Aligning Concerns in Telecare: Three Concepts to Guide the Design of Patient-Centred E-Health

Healthcare costs are escalating in all OECD countries due to, on the one hand, advances in medical science, technology, treatments and therapies, and on the other hand, a dramatic increase in patients with chronic diseases, such as diabetes and heart disease, associated with modern lifestyles and aging populations. At the same time, patients demand higher quality and more personalised care, putting extra pressure on governments and healthcare providers to increase access to quality care while maintaining or decreasing costs. It is often argued that one way of achieving this may be through an increased emphasis on ‘self-care’ and ‘patient empowerment’ supported by new patient-centred e-health services, such as personal health records (PHR), telehomecare, and remote monitoring (European Commission 2012; All Party Parliamentary Group on Global Health 2014).

There are great expectations for the potential of this approach to transform care, increase access, promote quality and lower costs (Holman and Lorig 2004; Street Jr et al. 2005; Koch 2006; Johnston et al. 2013). However, at present, these benefits remain largely hypothetical. Clinical efficacy and cost-effectiveness have not been documented (Dedding et al. 2011; Archer et al. 2011), and several authors warn that patient-centred e-health services may have potentially adverse effects on the patient-provider relationship, and that patients may be forced to take on tasks and responsibilities that they do not feel capable of coping with (Neuhauser and Kreps 2003; Dedding et al. 2011; Oudshoorn 2011; Chen et al. 2013).

The use of IT to improve healthcare delivery is nothing new, of course. Just think of electronic medical records (EMR), computerised physician order entry (CPOE), picture archiving and communication systems (PACS), and laboratory information systems (LIS). Patient-centred e-health services, however, are a significantly different breed of health IT applications, because they presume patients will be active users. It is therefore essential to take the perspective of the patient seriously and understand that patients have fundamentally different concerns and needs than physicians, nurses and other traditional users of e-health applications (Baron 1985; Toombs 1992; Carel 2011). These concerns and needs must be carefully addressed if patient-centred e-health services are to be successful. However, it seems that researchers and designers, in general, do not truly understand the scope and complexity of this task. Indeed, as Storni (2014) has recently pointed out, ‘the majority of self-monitoring technology and PHR (Personal Health Records) platforms are designed from the perspective of medical experts rather than patients’ (p. 407).

The objective of this article¹ is to help remedy this situation by, first, examining the difference between the way clinicians think about disease and the way patients experience their own illness, and, second, exploring the implications this has for the design of patient-centred e-health services. In doing so, we draw on insights from medical phenomenology (Baron 1985; Toombs 1992; Carel 2011) as well as on empirical insights from our own process of designing an e-health application for clinicians and patients with chronic heart disease. We argue that the dominant medical view of disease as a biological phenomenon or bodily dysfunction is at odds with patients’ subjective experience of serious chronic illness as a life crisis (Carel 2011), and that aligning the concerns of patients and clinicians is a prerequisite for the successful design and implementation of patient-centred e-health services. This does not imply that patients and clinicians need to have, or develop, identical perspectives on illness and treatment, but rather that their respective concerns and needs must be recognised, respected and reconciled. Otherwise, there is a high risk that either patients or clinicians will be reluctant or unwilling to adopt the service.

Our study offers three contributions to the literature on patient-centred e-health services. First, existing research rarely pays much attention to how patients’ perspectives, concerns and needs differ from those of clinicians and other health professionals. In response, we examine the experience of illness as an embodied phenomenon and highlight how it differs from the biomedical perspective of the physician. Second, we provide a detailed analysis of the design of an e-health service to support the treatment of heart patients with an implantable cardioverter-defibrillator (ICD). In doing so, we show that patients and clinicians have divergent perspectives and concerns that may prove difficult to align. Third, we develop conceptual guidelines for understanding and designing patient-centred e-health services that explicate three dimensions of alignment between patients’ and clinicians’ concern.

Early research and development of healthcare IT focused on administrative information systems such as hospital information systems (HIS) and laboratory information systems (LIS) intended to serve a single hospital or laboratory. Users consisted mainly of administrators, physicians and other health professionals (Bath 2008; Haux 2010). With the emergence of health informatics in the 1990s, the scope broadened to include systems spanning multiple hospitals, telehealth, global information infrastructures, and strategic information management (Shepherd 2007; Fitzpatrick and Ellingsen 2013). The users were still primarily hospital staff and administrators (Cho et al. 2009; Hara and Hew 2013).

