Open Access
CC BY-NC-ND 4.0 · Gesundheitswesen
DOI: 10.1055/a-2560-1100
Original Article

Experiencing digital technologies: The importance of feeling safe in healthcare – A qualitative participatory design

Erfahrungen mit digitalen Technologien: Die Bedeutung des Sicherheitsgefühls im Gesundheitswesen – Ein qualitatives partizipatives Design
Peter Minartz
1   Forschungsbereich Versorgungs- und Implementierungsforschung, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Carmen Vondeberg
1   Forschungsbereich Versorgungs- und Implementierungsforschung, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Laura Obert
1   Forschungsbereich Versorgungs- und Implementierungsforschung, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Christian Bleck
2   Sozial- und Kulturwissenschaften, Hochschule Düsseldorf, Düsseldorf, Germany
,
Bernhard Hemming
3   Allgemeinmedizin und Familienmedizin, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Mara Vöcking
1   Forschungsbereich Versorgungs- und Implementierungsforschung, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Claudia Ose
4   Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Matti Znotka
5   Stabsstelle Unternehmensentwicklung, Katholisches Klinikum Koblenz Montabaur gGmbH, Koblenz, Germany
,
Diana Cürlis
6   Münster School of Design (MSD), FH Münster University of Applied Sciences, Münster, Germany
,
Anne Karrenbrock
1   Forschungsbereich Versorgungs- und Implementierungsforschung, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Frank Oehmichen
7   Sozialmedizin und Ethik, Evangelische Hochschule Dresden, Dresden, Germany
,
Peter Heistermann
8   Program Physician Assistant, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
,
Silke Kuske
1   Forschungsbereich Versorgungs- und Implementierungsforschung, Fliedner Fachhochschule Düsseldorf, Düsseldorf, Germany
› Institutsangaben
 

Abstract

Background

Digital technologies are increasingly used in healthcare. In this context, perceived safety plays a critical role in their acceptance and implementation. Previous research had focused more on data security or specific digital technologies. There has also been a lack of participatory approaches to consider and empower healthcare recipients (and relatives), providers, and technology experts to broaden the phenomenon.

Objective

The aim of this study was to present a comprehensive perspective on the needs, influencing factors and related outcomes in the context of feeling safe with digital technologies in healthcare.

Method

A qualitative, exploratory, and participatory methodology was used with five guide-based focus group workshops. Each workshop explored one of five digital technology demonstrations provided at the beginning: (1) electronic health records (EHR), (2) robotics, (3) artificial intelligence (AI), (4) smart home, and (5) smart hospital. All participants were invited for a guide-based discussion. The study focused on the levels of ‘involvement’ and ‘collaboration’ by also empowering participation. The workshop target groups were actively involved in the development and execution of the workshops and were empowered. The data were analysed via a content analysis approach, with a mostly inductive procedure.

Results

Feeling safe was found to affect thoughts, emotions, and actions. For example, a higher level of perceived safety increased the acceptance of digital technologies, whereas a lower level of perceived safety decreased it. The corresponding needs and influencing factors differed in terms of their scope and focus depending on the context. The aspect of ‘control’ was most frequently addressed by all the target groups and was viewed as relevant in all the workshops. In general, digital technologies were viewed as supplements rather than substitutes for healthcare providers.

Conclusion

This study extends beyond the current state of research on perceived safety and the use of digital technologies in healthcare by providing a comprehensive overview of the corresponding needs and influencing factors at various levels, such as the individual, community-organizational, and system-society levels. The perceived safety of healthcare recipients and providers related to digital technologies should be taken into consideration to achieve positive implementation outcomes.


Zusammenfassung

Hintergrund

Digitale Technologien werden im Gesundheitswesen immer häufiger eingesetzt. Hierbei spielt die wahrgenommene Sicherheit eine entscheidende Rolle für deren Implementierung. Bisherige Forschung konzentrierte sich eher auf die Datensicherheit oder auf spezifische digitale Technologien. Es fehlte ein partizipativer Ansatz, der es Leistungsempfangenden (deren Angehörigen), Leistungserbringenden und Technologieexpert:innen ermöglichte, die Sicht auf das Phänomen zu erweitern.

Ziel

Ziel war es daher erfahrungsbasiert Bedürfnisse, Einflussfaktoren und damit verbundenen Outcomes im Kontext des Sicherheitsgefühls digitaler Technologien im Gesundheitswesen zu erforschen.

Methode

Es wurde ein qualitatives, exploratives und partizipatives Design, basierend auf fünf leitfadengestützten Fokus-Gruppenworkshops durchgeführt. Eine von fünf digitalen Technologien ((1) elektronische Gesundheitsakten (EHR), (2) Robotik, (3) künstliche Intelligenz (AI), (4) Smart Home und (5) Smart Hospital), wurde zu Beginn oder während des Workshops vorgestellt und/oder konnte ausprobiert werden. Alle Teilnehmenden wurden zu einer leitfadengestützten Diskussion eingeladen. Die Partizipation wurde auf den Ebenen der „Beteiligung“ und „Zusammenarbeit“ gefördert, indem aktiv in die Entwicklung und Durchführung der Workshops einbezogen sowie zur Teilnahme befähigt wurde. Die Daten wurden mit Hilfe induktiver Inhaltsanalyse ausgewertet.

Ergebnisse

Das Gefühl der Sicherheit beeinflusst Gedanken, Gefühle und Handlungen. Beispielsweise kann ein höheres Maß an gefühlter Sicherheit die Akzeptanz digitaler Technologien erhöhen, während ein niedrigeres Maß diese verringern kann. Entsprechende Bedürfnisse und Einflussfaktoren unterschieden sich je nach Kontext in ihrem Umfang und Schwerpunkt. Der Aspekt der „Kontrolle“ wurde von allen Zielgruppen am häufigsten angesprochen und in allen Workshops als relevant angesehen. Digitale Technologien wurden eher als Ergänzung, denn als ein Ersatz für Leistungserbringende betrachtet.

