Beyond Binaries: Reflections and a Suggestion on the Subject of Medical Professional Satisfaction

This article notes the tendency in the sociological literature to frame studies of medical professionals in terms of a series of binaries (e.g. control/resistance, powerful/powerless professionals, medicine/management). It suggests that moving away from this approach to acknowledge a more nuanced perspective would be helpful. The article draws on recent empirical studies to support this view.

This paper is written from the perspective of a sociologist who studies health policy and professionals, but whose research has never focused explicitly on the topic of doctor satisfaction (DS). The impetus for this paper was in part a realisation that there are many studies con ducted by sociologists that shed light on issues pertaining to DS. Yet there appears to be limited cross fertilisation of ideas and findings between these and other studies which have DS as an explicit focus. The aim of this short piece is to make a modest suggestion with a view to generating discussion and debate.
What follows is divided into three sections. The first of these discusses some studies raised by the literature on DS. This is not an exhaustive review, nor is it an evaluation of strengths and weaknesses of specific studies. Instead it is a preamble intended to highlight issues which are relevant to the arguments advanced in the rest of the paper. The second section draws on literature from the UK to discuss three sorts of binaries that the paper argues we should move beyond. Concluding remarks and a fourth binary are presented in the final section of the paper.

Learning from the literature
There are a range of approaches, terms and concepts used within the literature to explore what might broadly be termed DS. Some studies draw on one or more specific validated instruments to capture a range of concepts, others use their own questions developed for their particular study. This raises questions about conceptual definition and the ability to draw wider conclusions. For example, the Physicians Foundation (2014) in the US, asks whether individuals would choose the same career if they had their time again (over 71% would) and whether they would recommend medicine as a career (about half would not), as well as about intentions to "accelerate their retirement plans due to changes in the healthcare system" (almost 40% are doing this) and feelings about the current state of the profession (44% are positive about this). One interpretation is that physicians can display high levels of satisfaction (based on the 71%) at the same time as exhibiting fairly high levels of discontent. This suggests that the relationship between (dis)satisfaction and (dis)content is not necessarily straightforward. Another interpretation is that asking questions about having one's time over again tells us little about feelings today and may reflect experiences to date which are changing over time. Additionally, responses differ according to category of respondent, with younger doctors, employed doctors, females and primary care doctors more positive than others. So it is important to avoid drawing general conclusions about satisfaction on the basis of top level numbers.
As with the Physicians' Foundation study, much of the literature concerning doctors' attitudes to working life draws on large surveys, undertaken at a particular point in time. These often provide information based on the responses of a large number of individuals. However, the diversity of approach raises concerns about whether we can draw conclusions from diverse studies. Leaving aside the problems of attempting to do so, the findings raise questions about what constitutes a "high" level of dissatisfaction and whether we should be unduly concerned, especially as views on this may be largely in the eye of the beholder. In the US, where much of the research has been conducted, various studies highlight what might be regarded as high levels of physician dissatisfaction, especially in primary care settings (McKinlay & Marceau, 2011). The influence of managed care (Stoddard, Hargraves, Reed, & Vratil, 2001) and other factors that reduce "autonomy" is reported as having a negative impact on satisfaction (e. g. Hadley & Mitchell, 2002;Lammers & Duggan, 2002). However, the relationship between satisfaction and perceived autonomy is not straightforward (Tyssen, Palmer, Solberg, Voltmer & Frank, 2013). Several reviews highlight links between physician satisfaction and quality of care (e.g. Scheurer, McKean, Miller, & Wetterneck, 2009;McKinlay & Marceau, 2011), although there are studies which fail to identify such a relationship (e.g. Utsugi-Ozaki et al., 2009). As Casalino and Crosson (2015) note in this special issue, direct evidence of the relationship between physician satisfaction and the quality of care is relatively scarce and shows mixed results.
In terms of whether things are worsening over time Scheurer et al. (2009) found, based on a synthesis of longitudinal studies, that physician satisfaction in the US was relatively stable, although for primary care doctors a small decline was discernible. In contrast, Nylenna, Gulbrandsen, Førde, & Aasland's study of Norwegian doctors detected an increase in satisfaction over time (Nylenna et al., 2005; see also Carlsen and Bringedal, 2009) and Whalley, Bojke, Gravelle, & Sibbald (2006) found that satisfaction amongst English primary care doctors was higher in 2004 than 2001. However, this may in part reflect reforms which invested heavily in primary medical care and since then, a series of relatively unpopular (with the medical profession) reforms have been introduced which may have impacted on doctors' views.