However, over the last decade, patients have come to play an increasingly important role in their own treatment and care, and under the umbrella of e-health researchers and designers have begun to extend the network of users to include patients (and their relatives) (Ahern et al. 2008). Examples of e-health services that target patients include patient-provider portals (e-visit and e-booking), telehomecare platforms (Koch 2006) and PHRs (Lafky and Horan 2008; Piras and Zanutto 2014). Furthermore, there is currently a rapid growth in consumer-oriented health websites, mobile applications, and social software that enable patients to engage in self-management and exchange experience and advice to better cope with day-to-day personal health issues (Hartzler and Pratt 2011; Morgan and Trauth 2013). It is still too early to tell where this new technological trend will lead, but it is likely to have important impacts on the future of healthcare (Baird et al. 2011; Dedding et al. 2011; Oudshoorn 2011).

It is generally expected that patient-centred e-health services will facilitate patient participation, self-management, and empowerment (Dedding et al. 2011; Archer et al. 2011; European Commission 2012; All Party Parliamentary Group on Global Health 2014). Such services can, for instance, become a ‘supplement to existing relationships and forms of care’ (Dedding et al. 2011), improving behavioural outcomes through tailored communication and increased interactivity (Neuhauser and Kreps 2003). It is also suggested that e-health has the potential for creating ‘favourable circumstances for improvements or strengthening of patient participation’ (Dedding et al. 2011) and for ensuring ‘more transparency and empowerment’ (European Commission 2012, p. 14). For example, PHRs and various health sites may make patients better informed and empowered (Lafky and Horan 2008; Piras and Zanutto 2014) and thereby possibly better equipped for ‘self-management’ (Holman and Lorig 2004) and ‘shared decision-making’ (Neuhauser and Kreps 2003). E-health can also function as a lever for engaging health professionals in working together with patients (Tang and Lansky 2005), thus, promoting greater self-management and individual responsibility on the part of the patient (All Party Parliamentary Group on Global Health 2014; Waycott et al. 2014).

More recent studies examine how self-tracking and patient-generated data may support collaboration between patients and providers. Patel et al. (2012) describe how regular symptom tracking by patients who use a personal health information management tool in cancer care can support communication between patients and clinicians. Hartzler et al. (2016) show that patient-reported outcomes (PROs), trackers and dashboards can be a promising approach for integrating patient-generated data into prostate cancer care.

However, early experiences with patient-centred e-health services also point to many challenges and unintended consequences that should be taken into consideration when designing future applications and services. In CSCW, Bardram et al. (2005) described, more than a decade ago, the emergence of a new division of work in telecare. They studied home-based monitoring and treatment of hypertension using a self-inflatable blood pressure monitor and a PDA and found that unanticipated practices emerged. With remote monitoring, patients must learn to adhere to the frequency of the measurements and assume responsibility for using the medical equipment correctly. Dedding et al. (2011) warn that these services may ‘disturb’ patient-clinician relations and lead to more ‘sick work’ for the patients. Therefore, they advise, ‘more attention should be paid to the redistribution of tasks and responsibility to patients.’ Similarly, Piras and Zanutto (2014) describe how a PHR for paediatric diabetes management generated ‘tensions’ between patients, relatives and physicians because it was perceived as ‘an intrusion into the personal sphere’ of the family.

There can also be discrepancies between what patients want to share and what clinicians find useful. Jacobs et al. (2015) identify such misalignment when comparing what patients want to share in breast cancer treatment with what healthcare providers prefer. Miscommunication can happen when patients are given the opportunity to track health related data. Physicians may not find the data relevant, they may have different goals than patients, or they may simply lack the time for reviewing it. This leads to unmet expectations and makes patients dissatisfied and disappointed to an extent that it creates conflicts between the patient and clinician (Chung et al. 2016).

Another unintended consequence may therefore be added responsibilities and extra costs for providers (Dedding et al. 2011) – costs that are often not covered by existing reimbursement models. A related, but not as easily quantified consequence could be that patients develop new expectations; expectations that clinicians cannot meet (Nielsen 2015). Clinicians may also feel that their professional practice is challenged or even undermined by the new technologies. For example, Chen et al. (2013) highlight the challenges of e-visits, where patients and physicians communicate online, and question whether physicians can build trust and engage in diagnosis without being able to interview, observe, and examine the patient in person. Similarly, Oudshoorn (2011) describes how health professionals working with telecare need to learn new skills such as how to create a kind of intimacy at a distance. She also describes how telecare can have the unintended consequence of shifting responsibility onto patients so that self-care is forced upon them whether they like it or not. In a similar vein, Langstrup (2013, p. 7) describes how ‘family members become, willingly or unwillingly, parts of the chronic care infrastructure.’

