INTRODUCTION: DIABETES AND TECHNOLOGIES FOR SELF-MANAGEMENT ENHANCEMENT

This paper contributes to the debate on the patient–doctor relationship, by focussing on a specific chronic disease, type 1 diabetes (T1D), with the aim to highlight the fruitfulness of a relational approach. T1D is a metabolic disease characterised by absolute insulin deficiency and is treated with lifelong insulin therapy (ADA, 2019). In order to prevent possible complications, an intensive self-management is required (Daneman, 2006).

Even though self-management still lacks a universal definition (Ellis et al., 2017), in this paper we find Audulv's (2013) contribution to be particularly useful, because he described it as the set of strategies undertaken by people with chronic conditions in order to control disease, promote health and live well with illness. This makes it ‘a complex developmental process that takes place within the context of a disease trajectory, a health-care culture and a uniquely meaningful life’ (Thorne et al., 2003: 1349). Accordingly, self-management is a ‘total life experience’ (Price, 1993), which requires the individual to adapt to fickle and unique life situations and contexts (Audulv, 2013). Indeed, diabetes is a tricky disease, in which ‘something unexpected may always happen’ (Mol, 2008: 91). This demands an ‘attentiveness’ (Mol, 2008) about daily practices including glycaemic monitoring, carb and insulin intake computation, and lifestyle behaviours, which relies on personal logic, experiences (Minet et al., 2011) and lay expertise (Storni, 2015). However, self-management is not a stand-alone procedure in which the single patient acts autonomously. Rather it recalls the idea of ‘logic of care’, by which patients are active in dynamic shared doctoring: this entails experimenting, attuning to one another and tinkering together, as well as exchanging experiences, knowledge, suggestions and words of comfort (Mol, 2008).

Diabetes self-management has been enhanced by increasing technological innovation. Nowadays, the two main therapeutic devices (TDs) are those for continuous subcutaneous insulin infusion and those for self-monitoring of blood glucose (ADA, 2019). Insulin pumps, for continuous subcutaneous insulin infusion, are small, portable, insulin injection devices. The latest pumps have advanced dose calculation software that can simplify daily self-care, for example by recommending a bolus dose based on planned carbohydrate intake, blood glucose level, carbohydrate ratio, correction/sensitivity factor and also by considering the amount of active insulin in the body resulting from earlier boluses (Horwitz & Klonoff, 2017). For self-monitoring of blood glucose, patients may use either real-time continuous glucose monitoring or flash (or, intermittently scanned continuous glucose monitoring) sensors. They contribute to increase the number of daily measurements, which are important to set up the insulin intake. Moreover, by displaying the trend arrows, they can potentially support interventions with more accuracy (Galindo & Aleppo, 2020). Patients may choose to use just one or both of them. The combination of interoperable sensors and pumps may form sensor-augmented insulin pumps (SAP), and a hybrid closed-loop insulin delivery system, which is a step towards artificial pancreases. This further improves self-management, even though patients must check their devices, as they do not work completely autonomously (Galindo & Aleppo, 2020; Taleb et al., 2021).

These devices affect self-management in many ways, albeit not deterministically. For instance, these TDs allow huge data acquisition that potentially can be handled first and foremost by patients by allowing the practice of ‘clinical self-tracking’. This is considered to be ‘part of a relationship that defines reciprocal duties, expectations and roles’ (Piras & Miele, 2017: 51). Therefore, all of this may, in turn, affect the doctor–patient relationship, as highlighted by a body of literature.

THE PATIENT–DOCTOR RELATIONSHIP IN THE DIGITAL SOCIETY

The sociological debate on the patient–doctor relationship is, by and large, summarised by the juxtaposition between two opposite poles. On the one hand, we have the doctor-centred model, which is well represented by the conceptualisation of an asymmetrical patient–doctor relationship (Parsons, 1951). In a nutshell, Parsons' paternalistic sick role provided an authoritative physician and a passive and compliant patient (Crossley, 1998). This theory was criticised, for example, because it failed to explain diseases with longer temporal horizons, such as chronic conditions (Crossley, 1998). Yet, some scholars stressed the enduring power of the medical profession and its authority as a ‘remarkable persistence of asymmetry’ (Murdoch et al., 2020; Pilnick & Dingwall, 2011: 1374), therefore highlighting the contemporaneity of Parsons' conceptualisation (Timmermans & Tietbohl, 2018).

