Emotional and psychological safety in the context of digital transformation in healthcare: a mixed-method strategic foresight study
•,,,,,,,,,,,,.
...
Abstract
Background Perceived safety has received attention in the digital transformation of healthcare. However, the impact of perceived safety on the future of digital transformation has not been fully elucidated.
Aim To investigate perceived safety in the context of the digital transformation of healthcare while considering relevant needs, influencing factors and impacts, including crisis events, to provide recommendations for action based on a participatory, multiperspective, strategic 5-year foresight viewpoint.
Methods A strategic foresight study is conducted via a participatory mixed-methods design to understand the present related factors that are likely to be relevant to future developments in the digital transformation of healthcare.
Results We observed that feeling safe plays a complex role in the digital transformation of healthcare. How perceived safety is considered has and will continue to impact the individual, organisational and system levels. Regarding a potential crisis event, controversial consequences have been observed. At its core, digital (health) literacy related to equity of access and human support is one of the crucial aspects in the context of perceived safety related to the successful implementation of digital technologies in healthcare.
Conclusions The scenarios showed that a continuation of the current situation over the next 5 years may result in partly desirable and partly undesirable outcomes. Concrete key factors should be used in practice to support both education and healthcare quality development and research. The essence of the scenarios should serve as a starting point for research agenda setting and political decision-making in the future. However, additional research is needed to quantify the correlations among the relevant factors.
What is already known on this topic
Perceived safety has received increasing attention in the context of quality and safety research and the digital transformation of healthcare but remains poorly understood.
What this study adds
Emotional and psychological safety plays a critical role in both present and future healthcare in a complex and controversial way, including in the case of a crisis.
How this study might affect research, practice or policy
This study sheds new light on key factors and future projections that can facilitate the enhancement of perceived safety in digital transformation in the contexts of education, politics and research.
Background
Digital transformation (DTR) refers to an improvement process that involves facilitating changes based on the use of data, information and digital technologies (DTs).1 DTR can be viewed in context at the individual, organisational and societal levels,2 and it focuses on digitalisation and digitisation,3 which refer to the use of DTs to manage data and processes2 and to convert analogue data into digital data, respectively.3 DTs, for example, systems that focus on diagnostics, consulting, therapy4 and care,5 aim to increase healthcare benefits.6 In this context, perceived safety is of increasing interest because it is related to the healthcare recipient’s (HCRs) needs and trust, including the implementation and acceptance of DTs in healthcare.6 Emotional safety refers to a state of being free from threats of physical or emotional harm and is often conceptualised as part of a continuum of feeling more or less safe owing to the influence of internal and external factors.6 Recent studies have shown that emotional safety is closely related with (physical) safety, but that restrictive perceived safety can have negative consequences on basic care and patient safety.7 8 From a recent psychological perspective, the fulfilment of human needs, especially the need for social belonging, is necessary for feeling safe. Additionally, safety judgements are associated with social aspects.9 Similarly, psychological safety ‘as an interpersonal phenomenon, is most clearly understood to characterise small social systems like work groups in which individuals interact with each other’.10 In this context, studies have shown that the psychological safety of healthcare providers (HCPs) is related to several factors at the individual, team and organisational levels, such as teamwork, with consequences for patient safety.11 Research on this topic related to DTs in healthcare is scarce, with only a few studies to date having focused on this phenomenon.6 Generally, overarching societal and ethical perspectives are needed to reflect the benefits and challenges of DTs, also related to (perceived) safety12; however, knowledge concerning future scenarios in the context of feeling safe and DTR is lacking. Therefore, we aimed to investigate perceived safety in the context of the DTR of healthcare while considering the relevant needs, influencing factors and impacts, including crisis events, with the goal of developing recommendations for action on the basis of a participatory, multiperspective, strategic 5-year foresight approach.
