Introduction
In the UK, the governance of doctors’ performance has been a site of contestation and intense struggle for policy dominance for much of the past decade. The profession competes to maintain control over its historically strong system of professional self-regulation with an ambitious ‘command-and-control healthcare state’ (Moran, Reference Moran1999). This article explores how two sets of policies – the state's clinical governance policy and the profession's revalidation policy – have emerged in large part as a consequence of the politicisation of the profession's ability to regulate itself effectively. On the surface, and according to much of the literature (e.g. Harrison and Dowswell, Reference Harrison and Dowswell2002), the development of these policies and the new governance arrangements that they have brought about signal a further decline in the profession's dominance in defining clinical standards, and thus challenge the influential role played by professional self-regulation.
This article explores the emergence and conditions of possibility of the new governance of medical performance through clinical standards in the UKFootnote 1. Originally flagged up in the 1997 White Paper The New NHS, the formal inception of the Department of Health's (DH) clinical governance policy in 1999 has led to the creation of a plethora of arms-length regulatory agencies, including the National Institute of Clinical Excellence (NICE). At the same time, the General Medical Council'sFootnote 2 (GMC) revalidation policy calls for the periodic re-assessment of individual practitioners so as to determine whether they are fit to practise. After briefly describing the institutional context in which the two policies were developed, we consider how the new governance through clinical standards operates. We then address the processes of negotiation that produced each of the policies. Here, clinical governance can be interpreted as a swift response to the politicisation of medicine's self-regulatory apparatus following a series of high-profile cases of medical misconduct in the late 1990s. Unlike the relatively unified, hierarchically organised DH, the profession took considerably longer to put together its policy response. In contrast to the manner of the state's response, the profession was initially divided, with some of its members having to sacrifice their policy ambitions in the interests of expediency and political survival.
However, the nature and structural underpinnings of the state–medicine relationship mean that the clinical governance policy and, by extension, governance forms based on hierarchy lack teeth. We argue that this is the result of the state's dependence on medicine to meet the healthcare rights of citizens, and its consequently weak bargaining powersFootnote 3. Lastly, the article engages in a critical reflection on the arguments advanced thus far: what is missing from the analysis? What can the institutional context of the state–medicine relationship not explain? Here, we discuss the surprisingly divided state of the profession and consider the capacity of the new governance to alter the clinical autonomy of rank and file practitioners through clinical standards.
The article draws on in-depth, semi-structured interviews with 10 key informants and a range of published materials. A combination of purposive and snowball sampling was used to identify the informantsFootnote 4. We began by contacting an experienced and well-known policymaker in the profession, who provided the names of several other leading figures involved in the construction of the revalidation and clinical governance policies. Whereas the interviews offered valuable information about how these policies were negotiated, published materials helped to contextualise the interview data, to link the accounts to one another by establishing a record of events. The published materials employed in the analysis are diverse in type: press releases and the reports of professional bodies, the DH and clinical governance organisations, such as NICE and the Healthcare Commission (HC); a small number of articles and editorials published in leading medical journals; and newspaper accounts. The reports of professional and government agencies offered important information about how clinical governance and revalidation policies are intended to work in practice.
The institutional context
The governance of medical performance in the UK works through a largely unspoken agreement between the state's and the profession's respective regulatory domains: the former controls the performance of – and performances within – healthcare organisations as part of its remit to govern the consumption and provision of health care, while the latter assumes responsibility for the training, certification (via the Medical Royal Colleges) and licensing (through the GMC) of individual doctors. The latter also deals with poorly performing doctors via its ability – again through the GMC – to strip such doctors of their licenses following disciplinary proceedings.Footnote 5 This division of labour is reflected in the organisation of the state and the medical profession, respectively.
As a command-and-control healthcare state, the state component of the governance system operates through the DH and the National Health Service (NHS). Policies regarding the performance of organisations (that is, the setting, monitoring and evaluation of clinical standards) are formulated centrally through mechanisms like National Service FrameworksFootnote 6 (NSFs) and by arms-length regulatory organisations, such as NICE and HC, whereas responsibility for their implementation typically rests with managers working for the Strategic Health Authorities (SHAs), Hospital Trusts or Primary Care Trusts (PCTs). The assumption underlying such centralised governing arrangements is that policies can be written in one locale and translated into practices in another, often distant setting. Crucially, this presupposes that a strong and unbroken chain of command is in place and that managers have the authority over clinicians to enact policies, even in the face of resistance. The clinical governance policy has meant an expansion of the hierarchy-based governance system with the creation of bodies like NICE and HC.
The organisations of professional self-regulation are numerous and varied. Briefly, they include the Medical Royal Colleges, which certify members of the various specialities and oversee most professional development schemes; the GMC, which in addition to licensing doctors and maintaining a register for specialists, plays a major role in standards-setting in undergraduate medical education; the British Medical Association (BMA), which acts as a trade union, representing its members in contract negotiations with the state; and the specialist associations, which (like the Colleges) are learned societies but which also represent members’ interests in discussions with the state and other branches of the profession. The normative institutions of medical authority are based first and foremost on professional autonomy.