On the other hand, we have the patient-centred model. It took place by the end of the 20th century, as a consequence of some contextual changes (Heritage & Maynard, 2006; Nettleton, 2013), and in relation to the emergence of a consumerist approach in health care (Fraser et al., 2020). This new framework propelled a move from professional dominance to countervailing powers (Heritage & Maynard, 2006), or, in other terms, from patient compliance to a less authoritarian form of patient concordance (Nettleton, 2013). In such a context, the concept of patient-centredness arose in order to promote a more ‘egalitarian’ idea of the patient–doctor relationship (Fraser et al., 2020; Mead & Bower, 2000; Timmermans & Tietbohl, 2018). The basic idea was that patients are not passive recipients, but active subjects involved in health-care processes (Nettleton, 2013). However, some more drastic proposals also arose: van Olmen et al. (2011), for example, proposed the concept of ‘full self-management’ as an approach that maximises patients' autonomy and independence.

This debate has been further fostered by the advancement of health-care technologies. This field is dominated by the dichotomy between techno-critics and tecno-utopians (Lupton, 2016). Among the latter, some scholars (Swan, 2013; Topol, 2015) suggested that the revolutionary potential of technologies is capable of driving a paradigm shift from Parsonsian asymmetry to new assets, variously described as empowerment, active patient, expert patient and so on (Greener, 2008). To summarise, technologies magnify the ‘most vaunted of human commodities: choice, understanding, consciousness and freedom’ (Swan, 2013: 96), which conceptually push self-management to its highest point. Accordingly, the massive data acquisition brought about by sensors and related apps should lead to a new data-driven strategy and to the achievement of personalised and proactive self-management. Among techno-critics, Lupton (2014) highlighted the extension of the panoptical gaze of health professionals by means of technologies: since they create more ways of entering into people's private and intimate worlds, the role and power of physicians results to be strengthened. Another issue is the ambivalence often showed by patients as to whether they want to participate in their clinical decision-making in addition to wanting to trust and rely on their doctors, and to be guided by them in medical decision-making (MacArtney et al., 2020; Nettleton, 2013). In these contexts, scholars tend to stress the uncertain affective and digital atmospheres triggered by devices that are worn constantly, such as pumps or sensors, which might stimulate both sentiments of happiness or apprehension (Lupton, 2017).

The juxtaposition, outlined above, between doctor- and patient-centredness seems to suggest ‘substantialist’ (Emirbayer, 1997), static and reified (Kaspersen & Gabriel, 2008) conceptualisations of power, which might be conceived as ‘something that one owns’ (Elias, 2008: 136): either the power is held by physicians, or by patients. Actually, some authors stressed more nuanced patient–doctor frameworks that go beyond such static representations. For example, the dynamic tensions underlying the practice of goal-setting have been highlighted (Murdoch et al., 2020); or the ongoing conflict over power and control (Greenfield et al., 2012); or the ambivalence between logics of ‘being cared for’, or even ‘strongly guided’, and empowerment (Greener, 2008; MacArtney et al., 2020; Nettleton, 2013). In line with this more nuanced and dynamic analysis, we think that a relational perspective, which focuses on processes and flows of action (Powell & Dépelteau, 2013), as well as on transaction (Emirbayer, 1997), could be helpful in describing actual day-to-day patient–doctor relationships. As a matter of fact, it allows the idea to be overcome that both parties of this relationship are fixed, unchanging and independent entities. In contrast, it helps to stress dynamic, unfolding and ongoing processes (Emirbayer, 1997). The very idea of ‘centredness’ may be misleading, as actually there is no centre but rather an equilibrium or a balance. A vast dissertation about the relational ‘turn’ (Powell & Dépelteau, 2013) is beyond the scope of this work. Following previous studies (e.g. Brown et al., 2015), Elias' contribution seems to be particularly useful because he considers social life as a process of interdependent and interweaving social relationships (Kaspersen & Gabriel, 2008), in which the concept of figuration becomes paramount. For Elias (1978), human beings are homines aperti with a greater or lesser degree of relative, but not absolute, autonomy and dependency with other homines aperti (van Krieken, 2001). As highlighted by Dunning and Hughes (2013), humans are bound to others by fluid ties of interdependencies. Such interdependencies form dynamic figurations among people, which are described as ‘shifting networks of people with fluctuating, asymmetrical power balances’ (Dunning & Hughes, 2013: 52, Elias, 1978). In this sense, far from being ‘an isolated object in a state of rest’, power ‘is an attribute of relationships’ (Elias, 1978: 116); a ‘structural characteristic of human relationships’ (Elias, 1978: 74) since it lies in the interdependencies of people (Dunning & Hughes, 2013).

