From patient deference towards negotiated and precarious informality: An Eliasian analysis of English general practitioners' understandings of changing patient relations

This article contributes to sociological debates about trends in the power and status of medical professionals, focussing on claims that deferent patient relations are giving way to a more challenging consumerism. Analysing data from a mixed methods study involving general practitioners in England, we found some support for the idea that an apparent ‘golden age’ of patient deference is receding. Although not necessarily expressing nostalgia for such doctor-patient relationships, most GPs described experiencing disruptive or verbally abusive interactions at least occasionally and suggested that these were becoming more common. Younger doctors tended to rate patients as less respectful than their older colleagues but were also more likely to be egalitarian in attitude. Our data suggest that GPs, especially younger ones, tend towards a more informal yet limited engagement with their patients and with the communities in which they work. These new relations might be a basis for mutual respect between professionals and patients in the consulting room, but may also generate uncertainty and misunderstanding. Such shifts are understood through an Eliasian framework as the functional-democratisation of patient-doctor relations via civilising processes, but with this shift existing alongside decivilising tendencies involving growing social distance across broader social figurations. © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).


Introduction
This paper contributes to sociological debates regarding trends in the power and status of medical professionals, especially recent changes in doctor-patient relationships. Traditional models of interaction, characterised by patient deference, are typically depicted as being replaced by negotiation and consumer dynamics, yet few accounts exist of doctors' perspectives on such changes (Buetow et al., 2009). In this paper we present data on doctors' experiences of and views about their relationships with patients, drawn from a study of National Health Service (NHS) general practitioners (GPs) in England at a particular moment in NHS history characterised by professional turbulence and change.
Around the year 2000 General Practice was experiencing "an assault from all directions" (Gray, 2000, p.2), or a "meltdown" according to the chair of the British Medical Association's (BMA) GP committee (BMA, 2003). Multiple medical scandals had been accompanied by hostile mass media coverage (e.g. Charter, 2001) and various government policies seemingly circumscribing professional power (e.g. Harrison and Ahmad, 2000;May, 2007). A regular survey found levels of GPs' job satisfaction to be at their lowest since the survey began in 1989 (Sibbald et al., 2001), amidst complaints of increasing workloads, recruitment and retention problems, disruption caused by government imposition of a new NHS-GP contract, and increasingly demanding and disrespectful patients. Such complaints were not entirely new but they provide an insight into English doctors' social relations at this time.
One important angle of "assault" was reportedly via patient interactions. GPs' magazines published editorials deploring patients' lack of respect alongside letters from disillusioned, often recently qualified, GPs about experiences of unreasonable demands and complaints, rudeness, verbal abuse and threats; even physical assaults (e.g. Pulse, 22nd May 1999). Perhaps non-coincidentally, violence from patients also became a policy issue in NHS general practice (Elston et al., 2002). Professional interest organisations such as the BMA stressed the risk of abuse and violence, with doctors typified as innocent victims.
Drawing upon survey and interview data involving individual GPs collected in the year 2000, a pertinent moment of heightened attention to changes in GP-patient relations, the empirical contribution of this analysis is in exploring the under-researched experiences and perspectives of GPs regarding changing relational dynamics with patients. Theoretically, recent approaches have usefully located physician-patient dyadic encounters amidst wider contextual dynamics and structural change (e.g. May, 2007), yet these have largely been understood in terms of organisationalmanagement and professional-knowledge shifts whereas the influence of broader societal figurations and historical processes of change remain neglected. In discussing the implications of our findings for sociological understanding of doctor-patient relationships in the early 21st century, we explore the utility of Eliasian perspectives regarding concurrent civilising and decivilising processes (Elias, 1994). Before considering our data, we provide a brief account of relevant sociological literature on medical professionals, followed by an outline of some key themes within Eliasian sociology.

The "golden age" e nostalgic and nostaphobic perspectives
From the 1950s to the 1970s, it was sociological orthodoxy to consider doctors among the most powerful and highly trusted professionals, with interactions between doctors and their patients characterised by deference and social order. Empirical evidence from Britain to support this orthodoxy included observations of consultations involving NHS doctors, indicating the exceptionality of patients explicitly questioning, let alone overtly challenging or abusing, doctors face-to-face (e.g. Stimson and Webb, 1975;Tuckett et al., 1985). Whether such deferent patient behaviour was based on normative acceptance of medical authority or lack of power was sometimes queried. Nevertheless surveys in the 1960s and 1970s consistently indicated very high levels of patient satisfaction with NHS GPs (e.g. Cartwright, 1967).
Since the 1980s, however, this orthodoxy has been subjected to critical scrutiny. Two much-cited though contrasting American sociologists chose the same image to describe a perceived decline in medical authority at the end of the 20th century, suggesting that the "golden age" of doctoring was passing in both the organized profession's influence around policymaking and micro-level encounters with patients (Freidson, 2001, p.182;McKinlay and Marceau, 2002, p.381). This "golden age" metaphor seems equally apt when applied to changes affecting the medical profession in Britain and elsewhere in Europe (Kuhlmann, 2006;Elston, 2009). Given Freidson's and McKinlay's frequently critical approaches towards the American medical profession, it seems likely that 'golden age' had an ironic edge; but the complaints from the BMA and the medical press cited above clearly present a nostalgic tale, implying that, for NHS GPs at least, relationships with patients were much more deferential in the (unspecified) past.