The new governance through clinical standards
For the most part, the division of labour outlined above is reflected in the substance and scope of the clinical governance and revalidation policies: the former deals with the performance of organisations, while the latter concerns the performance and skills of individual practitioners as members of the medical community. However, there are a number of areas of overlap. For example, poorly performing doctors may be subject to NHS disciplinary procedures, as well as those of the GMC, although only the latter has the power to revoke their licenses. Along similar lines, a portion of NICE's clinical guidelines concern the actions of individual practitioners, instructing them on how to treat patients in the context of a specific set of contingencies. Thus, clinical governance necessarily involves the conduct of individual clinicians, even if its end goal is to establish, monitor and evaluate the clinical standards of organisations. Lastly, it is worth noting that the GMC must ultimately answer to Parliament, which gives the state a limited degree of leverage over the profession by allowing it to review the GMC and the system of professional self-regulation more generally when civil society places pressure on it to do so.
The clinical governance policy
As we will see below, when such pressures arose in the late 1990s, the DH responded with its clinical governance policyFootnote 7. In April 1998 it was announced that ‘for the first time in the history of the NHS hospital trusts were to be held legally accountable for the quality of the service they provided’, a requirement later given statutory force by the Health Act, 1999 (DH, 1998a). The consultation document, A First Class Service: Quality in the New NHS, accompanying the announcement defines clinical governance as ‘a framework through which NHS organisations are accountable for continuously improving the quality of their services and safeguarding high standards of care by creating an environment in which excellence in clinical care will flourish’ (DH, 1998b: 33). In terms of its underlying mechanisms, the policy divides the different activities of clinical governance (e.g. standards-setting and monitoring) into separate functions to be undertaken by distinct national regulatory agencies. NICE and the NSFs were to be responsible for standards-setting, while monitoring and evaluation were to become the task of a number of agencies: Health Authorities, Primary Care Groups (now Primary Care Trusts), the now defunct NHS Executive, the Commission for Health Improvement (now the Healthcare CommissionFootnote 8), the NHS Performance Assessment Framework, and the National Patient and User Survey. The Council for Healthcare Regulatory Excellence (CHRE), an independent statutory body, was set up in 2003 by Parliament to oversee the work and review the major decisions made by the health profession's regulators, including the GMC. The organisations listed above for evaluation, as well as provider hospitals and PCTs, were also responsible for intervention (NHSE, 1999a: 4; 1999b: 3).
Of all the agencies listed above, NICE has received considerable scholarly and media interest. Established in March 1999 as a Special HA, its remit includes approving and disseminating clinical guidelines produced by professional (i.e. the Medical Royal Colleges and specialist associations) and academic bodies, as well as appraising new technologies for their clinical and cost effectiveness and developing and promoting clinical audit. Although the establishment of NICE has given the state a more prominent – rather arms-length – role in the regulatory domain of standards-setting, the HC is one of the principal bodies responsible for monitoring whether healthcare organisations implement the guidelines provided by NICE and, more generally, whether they are abiding by clinical governance arrangements. Launched in April 2000 (also as a Special HA), the HC is an arms-length body that conducts annual performance reviews of NHS Trusts, which may be used by the DH in its review of funding arrangements for Trusts, and undertakes in-depth investigations in cases of severe service failure.
Although the standards-setting role of NICE and the monitoring function of HC are relatively well defined, mechanisms for sanctioning in cases of poor individual or organisational performance are less clear. In cases of poor clinical performance, it is not always clear how a given organisation or set of individuals are to be sanctioned, save for existing NHS disciplinary procedures in the latter case. As its investigation into 10 maternal deaths at Northwick Park Hospital in London shows, HC may identify a range of causes contributing to deaths, including managerial, organisational and resource deficiencies, as well as instances of clinical error (HC, 2006: 101–2). Thus, not only is there a lack of clarity regarding the degree of, and responsibility for, sanctions, but the identification of the causes of poor performance also involves considerably more than simply a search for the responsible clinician. It is just as likely to mean holding managers and managerial processes to account. As the case of Northwick Park also demonstrates, if being under-resourced is a contributing factor leading to compromised levels of patient safety, the sanction of advising the state to cut the funding of the Trust in question further is clearly inappropriate. Beyond the use of existing NHS disciplinary procedures for clinical and managerial staff, how then do clinical governance arrangements lead to sanctions in cases where clinical standards are not being upheld? The policy was uncertain about this issue when it was initially formulated in 1999 (Salter, Reference Salter2004); it remains so in its present stage of implementation.