For the most part, the literature does not attempt to link sociological concepts to the analysis and understanding of the issues involved. A notable exception to this is McKinlay and Marceau's (2011) paper, which reviews and classifies evidence before going on to suggest that the concept of alienation is helpful in explaining physician discontent. The paper outlines a number of factors related to the modern workplace which are seen as contributing to alienation. These include a loss of control over the practice environment, supportive colleagues and a perception of workload as reasonable. From this perspective, doctors are on the receiving end of policy changes which attempt to control them and this leads to dysfunctional impacts. Much of the analysis is informed by experiences of managed care in the US and of primary care physicians, whose relatively low status and remuneration may not be mirrored in other countries. However, the paper also draws on evidence from other countries highlighting unique and contributing factors which the authors see as explaining doctors' views.
www.professionsandprofessionalism.com Page 3 The authors' arguments about the scale, importance and causes of doctor dissatisfaction appear at times, to lead them to approach the evidence in a manner which seeks confirmation, rather than pursuing critical interrogation. This is evident in the discussion of studies measuring the relationship between physician discontent and quality of care. Findings showing that dissatisfaction impacts negatively on quality are accepted uncritically, but for studies which find no such relationship, a more critical approach is adopted (e.g. McKinstry et al.'s 2007 study finds no relationship between doctor's morale in English primary care and patients' perceptions of performance).
According to McKinlay and Marceau, "a professional norm against workplace disgruntlement" means that patients may not notice any difference between dissatisfied doctors and their less stressed counterparts. Furthermore, the authors suggest, quantitative techniques may be too crude to capture the "subtle association between physician dissatisfaction and the processes and outcomes of care" (2011, p. 320). The paper does not discuss the possibility that studies which present self-report (as opposed to data extracted from medical charts) by doctors in relation to, for example, propensity to commit errors and provide sub-optimal care (e.g. Melville, 1980) may be used an outlet for and reflective of "workplace disgruntlement" rather than reflecting care quality. Following McKinlay and Marceau's arguments about methodological weaknesses, it would presumably be preferable to use data extracted from patient records, rather than relying on self-report. McKinlay and Marceau do not suggest that this approach should be taken, but they do acknowledge that a Japanese study (Utsugi-Ozaki, 2009) which did this found no relationship between job satisfaction and quality of care.
McKinlay and Marceau (2011) identify encroachments on professional autonomy (such as requirements to pursue guideline driven care) as contributing hugely to dissatisfaction amongst medical professionals. Tyssen et al.'s 2013 study published after their review compared doctors' perceptions on quality of care, professional autonomy and job satisfaction in Canada, Norway and the US. They found that US doctors felt greater autonomy, but lower levels of satisfaction compared with doctors in Canada and Norway. Their paper raises questions about the way in which studies define and measure autonomy, as well as highlighting the complex relationship between autonomy and satisfaction.

Recognising complexity-beyond binaries
One way of exploring and understanding such complexity is to pay more attention to the insights from relevant, sociologically informed research. Sociological theories can provide a helpful perspective by offering concepts and frameworks locating doctors in the realm of the "social" to provide explanations of the views and behaviours of individuals which complement existing DS studies. There are a number of such studies which do not explore doctor satisfaction or discontent explicitly, but which shed light on doctors' behaviours and perceptions. These resonate with McKinlay and Marceau's (2011) paper since they are often focused on the impacts of reforms in healthcare settings and doctors' responses to changing contexts. Such research is frequently concerned to identify and categorize processes and outcomes in terms of control of professional work (Kitchener and Exworthy, 2008), though in many cases this results in a control-resistance framework, which leaves little room to accommodate more nuanced responses to reforms (Numerato, Salvatore & Fattore, 2012). In common with medical researchers in this area, there is a tendency amongst many sociologists to simplify and overemphasize findings which reinforce a priori assumptions. One of the problems with this type of approach is that it implicitly conceptualises power in terms of a zero sum game (Foucault, 1982), and linked to this, a control-resistance framework is an overly simplistic view of a complex context www.professionsandprofessionalism.com Page 4 (Bolton & Houlihan, 2009). A preoccupation with medical power can lead to a failure to recognise that doctors can and do feel powerless, at times (McDonald, 2002). Linked to this, there is a danger that reporting on "satisfaction" takes on an "either/or" quality which leaves little room to incorporate complexity.