Schlussfolgerung

Diese Studie geht über den aktuellen Stand der Forschung im Kontext wahrgenommener Sicherheit und digitaler Technologien im Gesundheitswesen hinaus, indem sie einen umfassenden Überblick über die Bedürfnisse und Einflussfaktoren von der individuellen bis hin zur System-/Gesellschaftsebene gibt. Die gefühlte Sicherheit sollte grundsätzlich berücksichtigt werden, um die Implementierung digitaler Technologien zu fördern.


Abbreviations

AI Artificial intelligence

BMB Federal Ministry of Education and Research

CDs Core dimensions

DT Digital technology

DTs Digital technologies

EHR Electronic health record

IFs Influencing factors

HCP Health care provider

HCR Health care recipient

PRISMA-ScR Preferred Items for Reporting on Systematic Reviews and Meta Analyses – Extension for Scoping Reviews

SteTiG Emotional safety as a condition for success in the digital transformation

WHO World Health Organisation

Introduction

Digitalization has advanced rapidly and influences a variety of aspects of society, including healthcare [1]. However, the implementation of digital technologies (DTs) in healthcare can be challenging, particularly for individuals who are less familiar with technology, and this process can have potentially negative impacts on their perceived safety [2]. Perceived safety as a component of patient safety is as crucial as physical safety [3] [4]. A few studies have highlighted concerns regarding safety in relation to DTs [5] [6], underscoring the importance of investigating individual safety in conjunction with DTs [6]. The World Health Organization (WHO) defines DTs as technologies that facilitate improvements in healthcare, encompassing areas such as artificial intelligence, robotics, and advanced data processing and specifies that DTs should prioritise patient safety among other things and place the individual at the core of the process [7]. Therefore, avoiding situations in which DTs cause physical or psychological harm to people and exploring the relationships between such technologies and perceived safety are even more important [8].

Perceived safety is defined as a state in which the individual feels free from danger [9]. The level of safety is dependent on both internal (e. g., the individual’s willingness to receive support) and external (e. g., the behaviour of others) conditions [10]. For example, an individual’s level of perceived familiarity with DTs is viewed as an internal factor, whereas the use of humanoid robots in healthcare is viewed as an external factor that can evoke an ambivalent level of perceived safety [11].

Our study distinguishes between the existing definitions of the terms “emotional safety”, which refers to perceived safety in general [2], and “psychological safety”, which is more commonly used in the context of (healthcare) providers and their work environment [12]. Studies have shown that DTs can have both positive and negative effects on perceived safety [13]. In this context, fair and person-centred DTs (e. g., telemedicine) require feelings of safety to be considered [14]. Conversely, feeling safe can influence attitudes towards technological innovations and their implementation [15], which are perceived differently depending on the target group in question [9]. Thus, limited perceived safety can negatively influence individuals’ willingness to implement these innovations [16]. Controversly, a higher feeling of safety when DTs are used can facilitate the successful implementation of DTs [17]. In addition, other factors can affect the implementation of specific DTs, including funding, infrastructure and health literacy [18]. However, evidence regarding the feeling of safety in the context of the digitalization of healthcare remains scarce. Previous research on DTs focused more on data security [19] without providing a comprehensive perspective on the phenomenon. Only a few studies have investigated perceived safety as a primary outcome. However, it is evident that there has been a notable shift in emphasis towards this aspect and perceived safety is gaining more scientific focus, with an increased focus on the individual as a key element in population-based care [13]. There has also been a lack of participatory approaches to consider and empower healthcare recipients (and relatives), healthcare providers, and technology experts to broaden and explore the phenomenon. Therefore, the aim of this study was to adopt a comprehensive perspective on the dimensions of people’s needs, influencing factors and related outcomes in the context of feeling safe and the use of DTs in healthcare.


Method

Design

The study employs a qualitative, exploratory and participatory research design [20] based on five participatory guide-based focus group workshops. These workshops were developed and planned in collaboration with the participating experts from the fields of technology and healthcare and were focused on (1) electronic health records (EHR), (2) robotics, (3) artificial intelligence (AI), (4) smart home, and (5) smart hospital. This study was part of a research project titled ‘Emotional safety as a condition for success of the digital transformation in healthcare’ (SteTiG), which was a strategic foresight study based on a participatory mixed-methods design [21]. The overarching study contained a substantial basis for a strategic foresight research design [21] incorporating a scoping review [13], a design ethnographic approach [22] and these participatory workshops. Its data were synthesized and triangulated with the aim of scenario field identification, containing the context of perceived safety and DTs, including needs, influencing factors and related outcomes of the current situation, as a prerequisite for the further steps of strategic foresight.

This study is registered in the Open Science Framework Registry: https://doi.org/10.17605/OSF.IO/UTSQN. Ethical approval: Ethical committee of Fliedner Fachhochschule Düsseldorf, Fliedner Fachhochschule Düsseldorf (04–2022).


Sampling design, recruitment and setting

A criterion-based convenience sample was applied to recruit a mixed sample of participants featuring multiple perspectives, including citizens who received healthcare treatment (healthcare recipients (HCRs)), healthcare providers (HCPs), (caring) relatives, and digital technology (DT) experts. The recruitment aimed to ensure the inclusion of participants of a wide variety of ages, education levels, diseases, types of vulnerability, and degrees of affinity for DTs. To represent a larger patient population, selected target groups of diseases with a high prevalence and/ or with a complex care situation, such as people living with dementia [23], were chosen. A comprehensive range of diseases was included in the entire project [21]. Participants were recruited (mid-April to the end of August 2022) by contacting professional associations for people with disabilities; the federal working group for self-help; various dementia research networks; and local associations, schools, youth centres, networks, social institutions, youth service offices, church institutions, and welfare organizations. Invitations were distributed via social media and institutional homepages and were publicized via a press release.


