected effort (effort expectancy). While reliability is one of the most prevalent feature in our analysis, it is not part of the current acceptance models. We thus chose to present it as an expression of effort expectancy as unreliable technologies increase the effort needed to use them (Figalová et al. 2022). Anxiety is a variable of TAM3 (Rondan-Cataluña et al. 2015; Venkatesh et al. 2012) and is used in our analysis to capture fears related to the use of technology. Furthermore, we modified the original TAM variables of “social influence” and “facilitating conditions.” Firstly, we use the term “social aspects” to capture other social factors that go beyond the perception that important others want the person to use the technology. Secondly, we employ the term “hindering conditions” to capture non-social factors that hinder the uptake of technology. Our results are divided into four sections, first discussing features that were common to all technologies (see Tables 2 and 3 presenting common barriers and facilitators for all technologies). Thereafter, we highlight features that were unique to each of the three types of technology studied (wearables, sensors and robots).

Table 2 Quotes for theme “Barriers common for all technologies”
Full size table
Table 3 Quotes for theme “Facilitators common for all technologies”
Full size table

Barriers common for all technologies

Performance expectancy

Although there was no consensus across all data, utility and necessity (or lack thereof) were themes that emerged across all stakeholder groups and in relation to all technologies. For the most common explanation given for the uselessness of technology was the fact that the participants can also live their life like their parents and grandparents who were fine without such technologies. Even where it was acknowledged that technologies could make life easier, this was not necessarily seen as a good thing. One participant noted that while technology might have utility, he would rather have the ability to function without it. For some technologies, participants weighed the benefits against the costs and often concluded that the negative impacts outweighed the positive consequences.

Effort expectancy

Participants from all groups expressed the view that technologies are not designed for older persons—particularly because of their complexity. Participants articulated what it meant for a technology to be sufficiently simple. For one older participant, “simple” meant having options that are not too many to be confusing and ease of using the technology. Underscoring the complexities of existing technologies, an informal caregiver discussed how difficult it could be even for their generation to keep up to date with technological nuances. Moreover, caregivers from both groups feared that their older care recipient would be overwhelmed by a new device, not being able to handle its complexity, and therefore rely on their caregivers for help whenever the technology would cause issues, thus increasing their caregiving burden. Robots that are meant to be used autonomously by the older persons especially received skepticism regarding increased caregiving burden, as caregivers were not convinced that older persons would be comfortable using the robot without any assistance.

Participants from all groups had concerns regarding the reliability of the discussed technologies, relaying concerns with each of them. There were, for example, worries expressed about false alarms and that the call for support in emergency may not function when needed. With regard to robots, caregivers expressed that a malfunctioning robot would upset their care recipient and that caregivers would be asked to fix issues immediately. In the case of monitoring technologies, all groups reported problems with charging the device, which may result in older persons forgetting to wear the device consistently.

Anxiety

Both formal and informal caregivers discussed how their care recipients could become anxious when using technology and that their confidence often remains low regarding the use of devices. Older persons did not explicitly mention that new devices would make them anxious or that they are not confident in using them. However, they often revealed general reluctance toward any technology that was presented to them as a first reaction. This was especially prominent for robots, where older persons disliked interacting with artificial beings and sometimes even imagined dystopian scenarios where robots would either replace all human caregiving or go rogue. Caregivers from both groups too imagined that older persons could be distressed or anxious when using robots. Furthermore, older persons were worried about technology’s reliance on electricity and the additional waste produced. At least one participant from each stakeholder group mentioned worries about radiation.

Social aspects

All participant groups mentioned that old persons may just be too old to introduce technology into their lives, suggesting that future generations will adopt caregiving technologies more easily because of previous exposure. Furthermore, the participants mentioned how they prefer human caregiving rather than caregiving facilitated by technology. This seemed especially true in regard to robots, as a few participants could envision robots as social companions. Formal caregivers worried that monitoring technologies would diminish the social, interactive aspect of caregiving and that the care recipient would receive less attention, as only data would be used to assess the patient.

Hindering conditions (digital literacy and cost)

Older persons especially complained that they were not informed about technological or digital innovations. Similarly, the caregivers were also surprised at their lack of awareness of technologies that could be used in a caregiving context. Another aspect mentioned by stakeholders that negatively influenced acceptance were most often cost of these technologies.

Facilitators common for all technologies

Performance expectancy

Participants from all groups felt that some technologies would alleviate caregiving burden. The possibility of technology relieving caregivers (both formal and informal) increased acceptance for all participant groups. Participants easily imagined scenarios where technology could offer a benefit, and many mentioned the shortage of human caregivers or their lack of expertise as a reason why technology should be used in caregiving.