What follows draws on recent UK studies to illustrate why moving beyond this approach is important if we want to make greater progress in relation to DS.

Beyond binaries-incorporating exploration of different types of medical work
In a recent study my colleagues and I compared doctors' responses in areas of activity that were the subject of "top down" reforms with those which were not (McDonald, Cheraghi-Sohi, Bayes, Morriss, & Kai, 2013). The former, the Quality and Outcomes Framework (QOF), involved guideline driven care underpinned by financial incentives (Martin Roland, 2004). In terms of the latter, the focus was on patients with medically unexplained symptoms (MUS) and the study found that doctors continued to practise as they had always done in this area. Whereas the introduction of guideline driven care appeared to embed new ways of working and thinking, with patients as groups to be managed as part of the organisational processes established for doing this, the organisational dimension with regard to MUS patients was largely absent from doctors' accounts. For patients with QOF-related conditions proactive call and recall systems were developed and facilitated by disease registers and customized software. For MUS patients, doctors described reacting to patients who consumed large amounts of their time. Often doctors felt unhappy with the process and unsure of their impact. No attempt was made by the doctors to work together as a group practice to identify patients who attended too frequently and/or to meet to discuss strategies for managing this group of patients. This contrasted sharply with QOF-related areas of activity, where doctors and other members of the practice team worked closely together to ensure that targets were met.
Whereas QOF resulted in standardised approaches to care, accounts of MUS patients were characterized by variations and individual and idiosyncratic practice. Doctors' accounts suggested that variations between approaches were seen as a "taken for granted" aspect of practice life, rather than something to be minimized in the pursuit of standardisation. Doctors gained little satisfaction from dealing with MUS patients. The absence of attempts to tackle this patient population could be attributed in part to the more complex nature of guidance and lack of computerized protocols for this group of patients compared to the various mechanisms which support QOF target delivery. However, it can also be interpreted in the context of the persistence of a logic (Scott, 2001) of medical professionalism, which involves doctors coping with whatever presents and focusing on interactions with individual patients, as opposed to conceptualizing this in terms of organizations and the populations they serve (McDonald et al., 2013). This example highlights that it is important not to approach DS from an "either/or" perspective, but to recognise the complexity of the context in which doctors are working. Examining different types of medical work undertaken by the same individuals helps to provide a more nuanced understanding.
McKinlay and Marceau (2011) view clinical guidelines as undermining clinical autonomy and contributing to physician dissatisfaction. Yet studies demonstrate that this is not necessarily the case and physician satisfaction has been shown to be positively associated with guidelines (Kerr et al., 2000). This suggests that doctors do not always view such guidance as unhelpful or overly managerial. Doctors did not express high levels of dissatisfaction in relation to guidelines in our study (McDonald et al., 2013). Furthermore, as Timmermans (2008) explains, in the US and elsewhere the growth in clinical practice guidelines depends to a great extent on the collaboration of medical professional organisations. Guidelines have been part of the www.professionsandprofessionalism.com Page 5 landscape of English primary medical care for some years and GPs as generalists do not necessarily see guidelines as threatening their professional autonomy (Checkland, 2004). In our study (McDonald et al., 2013) the doctors actively engaged with processes of organizational redesign and changes to ways of working. They welcomed being able to practise in a "well organized" environment and new structures did not undermine business ownership, which remained in the hands of the medical partners. Rather than viewing the reforms as running roughshod over traditional general medical practice, they can be seen as an attempt to build on and accelerate trends which were emerging pre-QOF, which may explain doctors' willingness to embrace them.
In a context where much of the sociological literature is concerned with powerful medical professionals and their ability to resist change (e.g. Currie, Lockett, Finn, Martin, & Waring, 2012) our findings were relatively unusual. The state of affairs described with regard to MUS patients is less attributable to the power of medical professionals than to feelings of powerlessness experienced by many of the doctors in our study. Whilst maintaining the status quo is often characterised in terms of the "political efforts of actors to accomplish their ends" (DiMaggio, 1988, p. 33), it can also involve less conscious, iterative, routine-based and routine-reproducing activities (Scott, 2008). These routinised behaviours continue in the context of MUS patients, where the logic of medical professionalism endures, despite the dissatisfactions of doctors in this area. At the same time, QOF activities were viewed in a relatively positive light, although some aspects (i.e. the potential to lose sight of the patient's agenda) were seen in more negative terms.