In terms of its underlying logic, the state-based governance of medical performance through clinical standards operates through hierarchy. The significance of the new governance arrangements brought about as a consequence of the clinical governance policy lies in the fact that they extend hierarchy-based forms of governance to clinical standards as an area of the governance of medical performance that has traditionally been dominated by professional self-regulation. At the same time, the governing arrangements are in keeping with existing practice; that is, they are premised on the idea that policies can be formulated centrally by the DH and regionally by the SHA bureaucracies and implemented locally by managers working for Trusts. However, as was the case with the internal market policy in the early 1990s, this assumption may prove naïve. When faced with the resistance of clinicians to NICE guidelines, managers may lack recourse, particularly those without medical training. By their very nature, clinical guidelines leave scope for interpretation. Interpretation in turn is predicated on clinical training and the cultivation of clinical judgement. Thus, the significance of medical expertise in both the making and the application of NICE guidelines should not be underestimated. In the case of PCTs in particular, the implementation of NICE guidelines is further complicated by the tendency for NICE recommendations on what is ‘cost effective’ to be centred on interventions that are also cost increasing. In such a situation, there is a strong temptation to fall back on medical authority to resolve the conundrum. In its attempt to become a more powerful player in the governance of doctors’ performance through clinical standards, then, the state and its managers must contend with medicine's entrenched power in the NHS and beyond (Salter, Reference Salter2004).
The revalidation policy
In contrast to the state, the profession governs itself by professional self-regulation. More specifically, as part of the negotiations of the new governing arrangements the profession makes use of networks, depending on the area of governance and the segment of the profession involved. When faced with intense political pressure, the profession may temporarily and intermittently put aside its differences. The revalidation policy was initially promoted by policy entrepreneurs in professional self-regulation to attach standards to a system of re-licensure, whereby practitioners would have to re-qualify for their license on a regular basis by providing evidence that they performed at a satisfactory level for each of the seven areas of performance identified in the GMC's Good Medical Practice. While the policy agenda was the outcome of a longer history, revalidation was launched as a political concept by the GMC in February 1999, when it announced its decision that, ‘to maintain their registration, all doctors must be able to demonstrate regularly that they continue to be fit to practise in their chosen field’ (GMC, 1999a: 1). The Report of the revalidation steering group on which this decision was based proposed a six-stage process incorporating all the three functions of governance: standards-setting, monitoring and evaluation, and intervention. The stages were local profiling of performance; periodic external peer review of the profiling process; providing evidence that would lead to revalidation of the doctor's entry in the register; (and where there were concerns about a doctor's performance) local remediation; referral to the GMC's performance procedures; action by the GMC on the doctor's registration (GMC, 1999b).
As we will see below, this proposal for revalidation was modified substantially as a result of the brokerage efforts of a network of specialist leaders a few years later, in many ways trying to reaffirm the basis of doctors’ traditional autonomy over controlling their performance. The second version of the policy, which the GMC introduced in 2003, calls for the appraisal of practitioners by peer review (i.e. review by other members of the medical community, typically a clinical manager)Footnote 9. In this version of the policy, the standards against which practitioners’ conduct is to be judged are typically vague, as the speciality-specific standards created by the Royal Colleges are in many cases to be taken only as suggestions, as opposed to explicit criteria against which to assess a clinician's performance. Moreover, evaluation lacks the degree of externality proposed in the earlier version of the policy, as a doctor's peers are responsible for his or her assessment. However, in contrast to the clinical governance policy, both the 1999 and 2003 versions are clear that when it comes to sanctioning, incompetent performance will mean that the doctor in question becomes subject to the GMC's disciplinary procedures, and may eventually lose his or her license. In sum, the revalidation policy has redefined professional self-regulation, defining more clearly standards for medical performance and the corresponding procedures for evaluation, monitoring and sanctioning of medical performance.
Negotiating clinical governance?
The clinical governance policy was not so much negotiated as imposed by the DH on the profession. However, given the political context, the profession was not in the position to resist. Strong conjectural pressures opened a window of opportunity for the state to act. Yet importantly, the conjectural pressures highlighted deeper institutional failures of existing governing arrangements and not least the relative isolation of the organisations of professional self-regulation. On 29 May 1998 the GMC ruled that two surgeons from the Bristol Royal Infirmary were found guilty of continuing to operate on children with heart defects when they knew their death rates were unacceptably high. In addition, a doctor manager was found guilty of failing to stop the operations after he had been alerted to the high mortality. Accompanied by intense media interest, these decisions sent what the British Medical Journal described as ‘nothing short of an earthquake through British medicine’, commenting with some foresight that ‘the reverberations are likely to be felt for years’ (British Medical Journal, 1998a). Public trust in doctors would, it seemed, never be the same again. All had changed, ‘changed utterly’ (British Medical Journal, 1998b: 1917). In the following month, faced with an unprecedented degree of public concern over the issue, the government announced that a formal inquiry, armed with a wide-ranging brief and considerable investigative resources, would be conducted into the management of complex paediatric heart surgery at the Bristol Royal Infirmary trust and its predecessor between 1984 and 1995. The policy window had opened. Other events were to ensure that it did not close.