These nostalgic sentiments are very apparent within recent research into perspectives of senior NHS doctors, especially regarding various disappearing terms and conditions of medical work (McDonald et al., 2006;Nettleton et al., 2008). Yet alternative accounts also exist, for example Australian doctors interviewed by Lupton (1997) did not necessarily regret emerging norms whereby patients no longer regarded doctors as omnipotent. Younger and female GPs, in particular, were likely to be in favour of patients having a more realistic appreciation of what doctors can offer, a finding we consider below in relation to our own data.
More nostophobic narratives have dominated accounts within policy-making since the early 1990s, emphasising the dangers of old-fashioned deference and blind trust in professionals (Calnan and Rowe, 2008). Successive UK governments have sought to make the NHS more responsive to patients, contrasting their proposals with negative framings of professional complacency. Although studies suggested resiliently high levels of trust in doctors despite the much-publicised medical scandals of the late-1990s (MORI, 2004), these failures were nevertheless used as political tools to challenge professional self-regulation and to promote consumer-oriented policies, undermining the 'producer'-oriented service of the past (Alaszewski and Brown, 2012). Accordingly, in recent years, NHS GPs in England have been faced with a succession of policy measures ostensibly intended to increase patients' ability to make informed choices about their healthcare and to augment GPs' accessibility, accountability and responsiveness. These have included the Patients Charter in 1991, setting out patients' rights in the NHS, amendments to GPs' NHS contracts to make patient registration more flexible, the fostering of a 'partnership' role for 'expert patients', and technological innovations to augment patients' choices when deciding about referrals to specialist care Gabe, 2001, 2009). More recently, groups of GPs have been given budgets to commission services for patients in accordance with the catch-phrase "No decision about me without me" (Secretary of State for Health, 2010). The extent of such policy intervention, alongside heightening external regulation, is therefore a subtle variation upon McKinlay and Marceau's (2002, p.382) more American-oriented theme of the 'shifting allegiance of the state' when describing 'the end of the Golden Age'.

Civilising and decivilising tendencies around patient-GP relations e an Eliasian framework
In contrast to much theorising within medical sociology, analyses drawing on the work of Norbert Elias (e.g. Goudsblom, 1986;de Swaan, 1981de Swaan, , 1988) are distinctive in their historical sensibilities e apposite when considering longer-term changes in patient-GP relations. Elias's seminal work (1994) seeks to explain the notable shifts in manners, interactional conduct and emotions which are visible when comparing prevalent social norms of the middle-ages with those of the nineteenth century. Central to this account is the 'deep-seated and iterative relationship' (Quilley and Loyal, 2004, p.10) between developments of the state, hierarchies and chains of interdependent relations within which citizens interact (sociogenesis), on the one hand, and an emerging habitus of selfrestraint, self-consciousness, empathy and identification with others (psychogenesis) on the other.
To provide a brief if schematic sketch, the monopolization of violence by more centralised states compelled increasing levels of self-restraint and affect-regulation to be exercised by individuals. In turn, this pacification facilitated interactions and exchange across society and the 'chains of action binding individuals together' (Elias, 1994, p.370) became longer and more tightly interwoven. Amidst these expanding webs or 'figurations' of interdependencies and the relative absence of violence, more attention is paid to the concerns of others e with deportment, civility and conduct of interactions increasingly important as a way of establishing 'respect' from others, while acknowledging respect for those in authority (Elias, 1994, p.425).
Within such growing proximity and interdependencies, social groups e not least the powerful classes e who earlier had cared relatively little about those from distant groups, were increasingly required to 'understand' these 'others' better, 'if only to better profit from their relations with them' (Flores, 2009, p.45;Elias, 1994, pp.177,381). Accordingly, empathy and understanding of others increased, gradually becoming internalised. Such shifts in emotional habitus (Elias, 1994, p.366) in turn facilitated yet further exchange and interwoven relations, fostering additional interdependencies and so cycles emerged. The NHS can be seen to play an important role in more recent phases of this civilising process in England, whereby state-funded healthcare provision partially emerges out of identification with the needs of others, in turn creating new interdependencies and thus a functional-democratisation e whereby reducing inequalities in social power relations facilitate growing convergences in attitudes and deportment, alongside increased mutual identification (Elias, 1994;Rodger, 2006;Brown and Flores, 2011).
Elias's figurational approach understands shifts in the more general patterning of social relations through changes, such as increasing interdependencies, within manifold smaller webs of relations. Indeed, various pressures towards growing interdependencies are perceptible in encouraging functionaldemocratisation within GP-patient relations. The traditional hierarchical nature of these relations has been largely based on a knowledge monopoly or asymmetry which has gradually been eroded through increased public access to information about health and medical care e through the mass media or, latterly, the internet (Haug and Lavin, 1983;Nettleton, 2004;Broom, 2005). Other broader changes in society, such as a late-modern exercising of lifestyle choices (Pickstone, 2000) and questioning of faith in modernism's grand (scientific) narratives (West, 2001) have arguably augmented patients' reflexivity and discretion towards 'doctor's orders' (Giddens, 1990). These developments, alongside explicit emphases upon the power and entitlements of the consumer patient within recurrent policy reconfigurations, have seemingly challenged earlier asymmetries (de Swaan, 1981;McKinlay and Marceau, 2002).