Beyond binaries-comparing different dimensions of autonomy/discretion
Various studies from the physician satisfaction literature make links between autonomy and satisfaction. However, autonomy is often measured and defined differently in different studies. Acknowledging developments over time which mean that nowadays professionals are much more likely to be located in employing organisations than was previously the case, Julia Evetts (2002) suggests that rather than using the concept of autonomy, the term professional discretion is more appropriate and highly relevant to modern professional organisational contexts. Professional discretion, enables workers to assess and evaluate cases and conditions, and to assert their professional judgement regarding advice, performance and treatment. To exercise discretion, however, requires the professional to make decisions and recommendations that take all factors and requirements into account. These factors and requirements will include organizational, economic, social, political and bureaucratic conditions and constraints. Thus, professional decisions will not be based solely on the needs of individual clients, but on clients' needs in the wider corporate, organizational and economic context (Evetts, 2002, p. 345).
This conceptualisation raises questions about the extent to which doctors are comfortable with a remit to "take all factors and requirements into account," as well as the relationship between discretion and satisfaction. It also suggests a more complex and multi-dimensional concept than the view of "autonomy," which features in much of the DS literature.
For Cheraghi-Sohi and Calnan (2013) it is important to recognise the multi-dimensional nature of discretion when exploring doctors' responses to changing work environments. The authors draw on Evetts' (2002) views to examine changes in English primary medical care and define various aspects of discretion (i.e. bureaucratic, social, organizational, economic and political). To some extent this resonates with McKinlay and Marceau's (2011) call to "change the approach to research on physician discontent […] to an organizational approach" (p. 322). But Cheraghi-Sohi and www.professionsandprofessionalism.com Page 6 Calnan (2013) go beyond this to examine how changes in each of these dimensions are perceived by the doctors involved. As their paper illustrates, conducting analyses on changes to discretion from a uni-dimensional perspective leads to, at best, a partial and at worse, an erroneous view of the situation. The authors find support for Evetts' (2011) assertion that modern professionalism is changing and "modern professional discretion is embedded in an organizational context and can only be understood on those terms with individual medical freedoms and latitude being tempered by increased collegiality, accountability and performance management" (Cheraghi-Sohi and Calnan, 2013 p. 59). This resonates with other studies which report the emergence of new forms of professionalism and an acceptance by medical professionals of such changes (e.g. Kuhlmann, 2008).

Beyond binaries-collapsing the medicine versus management divide
My final example is drawn from a recent study which I undertook with colleagues (McDonald et al., 2012) examining responses of hospital doctors to financial incentives which reward hospitals (as opposed to individuals) for providing care which is in accordance with a "best practice" pathway. As outlined above, guideline driven care has often been depicted as constraining clinical freedom. Furthermore, mechanisms such as financial incentives might be construed as part of the toolkit of management in a context where traditional notions of medical professionalism differ sharply in terms of ethics and orientation from the imperatives of management (Schlesinger, Gray, & Perreira, 1997;McDonald 2009). Many studies describe how managers and medical doctors inhabit different subcultures each with their own beliefs and practices (Morgan & Ogbonna, 2008;McDonald, Waring, & Harrison, 2006) languages and communication habits (Holtman, 2011). Numerato et al. (2012) describe how a whole raft of techniques and initiatives (e.g. New Public Management, governmentality) have been seen as part of a process whereby the tentacles of managerialism extend to control medical professionals. They conclude that: Management and professionalism are frequently framed as contradictory in doctors' views' and it is worth noting that concepts such as co-optation, adaptation, negotiation or resistance remain located within the cultural incorporation⁄opposition, in other words hegemony ⁄ resistance framework and tend to situate the result of the dynamics between professional practice and managerial logic on a continuum between resistance to and compliance with managerialism. We suggest that this conceptualisation tends to overemphasise the importance of a conflictual model (2012, p. 637).