Amid the plethora of cases of inadequate medical performance illuminated by legal action, GMC investigations, local Trust inquiries and independent inquiries in the period following the Bristol revelations, two stand out as having particular significance for public trust in the profession: the Royal Liverpool Children's Hospital Trust, Alder Hey, and the General Practitioner, Dr Harold Shipman. In September 1999, following the early findings of the Bristol Inquiry that parents had not given properly their informed consent for the hospital's retention of their children's organs, parents of children treated at Alder Hey Hospital initiated a series of actions, which resulted, firstly, in the CMO's investigation into the national situation regarding the retention of organs, body parts and tissues at post-mortem examinations and, secondly, in an inquiry chaired by Mr Michael Redfern QC into what had happened at Alder hey itself (Royal Liverpool Children's Hospital Inquiry, 2001). The Redfern Inquiry revealed that between 1988 and 1995 the pathologist at Alder Hey, Professor van Velzen, had removed the hearts, brains and other internal organs from all the 850 children on whom he had conducted a post-mortem, ostensibly for research purposes, without the consent of their parents.
In 2000, the CMO's national census of organs and tissues held by pathology services in England revealed the retention of approximately 10,500 organs, body parts, stillbirths and foetuses for research and teaching purposes, of which the relatives concerned were largely unaware (CMO, 2001: 22). The experience of parents at Bristol and Alder Hey was therefore not unique: it was established medical practice for children (and adults) to be interred having had their organs removed, but with the relatives believing they had been buried intact. The Bristol and Alder Hey inquiries transferred this customary activity from the private medical sphere to the public sphere, where the high-profile politicisation of pathology duly took place.
Amid the almost daily reporting of doctors failing their patients at the time, one other event in particular acted to sustain the political pressure for reform. On 31 January 2000, Dr Harold Shipman, a general practitioner, received 15 life sentences for murdering 15 of his middle-aged and elderly women patients by lethal injections of diamorphine. Subsequently, the first report by the Shipman inquiry established that he had in fact killed 215 of his patients (Shipman Inquiry, 2002). The Shipman trial reinforced the impression in the public mind that when it came to the control and dispersal of the human body at and after the point of death, doctors enjoyed a distinct social advantage. Shipman was able to illicitly stockpile large quantities of controlled drugs, kill his patients, avoid a referral to a coroner for a post-mortem in all but a very few cases, and arrange the cremation of his victims in the majority of cases, all undetected. As with Bristol and Alder Hey, the Shipman case illustrated with brutal clarity the large areas of discretion afforded to doctors by the public's trust in them as self-governing professionals.
Bristol, Alder Hey and Shipman – to name but three cases – irrevocably altered the territory of professional self-regulation, casting the capacity of the profession to regulate itself satisfactorily into serious doubt, and leaving room for the state (and, by extension, hierarchy-based forms of governance) to take on a more prominent role in the setting, monitoring and evaluation of clinical standards through its clinical governance policy. In response to these events and the consequent politicisation of professional self-regulation, the political instinct of the state and politicians was to take measures that would immediately attract the attention of the public and assuage its growing concerns, and to demonstrate that it was prepared to take action and introduce hierarchy-based forms of governance in what had formally been the exclusive domain of professional self-regulation. Institutionally, this instinct was expressed through the creation of the national regulatory agencies discussed above. However, the relationship that these institutions were expected to have with one another does not seem to have been considered carefully in the process of making the policy or establishing the various bodies. The fact that the contributions of NICE, HC and the other organisations responsible for clinical governance overlap was not addressed, nor was the type of horizontal or vertical relationships they were to forge among themselves made clear. Taken as a whole, the policy lacks an overarching framework for accountability (Salter, Reference Salter2004: 130).
Negotiating the revalidation policy: competition and compromise within a divided profession
The clinical governance policy was developed hastily as a response to the widespread concerns about doctors’ performance. On the one hand the policy reflects a medical profession under pressure, but on the other hand it also points to the continued strength of the normative institutions of medical authority. Among the countries discussed in this special issue, the UK is the only case where governance change includes a separate policy stream dominated by the medical profession. In contrast to the swift and hierarchical style of the DH's mode of policy making, the profession's networks dealt with the politicisation of self-regulation in a slower manner, working both horizontally and vertically across different branches of the profession to eventually put forward a proposal to the DH in 2002–2003 as part of consultant contract negotiations. Before discussing the 2003 version of the policy, however, we briefly consider the nature, scope and historical development of an earlier version of the policy pursued by a network composed primarily of GP politicians. Sir Donald Irvine, president of the GMC from 1995 until 2002, was instrumental in promoting the policy agenda of a standards-based re-licensure process.