The participatory approach

The participatory approach aimed to ensure the sustained involvement of all participants, taking into account the “Participation Choice Points in the Research Process”, which considers choices that meet the needs of the people involved [20]. This resulted in cooperative work between the participants and the researchers at the levels of ‘involvement’ and ‘collaboration’ in line with existing evidence [20]. To facilitate a collaborative approach, selected HCPs and technology experts were engaged in the design and development of the workshops. Furthermore, they assumed responsibility for moderating a subset of the workshops and presented preliminary findings to the various groups. A variety of techniques, such as video demonstrations, imagination exercises, selection of specific data collection environments and opportunities to see the DTs in action, were employed to engage and empower the participating individuals, especially HCRs and (their) relatives, during the DT workshops to collect thoughts, needs, influencing factors and related outcomes in the context of feeling safe with DTs in healthcare. Considering the participants’ diversity, two workshops were conducted during a project opening event hosted at a university of applied sciences (one workshop took place via Microsoft Teams). The two others were conducted at a ‘smart hospital’ and a ‘smart home’ area, considering the needs of people living with dementia, and their relatives, at the Fraunhofer Institute Duisburg.


Data collection

The data were collected through participatory guide-based focus group workshops [20]. The research results were subsequently validated via feedback discussions (member checking) [20] for each workshop. The entire data collection period spanned from June 2022 to February 2023. The data collection for each focus group workshop lasted 60–90 minutes and was conducted by four interviewers (SK, CV, AK, and CB). Sociodemographic data were collected prior to the workshops. To establish a familiar situation for people living with dementia, the sociodemographic data were collected by a local professional with whom the participants were familiar, in consultation with the project staff. The communication techniques of paraphrasing and narrating, alongside a communication board and card sorting [24], were used to collect, structure, and prioritize the data with the aim of obtaining a nuanced overview of the research phenomenon. An interview guide was developed (see Appendix A, online ) in accordance with the semistructured qualitative research interview method described by Qu and Dumay (2011) [25]. This guide was divided into three sections: 1. thoughts and perceptions about the DTs (using the think aloud method) [26]; 2. influencing factors (IFs) in the context of perceived safety and DTs; and 3. needs and recommendations. A thematic introduction of the DT (demonstration of the DT, partly video based, and haptic) included in each case required approximately 5 to 30 minutes to complete. The smart home/hospital workshops involved visiting DT areas, presenting and partly testing the DTs, and engaging in discussions about the corresponding experiences (approximately 90 minutes). The workshops were audiotaped and subsequently transcribed [27] by external transcription agencies. The member checking was recorded via field notes.


Data analysis and synthesis

A content analysis was conducted in line with Krippendorff (2018) [28], using primarily an inductive approach for exploration purposes. The transcript of each workshop was analysed completely, while explicit (perceived safety was mentioned explicitly) and implicit data (data were provided that were associated with a question or topic of perceived safety) to prevent interpretation bias were labelled [29] regarding our research phenomenon. First, the data were analysed in an iterative process of open coding and categorization of the meaning units at various levels of abstraction and enriched with each transcript separately. The main categories, subcategories and sub-subcategories were developed inductively. The overarching core dimensions (CDs) were developed and clustered into topical domains, according to our research aim of considering people’s needs, influencing factors and related outcomes in the context of feeling safe and the use of DTs in healthcare. According to the definitions of perceived safety both emotional and psychological safety were considered separately. Perceived safety outcomes, mentioned in the context of feeling (more) safe with the DT, were characterised as a higher level of perceived safety. Those that were mentioned as not feeling safe or only marginally feeling safe with the DT, were characterised as a lower level of perceived safety.

Second, for synthesis purposes, in line with our research question, common and complementary categories as well as their modifications were analysed [30], on a content basis, considering the different target groups, the technical affinity and DTs. The synthesized results were iteratively revised and updated according to the various rounds of feedback. The participants’ technical affinity was investigated on the basis on participant recommendations, and perceived safety outcomes were also synthesized as a result of the first step of content analysis. Third, the psychosocial, ethical, legal, political, and economic perspectives were analysed as part of the data interpretation process. The analysis and synthesis were discussed within the project team and collaborating project partners and were modified as needed. This paper represents results of one study that were triangulated in the overall project for strategic foresight purposes [21].



Results

A total of 34 participants, 25 of whom were women, were included considering diverse populations with a variety of sociodemographic backgrounds, including migration background, chronic illness or both acute and chronic illness, and different levels of education ([Table 1]). HCRs and HCPs (n=18 and n=11, respectively), (caring) relatives (n=2) and experts in the field of DTs (n=3) were included. The participants’ ages ranged from 15 to 83 years. A total of 25 participants took part in the feedback rounds (member checking).

Table 1 Participants’ characteristics and participation.

Total number of participants N=34 (35 participations)

Total n (%)

Nursing Robotics n

Eletronic health record n

Artificial intelligence n*

n*

Smart Hospital n

34 (100)

7

7

7

5

9

Gender

Female

25 (73.53)

6

5

5

3

7

Male

8 (23.53)

2

2

2

2

Divers

1 (2.94)

1

Age

in years R/A/±SD

15–83
37.44±20.34

25–57
34.0±10.2

21–76
38.3±20.2

26–62
41.9±13.5

62–83
72.4±9.0

15–27
19.3±3.6

Migrationbackground

Immigrational

3 (8.82)

1

2

Final school degree

No degree

2 (5.88)

2

Lower-level degree

7 (20.59)

2

1

4

Average-level degree

2 (5.88)

1

1

Higher-level degree

23 (67.65)

7

5

7

3

2

Other (e. g. primary school diploma)

Education

No educational qualification

Vocational (school or academy)

10 (29.41)

2

1

3

5

Technical/master (school/academy)

1 (2.94)

1

University or college degree

18 (52.94)

7

4

6

1

Missing value (no data available)

2 (5.88)

1

1

Other education

e. g. technician

3 (8.82)

3

Target group roles

Healthcare recipients

18 (52.94)

2

3

3

2

8

Of this group, people living with dementia

2 (5.88)

2

Healthcare providers

11 (32.35)

4

4

2

1

1

Informal caregivers

2 (5.88)

2

Experts with (work/research) expertise on digital technology

3 (8.82)

1

2

Disease

Chronic disease

7 (20.59)

2

2

2

1

Acute disease

Both acute and chronic disease

1 (2.94)

1

1

Neither

26 (76.47)

5

5

6

2

8

Technical affinity (a)

Highly tech-savvy

28 (82.35)

7

5

6

2

9

Moderate tech-savvy

Low tech-savvy

6 (17.65)

2

1

3

Member Checking

25 (73.53)

16, incl.: 13 new participants

3

7 incl. 3 new participants

n=number of participants; R=range; A=average; SD=standard deviation; (a) Extracted from interviews; *=Individual who has contributed to two workshops

The impact of perceived safety

A total of 32 outcomes were found to result from higher (n=17) or lower (n=15) levels of perceived safety in a variety of contexts. These outcomes were mostly reported for the target groups as HCRs (n=11), while fewer outcomes were reported for HCPs (n=8), and even fewer were reported for both HCRs and HCPs (n=3), the others addressed, e.g., the healthcare system.