In our study, doctors were generally in agreement with the "best practice pathways" rather than seeing them as a mechanism of control. This is hardly surprising since the national pathway development was for the most part underpinned by medical knowledge and input. Some doctors described working with local managers to exploit the potential of the incentive initiative in order to improve care. At the same time, different levels of "management" were reported as being more or less amenable to collaborative working on the issue depending on a range of factors including proximity to the service and history of working relationships. This reminds us that "management" is not a unified monolithic entity but a collection of managers at different levels of the organisation characterised by collaborative and competing objectives and impulses. Additionally, doctors reported using what might be seen as management tools in order to deliver "best practice" care as part of a process of persuading senior managers to allocate resources to their service. In the area of renal "best practice tariffs" (BPTs) this BPT is intended to encourage vascular access for haemodialysis. There are wide variations in vascular access use within Europe, which do not appear to be explained by differences in population characteristics (Pisoni et al., www.professionsandprofessionalism.com Page 7 2002) and rates in the UK are low. Consultant nephrologists identified a key issue for pathway compliance was access to surgeons and this could be problematic. Different doctors responded in different ways to this as illustrated below: …if we do this better, we get paid more… best practice tariff... has certainly given me some leverage with the Trust to say: "Actually, this is something that we do need to invest in."...what I've been in the process of doing ….is getting a business case together for bringing in a new member of staff, really, to coordinate the vascular access service. …The business case [was] approved because the best practice tariff will fund it. And without that, although it's something we wanted to do, actually, for a couple of years, we've never been able to because there's been no money for it. [Nephrologist A] In contrast, as the following quote illustrates, some nephrologists wanted to increase vascular access, but felt frustrated at their inability to influence factors which they saw as beyond their control: I can't do a lot about it if my surgeon doesn't do it... The problem is it's not really a best practice tariff for renal services but for vascular surgery because even if I'm running the best renal unit, if I don't have a proper vascular surgeon in the area I won't be able to increase my fistula rate… I would need a more enthusiastic vascular surgeon who would take it up as a priority. [Nephrologist B] It does not seem helpful to categorise these response using a control/resistance or medicine/management framework. The doctor in the second quote feels powerless and frustrated rather than actively resisting control. In the first case, the adoption of "management" tools and close working with local level managers has resulted in what the nephrologist sees as a positive outcome for patients. He does not appear to see this as threatening his medical identity, nor is it used to shore up medical power to defend against controlling managers. There is a literature describing how medical professionals attempt to maintain their clinical autonomy by selectively using "management" measures. However, my point here is that this is less about defending against encroachments on clinical autonomy than joining up aspects of care in a way that solves the sorts of problems that can leave doctors feeling powerless and dissatisfied. Furthermore, the example illustrates how differences and divisions between different types of doctors (in this case nephrologists and vascular surgeons) can be more important than divisions between doctors and managers.
In this example we can see how some doctors can incorporate what might be seen as management techniques and practices in ways which challenges static ideas about what constitutes doctors' roles. Sociologists have written much on the subject of identity (Jenkins, 2008) as a fluid and dynamic concept involving multiple identities (e.g. mother, doctor, wife etc.). Applying this perspective and locating individual accounts in a wider social and historic context, can help to understand doctors' views and actions and avoid simplistic assumptions about the relationship between medicine and management.

Concluding remarks
As the foregoing illustrates, exploration and understanding of DS cannot be reduced to simplistic binary categories. My argument is not that for example, medical power is a useless concept. Indeed the fact that vascular surgeon in the above example can choose to ignore some aspects of work (i.e. grafts for vascular access) and prioritise others suggests that consideration of power is often important. However, a focus on power can lead to assume that "power" at the macro level readily provides an important resource at the micro level for individual doctors. Similarly, control and resistance and/or the distinction between medicine and management may be important in explaining some aspects of professional work, but we do not have to assume that in all cases, they are the appropriate frames from which to view the action.
Sociologically informed approaches can add value to the study of DS. Their contribution goes beyond providing more detailed qualitative and/or ethnographic studies, to using and developing theory to inform our understanding of events. The sociological imagination suggests that "personal troubles" such as the accounts and views of individual doctors must be understood in terms of public issues, biography and history and the "intricate relationship between them" (Mills, 2000,). This suggests a wide remit and highlights the need to pay attention to the broader social and historical context in which individual experiences are located.
This brings me to another binary, which is the split between sociological research and medically led research into the topic of DS. Breaking out of the silos in which we work to combine insights from both of these areas would be a fruitful way forward. As with the other binaries discussed in this paper, there is a danger that some of us have become so accustomed to viewing things in terms of these binaries that they have a "taken for granted" quality that makes it difficult for us to reflect critically upon them. The intention is that this short piece will stimulate discussion about whether and how we should start to do this.