According to Irvine and other GP policy leaders, the push to define and – later on, enforce – standards of practice ought to be recorded in the history of the Royal College of General Practitioners (RCGP) and general practice more broadly. Indeed, Irvine maintains that the College was formed as a result of the concerns of ‘conscientious general practitioners about the large tail of poor performance in general practice’:
In 1950 Joseph Collings, an Australian then working at Harvard, published a major study assessing the fitness for purpose of UK general practice in the new NHS. His findings, which showed a substantial proportion of awful practice, caused consternation inside the profession at the time. One important result was that thoughtful doctors, who were outraged by the damage the minority seemed to be doing to their branch of medicine, formed their own college - the College of General Practitioners – in 1952. The object was to set professional standards for this branch of practice for the first time. So the RCGP, unlike the specialist colleges, started life as a protest movement against the serious problem of poor practice. (Interview with Donald Irvine)
Thus, the history of standards-setting in medicine is inextricably linked with general practice politicians’ quest to improve public and professional perceptions of their discipline. While Irvine emphasises the poor performance of GPs as the motivating factor, it is also worth bearing in mind that GPs and their leaders are, as one respondent phrased it, the ‘poor cousins’ of consultants. Pursuing policies that established and attempted to enforce standards can equally be interpreted as a means of bolstering the profile and securing the legitimacy of general practice within medicine. As we will see later, specialist leaders did not take kindly to Irvine and his network's efforts to transfer this policy agenda into the rest of medicine.
Policy entrepreneurs and their networks in the aftermath of Bristol
In The Doctors’ Tale, Irvine characterises significant policy change, such as that entailed by the revalidation policy, as a necessary response to threats to professional self-regulation being made by politicians at the time. Importantly, the use of networks was a key strategy for negotiating the revalidation policy and the corresponding redefinition of professional self-regulation. Reconstructing the history of the development of the GMC's early proposals for revalidation, Irvine notes:
Julia Neuberger, a lay member, had met the Tory and Liberal Democrat shadow teams at the King's Fund within the past few days. There had been a very strong message that they – the politicians – were ready to pick a fight with the medical profession. Ergo the need for fast action by the GMC and the profession. This view was reinforced by the recent lunch Finlay Scott and I had had with Ann Widdecombe, the then Shadow Secretary of State for Health. She started off the conversation by saying ’One more like that [Bristol] and you can say goodbye to self-regulation’ (Irvine, Reference Irvine2003: 142).
As president of the GMC, Irvine attempted to take on the role of a coordinator of different medical interests, utilising the policy window afforded by the politicisation of medical self-governance to forward his policy agenda in general and the revalidation policy in particular. This involved activating existing networks and forging new ones with like-minded policymakers in the profession. For Irvine, it was essential that the Medical Royal Colleges play a significant role in developing a system of revalidation, given that the Colleges were ‘the repositories of standards in their specialities, and they would need to indicate good performance in their members and possibly all those on the Specialist Register.’ Thus, ‘the need to strengthen the links with specialist societies and associations because of their expertise in their clinical subjects’ was clear (Irvine Reference Irvine2003: 140). Not surprisingly then, a key nodal point in Irvine's emerging network was the then Chair of the Academy of Medical Royal Colleges (the Academy), Sir Roderick MacSween. Shortly after attending an informal dinner at the Academy, Irvine sent a letter to MacSween in June 1998, which later became known within professional regulatory circles as the ‘The Dear Roddy Letter’, where he outlined the case for revalidation. The support of the Academy was duly won and, according to Irvine, this letter ‘effectively fired the starting gun for revalidation’ (141).
As part of an emerging network mobilising for the revalidation policy, in 1999 the Academy issued its own consultation document on revalidation, which supported Irvine's and the GMC's general direction, presenting five different organisational models of how revalidation might be implemented (AMRC, 1999). By April 2000, discussions had progressed to the point where the Revalidation Leads Group of the AMRC (representative of all the College and Faculty interests) was able to issue the tactfully titled document How the Royal Colleges and Faculties might contribute to the process of revalidation, which was a joint statement establishing a common format for the Colleges’ input to revalidation and thus a common political position regarding the regulatory territory that should remain in the Colleges’ hands.
However, as political entities the individual Colleges maintain their historic autonomy. Indeed, as one respondent remarked, when it comes to enforcing its will on the Colleges, the Academy is ‘effete.’ Although a few of the Colleges began to tailor Good Medical Practice to their specialities as the Academy had encouraged them to do, several colleges were slow to act. Here again, the dynamics of the specialist/general practice division in medicine proved to be significant, as arguably one of the most supportive Colleges was the RCGP. The leaders of the RCGP became another crucial set of actors in Irvine's emerging network, particularly its Chairman from 1998 to 2001, Mike Pringle.