In sum, delivering on the promise of patient-centred e-health is easier said than done. It requires an in-depth understanding of not only the technology but also the needs and concerns of all the relevant actors, in particular clinicians and patients. Paraphrasing Fitzpatrick and Ellingsen (2013), there is an urgent task for CSCW, and related fields engaged in patient-centred telecare research, of conceptualising and understanding how absent-yet-present patients and clinicians can collaborate.

We suggest that an understanding of the difference between the perspectives and goals of patients and clinicians, as well as the variation in needs and concerns within these two groups of actors, is key to successful design of e-health services. As S. Kay Toombs (1992) has pointed out in her seminal book on the meaning of illness, there is a ‘decisive gap’ between the way illness is experienced by the patient and the way in which physicians conceptualise disease in biomedical terms such as symptoms, diagnoses, pathology, treatment, and prognosis. ‘Consequently, rather than representing a shared ‘reality’ between them, illness represents in effect two quite distinct ‘realities” (ibid., p. 10). Generally speaking, patients may be said to experience illness as a unique, personal event that transforms their bodily awareness and disrupts their everyday lives, including their roles and relationships with others (Charon 1992; Pickard and Rogers 2012). In contrast, physicians may be said to understand disease as an entity in itself, a biological phenomenon that can be categorised and diagnosed as a known illness, for instance as a particular case of ‘diabetes’ or ‘ischemic heart disease,’ and treated according to scientifically tested procedures (Baron 1985; Toombs 1992; Mongtomery 2005).

The main findings presented in this paper derive from a large interdisciplinary research project, ‘Co-Constructing IT and Healthcare’ (2008–2012), carried out in a partnership between multiple hospitals and universities in the greater Copenhagen area in Denmark. The aim of the project was to understand the treatment and care for chronic heart patients with an implantable cardioverter defibrillator (ICD) and develop exploratory prototypes of a patient-centred e-health application to improve collaboration between patients and clinicians. An ICD is best understood as an advanced pacemaker, designed and configured individually to ensure a steady heart rhythm and, if needed, to deliver an electric shock to stop a fast and dangerous heart rythm.

In a follow-up project, the SCAUT project (2014–2018), we continued research and development under more commercial conditions (see Andersen et al. 2017). Two additional industry partners joined the consortium; Medtronic, the world’s largest medical device company, and Rehfeld Medical, a Danish healthcare IT start-up. This paper is mainly based on the CITH project, however, we use the empirical data from the follow-up project to illustrate how the conceptual guidelines that we propose, helped us increase alignment of concerns among patients and clinicians.

The care for ICD patients involves multiple actors. We investigated the two main settings where the care takes place: the hospital’s heart centre and the patients’ homes. Nearly all patients are monitored remotely. Data is sent from patients’ home monitor to the heart centre for follow-up every three months or whenever an ‘event’ that clinicians need to address occurs. An event could be that the device registers atrial fibrillation [irregular heart beat] and notifies the heart centre. For each remote follow-up, a lab technician interprets the data and, if needed, the technician asks for a second opinion from a cardiologist or, in some cases, hands the case over to the cardiologist. Often they consult the patient’s electronic medical record and a special device record. When needed, they make a phone call to the patient to get more specific information for evaluating the condition. If there are any problems, the patient is either referred to the local hospital for adjustment of his or her medication or the patient is called in to have the ICD adjusted at the heart centre. In most cases, however, no clinical action is needed.

We present our findings by analysing four empirical vignettes from our research. The idea is to illustrate the different challenges involved in aligning the concerns of patients and clinicians. There is a progression in presentation of the four vignettes, starting with a relatively simple example of a negative patient-clinician collaboration and ending with a more complicated, but also more positive example. For each vignette, we discuss how it illustrates a key challenge in the process of aligning concerns.

Based on these findings, we present a set of conceptual guidelines for designing patient-centred e-health services. The concepts are grounded in insights from medical phenomenology (Baron 1985; Toombs 1992; Carel 2011), in analytical reflections from the participatory design and in our empirical findings regarding the use of myRecord. These conceptual guidelines accentuate the importance of ensuring that e-health services align or reconcile the divergent needs and perspectives of patients and clinicians. Furthermore, they recognise the decisive gap in how patients experience illness and how clinicians conceptualise disease and suggest the importance of working towards designing e-health services that specifically seek to reconcile these, often divergent, concerns. Below, we present three properties that are essential for the identification and alignment of concerns in this context. The properties are meaningful, actionable, and feasible.