Shifting power figurations surrounding, alongside increasing interdependencies within, GP-patient relations have encouraged a more negotiated and mutually consenting interactions, replacing 'unilateral compulsion' (de Swaan, 1981, p.373), with interactions described as more patient-centred (Checkland et al., 2008;RCGP, 2011). Organisations responsible for the education, training and professional self-regulation of English doctors, have proposed a model of "new professionalism", explicitly endorsing the idea of doctors' acting in partnership with patients (Elston, 2009). Norms and expectations of patient involvement have, via policy, thus increased and exerted still more impetus towards change in interaction dynamics.
Yet caution is required before presenting too neat a picture of wholesale shifts in one direction. Elias (1994) emphasises perpetual and multi-level fluidity within social figurations, whereby decivilising processes exist alongside civilising ones (Mennell, 2001). Tendencies of consumerism (Calnan and Gabe, 2009;Newman and Vidler, 2006) and increased functional-democratisation towards negotiation-centred GP-patient interactions 'push against' (Mennell, 2001, p.32) other figurational changes marked by apparent functional-de-democratisation e increasing inequality, social division and divergent norms. Following Elias and Scotson (1965), Rodger (2006, p.129) notes that "where … sizeable minorities of people are socially and spatially excluded from full membership of society, then the progress of functional democratization is stalled". Accordingly, excluded groups come to be distinguished by their 'deviant' behaviour and lack of apparent 'civility' (Wacquant, 2004), which emerges out of and is characterised through a dearth of mutual understanding and empathy: "In those circumstances where marginality, social exclusion or sectarianism emerge, the sense of empathy for the other and the mutual restraint on behaviour which are built by frequent social interaction are absent" (Rodger, 2006, p.129).
Even though actors may have become more equal within specific forms of relationship (e.g. GP-patient), actors' positions and status amidst broader social figurations thus renders some better equipped to benefit from these changes than others (de Swaan, 1981, pp.374e376). As GP-patient contact becomes less hierarchical, structured and 'stiff' e and more informal and negotiated e so do the interactional performances required of both actors necessitate a 'controlled decontrolling of emotional controls [which] can turn sociability into an art' (Wouters, 2007, p.66). For those less well prepared for the complex subtleties of negotiation, more unpredictable interactions may lead to greater difficulties. Meanwhile, discourses of patient choice and doctor-patient partnership have not supplanted NHS GPs' continuing role as gatekeepers to expensive specialist services, working to achieve population health promotion targets and delegating routine activities to others (usually nurses) (Calnan and Gabe, 2009;Checkland et al., 2008). These shifting power dynamics, within the setting of English General Practice typified as "more efficient, but less personal" (Charles- Jones et al., 2003, p.,89), have important implications for GP-patient relations and the extent to which trust develops within, and therefore facilitates, these more informal relations (Mechanic, 2001;Wouters, 2007).
Below we briefly describe our study before considering the experiences and perceptions of our sample of NHS GPs regarding patient deference, respect and "consumerist" behaviour; alongside their experiences of aggression and abuse from patients. In particular we scrutinise insights into longer-term change in doctorpatient relations and interactions, in light of Eliasian theory.

The study
The data on which this paper is based were collected in 2000 for a study of violence against professionals working in the community . Two methods were employed, a postal survey and in-depth interviews. The survey questionnaire was sent to a one in three sample of the c.3000 GP principals contracted to provide services in the south east of England. A 62 per cent response rate (n ¼ 697) was achieved, high for a GP postal survey. The total respondent sample was representative of fully-trained GPs nationally in terms of age (30% < 40, 37% 40e49 and 31% > 50), sex (62% male, 38% female) and location (19% inner city/ large estate, 66% suburban/town and 14% rural), but had a slight under-representation of single-handed GPs (7%) and those born outside the UK (7%) compared to the national picture (NHS Information Centre, 2007). Reflecting recent trends in the development of NHS general practice, younger doctors were significantly more likely to be female, "white British", and to work in group rather than single-handed practices.
The questionnaire design focused on GPs' background, attitudes and experience and on practice organisation. Questions were asked about different types of transgressive behaviour by patients in the last two years, including verbal abuse, threats and physical assaults on doctors, and disruptive behaviour in practice premises. These data were analysed using SPSS for Windows.
In-depth interviews were conducted with a sub-sample of those GPs who reported that they had been threatened or assaulted in the past two years and who expressed willingness to be interviewed. A total of 26 interviewees were purposively selected to ensure inclusion of female (nine) and ethnic minority GPs (five), and doctors of different ages, ranging from under forty to over sixty. Reflecting the distribution of assaults and threats reported in the questionnaire (see Elston et al., 2002), more GPs were recruited from inner cities and areas of high density social housing (16) than from suburban or town practices (10) and none from rural areas. Interviews were conducted in the GPs' premises and normally lasted between one and two hours. GPs were asked about experiences of transgressive behaviour by patients and ways in which this was handled. The research was approved within the departmental ethics review framework at Royal Holloway and informed consent was provided by respondents in line with university ethics procedures. All interviews were tape-recorded and fully transcribed, with transcripts coded thematically using Atlas-Ti. Having completed the initial analysis e where long-term changes were central themes e critical discussions led us to consider the pertinence of Eliasian frameworks in interpreting the emerging patterns and themes in the survey responses and qualitative accounts.