Pringle accounts for why RCGP policymakers were, on the whole, especially supportive of Irvine and the revalidation policy:
The RCGP had something very particular to worry about, which was that although our metric are less obvious – we don't have baby deaths as our outcome, or no one can go back, like they could with the pathologists, to look at smears […] – we perceived ourselves to have by far the greatest quality problem. Our perception at the time [after Bristol] was that the biggest vulnerability in standards and quality assurance lay in General Practice. As it turned out, the only case that has really proven to expose our vulnerability was Shipman. (Interview with Mike Pringle)
As a result of general practice politicians’ belief in the political vulnerability of their discipline, then, Pringle did not find it difficult to persuade his Council to follow the Academy's suggestion that the College support the GMC by producing standards and criteria specific to general practice. After extensive consultation with its members and with the BMA's General Practice Committee (GPC), Good Medical Practice for General Practitioners was drafted jointly with the GPC in 2000 and published in 2002.
In 2000, however, RCGP and GPC representatives were effectively ‘stood down’ (as one of our interviewees termed it) at a meeting of professional leaders, and were informed that the GMC was shifting its policy direction towards an appraisal-based revalidation policy. With this new policy, which was officially forwarded by the GMC in 2003, the Colleges were to play a less formal role with the standards they developed, occupying the position of suggestions or recommendations. In what follows, we explore how another set of networks largely composed of specialist politicians mobilised to redirect the GMC's revalidation policy.
‘And then peace broke out’: specialist and consultant networks and the appraisal policy
While Irvine was trying to garner support for the revalidation policy, a network comprising members of the BMA's Central Consultants and Specialists Committee (CCSC) and the profession's Joint Consultants Committee (JCC) utilised the occasion of discussions with policymakers in the DH about the new consultants’ contract to broker a new version of the revalidation policy. Although the CCSC did the actual brokering with the DH, the idea for appraisal as a ‘pragmatic’ alternative to the earlier, standards-based model of the revalidation policy was formulated in conjunction with the JCC. This version of the policy championed appraisal as the profession's response to the crisis generated by Bristol and similar events. With the appraisal-based revalidation policy, the performance of every doctor was to be assessed by a fellow doctor, typically a senior doctor or a clinical manager, according to the criteria set out by Good Medical Practice. In the earliest version of the policy, a doctor who receives a satisfactory appraisal five years in a row is automatically re-licensedFootnote 10. Following criticism regarding the policy's lack of rigour, a further criterion for maintaining the license was added: clinical leaders are required to state whether data collected through clinical governance procedures demonstrate any cause for concern about the quality of the doctor's performance, or, as it is known among policy leaders, ‘clinical governance sign-off’. This is interesting as it establishes a connection between the otherwise relatively separate governing regimes.
However, Good Medical Practice is a necessarily vague document given that it was created for medicine as a whole. This leaves room for a great deal of interpretation on the part of assessors. Thus, the appraisal-based revalidation policy is much ‘softer’ than the version of revalidation put forwarded by Irvine and his allies, who wanted to see performance assessed according to the explicit standards set out by the Colleges and with a substantial amount of lay participation in the process. While the appraisal-based policy does call for a measure of lay input, its architects were attempting to preserve the basis of medicine's historic autonomy: the idea that only doctors have the relevant expertise and skills to evaluate the performance of one another, and that therefore the profession must be allowed to govern itself. That specialist leaders were able to broker the policy with the DH demonstrates the capacity of the former to mobilise the scarce resource of its expertise in its dealings with the state. In so doing, the specialist policy brokers leveraged the support of the rank and file, many of whom had become concerned about how a process of re-licensure would affect their day-to-day lives and even potentially pose a threat to their livelihoods.
The appraisal-based revalidation policy reformulates medicine's relationship with the state. From the vantage point of a number of our interviewees, the DH's acceptance of the appraisal policy signalled a re-emergence of the historically close relationship between profession and state. More specifically, through the entrepreneurship and brokering of a network of specialist and consultant leaders in the BMA and JCC, the BMA appears to have re-established (if temporarily) its historic power and close ties with the government. This signified a change in the government's response to the pressures of the late 1990s: it was no longer explicitly criticising existing arrangements, but accepted the policies of the profession in exchange for a quid pro quo in the negotiation of the employment contract.
Despite the state–medicine consensus in 2003, the revalidation policy was re-politicised as a result of Dame Janet Smith's intense criticisms of the GMC and the appraisal-based revalidation policy in the Fifth Report of the Shipman Inquiry, which was released in December 2004. The policy was subsequently suspended by the Secretary of State for Health, John Reid, and the Chief Medical Officer was ordered to review the GMC. A White Paper was published in February 2007. It calls for an increase in the proportion of non-professional members on the GMC, for appointments to the Council to be made by the independent Appointments Commission, and for the size of council to be reduced (CMO, 2007: 25–27). In relation to revalidation, it calls for the inclusion of regular summative assessment of a doctor's performance based on standards set out by the GMC, a ‘revised system of NHS appraisal’, and concerns raised by patients or colleagues (33).
