Dependency,trust and choice? Examining agency and ‘forced options’ within secondary-healthcare contexts

Introduction

A growing consensus regarding the appropriateness and effectiveness of patient choice within public healthcare systems is apparent among policy-makers across many post-industrial welfare states (Le Grand, 2007: 94), not least within Australia and England (Department of Health, 2012; Luxford et al., 2010). This trend continues despite important concerns regarding the popularity, equity and viability of choice within imperfect healthcare markets characterised by limited and asymmetric information (Brown and Calnan, 2012; Calnan and Rowe, 2008a; Greener, 2003; Taylor-Gooby, 2008). The promotion of patient choice emerges from, while also stimulating, a politicised awareness that services and professionals are of varying quality and safety; thereby shifting patients from a position of relative confidence (where professionals’ quality was more taken-for-granted) to one where contingency, risk and possibilities for regret make trust more salient (Giddens, 1991; Luhmann, 1988): ‘trust is only required if a bad outcome would make you regret your action’ (Luhmann, 1988: 98).

Greener (2003) argues that given the impossibility of calculative decisions between healthcare options – due to imperfect information, market structure and the complexity of medical practice – trust in professionals becomes a proxy for choice. In this sense trust is vital to understanding patients’ ‘choices’ but Greener also asserts the importance of choice for analysing trust. As noted already, emphasising healthcare choices renders trust more explicit and less taken-for-granted (Luhmann, 1988). Earlier models characterised by a less-questioning confidence or ‘hegemonic trust’ (Greener, 2003), wherein ‘the citizen is expected to give up all but the most humble rights, to put himself into the hands of the expert and trust his judgement and good intentions’ (Friedson, 1970: 355), are described as being increasingly replaced by more conditional, reflexive approaches (Calnan and Rowe, 2008b). Yet Greener (2003) is concerned that in many healthcare settings a lack of plausible alternatives shapes a ‘coercive’ form of trust, where patients are obliged to cooperate due to the impossibility of ‘exit’. In this manner different types of trust can be characterised in terms of choice or its absence (Greener, 2003: 84).

Meyer and Ward (2013) go further to suggest that where vulnerability or related power dynamics act to preclude choice then trust is no longer relevant and we should instead refer to dependence. As a logical continuation of Luhmannian (1988) arguments this would seem appropriate and compelling but further complications remain. Möllering (2005) and Barbalet (2009) consider, respectively, vulnerability and dependence as central to the purpose and nature of trust. And thus tensions emerge (Brown and Calnan, 2012) between, on the one hand, a recognition of the way vulnerability and related configurations of power may inhibit choices to trust (Goold, 2002; Greener, 2003; Meyer and Ward, 2013) and, on the other hand, arguments that emphasise that it is vulnerability (alongside uncertainty) which makes trust necessary in the first place (Möllering, 2005). After all, if one is not vulnerable to a future outcome then there is little need to trust.

Drawing on data from two qualitative interview-based studies – one in Australia, one in England – investigating trust in contrasting healthcare contexts, we interrogate these tensions. In doing so we aim to enhance understandings of the extent to which choice and trust are complexly interwoven and structured within power relations. Following brief sketches of these studies, we first move to problematise neat distinctions between choice and no choice, trust and no trust, as have thus far been defining of the literature (e.g., Davis and Rundall, 2000; Greener, 2003; Luhmann, 1988; Meyer and Ward, 2013). Pursuing this line of enquiry further, we consider structures of health system formats, and patients’ socio-economic background and related location within such contexts, as shaping divergent understandings and horizons of possibilities within patient lifeworlds (Scambler and Britten, 2001; Schutz, 1967: 59). We then explore chronologies of events and experiences (time), alongside habitus, emotions and related boundedness (consciousness), as further mechanisms by which agency to choose and to trust is configured.

The studies

By way of exposition and to ground our theoretical framework in the empirical, we consider data from two recent studies we have undertaken into trust in healthcare. The finer methodological details of these studies are available elsewhere (Brown, 2009; Meyer and Ward, 2013), however we offer brief synopses here in order to clarify the provenance of the data.

Both studies employed qualitative methodologies to explore the nature, extent and bases of trust via semi-structured interviews with patients in two distinct secondary-care settings. The England-based study, conducted in 2007, investigated trust among 20 cervical cancer patients across various diagnostic and treatment paths. Interviews with 20 healthcare professionals from a range of specialisms were also undertaken. The Australia-based study was conducted between 2008 and 2009, investigating the trust of 37 patients at various stages of treatment for heart disease (22 men and 15 women). Healthcare systems differ across these two countries, facilitating recognition of agency amid vulnerability and dependence in contrasting system contexts. In brief, the National Health Service (NHS) provides healthcare to all citizens in England, funded entirely from taxation. Australia, similarly, has a universal healthcare arrangement, Medicare. However the Australian government offers tax incentives to encourage those who can afford private health insurance to opt for this, in an effort to minimise strain on the public system (Harley et al., 2011). This creates a much higher take-up of private health insurance than in England (50% contrasting with 11% in 2012) – as approximately reflected in our samples (5/20 in the English study were private patients, 17/37 in the Australian study). Different system formats have implications for patients’ choice in accessing healthcare, as will be noted in later sections.