Social aspects

A social condition that seemed to increase acceptance was living alone. Participants from each stakeholder group mentioned that they would find the use of technology more acceptable if they or the care recipient were living alone, as they imagined themselves being more vulnerable to accidents and therefore needing a device to alert someone in case of a fall.

Facilitating conditions

Mirroring the hindering condition of not knowing about technological support and cost, participants from all groups wished for more education around technology as well as financial support. Trying out the technology or seeing the benefits first hand was often mentioned as important facilitators by all stakeholder groups.

Barriers to acceptance of wearables

Performance expectancy

We asked participants about wearables that would dispatch automatic alarms in case of a fall and that would monitor their health data (such as sleep, heart rate and steps and potentially blood pressure and other more sophisticated readings in the future) and monitor their location (via GPS). While older persons expressed general lack of interest in wearables, many caregivers specified that health data are not useful in and of itself, as more information is needed to assess how the care recipient is doing (Tables 4 and 5). Specifically, the older person should rather be asked herself how she feels, to get a proper assessment of their status. Informal caregivers felt that the health data would overwhelm them, and that they would not appreciate the added responsibility of checking the health data (Table 5).

Table 4 Quotes for theme “Barriers to acceptance of wearables”
Full size table
Table 5 Barriers and facilitators to acceptance of wearables
Full size table

Effort expectancy

Design came out as a specific barrier to acceptance for wearables regarding the effort to use them. In both caregiver groups, participants reported that older persons would not be able to handle complex devices with many functions and that font size, contrast and small letters may cause issues for people who have deteriorating abilities to see or hear (Tables 4 and 5). Additionally, vibrations may not be helpful in light of reduced sensitivity due to health conditions. Some older persons who had experienced more modern devices than familiar emergency buttons claimed that they are hard to use. Furthermore, all groups mentioned that the current aesthetic of wearables does not appeal to older persons, which they confirmed.

Facilitators of acceptance of wearables

Performance expectancy

The main expectation all participant groups had from wearables was reassurance (Tables 5 and 6). Both caregiver groups spoke of the reassurance of being able to locate a care recipient with dementia, as well as receiving health data from them when they cannot communicate their ailments well. Older persons appreciated the knowledge that they could get help when needed. Also, participants from all groups saw nudging as a useful mechanism to be reminded of appointments or healthy behavior, such as medication intake, thus enabling better self-management.

Table 6 Quotes for theme “Facilitators of acceptance of wearables”
Full size table

Social aspects

Some participants pointed out that wearables may appeal to older adults because wearing them would not be perceived as stigmatizing, because young and health conscious people are also using them (Tables 5 and 6). Caregivers suggested that wearables be fully customizable, regarding what values are monitored and what nudges are given, as well as when and how, so that older persons can adapt the device to their needs. Furthermore, customizability would ensure that older persons do not feel embarrassed because of unwanted nudges in public, as they could silence nudges according to their wishes.

Barriers to acceptance of ambient sensors

Effort expectancy

From all groups, participants mentioned that sensors would be hard to incorporate into the home of their care recipient, especially because some lived in very big apartments or houses (Tables 7 and 8). Both caregiver groups and older persons foresaw sensors as more useable in newly built apartments. When cameras were mentioned, caregivers from both groups worried that their caregiving burden would actually increase with the technology, given that someone needs to watch the produced footage.

Table 7 Quotes for theme “Barriers to acceptance of ambient sensors”
Full size table
Table 8 Barriers and facilitators to acceptance of ambient sensors
Full size table

Facilitators of acceptance of ambient sensors

Performance expectancy

While the idea of surveillance made participants in general uncomfortable and many were worried about data abuse, in each stakeholder group at least some participants appreciated the idea of surveillance for safety purposes or proof that others are fulfilling their roles and acting ethically (Tables 8 and 9). For example, informal caregivers mentioned situations where they wished to monitor the performance of formal caregivers as they were worried about misconduct, while formal caregivers thought of situations where cameras could actually prove their innocence.

Table 9 Quotes for theme “Facilitators of acceptance of ambient sensors”
Full size table

Effort expectancy

Despite expressing that sensors at home would be difficult to install, informal caregivers appreciated the fact that, once set up, the care recipient would be monitored passively and without any effort (Tables 8 and 9). Issues that were mentioned in relation to wearables, for example, the need to remember to wear the device or to charge it, were resolved in the case of installed sensors.

Social aspects

Acceptance increased when study participants thought about the care recipient as being very ill or suffering from cognitive decline (Tables 8 and 9). Once the relationship between the caregiver and the care recipient changes into a more dependent stage, monitoring becomes more acceptable, possibly because of an increased worry for caregivers and thus an increased need for reassurance. Nevertheless, caregivers wanted their care recipients to live their life as normally as possible, despite their increased dependency, and therefore preferred sensors without cameras over more intrusive monitoring methods. Furthermore, older persons themselves thought that video surveillance is only acceptable for people suffering from advanced cognitive decline, and never imagined themselves as being in such a situation in the future.