Following the White Paper, the Health and Social Care Bill was introduced in Parliament in November 2007 and became law in July 2008. The legislation contains a set of key measures that will initiate a new system for regulating doctors and allied health professionals. These include the requirement that the GMC and its non-medical counterparts adopt a civil rather than criminal standard of proof in fitness to practice cases; the creation of the Office of the Health Professions Adjudicator (OHPA), an independent adjudicator that will take over adjudication responsibilities from the GMC; and the establishment of the post of Responsible Officers, who are individuals appointed by healthcare organisations to identify and manage cases of poor medical performance and to make recommendations to the GMC for revalidation. Combined, these three measures signal a significant shift in the political landscape of professional regulation away from self-regulation to regulation by an arms-length organisation which answers to the government, and to the officers who answer to healthcare organisations. In short, when it comes to adjudication and the management of poor performance, the policy's intention is to ensure that professional regulation will no longer be the sole political domain of the profession. However, the implementation of the policy will depend on the cooperation of the profession, and it can be anticipated that there will be lengthy negotiations over detail.
The institutional context at work: institutional frameworks, the state–medicine relationship and the persistence of professional self-regulation
To understand the policy responses and new governance arrangements developed by the state and the profession, it is necessary to examine the institutional frameworks within which each operate, the dynamics of the relationship between them, and their respective relationships with the civil society. The organisation of the state in the health sector relies on a centralised system of command and control, and makes use of hierarchy-based forms of governance. Health policy is formulated in the centre, with regional and local organisations taking orders to implement policy. The clinical governance policy, together with the setting up of bodies like NICE and HC, follows this approach almost to the letter. Although an independent organisation, the establishment of NICE in particular took a good deal of the responsibility for designing clinical guidelines out of the hands of the profession. In terms of governance change, it is significant that the state extended hierarchy-based forms of governance to governing medical performance, which has traditionally been an area exclusive to professional self-regulation. In terms of the implementation of governance change, however, it has to be taken into account that hierarchy-based forms of governance are predicated on the expectation that NHS organisations (and, to a lesser extent, healthcare organisations in the private sector) would follow orders or face the consequences of not doing so. As was also the case with its internal market policy in the early 1990s, in creating policy the institutional structures of the state ignore the constraints imposed by medicine's entrenched power in the NHS and the private sector. These are the result not only of the profession's monopoly over the scarce political resources of medical expertise, but also of the difficulties faced by managers in implementing policies that are unpopular with the profession.
In contrast, the organisation of the medical profession operates through professional self-regulation, and as part of the negotiations of the revalidation policy the medical profession made extensive use of networks to bring about governance change. In substantive terms, the policy signals a redefinition of professional self-regulation with somewhat more explicit setting of standards for medical performance and associated procedures for evaluation, monitoring and sanctioning. While the policy chosen by the profession in 2003 has since been rejected by the state and replaced with a policy more akin to the version championed by Irvine, AMRC and RCGP, the fact that the revalidation policy originated with the profession – and that this occurred while professional self-regulation was under a sustained political attack – is testimony to the profession's ability to organise itself internally to address opposition.
The ‘triangle of forces’
A question remains: how did a series of cases of medical misconduct excite the DH into a fervent state of policymaking and institution-building? How did it lead to the profession accepting proposals to subject practitioners’ licenses to regular review? Addressing this issue requires a closer examination of the dynamic of the relationship between the state and medicine, which is embedded in the institutional structures of each. It also means considering their respective relationships with civil society. What has been termed the ‘triangle of forces’ constituting these relationships sheds considerable light on why professional self-regulation became as politicised and vulnerable to change from without as it did in the late 1990s (Salter, Reference Salter2004).
With the 1948 NHS Act, access to quality medical care became a right of citizenship. The state was henceforth required to meet the medical and healthcare ‘needs’ of its citizens. However, the limits to what constitutes ‘need’ were not specified, beyond the notion of treatments or services deemed by healthcare personnel to be ‘medically necessary.’ Moreover, explicit forms of rationing were not acceptable because there would be high political costs to pay. The state thus depends on the medical profession to meet the healthcare ‘needs’ of its citizen and to undertake covert forms of rationing (e.g., by constructing waiting lists or controlling the range of services defined as medically necessary). In exchange for meeting this demand, the profession was granted the right to keep its system of self-regulation and its control over the education, certification and licensing of its members. As a result of this regulation, civil society maintained its trust in the medical profession.
With cases like Bristol, Alder Hey and Shipman, the triangle of forces underpinning the relationship between the state, medicine and civil society became destabilised. Although only temporarily, the state could no longer depend on its citizens’ belief in the legitimacy of medicine's self-regulating institutions. Not only was there growing concern about the safety and quality of medical performance, but the legitimacy of the state's trade-off with medicine – professional self-regulation – also became the target of intense scrutiny. The activation of the civil society portion of the triangle made it desirable (if not politically necessary) for the state to act by introducing clinical governance arrangements.