Using the three concepts – meaningful, actionable and feasible – we recommend that designers of e-health services consider a number of important questions at specific times in the process (see Figure 6). The three-step e-health design process, described below, provides inspiration for how to proceed and which questions to ask at what time.

Designing for alignment of concerns is a challenging undertaking that ideally requires prolonged interaction with both patients and clinicians. It requires more than a traditional requirements specification process, because it calls for an extended commitment to understanding the different types of users’ perspectives (Fitzpatrick and Ellingsen 2013) through collaborative design activities. Working to align concerns is particularly demanding in terms of ensuring the articulation and negotiation of both patients’ and clinicians’ concerns. Seeking compromises and establishing a consensus can be highly contentious, and not something that is easily achieved in one go. As we have shown above, patients and clinicians have inherently divergent goals and focus on different aspects of illness and disease. Aligning concerns is an active, collaborative process that requires continuous translation and adjustment between what is regarded as meaningful and feasible by both patients and clinicians. We, therefore, recommend applying a participatory design approach where prospective patients, informal caregivers and clinicians are actively involved throughout the process.

Participatory design methods (Bødker et al. 2009; Simonsen and Robertson 2013) can be tailored to identify and reconcile divergent needs and concerns by co-designing. Within health informatics, co-design is increasingly suggested as a way forward (Clemensen et al. 2007; Pilemalm and Timpka 2008; Hardisty et al. 2011). This involves research and design techniques such as ethnographic field studies, design workshops, scenarios, mock-ups and prototyping. We suggest organising the participatory design process in iterative cycles involving three different modes of inquiry to support alignment of patients’ and clinicians’ concerns. The three modes are exploration, experimentation, and evaluation. Figure 6 illustrates how designers can design patient-centred e-health that is meaningful, actionable, and feasible in a three step, iterative process.

While the conceptual guidelines and methodical approach presented in this paper are aimed at designing patient-centric digital services to support treatment of patients with a specific chronic condition, there are strong indications that it will become increasingly important to focus on the alignment of concerns between patients and clinicians in a wider context of healthcare. In this section we therefore discuss the extended implications of our conceptual guidelines for research.

As mentioned in the introduction, e-health research and development has for a long time focused on supporting different clinicians and other health professionals in their work. Although clinicians cannot be considered a homogenous group with entirely shared concerns, they do, after all, to a larger degree share perspectives – and concerns – with each other than with their patients. However, much of e-health research and development appears to view e-health design as a matter of supporting collaboration between people with relatively similar concerns, even when designing patient-centred e-health services, while often failing to recognise the complexity of the challenges at hand. This approach is perhaps further fueled by the strong associations between e-health and idealistic discourses about shared care, patient empowerment and so-called ‘participatory medicine’ (Weitzel et al. 2009; Gallant et al. 2011). We suggest that moving beyond this one-sided approach requires a more ‘holistic’ theoretical approach – one that is less prescriptive and more analytical in its understanding of patient-clinician relations. Sparked by empirical insights and our experience with the difficulties of designing e-health systems that align the often divergent concerns of patient and clinicians, we have found medical phenomenology inspiring (Baron 1985; Toombs 1992; Carel 2011) for its attention to fundamental differences between patients and clinicians. Furthermore, we have translated this phenomenological perspective and made it relevant for designing patient-centred e-health by devising conceptual guidelines.

Several societal trends underline the importance of focusing on the alignment of concerns among patients and clinicians. First of all, multimorbidity has become the most common chronic condition in itself (Tinetti et al. 2012). This means that more patients are at the same time treated for multiple, chronic diseases and often by different, specialised health professionals (Bussche et al. 2011; Agborsangaya et al. 2012; Moth et al. 2012; Barnett et al. 2012). In practice, the challenge is therefore to ensure the meaningfulness, actionability, and feasibility of a patient’s concerns, not just with one type of clinician, but with a large, heterogeneous ensemble of organisationally distributed health professionals, and vice versa. This emphasises the importance, yet also the complexity, of supporting patients and clinicians in aligning their concerns. To fully anticipate the consequences of multimorbidity, some questions regarding alignment of concerns remain to be explored: In cases of multimorbidity, the patient will often interact with several communities of highly specialised health professionals, potentially with different (sets of) concerns. What is meaningful, actionable, and feasible may therefore differ from one context to the next, making alignment of concerns even more challenging. Also, as traditional health management programs typically focus on single conditions and do not stipulate interactions between combinations of chronic disease, it is common that care interventions intersect and produce adverse and some times unforeseen effects (Boyd et al. 2005). Therefore, it is of crucial importance to investigate the need and potential for aligning concerns among different groups of clinicians as well as to explore how this can be accounted for in our conceptual guidelines.