Both studies were theoretically driven and designed to critically examine social theories of trust. Data analysis for both studies was carried out via the coding and connecting of reoccurring themes, in light of theories of trust. The use of abductive and retroductive inference facilitated a more comprehensive analysis of the data while concurrently questioning and refining theory in light of data (Meyer and Lunnay, 2013). Although theoretically driven, the analysis was informed by considerations of the data within the wider context of each participant, including their illness and healthcare system experiences (Coffey and Atkinson, 1996).

The impossibility of choice?

A defining characteristic of sociological analysis is an understanding of the way actors’ thoughts and actions are constrained by, enabled through and a reaction against (Giddens, 1984) a complex tapestry of cultural norms and material structures (Archer, 1996), taken-for-granted assumptions (Schutz, 1967), tastes and dispositions (Bourdieu and Wacquant, 1992) and acquired embodied understandings and emotions (Barbalet, 2001; Merleau-Ponty, 1968). The shaping of preferences within social environments is by no means a purely sociological perspective (e.g. Simon, 1982), however sociological theorists have devoted particular attention to interrogating the manner by which decision-making tendencies are bounded and supported by ‘the social’ and, in particular, how the ‘power of the individual is confined by a range of specifiable circumstances’ (Giddens, 1984: 14–15).

Tensions between constraint and freedom lie at the heart of Giddens’s (1984) ‘duality of structure’, providing a useful departure point for considering choice, trust and their mutual relationship. The choice to trust can be said to never exist, in a wholly unfettered sense, but moreover as never fully ceasing to exist – in that a certain agency always endures: ‘it is of the first importance to recognise that circumstances of social constraint in which individuals “have no choice” are not to be equated with the dissolution of action as such’ (Giddens, 1984: 15).

Although stressing a rather stark ‘choice’/‘no choice’ distinction in parts of his 1988 chapter on trust, a recognition of these finer subtleties around choice is very much evident elsewhere within Luhmann’s writings on complexity and contingency (Luhmann, 1985: 25). Due to the problems of stress, disappointment and fear (1985: 32) which complexity and contingency otherwise give rise to, not to mention the ‘impossibility’ of making fully conscious selections ‘at every point in time for everyone’s every expectation’ (1985: 31), assumptions are inevitably resorted to. Assumptions which become commonly shared and adopted over time can develop into social structures, as patterns of understandings and expectations become routinised. These structures necessarily ‘limit the scope of options by the act of choosing, which is almost not consciously conceived of as such’ (1985: 31). ‘Choices’ in healthcare settings, including those involving trust, should be understood in this light.

Things become further complicated however when acknowledging that trust, as a format of expectation which reduces complexity (Luhmann, 1988), also fits a number of Luhmann’s (1985: 32) criteria of ‘structure’. Thus a ‘choice’ over whether to trust is shaped by prevailing assumptions and structures but in turn becomes a ‘structuring’ force itself, so reducing the complexity and contingency of later ‘choices’. This setting in motion of pathways of structured decisions which play out over time will be returned to (see section on time). For now it is sufficient to acknowledge that complexity and contingency are sufficiently problematic to render the development of solutions, such as trust, a ‘life necessity’ (1985: 40) in limiting choice.

This inherent need to overcome complexity and contingency should not be misinterpreted as necessitating trust however (cf. Greener, 2003). Luhmann (1979) underlines that distrust is also a viable solution to complexity. Yet this latter approach precludes cooperation with a potential trustee that would enable particular positive outcomes to be realised: ‘either A trusts B to achieve C, or A cannot have C’ (Barbalet, 2009: 372). This leads Barbalet (2009) to describe trust in William James’s terms as a ‘forced option’, yet further potentialities for agency and choice endure. Such other options include pursuing C but with significant doubts and anxiety, in the relative absence of trust.

This latter path, as well as various forced options to trust, were apparent across our two studies. In the example below it seemed that knowledge asymmetry created a certain form of dependence upon the expert through which trust was compelled (Meyer and Ward, 2013):

Moreover the fundamental vulnerability of the patient amid the illness experience and the need to fend against anxiety similarly ‘forced’ the patient quoted below towards focusing on positive expectations:

Yet a small number of respondents in the English study referred to distinctly distrusting. This suggested that a choice remained around trust, even when choice over treatment was not envisaged. For these patients, although they were indeed obliged to go down the route of treatment for their cancer due to no perceived alternative, agency was still apparent in forming negative expectations and attitudinal-evaluative positions towards particular professionals. The patient below, having said more generally, ‘Doctors: [I] don’t trust them’, went on to explain her distrust of the main doctor overseeing her care, due to poor communication:

Similarly, another patient expressed how she did not trust the doctor who had recently performed her colposcopy examination. She went on to refer to her negative concerns regarding the possibility of another appointment with the same clinician a few months later:

And when he said – ‘have you ever smoked?’ And I said, ‘yes’, and I told him that I’d smoked for a couple of years, like six years ago. And I’d stopped, but I sometimes smoke … you know … socially. And he said – ‘why do you do that?’ And I thought, ‘well, fuck you, it’s none of your fucking business’.