Method
In this strategic foresight study, we employed a participatory, mixed-method (qualitative and quantitative), parallel and sequential research design.13 Strategic foresight focuses on understanding the present with the goal of systematically obtaining ideas regarding the factors that are likely to be relevant to future changes, thus shaping present-day decisions.13 The participatory approach14 entails continuous structured participant involvement and member checking,15 including reflection, adaptation to study results, and decision-making performed in collaboration with the research team. The scientific standards of research involving mixed methods and strategic foresight are considered.16 This study is registered with the Open Science Framework at the Centre for Open Science: https://doi.org/10.17605/OSF.IO/UTSQN.
Sample design and recruitment
A multiperspective mixed sample of HCRs (eg, patients and residents), HCPs (eg, physicians) and (caring) relatives was recruited for this study, which provided a wide range of ages, school degrees and levels of education. A criterion-based sample of DT experts from the field of strategic foresight and safety research, as well as experts with research expertise and/or (structural) responsibilities in ethical, health-related, legal, political and economic settings, was also invited. All participants were invited to take part in all phases repeatedly, and a sample size of 30 participants in each of the scenario rounds was calculated. The broad recruiting approach used in this research (from 20 February 2023 to 20 June 2023) involved public relations strategies (social media, institutional homepages and press releases) that included contacting the public, social services, healthcare institutions and scientific networks.
Data collection and analysis
The strategic foresight process included five core phases16: (1) scenario field identification; (2) key factor (KF) identification; (3) KF projection; (4) scenario development and (5) scenario finalisation and interpretation/transfer (see figure 1). Several data collection methods were used, namely, guide-based workshops, interviews and an online survey. The data collection process was based on a combined quantitative systematic-formalised and qualitative creative-narrative approach.16 Sociodemographic data and expertise were initially collected. In each phase, previous results were presented, reflected on, discussed and approved or disapproved.
The strategic foresight approach (process and methods) based on Kosow and Gaßner.16
Scenario field identification
The first phase (22 June 2022–30 January 2023), which involved a description of the situation, focused on synthesised data (domains, key dimensions and categories) pertaining to the context of perceived safety and DTs, including needs, influencing factors and related outcomes. A scoping review,6 5 participatory workshops and 10 design ethnographic use cases17 were used. The ‘Classification of Digital Health Interventions V.1.0’18 was applied for selecting a broad range of DTs, for example, artificial intelligence, robotics, electronic health records, DTs and information systems used in smart homes and (smart) hospitals, health apps, virtual reality, telemedicine, digital simulation training in emergency care, mobile ECG cases and a closed-loop system. The data were subjected to content analysis19 and then synthesised and triangulated20 for the subsequent phase.
KF identification
The second phase (from 20 February 2023 to 8 March 2023 (reminder 28 February 2023), involved the systematic selection of KFs via a voluntary cross-sectional online survey that considered scientific standards.21 A structured questionnaire based on n=105 multifactorial dimensions and subdivided into n=16 overarching interrelated domains of the scenario field identification results was used (see online supplemental appendix 1a). Two core questions based on a 5-year perspective of DTR focused on factors that are very likely to be related to feeling safe and factors that are particularly relevant to feeling safe in future. These factors were rated as ‘yes’, ‘no’ or ‘not ratable’. A pretest was also conducted (n=2) that focused on the consistency of the document, the wording of the questions, the level of difficulty and the time required to complete the online survey. The quantitative data were examined for missing data (without replacement) and plausibility, that is, data sets indicating that a factor was unlikely to be related to perceived safety were excluded from the relevance analysis. The final eligible KFs were selected on the basis of the following inclusion and exclusion criteria, considering qualitative and quantitative data equally. At a minimum, one factor was required to be included per domain, the most frequent factors were included, no factors were excluded beforehand on the basis of probability and a feasible number of KFs (n=10–25)16 was required.