Furthermore, the appropriateness of traditional biomedical treatment goals is increasingly being discussed. For instance, in the emerging discourse on goal-oriented care (Mold et al. 1991), the traditional biomedical model of problem-oriented care is criticised for a lack of concern for the patient’s desired outcomes and for a lack of ability to support coordination between different treatments. In the problem-oriented approach, therapy is mandated using condition-specific indicators and guidelines designed to achieve a certain target (Singer et al. 2002). In contrast, a goal-oriented approach to care is mandated by functional goals defined by the patient, for instance regarding symptoms, physical functional status, and mobility (Mold et al. 1991; Reuben and Tinetti 2012). A goal-oriented approach will require more attention to patients’ perspectives, and it will become even more important to render their concerns meaningful, actionable and feasible for the clinicians.

Taking these observations a step further, it is important to highlight that many other actors, and thereby concerns, play a role in healthcare practices and innovation and thus should be acknowledged as relevant participants. This is increasingly recognised in participatory design of health IT (Kanstrup et al. 2017). These actors include hospital managers, administrators, policy-makers, relatives, and commercial companies (in our case, ICT manufacturers, see Andersen et al. 2017). This further complicates the task of aligning concerns. While it is beyond the scope of this article to further flesh out and exemplify how this complexity of concerns can be accounted for in design of digital services, we recommend the methodological approach outlined above as a guideline for further investigations. Moreover, and as a final note, the growing interest in frameworks like value-based healthcare and patient-reported outcomes may prove valuable for the alignment of concerns among multiple actors, with patient experiences being the privileged starting point.

In this article, we have sought to provide reflexive analytical concepts for informing the design of patient-centred e-health services. We started out by recognising the shift from hospital information systems to e-health and the new challenges that arise when extending the healthcare network to include patients as active, participating users. We then presented important insights from medical phenomenology regarding the general differences between patients’ and physicians’ perspectives, while also recognising the variation of concerns within both groups, suggesting that it is crucial to take these insights into account when designing and evaluating e-health services. With this overall conceptualisation as our analytical point of departure, we unfolded four cases from the design and user tests of a working prototype, myRecord, demonstrating the implications of patients’ and clinicians’ different concerns in the specific contexts of ICD-care and the need for appropriate alignment of these. That is, when designing e-health services to support collaboration between patients and clinicians, it is important to identify the concerns that are meaningful to both patients and clinicians and, moreover, which concerns are or can be made actionable and feasible.

In our analysis, we identified the consequences of introducing e-health into existing care networks and how this complicates care. However, we have also demonstrated how e-health provides opportunities for new forms of care where active patient participation improves clinical decision-making and the quality of healthcare. Introducing a personal health record for ICD patients and the involved professionals turned out to reproduce, and even underpin, existing challenges of empowering patients to engage more actively in chronic care. The version of myRecord, featured in the first three cases, made the decisive gap between patients and clinicians more tension-filled and dissatisfactory for at least one party, often the patient. In these cases, the e-health service did not succeed in tapping into or supporting the alignment of concerns between patients and clinicians. More specifically, the prototypes of myRecord failed due to a persistent mismatch between the patients’ and clinicians’ respective perceptions of what constitutes a meaningful concern, and subsequently an appropriate action. In the final case, however, we showed how the service succeeded in aligning concerns by facilitating a translation of the patients’ concerns into clinically meaningful issues and making them clinically actionable and organisationally feasible.

The somewhat contradictory outcomes of the design and use of myRecord illustrated by the different cases only underpin the necessity of developing theoretically grounded and empirically informed concepts for e-health, as we believe the concepts put forth in this article – alignment of concerns – to be. Echoing research that suggests participatory design and co-design as a way ahead (Clemensen et al. 2007; Pilemalm and Timpka 2008; Hardisty et al. 2011), we recommend operationalising the concepts by engaging in participatory design activities that employ a strategy of iterating between exploring, experimenting, and evaluating to enable the necessary collaborative process of identifying and negotiating concerns. The fact that the heterogeneity of patients’ and clinicians’ perspectives in practice may be even more complex than indicated by classic medical phenomenology only underlines the importance of approaching the alignment of concerns as a participatory and iterative process. This is also an encouragement for CSCW researchers to experiment with collaboration, guided by well-founded analytical resources and an awareness of the intricate dynamics between such different actors as patients and clinicians, yet without losing sight of the ambition to design.