GPs' views on changes in general practice and patient attitudes
We first asked GPs to rate their agreement about the work of a GP currently, compared to an unspecified time in the past, regarding three statements: "Working in general practice tends to be more stressful nowadays"; "The general public are less likely to respect the knowledge and advice of general practitioners nowadays"; and "The public tend to have more reasonable expectations of the services they can obtain from general practitioners nowadays". A clear majority agreed or strongly agreed with the first and second statements (88% and 69% respectively), with only a minority (20%) agreeing that patient expectations had become more reasonable (with 49% "disagreeing", and 17% "strongly disagreeing"). There was no statistical relationship between these statement ratings and GPs' gender, location or size of practice, or ethnicity, but there were age differences for two of the statements: Older doctors were the least likely to agree that patients were less respectful nowadays (74% age < 40, 66% 40e49, 65% age > 50, c 2 : p ¼ .06) and the most likely to agree that patient expectations were becoming more reasonable (11% age < 40, 15% 40e49, 34% age > 50, c 2 : p < .001).

GPs' reports of transgressive and disruptive behaviour from patients
Seventy eight per cent of our respondents reported at least one specific incident of transgressive behaviour from a patient or member of the public in the previous two years, including: physical assaults (10%); at least one incident of "verbal abuse" (75%); and more than one such verbal attack (15%). Not all GPs were equally at risk. Being aged less than 40, being "white British", and working in an inner city/large estate were all associated with higher incidence of verbal abuse (c 2 : p < .006 in each case). There was no association with GPs' gender. For all age groups, inner city doctors were more likely to report verbal abuse than GPs working elsewhere. Doctors aged 50 years and older, in suburban/small town and rural practices, were less likely to report verbal abuse.
GPs who reported experiencing verbal abuse in the past two years were significantly more likely to agree that patients "nowadays" have less respect for GPs' knowledge and advice than those who did not report any abuse (71% compared to 53%). The relationship between age and GPs' ratings, however, remained statistically significant when verbal abuse was controlled for (although numbers in some cells were small).
We also asked questions about GPs' experiences of "disruptive behaviour" from patients on practice premises, including "impoliteness", "insulting behaviour" and "complaining loudly". As with verbal abuse, across all age groups, GPs working in inner city or large estate locations were more likely to rate these behaviours as at least "something of problem" than those working in other locations (impoliteness p ¼ .007; insulting behaviour p < .001; complaining loudly p < .001). Outside the inner city, doctors aged 50 years and over were significantly less likely to experience these problems.

Summary
It is possible that response bias led to some over-representation of those who had experienced transgressive and disruptive behaviour among our respondents. But our data do suggest that, in south east England around the turn of the millennium, most GPs agreed that patient respect had declined while disagreeing that patients' expectations had become more reasonable. Intriguingly, our analysis identifies age differences in doctors' perceptions of changes in patients' expectations and respect, and age and locality differences in reports of verbal abuse and disruptive behaviour.

Qualitative data
Using cross-sectional survey data regarding perceived changes over time raises concerns around recall bias, for example considering a romanticised past among older doctors or a mythical one among their less experienced colleagues. However, potentially important changes in relationships over time can also be inferred from the age-cohort differences across doctors' far more recent experiences of patient interactions. In this section we corroborate, elaborate and extend these findings through qualitative analysis of GPs' understandings of change that emerged within interviews. Differences in GPs' accounts of their current approaches towards patients (coded as egalitarian, pragmatist or traditionalist) across age-cohorts and, similarly, the commonly described generational differences in patients' comportment and deference, were read in light of our survey findings. Together, this mixed-methods analysis offers a more robust, if still constrained, purchase on tendencies in GP-patient relations over time.

Generational differences in patient deference, respect and doctors' responses e indicators of functional-democratisation and informalising tendencies
Early in each interview we asked GPs about how patients viewed them and whether this had changed over time. Given that our interviewees had all been assaulted or threatened with physical assault (and also verbally abused), it is perhaps not surprising that most were of the definite opinion that patients were becoming less deferential. Older doctors' narratives, interpreted as offering somewhat more (although far from perfectly) reliable accounts of longer-term changes, similarly supported understandings of reduced deference. Several described changes in patient comportment, (lack of) self-restraint and related behaviours as being fuelled, at least in part, by the mass media or by government policy, as the GP below suggests: 'There's been a big change in the last few years in terms of patient anger and demand, a huge change, particularly in the last five years … we get a lot more complaints than we used to do. Complaints were incredibly rare 30 years ago and now we get a complaint once a fortnight probably. You know a proper complaint … I think they are angry people who've got to offload some frustration and anger somewhere. I don't think it's been helped by the Patient's Charter and encouragement to complain about everything in every walk of life. There's no Doctor's Charter! (Int.17. Female. 50)' More often, however, changes in attitude were related to more general social changes and longer-term social processes, with an experienced loss of authority not seen as specific to GPs. Generational differences were a recurring theme, with older patients reported as more likely to defer to doctors' authority than younger ones. This view was particularly strongly expressed by some of the younger GPs, supporting the survey findings and suggesting a particular incongruence of interaction norms as these had changed across generations: 'The elderly, I find some of them will be very stoical and deferential, inappropriately almost … which I've been embarrassed about almost. Whereas the younger ones will sometimes be very demanding. It's their rights … "bloody doctor now", you know they're going to complain, that sort of attitude. I just take it on the chin. What else can you do? (Int. 1. Male <40)' Implicit here were traces of longer-term social changes, indicated by generational differences, (sociogenesis), as were associated with differences in conduct, interaction norms and self-control (psychogenesis) (Elias, 1994). More demanding, less stoicaldeferential conduct represents an 'informalisation' (Wouters, 2007) of patients' interactional comportment, as experienced by these doctors.