Additionally, an impact analysis was performed by the research team using an impact matrix with the goal of determining the extent of the reciprocal relationships of the selected KFs via a Likert scale: ‘0=no influence; 1=weak relationship; 2=medium relationship; 3=strong relationship’16; and 99=not rateable. For this purpose, the KFs were rated in two directions regarding the pairs of KFs: (1) the extent to which one factor can influence the other factor, that is, the calculated active sum (AS) and (2) the extent to which one factor can be influenced by the other factor, that is, the calculated passive sum (PS).16 The AS and PS were calculated by summing all the entered values for each KF of each row of the impact matrix such that each KF resulted in an AS and PS.16 The KFs were determined by their type, namely, active—impulsive (high AS and low PS); balanced (neither high nor low AS or PS); or passive—reactive (high PS and low AS); they were then calculated by the quotients of the AS and PS of each KF. Based on these results, the KFs were examined for further exclusion.
The aim of the final three phases was to develop a concrete best-case scenario (BS), worst-case scenario (WS) and trend scenario (TS) while also considering crisis events. The BS, defined as the desirable scenario, reflects the participants’ needs; the WS, defined as the least desirable scenario, reflects unpleasant traits; and the TS defined as the scenario that represents a continuation of the status quo.
The scenario data collection involved two guide-based scenario rounds that were held in situ on 14 March 2023 (phases 3 and 4) and 20 June 2023 (phase 5). A qualitative creative-narrative workshop approach16 was applied and divided into two steps on the first day of the workshop, according to the scenario guide. First, the identified KFs were approved, analysed and then projected in positive, negative and trend directions based on a 5-year perspective. These projections were required to be probable, realistic, distinct, precise and complete.16 The projections developed were documented in a structured form during the moderated subgroup sessions and considered a mix of experiences. A final analysis was subsequently conducted. Second, probable and realistic raw scenarios and drafts of scenario cores were developed on the basis of these projections. Therefore, before the second group session was held on the same day, the KFs and their projections were bundled by the project team to develop raw scenarios. These bundled KFs and the corresponding projections were discussed, validated and concretised by the participants. Additionally, the first drafts of the defined scenario cores were developed, and the discussions were audiotaped, transcribed and analysed using deductive content analysis featuring inductive subcategory development.16 Based on participant feedback, the research team subsequently accepted the projections and confirmed their discriminability.
Additionally, a consistency analysis16 22 of all scenarios was performed via Parmenides EIDOS software23 (Parmenides AG, Poecking, Germany) for scenario calculation. We calculated the most consistent bundles of the raw scenarios with the KFs and corresponding projections that exhibited the best fit. The number of KF bundles is a result of the product of the number of projection alternatives per KF.16 In our case, 3 projections, given 19 KFs were calculated by 319 and resulted in 1 162 261 467 possible KF bundles. The Likert scale developed by Hinkeldein (2009)24 was used by two researchers to rate the consistency in the software consistency matrix as follows: −3=mutually exclusive; −2=contradictory; −1=bad fit; 0=unrelated; 1=good fit; 2=supportive of each other and 3=belong together. The consistency of a scenario is expressed in terms of the average consistency,16 which is calculated by summing one consistency (C) value (ranging from −3 to 3 on the Likert scale) from each pair of KF projections (per KF column of the consistency matrix) and dividing that by the sum of the investigated values per column, in our case, n=171 per calculated raw scenario. For example, in column 1, the C values of K1a and K2a were summed with the C values in column 2 of K1a and K3a and then summed with the C values of K2a and K3a, and so on. The final sum was then divided by the sum of the n=171 investigated values. The most consistent BS and WS were chosen for final inclusion. The TS, which exhibited lower consistency on its own terms, was an exeption. Minorly consistent bundles were excluded following the suggestions of Tourki et al.24
A contribution analysis was performed to determine whether the scenario components were consistent with those pertaining to one selected scenario (internal consistency).16 In this context, the C values, which are associated with each row of the KF projections pertaining to one selected scenario, were each multiplied by a proportion of 2/19 KFs and then summed. For example, 2/19×(C value of K1a and K2a)+2/19×(C value of K1a and K3a)+2/19×(C value of K1a and K4a), etc. These sums were subsequently normed by calculating the quotient in relation to the maximum or minimum internal C value associated with the projections of a selected scenario.