For these reasons, Elias' contribution helps to investigate in greater depth the doctor–patient relationship by allowing patients' possible behaviours and choices to be highlighted as the outcome of interdependent figurations and power balances. Consistently with this aim and theoretical background, we adopt Hirschman's exit-voice model (1970) with a relational lens: the purpose is to understand how patients manage their satisfaction and dissatisfaction about the received care within interdependent figurations that are characterised by technology (TDs) and intensive self-management requirements.

In Hirschman's theory, voice is defined as any conceivable strategy that changes an objectionable state of the art, with the aim of improving the situation (Hirschman, 1970). An effective voice requires at least three aspects (Pickard et al., 2006): users must be confident of being able to influence decision makers; there must be channels for communicating users' demands, such as in our case, the possibility for patients to express discontent directly to their diabetologists, or to access the Diabetes Centres through the public relations office; ‘voicing’ users cannot face penalties which might limit this option, such as being given worse treatment or being dismissed by doctors or Diabetes Centres. On the contrary, exit means that users abandon an organisation, ceasing to turn to it to receive their services without giving providers any feedback. For it to be a feasible option, other providers must exist and be reachable by exiting users. In the case hereby presented, the Italian National Health Service allows disappointed patients, aiming to change their current providers, to choose among all the existing Diabetes Centres, even those in other Regions. Of course, for the purposes of this paper, we only intend ‘active’ exit, namely when patients intentionally choose another professional, excluding the ‘passive’ exit, that is patients who are assigned to another diabetologist because they have moved (Vengberg et al., 2019). The third concept in Hirschman's archetype is that of loyalty. It is defined as a particular attachment to an organisation which supports their voice and lowers the tendency to exit. This is a key concept in the interplay between voice and exit, because without it, exit would become a costless option (Hirschman, 1970).

METHODS

MAIN FINDINGS

The main characteristics of participants are outlined in Tables 1 and 2.

Increasing patient competence both on clinical and technological grounds within a framework of self-management and shared doctoring creates high expectations regarding the figurations with their diabetologists. They are required to have clinical competence, technological skills and also the capacity to share the doctoring, which entails relational traits such as exchanging knowledge, suggestions and experiences, as well as giving words of comfort (Mol, 2008) and empathy (Brown et al., 2015; MacArtney et al., 2020). These requirements reject asymmetrical and paternalistic relationships, as well as seeing patients as independent figures sublimated by concepts of empowerment and patient-centredness. Rather, they help shape different interdependent figurations, which in turn may facilitate specific behaviours.

DISCUSSION

By investigating the patient–doctor relationship in the milieu of diabetes care, this paper highlighted its relational and interdependent characteristics, which form specific dynamic figurations. Patients' management of satisfaction and/or dissatisfaction takes place within such figurations and allows them to enact four different behaviours. We also showed that these behaviours are not deterministic, but they are the outcomes of the kind and quality of particular patient–doctor figurations.