As will be explored further in the following section, the role of social proximity between GPs and their patients was emphasised and was referred to as potentially mitigating the effects of wider societal change. Some interviewees suggested that, although there had been a general decline in deference and respect for GPs, they themselves were respected because of their particular personal relationships, esteem and empathy with their local community. For example: 'I think it is nearly thirty years I've worked in this area and twenty years as a full partner. I have got to know the practice, the patients, extremely well. And I think that fosters a degree of … you have a different relationship with your patients if you are a pillar of the community I suppose, a well-known member of the community which I am; I live locally.' (Int. 18. Female. 50).
But not all interviewees sought to be "pillars", or even to be highly visible, in their practice locality. For these latter GPs, a decline in the exalted status of the doctor or in patient deference was explicitly welcomed, as bringing an end to inappropriate patient subservience; fostering the development of a more equal, somewhat removed yet less formal partnership. This type of egalitarian self-presentation, and its critique of past norms, was mainly expressed by younger doctors, and is captured in the following account from a GP in her thirties: 'My Dad worked in a rural practice … He was the centre of the community, you know. They all thought he was wonderful and he'd get Christmas presents flooding through the door. And that's not nearly the same … I think that sort of deference and subservience that the people adopted, it's a certain respect, people have lost that a lot, which I think is good. But it doesn't mean they're rude or bossy all the time. They just have a different kind of respect. It's like they respect me like I respect them.' She continued: 'We don't have the same power that doctors used to have. So I think if people have respect for me as a GP it's partly as a person and partly for my knowledge, but much less for my authority … People aren't afraid of me … of their doctors anymore. They might respect them but it's not fear, whereas I think there used to be a lot of that. It was like the Priest, the Policeman and the Doctor, and you were going to be caught by any of them doing anything … ' (Int. 13, Female. <40) This doctor's narrative was thus characterised by a shift away from hierarchical power towards mutual respect. While such apparent functional-democratisation (Elias, 1994) was depicted positively by a number of participants, a small number of (mainly older) GPs gave what might be termed consistently traditionalist accounts: no longer being looked up to was regretted, partly because, in their view, this had led to more instrumental, strained or even confrontational, doctor-patient interactions. Such nostalgia for a lost 'golden age' is illustrated in the next quotation from a male GP, in his fifties: '…when I first came into practice a doctor was a respected member of society, and people sort of looked up to you, you had some degree of respect and people would accept what you say. Now you're actually a civil servant, it's an 'us' and 'them' situation. You're seen as part of the establishment and they don't believe what you say … they are less likely to accept any reasoned argument or look at the thing in perspective. People don't respect you, they see you as a means to an end…' (Int. 10.

Male. 50)
Most of our interviewees, however, presented themselves as reflexive pragmatists, adjusting to what they saw as inevitable but not wholly unwelcome change. "Paternalism" (many used this term) was consigned to the past. They did not strongly mourn the passing of an era when GPs were put "on a pedestal" just because they were GPs. Respect, they suggested, had to be earned by GPs and should be mutual.
This reduced asymmetry and heightened informality within interactions did not however mean always complying with patients' wishes or even negotiating with patients, especially when such demands were expressed in abusive or disruptive ways. According to one male inner-city GP, far from undermining mutual respect, his forceful, activating approach aimed to foster interdependency (mutual recognition, as opposed to asymmetric dependency) among troublesome patients: 'By me having shouted at them and saying "No, you're not getting the other thing", and "no, I will not see you if you're drunk, under any circumstances" … they will come and see me sober. They won't come in here drunk … And that does make a difference. It means that, number one, they have respect for me, but they have more respect for themselves as well. And I think that's the bottom line in trying to get people better, is to get the respect back in themselves, and if you play games with them and just dish out pills to them, whenever it suits them, you're not showing any respect for them.' (Int. 26. Male 40e49.) A rather different strategy for displaying respect for patients was reported by other pragmatists, especially when encountering highly educated, internet-informed patients "who know more than I do" about "some kind of spurious or rare disease": "I unashamedly say to them, 'I'm no expert in this, I'm listening to what you say' 'cause you're also hearing all the latest from the experts in the hospitals." (Int 2 Female 50).
Doctors' negotiation of "respect" within more informal interactions could thus take various forms (de Swaan, 1981, p.379), either demanding respect as a means of encouraging mutual recognition, or through their own displays of deference to expertpatients. Both approaches point towards experiences of functional-democratising shifts (Elias, 1994) between GPs and their patients, driven by reducing knowledge asymmetry and government policies amongst other factors (Buetow et al., 2009), which have developed since an earlier 'golden age' where intrinsic asymmetric respect was the norm.
These more informal modes of interactional comportment e as advocated by the (generally younger) egalitarian doctors and commonly experienced in interactions with younger patients e were seemingly more fragile however (Wouters, 2007). Importantly, these changes within GP-patient interactions were not just about reducing hierarchy and formality. The traditionalist quoted earlier (interview 10) did not simply bemoan a loss of respect but also described a growing divide between 'them' and an 'establishment' 'us' (Elias and Scotson, 1965). Such broader societal changes and growing social distance were also salient and common features of GPs' interview narratives e as we now move on to explore further.