In the final scenario round (phase 5), the final scenarios and their cores were approved plenarily. The corresponding consequences at the individual, organisational and system levels of each scenario were subsequently determined while also considering participant expertise. Recommendations for action at the educational, research and political levels were developed according to the scenarios and on the basis of three randomly formed subgroups. Realistic crises that impacted the scenarios were reconciled, and the consequences of each scenario as well as the recommendations for action were developed by planners who considered education, research and politics.
Results
In summary, 19 KFs and the corresponding 57 projections formed the basis of the final BS, WS and TS (scenario cores) when considering the selected crisis event, that is, war, and in the context of perceived safety and DTR in healthcare. This process was based on a 5-year perspective. The TS represented a realistic future direction, whereas the BS represented additional future needs and objectives. Finally, the WS focused more on the future consequences of a restricted consideration of perceived safety in the context of DTR. In total, 92 participants participated in multiple phases, and 130 participants participated in all phases (see table 1).
Table 1
|
Participant characteristics regarding the scenario phases
The scenario field
The scenario field was characterised by 16 overarching interrelated domains (see figure 2) based on 105 multifactorial dimensions of the influencing factors (see online supplemental appendix 1a) and 216 main categories that focused on several context levels, that is, the individual and the DT itself, the community/organisation and the system. Apart from the domain ‘equity’, which was only related to emotional safety, all domains included dimensions of both emotional and psychological safety. We observed that the scenario field was enriched by data collected via our various methods. Compared with our scoping review,6 almost two-thirds of the new main categories came from the workshops and use cases. The data indicated that perceived safety was related to positive and negative outcomes, among other outcomes, regarding DT implementation. A lower level of perceived safety could lead to avoidance of DT use, for example, when DTs do not meet the participant’s needs, when competences are limited, when negative feelings result in the continued use of analogue alternatives or when DTs are used differently, partially or only under certain conditions. In contrast, a higher level of perceived safety was associated with, for example, DT acceptance, adoption, adherence and positive feelings of well-being, trust and self-confidence. However, increased DT use can also promote refusal behaviours related to the unknown (see online supplemental appendix 1b).
Domains of emotional and psychological safety resulted from our data sources on system level; community/organisational level; individual level and digital technology level. Black sphere=covered by most frequently by 12 dimensions (3 data sources); grey sphere=covered by 3–11 dimensions (3 data sources); light grey sphere=covered by 6 and 3 dimensions (2 data sources); *Only emotional safety.
The domains contained several dimensions (range: n=3–12). Considering these results, the domain ‘knowledge and competence’ may be considered essential as it was emphasised most often (n=12) and contained most of the main categories. Here, perceived safety was related to, for example, digital (health) literacy and its promotion, empowerment, technical understanding and capabilities pertaining to DT use, and readiness. The domain ‘control’ may be also considered essential as it was emphasised very often (n=11) and from several perspectives, that is, being controlled by DT, having control due to DT and gaining control over DT. ‘Experiences and attitudes’ (n=10) were related to, for example, previous experiences, openness and the individual’s own habits/rituals. The other domains, for example, ‘perceived benefits in healthcare’ (n=8) and ‘perceived disadvantages in healthcare’ (n=7), were associated with positive and negative feelings of safety, respectively. In contrast, positively addressing ‘enhanced treatment options by DT’ could also result in negative impacts in cases featuring implementation concerns due to a lack of concepts regarding digital care. Although other domains were also associated with 7 or a lower number of dimensions, they were not of minor importance because they appeared in all data sources or in two data sources as in the case of ‘communication’ (n=6) and ‘professionality’ (n=3).