Here we see clear expressions of distrust, despite apparently limited choice over undergoing further treatment with that particular clinician. As will be explored below, an array of choices and moments of agency thus emerge within the process of (dis)trust – which is best seen as an evolving sequence of embedded agency rather than a single momentary decision (Greener, 2003; Rapley, 2008; Schutz, 1967).

Social structures, horizons of possibilities and the limiting of choice

Time – the chronology of decision-making pathways

The preceding section denoted interwoven features of the patient experience – including system perceptions, societal discourses on health, individual stocks of capital and experiences of vulnerability – which led to the structuring of patients’ horizons of possibilities or lifeworlds (Scambler and Britten, 2001). This shaping of lifeworlds took place both more deliberatively – by actively bracketing-off uncertainty and negative futures – and/or passively – through taken-for-granted assumptions and related understandings (Schutz, 1967). The focus thus far has been upon how choices and trust were shaped via structures, experiences and interpretations in the present, but less about processes already set in motion in the recent past. This section aims to draw out the fundamental role played by time as a further basis for considering the dialectical interaction between structure and ‘choice’. Trust, as a format of linking the present to the future, by drawing on knowledge derived from the past, is inherently formatted by the lived chronology of the truster and time can be seen to structure choices through a number of mechanisms.

Reconsidering the consciousness of trust – the influence of socio-biographies upon trust

Influences of power in constraining ‘choices’ to trust, via wider socio-structural and chronological processes, are regularly overlooked within the social-scientific literature, due to a focus on the distinctive features of trustees and systems and how these are appraised by the truster. Trust furthermore tends to be depicted as a largely conscious and highly reflexive ‘decision’, yet we have already stressed that many ‘choices’ to trust are less-than-consciously conceived (Brown, 2009; Giddens, 1984; Luhmann, 1985) and that prior trust (in systems and individuals) encourages a ‘docility’ (Simon, 1982) which bounds or brackets reflection. Starting from these limits to the ‘consciousness’ of trust and choice, this final main section explores how mechanisms of habitus, emotions and bounded rationality, as are inherent to trust (Misztal, 1996), constitute further dimensions by which healthcare decisions are embedded within socio-historical contexts and the power relations pervading therein.

Conclusion

Within the preceding sections we have sought to illuminate an array of mechanisms by which choices – to receive treatment and/or to trust – are embedded. Considering our data, trust was seemingly a sufficient but not necessary proxy for choosing healthcare options, forming one important mechanism among others which acted to bound reflection upon pursuing particular healthcare paths (cf. Greener, 2003; Rapley, 2008). Various features of system understandings, socio-economic structures, illness vulnerabilities and chronology interacted to constrain but not annihilate choices to trust (cf. Greener, 2003), with trust usefully understood as an emergent enactment of structure from the (near or distant) past which in turn structures, but does not determine, ‘choice’ in the present.

Continuing agency is apparent amid each of the various structuring mechanisms addressed in our analysis. Emotions, for example, while rooting decisions in past experiences, are not an overwhelming force which removes agency. A swathe of literature following Hochschild (1979) among others notes various ways in which emotions are ‘worked on’ within social contexts and therefore the influence of affect is open to shaping by the individual. Similarly, the manner by which a ‘will to trust’ impacts upon individuals in vulnerable circumstances still requires a positive construction of expectations and a wilful ‘bracketing away’ of negative possibilities, emphasising the ‘highly active role’ played by the truster within these situations (Brown, 2009: 394). Of course both these forms of ‘agency’ are constrained, respectively, by feeling rules (Hochschild, 1979) and by limited alternatives and vulnerability when in need. The perpetual tension around the ‘duality of structure’ thus endures (Giddens, 1984).

Narratives of distrust were more apparent among the small English gynaecological-oncology patient sample than from the Australian cardiac-care case study. Rational-actor understandings of healthcare choice would predict that such distrust would lead to patients exiting care, or at least voicing their concerns and demanding alternatives. It was striking however that distrust seldom led to exit or even to voice, but importantly influenced the meaning and experience of healthcare for patients – often in terms of heightened vulnerability and anxiety. Possibilities for exit or voice were importantly shaped by varying stocks of capital (especially socio-economic and cultural-educational), with these impacting via private insurance possibilities, understandings of choice and more or less deferential interactions with professionals. It is here that the power dynamics of choice and trust become most palpable, with the economically and/or culturally privileged, alongside the less unwell, more able – via exit or voice capacities – to afford not to trust. That the loyalty of the less privileged and/or more seriously ill may be eased by trust may therefore oblige such a bracketing of uncertainty (Brown, 2009). Examples of distrust in these contexts suggest, nevertheless, that trust is obliged amid continuing agency rather than coerced (cf. Greener, 2003). Meanwhile those who are ‘loyal’ and distrusting may be most vulnerable of all.