GPs' accounts of aggression from patients e strained interactions across social distance
As noted in the preceding section, a number of GPs (especially younger ones) described being uncomfortable if patients were totally trusting, but saw themselves as increasingly having to put up with some, usually younger, patients being excessively demanding. Below we see further reference to inter-generational shifts in comportment and emotions within the doctor-patient encounter, contrasting encounters characterised by formality and deference with less hierarchical negotiations and mutual demands (de Swaan, 1981) e as usefully captured by the concepts of informalisation (Wouters, 2007) and functional democratisation (Elias, 1994): 'There are different attitudes from different groups. There's one group who have been with this practice for 20, 30 years with us … and wouldn't say anything. And there is another group of patients who do not have that sort of respect or attitude at all, especially the young age group … there is a difference in attitude among some of the youngsters belonging to the low socioeconomic classes where they can be very demanding and can lose their patience or control or respect … they can explode on any trigger … ' (Int 5. Female. <40) As captured here, narratives regarding younger age groups were in many cases interwoven with socio-economic characteristics. Interactions with young working-class adults were commonly referred to as more prone to 'losses of control' e outbursts of 'instant enmity' (Wouters, 2007, p.88).
This interweaving of age-cohort-related and socio-economic features was especially prominent when GPs were asked to discuss experiences of verbal abuse, rudeness and disruptive behaviour. Interview accounts corroborated the general picture shown in the survey: verbal abuse and disruptive behaviour were occasional but not exceptional experiences for GPs and their practice staff, particularly for those working in inner-city practices with marginalised populations. While self-reported experiences and sense-making narratives should not be confused with social facts, doctors' qualitative accounts of their experiences e triangulated with geographical variations in experienced violence reported through the survey e can be considered as providing some salient insights into the underlying characteristics of interactions experienced as violent.
Patients' health-related personal troubles were generally perceived as the most common proximate causes of incidents of violence and aggression, rather than frustrated patient expectations alone (Elston et al., 2002). But these two factors were also combined within explanations. For example, a female GP in her fifties described the patients that caused most difficulties, as follows: 'Drug addiction … the constant nuisance, I would say … Oh [and] alcoholics … They come and demand something and the girls [receptionists] can't cope and they won't leave the premises. They're drunk, it's usually the case of the local winos. They want to be seen now, and they have got a bad toe or a bad foot.' (Int 12. Female. 50.) Aggression and rudeness were generally denoted as increasing, with this trend primarily ascribed to increasing prevalence of alcoholism and other forms of substance abuse, but also to social disadvantage and community breakdown, as noted earlier and as reflected again here: 'There's no doubt that patients who live in what you might loosely call council estates, the less well-off patients or disadvantaged patients, whatever you want to call them, social class four, five, they often … they seem to have a lower threshold for becoming violent somehow. I don't know whether it is lack of education, whether they don't understand the system, I don't know but they would be the first ones to fly off the handle and get agitated with you. Not always; tends to be. They are often unemployed, unmarried mothers or unmarried mothers' boyfriends … ' (Int 24. Male. 40e49.) This narrative, from a traditionalist GP, provides further glimpses of common understandings of violence amongst our participants. Such lay theorisations of violence reflected social scientific understandings of the lower 'sensibility thresholds' for the 'perpetration of violence' associated with broader societal processes of exclusion (Fletcher, 1997, p.179). A lack of integration and familiarity with, and therefore understanding of, 'the system' was seemingly related to the inculcation of certain emotional and interactional tendencies (Elias, 1994;Rodger, 2006, p.129). An alternative but complementary perspective would consider this more 'instant enmity', alongside 'instant intimacy' (Wouters, 2007, p.85), as common features of interactions with anonymous-others within the informalising interactional contexts referred to in the preceding section. Patients 'flying off the handle' (instant enmity) may be seen as resulting from doctors' failed attempts at constructing instant intimacy with 'others' rendered more distant from the mainstream 'system' by socio-economic divides (Elias and Scotson, 1965;Wouters, 2007).
Difficulties in mutual understanding were also noted by those presenting more egalitarian professional selves. These participants usually offered more empathetic, less judgemental, accounts of their more troublesome patients, deploying patient-centred discourses in describing attempts to understand aggressive behaviour from the patients' perspective: '(I've had) people saying they will do certain things if you don't do certain things for them. And basically people will say "Well, if you don't do this I am going to sue you", and [then] I've got problems. But again it's a case of negotiating with them, I think. You know I'm sure a lot of threats are made because they are afraid of something. They have something they want to get out of it and they they're afraid they're not going to get it. And it's very difficult to get out of them what the problem is.'(Int.3, Male GP, 40e49).
This account contains different emphases from that of the preceding traditionalist, however themes of emotional tendencies and comportment styles emerging due to social distance from, and a lack of understanding of negotiating with, 'the system' are again visible. These problems seemed to emerge in this instance within a negotiated (more informalised) interaction and involved 'outsiders' who experienced great difficulty in conducting successful encounters with the more 'established' GPs (Elias and Scotson, 1965). The interactional 'tensions' apparent here were latterly linked to deeper underlying fears, especially those related to 'unpredictable exposure to those in power' (Elias, 1994, p.444), with these fears appearing to 'shape' emerging patterns of conduct within shifting interactional orders (Elias, 1994, p.444).