For example, for HCRs, a lower level of perceived safety was associated with anxiety and discomfort, a situation related to a DT design that was not individualized to the target group’s needs. Otherwise, increased perceived safety can lead to increased trust in DTs. For example, for HCPs, a higher level of perceived safety can motivate the implementation of DTs, whereas a lower level of perceived safety resulting from a lower level of transparency in DTs can result in a lack of trust in DTs, thus increasing the corresponding burden. We observed that perceived safety can affect not only the implementation of DT but also the healthcare system in general as well as the economy (see Appendix B, online ).


Needs and influencing factors

A total of 12 domains based on 25 CDs were developed in the context of feeling safe and the use of DTs in healthcare. The 25 CDs contained 74 main categories and 85 subcategories (36 needs, 82 IFs and 13 factors covering both IFs and related needs) (see Appendix C, online ). All the domains varied and covered different levels: the individual, DT, organization, and/or system-society levels. Most of these domains were focused primarily on the (individual) HCR level, with a focus on the DT level (see [Fig. 1]). Three domains were described exclusively in the context of emotional safety, and only one domain was mentioned as an exclusive need related to psychological safety. Eight domains pertained to both emotional and psychological safety.

Zoom
Fig. 1 Domain levels in the context of feeling safe and the use of DT in healthcare Legend: largest black sphere=most-addressed domain; bottom layer=DT level; left layer=HCR level; right layer=HCP level; upper layer=system-society level; rear layer=community-organizational level).

The most frequently addressed domain was ‘control’, which played a crucial role in all five workshops. ‘Control’ encompassed various dimensions that offered different perspectives on DTs and the corresponding contexts. Most of these dimensions covered both needs and IFs. ‘Control’, which was related primarily to the level of DT from the perspective of HCPs, was viewed predominantly as a facilitator of perceived safety and a requirement for data sharing and, moreover, DT usage. However, all the target groups addressed this CD, which was also related to the data authority of the DT user (the system/HCR level). Having control over DTs or over health data was important for perceived safety, whereas being controlled by DTs was discussed contoversly. Having control over the DT was more frequently emphasized by HCPs, whereas having control over one’s own health data was more frequently emphasized by HCRs, especially by those who showed an affinity for DTs. The HCRs and DT experts demanded that AI should not make any independent health decisions and indicated that it must always be monitored by medical professionals. Across the workshops, the opportunity to control the DT (e. g., using an off button) played a significant role in the context of robotics and smart hospitals, which were perceived differently.

Robotic: “So that there is this mechanism at all, that this person still has an instant off button at the end of the day or at the end of the action (…). But what dangers do you have or do you suspect so that you are installing such an off button in the first place?”

In particular, the controllability of DTs by HCPs was related to the need for psychological safety. In the context of EHRs and AI, a sense of control was perceived in cases featuring an analogue or manual backup of data. In contrast, the aspect of being controlled, e. g., monitoring the person receiving care or control of one’s health status, was particularly significant in the smart home and smart hospital workshops, although this topic received less attention. Additionally, seven domains were addressed in four of the five workshops with varying degrees of frequency.

The domain ‘trust’ was exclusively related to emotional safety, encompassing a variety of viewpoints and beliefs held by HCRs regarding their HCPs and the DTs. For example, a positive impact of DTs on HCPs availability was reported. HCPs can establish confidence in HCRs by establishing reliable relationships, facilitating initial contact, and introducing a DT workflow to HCRs. This issue is addressed in the context of robotics and AI, particularly among individuals who exhibit lower levels of digital (health) literacy. Existing trust in HCPs is viewed as an essential prerequisite for access to EHRs. ‘Trust’ was viewed as positively related to emotional safety, but ‘trust’ was viewed as challenging in cases featuring reduced personal contact during the usage of DTs, including decreased eye contact or in emergency situations.

AI: “(…) emotional safety is an issue, the question of when do I actually trust such a system in emergency situations and (…) you have to take ethical issues into account, consider them, don’t go into it naively, the issue of big data […]”

The domain ‘design & handling’ focused primarily on emotional safety and was almost exclusively related to the DT level. This domain encompasses aspects such as the customizability of DTs, practical handling, and life-oriented design. For example, user-specific customization and the human-like appearance of the DT were described as factors that could promote perceived safety, especially when it was titled as a need.

Smart Home: “(…) the feeling of safety could possibly be increased if [the voice of the DT] (…) is a familiar voice [and not an artificial one]”

In contrast, impractical treatment may decrease perceived safety. However, the IFs and needs of this domain were not explicitly mentioned in the EHR workshop. This domain was particularly important for people who exhibited a higher affinity for DTs, although it was also one of the domains most frequently addressed by individuals with a lower affinity for DTs.

The domain ‘experiences and attitudes’ was classified as follows: the attitudes expressed were often shaped by past experiences but also partly reflected basic attitudes in the form of IFs. For example, past contact with DTs or experienced media reports may have potential impacts on perceived safety. Basic attitudes towards DTs and users’ perceptions that DTs cannot substitute for humans can negatively affect perceived safety. This aspect was considered by individuals who showed varying levels of digital literacy. Their experiences were expressed at the DT level, whereas their attitudes towards DTs were observed at the society level. The described domain was not addressed in the EHR context; however, this factor was highly prioritized in the other workshops. In this domain, individuals who showed a lower affinity with DTs cited concerns that they might destroy the DT as a factor that inhibited their ability to feel safe.