Barriers to acceptance of robots

Performance expectancy

For both robots presented, participants in all groups expressed the view that they were not advanced enough yet to provide any real benefit (Tables 10 and 11). For the humanoid robot, they thought it would not be able to do any useful chores, both regarding household tasks and caregiving tasks (referred in the table as instrumental tasks). For the pet-like robot, they suggested that it should at least simulate a will of its own and ask for attention, in order to provide a more interesting, real interaction that would make the older persons feel needed and useful. Interestingly, while some caregivers considered robots as being entertaining for older adults, some caregivers from both groups even rejected that use, stating that older adults should not be stimulated and entertained all the time.

Table 10 Quotes for theme “Barriers to acceptance of robots”
Full size table
Table 11 Barriers and facilitators to acceptance of robots
Full size table

Social aspects

Participants from all groups expressed that older persons may feel judged when seen interacting with a robot (Tables 10 and 11). Robots were perceived as childish by some participants, and more appropriate for people with dementia. Older persons thought that they did not provide any form of useful stimulation or education, alluding to the fact that they still wanted to learn and expand their horizon, rather than being just entertained or occupied.

Facilitators of acceptance of robots

Performance expectancy

Some participants from each group saw robots as entertaining and stimulating for older persons (Tables 11 and 12). Humanoid robots were deemed as an interesting, exciting experience, a new form of interaction. Pet-like robots could provide companionship without the burdens that come with taking care of an animal, according to caregivers from both groups. They mentioned how some older persons would still like to have pets as companions, but are unable to take care of them, thus seeing an opportunity for robots to fulfill the needs of companionship without pushing older person’s beyond their capabilities.

Table 12 Quotes for theme “Facilitators of acceptance of robots”
Full size table

Social aspects

Participants from all groups expressed greater acceptance of robots if caregivers were unavailable or, in the case of older persons, unpleasant. Furthermore, some participants from all groups imagined that using a robot for tasks related to hygiene may actually be less shame inducing than interacting with a human caregiver. Participants from both caregiver groups saw the use of robots as being more appropriate in a nursing home setting rather than for older persons still living at home. Some older persons living at home expressed the same preference. In all participant groups, suggestions were raised about how exactly Pepper would be interacting with residents in a nursing home. For example, professional caregivers thought that robots could provide prompts to participants, reminding them of mealtimes or activities that were about to commence. Informal caregivers imagined the robot being used in group activities, discussing a certain subject prompted by Pepper and thus facilitating group interaction.

Discussion

As mentioned in the introduction, the goal of this paper is to add qualitative research to the more quantitative models of acceptance, such as UTAUT (Venkatesh et al. 2003) and TAM (Davis 1989) in order to provide more depth and context to the main facets of technology acceptance discussed in both models, as well as to potentially reveal additional important factors that could improve these models. Thus, we categorized our data similarly to the main variables of these acceptance models. Our study revealed many barriers and facilitators that are similar to the facets of TAM and UTAUT. Nevertheless, we also found additional facets that may be worth including in future acceptance models.

In the context of the current literature, our study substantiates the acceptance models’ assumption that the easier a technology is to use and the more useful it seems to users, the likelihood of acceptance increases (Cimperman et al. 2016; Mao et al. 2015; Tubaishat 2018). Furthermore, there is evidence that the worry of caregivers that older persons would get confused by new technologies is justified and that technologies need to be simple (Chung et al. 2021; Cullen et al. 2022). Reliability is rarely mentioned as a factor in TAM or UTAUT studies (Zeng et al. 2023) and more often mentioned in studies investigating acceptance outside of the TAM and UTAUT models (Dorsten et al. 2009; Gagnon-Roy et al. 2017). Given the explicit and recurrent mention of reliability issues by our stakeholders, we suggest to include it as a predictor for effort expectancy.

Anxiety emerged as a variable in TAM 3, trying to capture the nervousness, apprehensiveness and general discomfort users experience when starting to use a technology and the variable has gained attention in research employing TAM 3 and other models (AlQudah et al. 2021; Dai et al. 2020; Khaksar et al. 2019; Meuter et al. 2003; Rajak and Shaw 2021). The caregivers in our study noticed anxiety in older persons, thus supporting the inclusion of that variable into the acceptance models. Not included within the existing models, our study uncovered that many participants expressed concerns regarding radiation, use of electricity and the production of waste (Li et al. 2019). The artificial character of robots and the fear of them disobeying also seemed to induce anxiety in study participants. Another study found that older persons prefer to limit the degree of autonomy of robots (Scopelliti et al. 2005), a preference that could be related to the fear of disobedience by robots. Furthermore, a fear that was mentioned often and by all participant groups was the replacement of human caregiving through technology. This fear has been found in other studies (Felber et al. 2023; Vandemeulebroucke et al. 2021; Wangmo et al. 2019). In order to gain more understanding of the variable labeled “anxiety” in the acceptance models, we suggest examination into new factors that we found such as radiation, fear of disobedience and replacement of human care.