Therefore, our findings do not support patient-centred approaches that aim to highlight characteristics such as autonomy, independence and ‘full self-management’ (Swan, 2013; Topol, 2015; van Olmen et al., 2011). For instance, Armstrong (2014) postulated the emergence of a ‘new patient identity’ in connection with the increasing relevance of patient autonomy and her/his centredness: namely, a patient who exercises agency in terms of preparedness, vigilance and capability of making autonomous choices. However, in this paper, respondents who opted for loyalty and resignation indicated a much more tangled relationship, based on interdependency. Besides, we show that even when choosing the exit strategy, which is the most disruptive scenario, patients do not seek autonomy and independence, but on the contrary a new diabetologist to trust and with whom to build a new figuration. These findings are more in line with Fraser et al. (2020), who described patients' desire to have ‘an honest and productive relationship with a professional health practitioner’. So, we could say that ‘patient agency’ is exercised to a greater extent within a patient–doctor figuration rather than in autonomous and independent assets.

On the other hand, our findings do not support doctor-centred or traditional asymmetrical frameworks either (Murdoch et al., 2020; Parsons, 1951; Pilnick & Dingwall, 2011). In particular, Pilnick and Dingwall (2011: 1381) postulated that the persistence of asymmetry is embedded within a wider functionality of the institution of medicine in society. We also reported the existence of asymmetrical figurations, as described in the excerpts related to resignation, for example. However, both ‘voice’ and above all ‘exit’ allow us to state that asymmetry is not static, but a dynamic and fluid characteristic of figuration which may entail different scenarios.

Both these static and substantialist ideas of relationship and power, centred and focussed only on one actor, are incapable of explaining actual nuances in the patient–doctor relationship. On the contrary, our findings support the idea that interdependent and figurational perspectives seem to be more consistent and useful in shedding further light on the ambivalence existing in doctor–patient relationships. Some of these contradictions have been pinpointed in the sociological literature. For instance, Greener (2008) stressed the co-existence of the logics of caring and empowering, whereas Phillips and Scheffmann-Petersen (2020) stressed the co-existence of both models of doctor- and patient-centredness and considered the relationship as the unstable product of dialogic meaning-making processes that are inherently complex and full of tensions. By the same token, MacArtney (2020: 855) underlined the ‘limitation of dichotomised framings of the patient–doctor relationship’, by stressing both the asymmetries of knowledge and social capital, on one side, and patients' empowered choices and individualised care, on the other side. Our findings corroborate the need to go beyond ‘dichotomised framings’. As a matter of fact, by using a figurational framework, we could explain that such ambivalences mentioned above are shaped by interdependent and dynamic figurations that rely on unstable power balances. In turn, this unstable equilibrium can lead people with T1D to manage their personal (dis)satisfaction with doctors by choosing one among loyalty, voice, exit, resignation. As purported by Murdoch et al. (2020), patients and doctors exercise their own power in a subtle interactional tug of war, in which there arises ‘a covert conflict over power and control, where each party plays its own cards, striving to preserve autonomy and control’ (Greenfield et al., 2012: 1208). Consistent with this, we describe power as not being a static possession, as well as transactors not being pre-determinedly fixed and independent entities. In this way, we can state that patient agency is interdependent with doctor agency: accordingly, consultations and figurations take shape in ongoing processes and unstable equilibriums that rely on dynamic and fluid power balances among transactors.

Power balance is the mechanism that explains how people with T1D manage their own (dis)satisfaction within patient–doctor figurations. Our respondents showed that this balance is affected by three relevant variables: one basic interpersonal dimension and two technical skills. The first one, the interpersonal dimension, allows shared doctoring (Mol, 2008). Our findings show the importance of this issue, according to which doctors are expected to be empathetic and open (Brown et al., 2015; MacArtney et al., 2020), as this can enhance patients' satisfaction. However, we also show how enacting such interpersonal attitude does not exempt diabetologists from deciding (MacArtney et al., 2020) about, for example, therapy or TD prescriptions. The technical dimensions entail clinical-therapeutic skills and technological skills. Our findings show that a certain degree of expertise attained over time by people with T1D reduces the sharp epistemic asymmetry, which relies on a knowledge gap (Montenegro & Dori-Hacohen, 2020), sustaining self-management tasks. They also allow us to go beyond the concepts of experiential knowledge and lay expertise (Maslen & Lupton, 2019). Indeed, our study described that, by using TDs, people with T1D ‘may also have unique epistemic access (knowledge), and even epistemic primacy and authority, over their day-to-day diabetes-related behaviours’ (Montenegro & Dori-Hacohen, 2020: 2). This happens when, as demonstrated, people with T1D can use and (self)manage data produced by TDs to act on their own therapies through clinical self-tracking (Piras & Miele, 2017). This contributes to make figurations further ‘complicated by potentially competing knowledge between physicians and patients’ (Montenegro & Dori-Hacohen, 2020: 2). Of course, these variables, while affecting power balances, do not entail deterministically one specific outcome, as it will evolve only within each particular figuration.