In contrast, in the domain of ‘technical functionality and quality’, the need for standardized certification systems, as a quality criterion, was viewed as a prerequisite for feeling safe.

AI: “I would wish for emotional safety if the AI applications in some form, perhaps undergo a certification that simply ensures that certain, certain standards are adhered to”.

In this domain, the perceptible quality of the DT as well as its (un)reliable functionality must be considered, as these factors tend to attract people with higher technical affinity. The reliability of warnings, functions and diagnoses is highly important. Conversely, an unreliable DT was associated with decreased perceived safety. This domain had a stronger influence on emotional safety but was also a need for psychological safety by HCPs. In the context of the EHR, the domain was not addressed.

The domain ‘knowledge & competence’ addressed both emotional and psychological safety. Competence with and knowledge about the DT were stressed both by HCPs and HCRs, but focussed on HCRs’ specifically on the impacts of uncertainties in managing DTs and reduced perceived safety ability. The domain was also considered at the system-society level, as the establishment of the general conditions necessary to support such development is a governmental task. Furthermore, for HCPs the issue of financial support for competence development was viewed as important. The participants with lower levels of digital affinity identified a need for competence as a factor that facilitates emotional safety.

EHR: “So I think that people who are less related to digital aspects find it much more difficult to deal with them (…) We older people learn too. Only you younger ones have to bring time with you. We can’t process it that quickly (…) Yes, if I can do that, for example, then I feel safe. No in this case. Because then I can also, I say, get influence, because I can’t get influence like that and when I have influence, I feel safe”.

For HCRs, this domain was discussed both in the context of EHR and in the context of smart home. In contrast, psychological safety was discussed in relation to robotics as well as smart hospital and smart home. HCPs, especially those who showed a higher level of technical literacy, emphasized the need to develop competence in dealing with DTs and viewed this competence as a determining factor.

The domain ‘efficiency & effects’, not mentioned in the EHR workshop, pertained to the perceived efficiency of HCPs’ use of DTs and was primarily observed at the system-society level; furthermore, it was relevant mainly for HCPs. This domain included not only greater efficiency in healthcare provision from a system perspective but also the potential increase in health impact resulting from the use of DTs at the individual level. Overall, this domain was predominantly discussed by participants who exhibited a higher affinity for DTs in the smart hospital workshop.

The domain ‘prerequisites’, not mentioned in the AI workshop, included the need to ensure equal access to DTs at the system-society level and was of the greatest relevance. The objective of providing equal access to all user groups was mentioned by both HCPs and HCRs, who almost exclusively exhibited a higher affinity for DTs. However, uncertainties were associated with the legal and financial conditions of DT services. These aspects were observed at the individual HCP level and implementation conditions at the organizational level were identified to ensure the psychological safety of HCPs. Additionally, readiness-related factors of HCRs were relevant.

The domain ‘support’ was addressed in three of the five workshops (i. e., the EHR, smart home, and smart hospital workshops). Most related discussions focused on emotional safety, with some consideration of psychological safety. This domain was related to the provision of healthcare using DTs, e. g., assisting HCPs in activities extending beyond their everyday tasks. This type of support was related to the DT level and was also discussed with regard to individuals who exhibited limited DT affinities at the system-society level. With respect to psychological safety, DTs support HCPs in the task of providing healthcare for HCRs, e. g., on the basis of the the life-sustaining features of DTs. We observed that emotional safety, in contrast to psychological safety, was evidently more strongly influenced by support.

The domains ‘data protection & security’ and ‘need for recognition’ were covered in two workshops. ‘Data protection and security’, which was mentioned in the EHR and smart hospital workshops, emphasized expectations related to restricted privacy, anonymity, and data security while using DTs, which were identified as a requirement for emotional safety by most of the participants. However, the advantages of DTs were also viewed as potentially leading to the risk of compromised data protection.

EHR: “And the whole thing also makes me feel not very safe, because you keep reading about data theft, you keep reading about server failures; (…) Will I still be able to access my health data in an emergency (…)?”

This domain was addressed only by people who showed an affinity for DTs and was often identified as a concern at the system-society level. The domain ‘need for recognition’ was also considered at the system-society level and was mentioned in relation to robotics and AI, highlighting the political and professional recognition of HCPs in their approach, als well as the central role they play in the implementation of DTs. The domain was expressed solely by individuals with a higher affinity for DT at the level of psychological safety needs. The less frequently addressed domain, ‘autonomy of recipients’, pertained to the potential to maintain autonomy by using DTs in the smart home context, by enabling the HCRs to remain in their home and preserve their independence. Notably, caring relatives notably highlighted the significance of this factor, which is also reinforced at the community-organizational level when individuals can remain in their homes.


The core dimensions of needs and influencing factors pertaining to different DTs

The results regarding DTs, such as robotics, EHRs, AI, smart home, and smart hospital, varied. AI was linked to a discernibly high-quality DT and the optimization of healthcare, and it offered advanced medical diagnosis and treatment capabilities. The participants highlighted the importance of regulating AI to ensure safety, whereas care providers highlighted the need for control measures and transparency throughout the process of AI development and implementation.

AI: “(…) So, if the person then just, I’ll say, blindly trusts, but can’t know or isn’t given any real [factual basis], then I think that’s a bit more difficult. So - I wouldn’t feel so safe now. So, if a self-driving car suddenly initiates emergency braking but doesn’t tell me why (laughs), that’s where we get into this area. Yes. I think that’s important.”

With respect to robotics, the participants highlighted the significance of HCP oversight of its use, including the option to terminate the process during emergencies. Positive feedback concerning the human design of robotics, its potential for resource preservation, relief for caregivers, and the efficiency of care was provided. In the realm of smart home technology, attitudes towards this factor and HCRs’ competence and interest played crucial roles in the participants’ perceived safety. In the smart hospital workshop, negative effects and inappropriate applications of DTs in healthcare were viewed as compromising perceived safety, thereby potentially increasing dysfunctionality. Limited privacy expectations were perceived as inhibiting factors. The EHR has encountered issues regarding the equity of access to DTs, limitations in terms of data protection and security, the centralization of organization and communication, and the education of HCPs and HCRs.