Identified KFs and projections
21 KFs were identified by the participants as very likely (probability, range=52.94%–91.42%) and particularly relevant (relevance, range=68.75%–96.97%) in the contexts of perceived safety and DT. The lowest probability was associated with the factor ‘research gaps regarding DT (…)’, whereas the highest probability was associated with two factors, that is, ‘understandable language/communication between humans and DT’ and ‘reliability of the DT (…)’. This factor was also rated as the highest in terms of relevance. The lowest value was associated with ‘professionality due to DT (…)’. The number of KFs could not be reduced by impact analysis (see online supplemental appendix 2a,b).
Final scenarios and scenario cores
On the basis of participant feedback, a qualitative synthesis of four factors resulted in a final set of 19 KFs, each of which was associated with the three (positive, negative and trend) projections. Specifically, (n=57) represents the (raw) BS, WS and TS (see table 2).
Table 2
|
19 final key factors and their projections according to the development of best, worst and trend scenarios
Although we observed a mix of BS, WS and TS projections, considering the consistency of the raw scenarios, the following exhibited a good fit: BS (C=1.80), WS (C=1.60) and TS (1.00). During the scenario process, the TS projections that appeared in the WS or BS could result in corresponding developments. The WS, which included trend projections, showed slightly better consistency than did the alternatives (1.63 vs 1.60). Thus, the more consistent scenario was chosen. The TS projections included were as follows: ‘Human resources to support the implementation of DT remain scarce (with a negative effect on perceived safety)’ and ‘easy, efficient handling of DT is limited by poor underlying analogue processes and structures that are copied into DT without reflection (…)’. The qualitative data also indicated that the latter projection was equal to the WS/TS projection. The included trend projections in the BS were not chosen due to their lower consistency. Finally, the contribution values provided an overview of the core elements of each scenario (see online supplemental appendix 2b).
These scenarios and the final scenarios were approved by the participants in light of the qualitatively developed scenario cores. For all scenarios, especially those pertaining to the BS, the projection of ‘demand-oriented, flexible availability and usability of DT’ (value of 1.0) was critical. With respect to the WS, three further projections (each 0.9) offered the greatest contributions: ‘knowledge concerning the opportunities and limitations of DT (…)’; ‘consideration of (HCRs’) health status and resources when using DT (…)’; and ‘professionalism due to DT (…)’. With respect to the TS, ‘self-confidence in addressing DT (…)’ was associated with the highest contribution value (0.9). Although the quantitative data focused on slightly different aspects than did the qualitative data, the participants completed the following three scenario cores, captured the KFs and, to some degree, the consequences: (BS) ‘the user in focus (…) reliability, user-friendliness, equity of access, suitability of DT for everyday use’; (WS) ‘lasting damage to the individual, the organisation and society’ and (TS) ‘act of establishing a balance among various demands and realities pertaining to digitalisation in healthcare - self-confidence/trust (as key) - so that the gap between people who are digitally ‘left behind’ and those who are digitally competent’ can be considered. At the core, the TS emphasises ‘tensions between requirements for change and reality’ (see figure 3 and online supplemental appendix 3a, b).
Ethical, psychosocial, health-related, legal, political and economic consequences were determined, and controversies related to digital (health) literacy or affinity were identified. For example, in the case of the BS, many KFs could be improved over 5 years from an ethical perspective, with the exception that not every target group could be reached. This situation might be characterised by ambivalence towards self-determination options in the context of digitalised healthcare (technology open vs technology denying) by questioning the self-determination options of technology-denying people. Hence, in the case of the WS and TS, social divisions may occur due to the inadequate and unequal distribution of digital healthcare services. Political radicalisation and a lack of solidarity resulting from a low-needs orientation are expected, but an increasing aversion to (digital) healthcare is also expected due to inhibitory attitudes among the population. In the case of the TS, negative consequences were observed regarding perceptions of inequalities in healthcare depending on digital competence given that being digitally competent could be related to loss perceptions resulting from inequalities due to equalisation with respect to digital offers, whereas being less digitally competent may result in a feeling of need and being left alone.