More middle-class and/or highly-educated patients scarcely figured in accounts of challenging or "demanding" patients, despite many such patients being registered with our GPs; although some respondents referred to such patients when describing interaction styles (see latter quotation in preceding section). Any problems caused by the professional classes and/or the highly health-literate demanding instant services were presented as trivial (c.f. Buetow et al., 2009). Socio-culturally, GPs shared much in common with these latter more 'established' groups (Elias and Scotson, 1965) and we infer that this proximity eased these encounters, through a more straightforward 'mutual consideration' which is fundamental to the effective handling of more informal interactions (de Swaan, 1981, p.376).
In contrast, 'outsider' patients e such as those more likely to live in the inner-city 'estates' described by the GP participants e dominated accounts of difficult patients and violent or aggressive encounters. Due to more limited interdependencies, interactions and embeddedness within mainstream society (Rodger, 2006;Elias, 1994), outsider patients may correspondingly have been more likely to express and comport themselves through manners which were (mis)interpreted as uncouth or even threatening by typically 'established' GPs (Elias and Scotson, 1965).
These difficulties of conducting more informal interactions across social distance may, as referred to in the preceding section, have been partially mitigated where GPs were well known within, or at least familiar with, the local community. Older GPs more commonly referred to such an embeddedness within local social contexts and this may help explain their lower reporting of aggression. Meanwhile younger GPs e which as we saw in our survey data were more likely to experience rudeness and violence e commonly expressed in interviews a desire to live outside their practice community. So whereas the geo-social location and related familiarity of older GPs may have helped bridge social divides, the geographical distance of their younger counterparts from their work may have amplified their social distance from more marginaloutsider patients.

Discussion
Drawing upon our study of GPs' experiences and perceptions of patient deference, respect and consumerist behaviour, alongside those of abuse and violence, we have explored the extent to which doctors felt that they had lost authority within relations and interactions with patients, in comparison with the past, and the consequences of this lost authority. Our qualitative data broadly confirmed the picture provided by the survey data, while refining understandings in ways relevant to claims about the passing of a "golden age" of deference. Longer-term changes in social relations (sociogenesis) around, and interaction norms (psychogenesis) within, patient-doctor encounters were two central and defining themes.
Developing inferences of longer-term changes from crosssectional data raises, as already noted, various potential difficulties (Annandale and Hunt, 1998). However the inferences made here are not simply based on recollections of potentially romanticised or mythical pasts, but through age-cohort differences within the recent/current experiences and practices of GPs, understood through both large-scale survey responses and in-depth interview narratives. Integral to interview accounts, moreover, were the repeated references to patients' comportment, also emphasising generational differences. That those GPs espousing more egalitarian-negotiating approaches were mainly aged under 40, whereas the traditionalists were mainly over 50 e alongside the contrasting of the (sometimes extreme) stoical-deferential conduct of older patients with a less inhibited and more demanding (and occasionally highly aggressive) behaviour of younger patients e combine to indicate strong inter-generational tendencies towards the informalisation of GP-patient interactions (Wouters, 2007). Triangulating insights from our qualitative interview analysis with our survey findings assists us in the corroboration, elaboration and extension of the latter quantitative analyses, while corroborating and extending some of the thematic patterns apparent in the qualitative data (Bryman, 2006, p.105).
There is much to be gained from further exploration and theorisation of the age-cohort/diachronic and geographical patterning apparent in our findings. Future, more 'Eliasian', methodologies (Elias, 1994;Wouters, 2007) might invoke historical analyses of change apparent within documents such as medical schools' teaching materials. Alternative approaches might include comparing very fine-grained analyses of the micro-dynamics of very recent GP-patient interactions with those apparent through older studies (e.g. Heath, 1986). However both documentary and interactionist methodological approaches would also enable only partial understandings, particularly regarding the socio-economic and geographic (sociogenesis) processes explored here.
Our quantitative analysis identified that most GPs, particularly younger ones, saw patient respect for doctors' expertise as having declined. Whereas younger GPs may have been comparing their current experiences with a more mythical past, older GPs could refer to specific memories, even though the potential for romanticising these memories in relation to moral scripts remained (Annandale and Hunt, 1998). Our qualitative analysis suggests the possibility that differences in survey responses may also have been shaped by older GPs' heightened understanding and empathy of diverse local citizens through years of interacting and building relations with them, and/or enduring structures of deference towards older professionals who may have conducted interactions with patients less informally than their (more egalitarian, less traditionalist) younger colleagues.
Generational differences, alongside those of locality, were also apparent in our survey data regarding experiences of violence and abuse. While most participants experienced at least occasional incidents of verbal abuse and disruptive behaviour, younger GPs and those working in inner-cities reported such experiences more commonly. GPs possess many of the characteristics associated with higher risk of violence and aggression against small businesses, particularly in areas of social deprivation; most obviously, constant contact with the public, some of whom are distressed or under the influence of alcohol or psychoactive drugs (Hopkins, 2002), within practice premises, patients' residences or the streets they walk when on duty. Urban and more unequal settings are also more prone to expressions of 'incivilities' (Phillips and Smith, 2003;Sennett, 2003).