The overarching target group and specific perceived safety

The twenty-five core dimensions (CDs) were strengthened by 178 participants’ contributions. Most of these contributions (n=81) were made by HCPs, followed by HCRs (n=66), thereof 12 were made by people living with dementia, 20 by caring relatives, and 11 by DT experts. We observed that the different target groups had different perspectives on feeling safe, and they also emphasized the perspectives of the other groups. For example, the HCPs addressed emotional safety from the perspective of HCRs and vice versa. Overall, the participating groups largely addressed similar domains, except for the DT expert group, who addressed a narrower set of domains. The data facilitated the identification of the participant groups’ priorities. For example, the effectiveness and health impact of DTs on HCRs impact HCPs’ perceived safety, whereas HCRs’ perceived safety is influenced by their trust, attitudes, and experiences with DTs. Notably, HCPs were the only group that addressed all twelve domains (see [Fig. 2]).

Zoom
Fig. 2 Core dimensions according to target groups Legend: DT=Digital technology.

Several participants had a higher affinity for technology as well as greater interest in and experience with the use of DTs (n=28). People who showed affinity for DTs covered the domains significantly more often (n=120) than did people who lacked such affinity (n=28). The participants who showed a lower DT affinity mostly addressed aspects of the domain ‘knowledge & competence’, followed by ‘support’ and ‘design & handling’. The participants who showed a higher affinity for DTs frequently mentioned that perceived safety was influenced by DT-related improvements in healthcare and outcomes.

HCRs identified ‘knowledge & competence’ as particularly important while also mentioning the importance of ‘trust’, ‘prerequisites’, ‘efficiency & effects’ and ‘control’. Perceived safety was influenced by the uncertainty of DT use and the possibility of gaining competence in this practice. DTs cannot replace interpersonal interactions such as eye contact because of the corresponding loss of trust on the part of HCRs. The participants in the smart home workshop (mostly people living with dementia and their (caring) relatives) predominantly focused on aspects of ‘design & handling’, which were especially crucial in their immediate living environment. Therefore, DT individualization, which can be utilized in a discrete manner and function covertly to prevent discomfort associated with visible DTs, is important. The participants mentioned that DT is a valuable instrument for controlling and preserving their own health. Monitoring features facilitated continuous surveillance of the individual under their care. HCPs and caring relatives emphasized the ability of DTs to maintain the autonomy of HCRs and the importance of ensuring enhanced physical safety in the home environment via DT support.

Regarding the smart hospital workshop (which included mostly young adults), various domains, including ‘efficiency & effects’, ‘data protection & security’, ‘trust’ and ‘support’, were relevant. The participants specifically mentioned that DTs should be able to be handled in cases of dysfunctionality and potential negative effects.

Smart hospital: “(…) you just don’t notice everything, then you tend to feel safer (…) perhaps that problems occur and of course it always has a better effect on the patient when modern aids are everywhere in the room (…) when you know that you can, you simply have the opportunity to intervene and that it’s not just technology, but that you can still act, that gives you a feeling of - yes, safety and dependability.”

They highlighted the risks related to privacy restrictions and data protection that could arise from the use of DTs. DT experts focused mainly on the domain ‘efficiency & effects’ followed by the domains ‘experiences & attitudes’ and ‘control’. They identified the potential of optimizing the diagnostic process via DTs and extensive AI data access as factors that could promote safety. A lack of transparency in the decision-making process related to DTs and the corresponding diagnostic pathways alongside cases of incorrect suggestions could undermine perceived safety. As a result, these experts claimed that it was crucial to have control over DTs, in which context data verification and the option to terminate DT use were particularly critical factors. Additionally, according to the recommendations of DT experts, the responsibility for decision-making should be with the HCPs, and the DT must not be granted autonomy in terms of decision-making. For HCPs, ‘efficiency & effects’ was also the most frequently addressed domain, closely followed by ‘knowledge & competence’; HCPs thus overlapped with DT experts in this respect. HCPs prioritized improved and efficient healthcare by acknowledging that the use of DTs impacts HCPs’ perceived safety. In this context, the opportunity to gain additional competencies via DTs was viewed as particularly important.


Political, legal, economic, ethical, and psychosocial perspectives on perceived safety

Overall, the participants focused on ethical and psychosocial factors and needs. For example, participants emphasized the importance of enhancing the professional image of HCPs within society and recognizing the significance of nursing in light of digitalization. In addition to this psychosocial perspective, HCRs mentioned that using DTs enabled them to move freely within their home without further assistance, and consequently participate in social activities, which in turn positively affected their perceived safety. However, the excessive use of robots, alongside the corresponding decrease in social interactions, was associated with a negative impact on perceived safety. Equal access to DTs, regardless of income or age, was often cited as an ethical imperative. Another ethical consideration involved striking a delicate balance between safeguarding privacy and protecting confidential data while providing healthcare in the context of emergencies or monitoring patients’ health status. Although ethics were not explicitly mentioned, relevant issues were paraphrased in this context. The political and legal perspectives were addressed less frequently overall but were mentioned directly on numerous occasions pertaining to the tasks of implementing and structuring DTs in the context of healthcare, educating relevant actors about DTs, establishing a framework and the necessary conditions for DT implementation, and emphasizing the importance of the political recognition and prioritization of psychosocial care. The legal discussion of DTs focused on the tasks of implementing and creating a legal framework for DTs, including quality assurance of DTs in accordance with legal regulations and the creation of laws to promote the safe use of DTs (including data protection). Financial aspects, such as funding for necessary DT training, were also regularly addressed from an economic perspective. Additionally, the economic standpoint was viewed as promoting safety, which can lead to cost savings in the context of healthcare.