However, once civil society discontent had stabilised and the failings of self-regulation were no longer headline news, the state was obliged to recognise that the situation it had faced since the inception of the NHS had not changed: it must still negotiate with medicine to meet citizens’ healthcare rights. The contract negotiations with the BMA in 2002 and 2003 illuminate this scenario clearly. In England and Wales, consultants rejected an earlier version of the contract on the grounds that it gave too many new powers to managers, thereby reducing practitioners’ clinical autonomy (Salter, Reference Salter2004). The DH was therefore forced to renegotiate the terms of the contract. The version that was eventually approved did not include such powers, but it did include an improved pay package and the state's approval of the appraisal-based revalidation policy. Thus, what initially appears as a set of conjectural factors (that is, cases of malpractice and the media's attention to them) can be explained via the contract between medicine and state, a contract that is embedded in their respective institutions.
Critical reflections: what the institutional context cannot explain
One interesting finding from the research undertaken for this case study was the degree to which the profession is divided. At times, ‘networks at war’ made for a more accurate metaphor than the type of strong unity traditionally associated with a profession. The 2003 version of the revalidation policy squeaked through, but things could have unfolded otherwise. And not too long afterwards they did: the Fifth Report of the Shipman Inquiry put the Government once again in the position of having to act, in this case to suspend the policy. The result has been the re-birth of the type of revalidation policy – a standards-based approach – advocated by Irvine and supported by AMRC and RCGP. What Irvine and others may not have counted on, however, was the creation of yet another arms-length regulatory body to take over the powers that were previously exercised by the GMC.
The analysis of the organisation of professional self-regulation offered in this article accepts the principle of its diversity, but cannot completely account for why the profession appears to no longer unite when it would seem to be in its collective interest to do so. It is possible that divisions centred around function (learned society, licensor, trade union, etc.) and speciality have taken on increased significance in the context of state policies that make the governance of medical performance an area for statutory regulationFootnote 11. Moreover, the divided state of the profession is perhaps part of the reason why the government was able to introduce the current reforms to the GMC and revalidation: it was not confronted by a unified opposition. To an extent it can be argued that the state's entrance into areas formerly regulated by the profession has succeeded in a strategy of divide, though not conquer.
As a result of both the clinical governance and revalidation policies, a growing source of division may be between the rank and file practitioners and the medical elites who control self-regulation. As the literatures on the rise of scientific–bureaucratic medicine (Harrison et al., Reference Harrison and Dowswell2002) and evidence-based medicine (e.g. Coburn et al., Reference Coburn, Susan and Ivy1997; Rappolt, Reference Rappolt1997; Rodwin, Reference Rodwin2001) suggest, the longer-term consequences of clinical governance arrangements remain opaque, particularly their effects on the working conditions and the clinical autonomy of rank and file practitionersFootnote 12. What is clear is that the institutional and normative landscape of medical governance has changed. In terms of broader-scale governance arrangements and the state–medicine relationship, it is uncertain whether placing the expertise of professional and academic bodies at the disposal of state bureaucracies, and thus managers, will slowly erode professional self-regulation. Such a shift, not the least, relies on the effective implementation of the clinical governance policy at the regional and local levels, where the centralised DH and NHS bureaucracies have had very uneven levels of success in the past, especially in the face of medical opposition. The state's experience with the internal market policy of the late 1980s and 1990s is a case in point.
Conclusions
Clinical standards are a key dimension of the control of medical knowledge. Bodies that are external to the universe of professional self-regulation have entered the clinic in new and important ways, charged with the responsibility of ensuring that practitioners implement guidelines. In this way, one may argue that medical expertise has been placed at the disposal of state bureaucracy, to be used by the latter to reward and punish as it sees fit. However, the influence of hierarchy-based forms of governance is limited by the absence of a clear system for sanctioning poor-performing organisations and by its reliance on organised medicine to provide it with the knowledge and know-how to create clinical guidelines in the first place. Furthermore, the model of centralised decision-making with implementation via a lengthy chain of command is particularly vulnerable in the face of professional resistance.
What is perhaps surprising is that none of the medical politicians and policymakers interviewed for this study viewed the clinical governance policy or organisations like NICE and HC as posing a challenge to professional or clinical autonomy. Rather, clinical governance was interpreted as part of an attempt to improve the quality of care and clarify the channels of accountability in NHS organisations. Respondents perceived little to no incursion into the domain of professional self-regulation, which they saw as being concerned with the education, certification and licensing of individual practitioners. However, interviews were conducted prior to the 2007 White Paper. The BMA's response to the introduction of the Health and Social Care Bill suggests an increasingly disillusioned profession (BMA, 2007), although this feeling is unlikely to be shared by networks in the profession calling for a standards-based approach to revalidation.