We included social aspects as a variable in our analysis to capture conditions related to the living situation of older persons, their social image or interactions with others which would influence acceptance of technology. We found that living alone was deemed to foster acceptance of technology. As social influence can also be characterized as understanding the importance of adopting the technology by the targeted user (Rajak and Shaw 2021), living alone may be a factor that positively influences that understanding. Regarding the issue of feeling stigmatized by others when using technologies, especially robots, has been found by other researchers, who used also proposed to add stigmatization under social aspects influencing acceptance (He et al. 2022). We also saw that acceptance of monitoring technology increases for caregivers when the care recipient’s cognitive capabilities decline. This relationship has been found in other studies involving caregivers and people with dementia (White and Montgomery 2014; Williamson et al. 2017). As with the variable of anxiety, we therefore propose to additionally study these dimensions of social aspects so that they can also be added to the acceptance models.

Facilitating conditions such as increased education for technologies was also found by other studies using TAM and UTAUT (Liu et al. 2015; Prayoga and Abraham 2016; van der Waal et al. 2022). Turning this aspect around, the lack of awareness is an issue for the uptake of innovations in healthcare in general (Watkinson et al. 2021). Lastly, cost has been found to be an important barrier in other studies (Dai et al. 2020; Dorsten et al. 2009), and it is part of the updated UTAUT model (Venkatesh et al. 2012). However, lack of awareness has not yet found its way into the acceptance models, although some studies ask for the degree of familiarity with technology (Harris and Rogers 2023; Scopelliti et al. 2005), which is a related question. If a person is not familiar with a technology, she is also not aware of the possibilities offered by the technology.

Future research and implementation

A next potential step could be to investigate how these barriers and facilitators can be employed to provide better caregiving with the assistance of technology. The framework of responsible innovation (RI) deals with a similar question, as one definition of RI describes it as “a transparent, interactive process by which societal actors and innovators become mutually responsive to each other with a view to the (ethical) acceptability, sustainability and societal desirability of the innovation process and its marketable products (in order to allow a proper embedding of scientific and technological advances in our society)” (Schomberg 2013, p.19). Under this definition, society defines, at least partially, which technologies are acceptable and desirable for its participants. While the detailed decisions will come down to the individuals and will differ case by case, the idea of individualization and autonomous decision making (conditions that were affirmed by study participants) need to be agreed upon on a societal level, and then respected by the designers of said technologies. What the different definitions of RI have in common is the assumption that societal participation in innovation will lead to outcomes that are not solely beneficial in an economic sense, but also promote “goodness” overall (for example, for the environment or society in general), which in turn means the involvement of societal actors to determine what this goodness exactly is (Timmermans and Blok 2021). Our study has provided a first step regarding said determination.

Strengths and limitations

Overall, our study has shown that qualitative research investigating acceptance of technologies can add valuable information to established quantitative measures, such as the TAM and UTAUT models and their modifications. Qualitative findings are often rich and nuanced due to its qualitative nature and can provide deeper insights on the understandings of certain terms widely used in a quantitative method (Lipworth et al. 2010; Mallinson 2002). The large scope of our research ensures that our paper lays important groundwork for future quantitative research. Despite our relatively high sample size, our findings are not generalizable. Furthermore, social desirability bias may have influenced our results, as study participants may have answered questions in accordance to what they thought the authors would want them to answer (Bergen and Labonté 2020). While interviewers made an effort to avoid language that would encourage such bias, its presence cannot be excluded.

Conclusion

This paper presents the barriers and facilitators of new technologies in caregiving as perceived by the three main stakeholders—older persons, professional caregivers and informal caregivers. Our qualitative approach found similar variables like those used by TAM and UTAUT, such as performance expectancy, effort expectancy, social influence and facilitating conditions, thus strengthening the models. It also found important nuances to these variables, such as reliability regarding effort expectancy or the fear of radiation regarding the variable of anxiety, which is a variable of TAM 3, thus suggesting how these models could be enriched. Furthermore, as our study included a variety of technologies, ranging from wearables to sensors to robots, we uncovered details regarding the individual features of the technologies that foster or hinder acceptance, such as the issues of charging regarding wearables or the perceived high effort to installing sensors at home. The results provide important insight into the acceptance of technology in aged care and can inform future research and design of technology in order to increase acceptance among the stakeholder groups.