More basic understandings and interpretations of our transgression-related survey findings are enhanced and nuanced when read in light of in-depth interview accounts. Generational patterns in patients' propensity for violence were again frequently referred to, with younger patients e particularly the socially disadvantaged, less educated young e depicted as much more prone towards uncontrolled outbursts and aggression. It is the combination of diachronic, generational and socio-economic/ geographic patterns and themes in our data, as interpreted through the enhanced mixed-methods lens, which directed our theorisation of these findings towards an Eliasian framework. Core themes in the qualitative analysis e more negotiated and less inhibited interactions (informalisation), reduced deference and mutualised respect (functional-democratisation), and recent interactional tensions associated with growing social distance (decivilising processes) e resonated strongly with features of 'civilising processes' (Elias, 1994) and more recent refinements within this theoretical tradition (especially Elias and Scotson, 1965;de Swaan, 1981;Rodger, 2006Wouters, 2007. In short, changing features within these dyadic doctor-patient figurations are only understandable through attentiveness to changes beyond, involving these actors' roles and positioning within wider societal figurations (de Swaan, 1981). Indeed if the earlier 'golden age' figurations were defined or implied by our participants as formal-hierarchical-dependent dyadic GP-patient relations within relatively homogenous and familiar communities, more recent figurations were characterised by more informal-equal-interdependent interactions amidst inequality and heterogeneity. As de Swaan (1981, p.376) suggests, regarding a declining 'management' of interactions based on earlier established 'canons of authority': 'where these orders have lost their sway, people do not abandon themselves to inconsiderate indulgence [outbursts] but negotiate some kind of arrangement with those close to them'. However an emerging social distance, which has coincided with the decline of the golden age canon, has seemingly inhibited such a closeness towards those whose social-location has become more marginalised from the 'established' position of GPs (Elias and Scotson, 1965). Managing interactions through 'negotiation' may therefore more commonly be rendered dysfunctional in the relative absence of social proximity (de Swaan, 1981).
Social distance was most problematic for younger GPs, for whom living outside their practice area was often a deliberate work-life-balance choice to avoid interacting with patients when off duty, as well as to achieve a suitable quality of life for a professional family. Moreover if, as our interview data suggested, younger doctors tended to adopt more informal consultation styles then they might have been more likely to be abused e when negotiations broke down (de Swaan, 1981) e especially if they had a higher proportion of younger (and perhaps less deferent) or substance abusing patients in their caseload than their older colleagues. In contrast, many years of visibility and relationship building may have rendered some older doctors more familiar, respected and/or equipped with an experienced deftness in dealing with "the many and intricate ways class position and employment futures impact on psychological functioning and conduct" (Rodger, 2006, p.132).
Growing inequality and distance, in contrast, destabilise the practices of 'mutually expected self-restraint' (Wouters, 2007, p.188) amidst 'fleeting encounters' (McKinlay and Marceau, 2002, p.402) with anonymous GPs e where the term doctor-patient 'relationship' is often a misnomer (May, 2007, p.40). Changing interaction norms, facilitated through reduced inequality between those located more centrally within mainstream societal figurations (de Swaan, 1981), may accordingly be experienced as 'alienating' when marginalised citizens encounter privileged professionals (de Swaan, 1981, p.369). Income inequality and divisive societal tendencies, as increased across the decade before 2000 (Taylor-Gooby, 2013), help explain such alienation (de Swaan, 1981;Rodger, 2006) which was described within GPs' accounts as a more recent phenomenon. GPs, rooted firmly within established groups via educational experiences and often relatively privileged socioeconomic backgrounds, may therefore be increasingly exposed and sensitive to 'instant enmity' (Wouters, 2007, p.88) within less structured, more informal, interactions with socially-distant outsiders.
The historical moment of our study can thus be understood as involving a 'double movement' (de Swaan, 1981, p.369) between informalising (civilising) and distancing (decivilising) processes; where the management of GP-patient relations through negotiation (rather than authority-by-command) leads to unpredictability amidst growing social estrangement (de Swaan, 1981, p.369). Neither stable nor continuous, (in)formalisation processes ebb and flow, occasionally surging in waves (Wouters, 2007, p.169) whereby "the older standards have been called into question but solid new ones are not yet available" (Elias, 1994, p.440). A surge in informalising pressures within GP-patient interactions during the 1990s (government policies driving consumerism; internet making information more available), alongside shifting figurations across the 'encompassing society' (de Swaan, 1981, p.364), is useful in analysing the unusually heightened state of 'flux' (Wouters, 2007, p.20) apparent around general practice in England in 2000.
In identifying various longer-term tendencies, our findings do not wholly support Jones and Green's (2006) optimistic view of the impact of changes in general practice on doctor-patient relations. Our main causes for concern are wider social and policy changes through which GPs may be gradually becoming more distant from some (especially younger) groups e note that older working-class (i.e. more integrated) patients were absent from negative narratives e while nevertheless expecting and being expected to work more in partnership with their patients during fleeting encounters. Doctor-patient communication, when impeded by social distance, can lapse into mutual misunderstanding and fear which is experienced by both parties as rudeness and confrontation.
These resulting frictions, rooted in (de)civilising and (in)formalising processes 'pushing against' one another (Mennell, 2001), may have intensified since these data were collected. GPs' professional status, based on a monopoly of knowledge (Abbott, 1988), had already been compromised before 2000 and arguably this process has continued, as has growing inequality (Taylor-Gooby, 2013). Meanwhile, recent policies have strengthened practitioners' position as gatekeepers through new commissioning roles. Subsequent tensions between authority, knowledge and respect render interaction skills more vital than ever amongst a range of 'discriminating customers' (Newman and Vidler, 2006); not least those experienced as 'outsiders' by the professional 'establishment' (Elias and Scotson, 1965).