Considering the relevant legal aspects for both the BS and TS, new requirements for legislation regarding DTR are necessary to consider the context of perceived safety. In the case of the TS, the delegitimisation of the system (increasing the (legal) probability of legal action), for example, due to unequal opportunities, could be expected. With respect to politics in the case of the TS, perceptions of responsibility are emphasised, whereas with respect to education in the case of the WS, a loss of competencies of care is expected. With respect to the research, additional private research findings are expected to be forthcoming. From an economic perspective, in the case of the BS/TS, increased efficiency regarding the provision of healthcare via DTs is expected, but in the case of the TS, this expectation holds only among HCPs and digitally competent people. In contrast, digital and analogue healthcare inefficiencies are expected in the case of the TS, especially among people with a lower level of digital competence. Regarding the WS, setbacks in DT development and usage as well as negative effects regarding patient safety are expected.
In general, disadvantages in healthcare are assumed to exacerbate or preserve the inequality of opportunities and the inequities regarding healthcare access. In contrast, a newly informed culture of digitalisation that is enlightened by politics may be implemented in society. Alternatively, however, a balanced relationship between humans and DT in healthcare that considers the importance of people’s concerns, replacement and dehumanisation may emerge or increased healthcare quality may result.
The crisis war was chosen after two election rounds (pro a12/b10), followed by climate change (pro a11/b5) and cyber-attacks (pro a11/b1). All scenarios were characterised by a point in time at which the crisis would occur and be associated with controversies. If a war were to occur instantly, the insufficient perceived safety associated with DTR would have consequences concerning the resilience of the healthcare system. In other words, the scenarios lose relevance (see online supplemental appendix 3c). Such a crisis could be characterised by a general shift in priorities and the threat of digital warfare with respect to health data. However, it was also expected that a war could lead to innovations in healthcare through transfer actions.
In cases in which the different scenarios were reached over 5 years, the TS was determined to be the most vulnerable because of the high degree of uncertainty resulting from persistent and exacerbated disadvantages in healthcare, for example, a reduction in equal opportunities and access due to the absence of backup systems (analogue/digital) or the absence of advantages resulting from the other scenarios. Knowledge, however, can be drawn from other parties. For example, the BS and WS were simultaneously viewed positively allowing the digital communication could be performed (BS) and analogue’s resilient backup of healthcare structures remains available (WS). Although perceived safety decreases in the case of war, acceptance of DTs could increase.
Participants’ overarching recommendations for action
The overarching recommendations for action provided by participants were related to education, politics and research and did not differ even when a crisis was considered. In summary, promoting digital (health) literacy throughout society (global), establishing a low threshold for access to care, reducing inequalities and offering consults and teaching in digitalisation were emphasised. Furthermore, guaranteeing technical standards, enhancing the credibility and legitimacy of healthcare policy, and facilitating the efficient implementation of DTs were viewed as necessary. Finally, open and transparent scientific communication, which can also ensure high-quality and multidimensional research, was recommended.
Discussion
To the best of the authors’ knowledge, this research represents the first 5 year foresight study on perceived safety in DTR and shows that perceived safety plays a critical role in both present and future healthcare in a complex and controversial way, including in the case of a crisis. Depending on how perceived safety is considered, this factor continues to impact DTR at the individual, organisational and system levels.