Discussion

The IFs of perceived safety varied by their scope and focus. They were discussed in the context of different DTs, with some specific to a particular DT. The results were comparable among the participant groups, with some groups addressing identical domains and others expressing different priorities and specificities. The key findings showed that perceived safety is a complex and multidimensional phenomenon, including its context and varies with the type of DT. For robotics, for example, there were triggered discussions around control mechanisms, wereas at the AI workshop the participants raised concerns about the reliability and accuracy of diagnostic algorithms. Smart hospitals were seen as beneficial for maintaining independence, but participants voiced concerns about privacy and data security​ as influencing factors for perceived safety. Overall, the findings suggest that individual factors impact an individual’s feelings of safety when interacting with DTs. This claim is supported by the research of Akalin et al. (2022) [8] on perceived safety in the context of human-robot interactions, which identified unique human character traits as predictors of perceived safety. Most IFs and needs are observed at the individual level, strengthening previous research by Nyholm et al. (2021) [11], who reported that perceived safety is strongly dependent on the fulfilment of the individual’s needs and can be highly variable. Our findings indicated the existence of variance in perceived safety among different target groups, as well as specific variations across different levels of influence. Notably, an individual’s level of DT affinity also influences their level of perceived safety, with participants with a lower affinity for DTs being more likely to identify IFs at the DT level. These findings suggest that people with a lower affinity for DTs perceive them on a direct level and that their feelings of safety can be influenced primarily by the level of DT in question. These results are consistent with those reported by Schirmer et al. (2023) [31], who emphasized the importance of user-centred designs and the usability and applicability of DTs for people with lower levels of DT literacy. In this context, Johannessen (2021) [32] suggested that future users should be involved as codesigners to prevent this issue in the implementation of DTs. In contrast, individuals who exhibit a higher affinity for DT tend to focus more on the processes underlying DTs and are more susceptible to the impacts of the system-society and community-organizational levels on their perceived safety.

HCPs also discussed design and usability in terms of how these factors affect their perceived safety. As described by Ekman and Halpern (2015) [33], this shift in perspective among HCPs may be attributed to their ‘professional empathy’, which leads them to view themselves as primarily accountable for the well-being of HCRs. This fact may also be related to the predominance of emotional safety rather than psychological safety as an IF. All the target groups, in turn, expressed concerns that the use of DTs at home may decrease personal interactions with others and subsequently decrease perceived safety, a situation that was also described by Åkerlind et al. (2018) [9]. Additionally, concerns were expressed regarding the negative effects of increased DT use, including the risk of losing social contact. These fundamental attitudes towards DTs, alongside people’s initial experiences using them, can influence perceived safety. Furthermore, the underlying ethical framework in which DTs are introduced can also impact such perceptions. A potential unequal distribution of access to DTs, for example, may lead to negative attitudes towards DTs and thus affect perceived safety. Unequal access to DTs was previously described by Kim et al. (2009) [34] and, in relation to perceived safety, it was also identified by Lounsbury et al. (2021) [35] as a major challenge for healthcare in the future.

Perceived safety can be affected either positively or negatively by various factors and consequently influence the thoughts, emotions and actions of the people involved, as well as the implementation and provision of healthcare. This finding is in line with the conclusions of Nyholm et al. (2021) [11], who claimed that reduced knowledge about DTs can result in a decrease in perceived safety, leading to further fears that DTs can dehumanize the healthcare system, which ultimately decrease the acceptability of DTs.

We identified 12 domains of IFs and needs. These results are partly comparable to those reported by Akalin et al. (2022) [8], but our findings extend beyond those of previous studies by adopting a broader perspective on relevant DTs and target groups. Using an experimental research design featuring 27 younger participants, Akalin et al. (2022) [8] investigated the relationships among the factors influencing perceived safety in human-robot interactions. These authors focused on one DT and one target group and described six domains. Our domains were based on five different DTs and four different target groups and thus constituted an appropriate extension of previous research results. The present study also differed from that of Akalin et al. (2022) in terms of method [8]. All the other IFs that Akalin et al. (2022) [8] described, such as comfort, predictability, transparency, trust, experience, and familiarity, were also mentioned by our participants, albeit to varying degrees.

Across all the workshops, control was identified as an important factor influencing perceived safety, which can operate at several levels, such as ensuring control over the DT and gaining control by using the DT. As a result, both levels represent an increase in the feeling of control over one’s everyday life, a so-called ‘sense of control’, which was previously described by Akalin et al. (2022) [8]. Against this backdrop, vulnerable participants and their relatives in particular perceived DT as a means of maintaining control by handing over control to DTs with the aim of strengthening their own autonomy, as also described by Åkerlind et al. (2018) [9]. In particular, trust was impacted by prior experience and by other actors involved in the use of DTs, such as HCPs. Christoforou et al. (2020) [36] explained that when HCPs provide a comprehensive introduction to the use of DTs, this approach can positively influence HCRs and increase trust. Similarly, Lynch et al. (2022) [37] highlighted the importance of incorporating DTs as an integral part of the clinical care team to increase trust.

Our data reveal notable discrepancies between the target groups with respect to their needs and the factors that influence their perceptions of safety. For example, HCRs stated that they must develop competence and trust in using DTs to feel safer, especially considering concerns about data security and the loss of human interaction. It seems essential to ensure that DTs supplement, rather than replace, personal contact in healthcare settings to foster adoption among this group and strengthen their perceived safety in the use of DTs. This is in line with the findings of Minartz et al. (2023) [13], who indicates that the interpersonal interaction between the HCP and the HCR is an important factor for ensuring emotional safety [13] and has an impact on decision-making [38]. For HCPs, digital literacy is critical to ensure that they can confidently integrate DTs into their practice while maintaining control over clinical decisions, which with the findings of Wosny et al. (2023) [39], who studied HCPs' experience of using DTs in hospitals and reported new DTs can often cause anxiety during usage and can threaten their professional autonomy [39]. Our findings underscore the necessity for DTs to facilitate, rather than overrule, professional autonomy. They also underscore the importance of tools that enhance workflow and patient outcomes without undermining the role of healthcare professionals; this is in line with the views of DT experts, who emphasised the importance of transparency, control and data security in the development of DTs. They also highlighted that HCPs should retain decision-making authority when using DTs, alongside the ability to review and manage technological processes, which is essential to maintaining trust and perceived safety in the diagnostic process. As defined by Lysaght et al. (2019) [40], from an ethical perspective, AI technologies in particular are only fit for use to assist in clinical decision-making and should demonstrate a high degree of transparency when they recommend, for instance, a decision contrary to that of the HCP [40]. There is a need for targeted support and user-friendly designs for caring relatives to strengthen their perceived safety and allow them to effectively utilise DTs in home care settings. The discrete and nonintrusive design of DTs is particularly important in maintaining comfort and ensuring usability for both the caring relative and the HCR, as seen in the smart home context, especially for people living with dementia. The visibility of DTs is a factor that is perceived differently by individuals, which can lead to varying perceptions of safety depending on the individual and whether a visible or invisible technology is preferred [13]. We showed that perceived safety can affect feelings and attitudes towards DTs in healthcare and influence anticipated implementation.