LIMITATIONS

This study has some limitations. First, the recruitment of a sufficient number of diabetologists was a difficult and challenging operation. Some characteristics, discussed above, rendered them akin to closed élites. Besides, we had to take into account the time constraints regarding the research (the fellowship was programmed to last only one year). Therefore, in such a context, snowball sampling was considered to be the most efficient method to add further participants, even though several shortcomings and biases make it only a second-best methodology (Cohen & Arieli, 2011). Second, in this study we checked, but did not, find specific differences according to personal variables (gender, age, duration of disease): their potential relevance should be analysed in future research with bigger samples. Third, the study was conducted in only one Region: future studies should investigate at a national, if not international, level, in order to check how different health-care systems might shape both figurations and outcomes. Last, in our sample, basically all respondents were using at least one TD; given the low percentage of their use (e.g. Bruttomesso et al., 2015), our sample represents a niche, and this might have affected our findings. Nonetheless, this study allows light to be shed on a particular kind of patient–doctor figuration, when high levels of self-management are required to patients and simultaneously when TDs are used for sustaining the daily patient's management. Given that the relevance of both chronic conditions, such as diabetes, and technological development are going to increase, these findings seem to be useful.

CONCLUSION

This study suggests that relational perspectives are better equipped for analysing patient–doctor relationships, by describing them as interdependent figurations. Indeed, this seems to be important in the field of chronic conditions, where such transactions last for the entire lifetime of patients. Besides, this approach seems to be particularly promising when considering milieus in which both self-management and TDs are becoming increasingly important. In such figurations, smarter and knowledgeable patients have high demands about three paramount variables that affect one's own self-management, as well as the figuration itself: interpersonal, clinical-therapeutic and technological skills. They are the three cornerstones for dynamic balances of power and also for the figurations themselves. The power balance may help explain certain behaviours adopted by patients, which are naturally fluid, and can be framed according to a fourfold exit-voice model, which stems from Hirschman's idealtype (1970). Even though loyalty prevails, when one or more of the abovementioned cornerstones falter, exit, voice or resignation might arise. Thus, the study contrasts both patient- and doctor-centred models, above all in their drastic forms of doctorless full-self-manager patients (Swan, 2013; Topol, 2015; van Olmen et al., 2011), and asymmetrical patient compliance and passivity (Parsons, 1951). As a matter of fact, these models seem to be too static, and not suitable to stressing dynamic and interdependent transactions, which actually do exist and, at least in part, account for patients' different choices in their management of (dis)satisfaction.

Future research should broaden the understanding presented here by following two directions based on relational perspectives. On the one hand, by considering the role of social capital and different health-care systems in contributing to shaping power balances and choices-behaviours. On the other hand, by studying diseases characterised by lower degrees of self-management requirements, and TD availability and use: this would allow further understanding as to whether this fourfold framework can be applied in general, or alternatively to assess the specific role of these two issues. Moreover, it would also facilitate understanding as to how the dynamic power balances may be shaped in such milieus.

AUTHOR CONTRIBUTION

Alberto Ardissone: Conceptualization (lead); Data curation (lead); Formal analysis (lead); Investigation (lead); Methodology (lead); Writing-original draft (lead); Writing-review & editing (lead).