Many factors were determined to be related to perceived safety and DT, and the identified needs and influencing factors impacted the healthcare system, including successful implementation, in several ways. Consistent with the psychological view of perceived safety,9 it was further observed that social aspects are relevant for perceived safety and that, according to recent publications,8 patient safety is interrelated. However, similar to models of perceived safety in inpatient care,25 it was noted that perceived safety is a complex phenomenon. When considering DTs in this context, we were able to show that multiple dimensions are essential for emotional and psychological safety and that these dimensions extend beyond previous studies6 that have been restricted by a lower number of selected DTs, such as those associated with robotics and the target groups on which they have focused, that is, older people and their relatives. Nevertheless, depending on the context of the selected DTs, stakeholders and settings, the results herein are somewhat comparable to the extant research. That said, we were able to obtain a broader and more detailed picture because we used several methodical approaches that considered a wide range of DTs and target groups. Central to our research, we determined that, on the basis of all the knowledge acquired from the scenarios and related discussion points, our findings are in line with existing reflections on DT implementation26 and interventions aimed at facilitating improvement, such as strengthening digital (health) literacy through education in the context of DTR while also taking patient safety into account.1 2 Other published scenarios regarding digital health also indicate that health and data literacy is essential.27 However, the degree to which perceived safety is considered is low, in contrast to our study that showed that knowledge and competence are essentially related to perceived safety.
Furthermore, reflections on the implementation of DT and considering crises such as COVID-19 indicate that thoughts of DT focus on the context of the needs of HCRs, the roles of HCPs, the benefits of DT and the issues pertaining to implementation.26 Most interestingly, this discussion is in line with our results regarding the transfer of analogue-related and DT-related healthcare, and it highlights the need for the fundamental reorganisation of processes and structures.
In our study, the prioritisation of factors that are crucial for the contemporary world differed from the KFs that were most relevant in the scenarios; they also differed slightly when examining the quantitative and qualitative data. These differences could be related to the processes of negotiation that characterise the use of data collected via mixed methods, including member verification, which captured both the present and the future. These methodological challenges have been identified as representing one of the core challenges of mixed-method research.15 28 However, the fact that various degrees of data abstraction16 were used, ranging from concrete data of use cases to abstract projected KFs, may explain this situation. Moreover, these challenges can be viewed as the keys to DTR. From the perspective of complexity theory, the whole system should be considered while simultaneously considering its parts. Nonetheless, more attention should be given to the system process than to its state. This process focuses on (implementation) learning29 30 and identifying interacting factors rather than making a decision for one determinant (interrelationships).30 Thus, this process enables us to obtain a differentiated picture of the various factors that are relevant in this context and the reasons underlying that relevance.30 Taking these approaches into account, the data collected for this study have several uses. The dimensions that characterise the present situation include concrete examples of events occurring in real-world healthcare settings, which may enhance the intelligibility of practice. The overarching scenario cores, for example, digital (health) literacy as expressed in terms of ‘(…) the gap between people who are digitally ‘left behind’ and those who are digitally competent’, allow us to obtain an overview of future educational, political, and research-related trends pertaining to DTR and decision-making.
Limitations
We employed a comprehensive, multiperspective, strategic 5-year foresight approach. In general, the scenario approach is perceived as acceptable considering current discussions regarding consistency and traceability. Consistency is perceived as a ‘substitute for empirical validation’31 in light of the inherent problem of anticipated objective accessibility and the corresponding complex and adaptive characteristics.31 However, our research followed the usual standards of foresight research.16 Although participatory research14 offered a high level of added value to our study, we concluded that participants’ engagement in a study over 2 years is very time-consuming and demanding for the participants given the usual demands of daily life.
Conclusion
This foresight study, which was conducted from a 5-year perspective, represents the first study to offer a broad and multifaceted picture of perceived safety in the context of DTR. The findings indicate that perceived safety in a DTR environment is influenced by several factors and has substantial impacts on DT implementation and healthcare. The scenarios developed as part of this research suggest that continuing the current state over the next 5 years may result, to some degree, in controversial outcomes. Therefore, concrete results could be used in practice to support education and healthcare quality development and research. These scenarios could further serve as additional starting points for research agenda setting and political decision-making for future healthcare. However, future research is needed to quantify the factors’ correlations with perceived safety and the corresponding impacts on DTR.