Strengths and Limitations

To ensure a representative and balanced sample with a variety of experiences, we recruited participants from a broad range of settings. Our participatory approach, aimed to facilitate the active involvement of stakeholders and DT users to enhance the quality and sustainable benefits of the research [41]. The participatory approach allowed us to involve several target groups including vulnerable people such as people living with dementia. However, not all participants such as HCRs were involved in all stages of the research, such as the workshop design, moderation or content analysis of the results, to avoid overburdening them regarding time resources or mental overload. On the one hand, it was difficult to capture the participants’ perspectives because some HCPs changed their perspective to take account of the HCRs, and on the other hand, member checking led to a coordinated interpretation of the data. The intensity of the preferred involvement in selected phases of the study should have been discussed more with all participants before the study started. Although approximately half of the participants had a university education, some had only a vocational or technical/master education. We were only able to enlist a minor number of individuals with diverse diseases and family members. However, this limitation was compensated for in the overall study by triangulation. Furthermore, our overall study [21] showed that this methodology allowed us to capture a substantial part of the overall picture of the phenomenon, which was further enriched by triangulating data from the Scoping Review [13] and the design ethnographic approach [22]. Therefore, limitations in terms of data generalizability affect this research because of the design we employed. However, this approach is based on enhanced triangulation dependability. Before data collection, the participants were informed of the study phenomenon via informed consent, which might have influenced their responses. However, a think-aloud method was used to ensure considerations other than those related to perceived safety.


Conclusion

This study extendes beyond the extant research on perceived safety and DTs in the context of healthcare and provides a comprehensive overview of the factors that impact perceived safety, demonstrating differences on both their scope and focus, such that some factors are relevant across multiple DTs, whereas others are specific to certain DTs. The results emphasize the significant influence of individual factors on feelings of safety in the context of DT use. The degree of DT affinity emerged as a significant factor, as individuals with a lower affinity for DTs tended to focus on DT levels, whereas those with a stronger familiarity with technology considered the corresponding community-organizational and system-society impacts. Further research is needed to examine the applicability of the dimensions of perceived safety in a variety of alternate DT healthcare settings and with reference to different target groups. Our results can serve as a starting point for the development of a framework to assess perceived safety in the context of DTs in healthcare. In the future, it is crucial to emphasize the perspectives of HCPs more strongly and to explore psychological safety in the context of the provision of DT services. To measure the degree to which DTs and their use it are correlated with feelings of safety, quantitative research is needed.

Ethical approval and consent to participate

This study is part of a research project titled ‘Emotional safety as a condition for success of the digital transformation in healthcare’ (SteTiG), which is registered in the Open Science Framework Registry: https://doi.org/10.17605/OSF.IO/UTSQN. The protocols for the entire project were approved by the Ethical Committee of Fliedner Fachhochschule Düsseldorf (University of applied sciences) (04–2022). The study was conducted in accordance with the Declaration of Helsinki.


Consent for publication

All study participants and/or their legal guardians provided informed consent and agreed to publication.


Data Availability

All supplementary data can be provided upon request from the corresponding author.


Contributions

All the authors made significant contributions to this manuscript. SK, CV, AK, CB, BH, MV, CO, MZ, DC, FO, and PH were responsible for the study design. All the authors supported the participant recruitment process and were involved in the workshop conception. CV, AK, and SK conducted the data collection. CB supported the data collection in the smart home workshop, while BH advised the data collection regarding selected DTs. CV analysed the EHR, robotics, AI, and smart hospital workshops; LO analysed the smart home workshop; and CV conducted the total analysis. PM performed a profound synthesis to facilitate data interpretation. SK supervised the research. The fundamental structure shown in [Figs. 1] and [2] were developed by SK and were discussed and refined by SK, CV, and PM, who designed the figures. PM, CV and SK were responsible for preparing the manuscript, while all the authors reviewed the data used in this study and edited and approved the final version.


Funding Information

This research was funded by the Federal Ministry of Education and Research Germany (BMBF - Bundesministerium für Bildung und Forschung): Grant 16INS108. The responsibility for the content of this publication lies with the authors.



Conflict of Interest

The authors declare that they have no conflict of interest.

Zusätzliches Material


Correspondence

Prof. Silke Kuske
Fliedner Fachhochschule Düsseldorf
Forschungsbereich Versorgungs- und Implementierungsforschung
Geschwister-Aufricht-Straße 9
40489 Düsseldorf
Germany   

Publikationsverlauf

Eingereicht: 22. Juli 2024

Angenommen nach Revision: 07. März 2025

Artikel online veröffentlicht:
07. Mai 2025

© 2025. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Georg Thieme Verlag KG
Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany


Zoom
Fig. 1 Domain levels in the context of feeling safe and the use of DT in healthcare Legend: largest black sphere=most-addressed domain; bottom layer=DT level; left layer=HCR level; right layer=HCP level; upper layer=system-society level; rear layer=community-organizational level).
Zoom
Fig. 2 Core dimensions according to target groups Legend: DT=Digital technology.