SEARCH

SEARCH BY CITATION

Keywords:

  • National Service Frameworks;
  • general practice;
  • street-level bureaucrats

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

This paper argues that the past decade has seen significant changes in the nature of medical work in general practice in the UK. Increasing pressure to use normative clinical guidelines and the move towards explicit quantitative measures of performance together have the potential to alter the way in which health care is delivered to patients. Whilst it is possible to view these developments from the well-established sociological perspectives of deprofessionalisation and proletarianisation, this paper takes a view of general practice as work, and uses the ideas of Lipsky to analyse practice-level responses to some of these changes.

In addition to evidence-based clinical guidelines, National Service Frameworks, introduced by the UK government in 1997, also specify detailed models of service provision that health care providers are expected to follow. As part of a larger study examining the impact of National Service Frameworks in general practice, the response of three practices to the first four NSFs were explored. The failure of NSFs to make a significant impact is compared to the practices’ positive responses to purely clinical guidelines such as those developed by the British Hypertension Society. Lipsky's concept of public service workers as ‘street-level bureaucrats’ is discussed and used as a framework within which to view these findings.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

The National Health Service (NHS) was established in the UK in 1949. At this time, primary health care was provided by general practitioners (GPs) who usually worked alone, often from their own homes, with minimal administrative support. The subsequent five decades have seen enormous changes both in the organisational context in which GPs work and in the work that they are expected to do. From the beginning of the NHS everyone was entitled to be registered with a GP, and, as well as providing first-line care, these doctors acted as gate-keepers to the hospital sector, a role that they still retain. Rather than being employed by the health service, GPs acted as ‘independent contractors’, providing services to the NHS according to a contract. This contract was initially very simple, requiring a GP to ‘render to his [sic] patients all necessary personal medical services of the type usually provided by general practitioners’ (British Medical Association 1966); payment was largely according to numbers of patients served. From 1949 until the early 1960s, general practice was the ‘poor relation’ of the NHS, with little prestige and poor premises. The threat of mass resignations in the 1960s resulted in a new ‘charter’ for general practice and much-needed investment (Handysides 1994). Practice premises were improved, groups of doctors were encouraged to join together into group practices and employ ancillary staff (Laing et al. 1997) and the College of General Practitioners obtained its Royal Charter, providing the profession with increased academic respectability. Through the 1970s and 1980s practices became more complex, with the employment of practice nurses, the purchase of computers and the setting up of screening programmes for blood pressure and child development (Baker and Thompson 1995). GPs and practice nurses worked together with their community colleagues to form ‘primary healthcare teams’.

In 1990 the then Conservative government imposed a new contract on a reluctant profession (Lewis and Gillam 2002) and the trend towards increasing complexity accelerated, as practice income was made partially dependent on the provision of services such as cervical screening, childhood immunisation and the screening of well patients rather than simply being based upon the number of patients served (Handysides 1994). In the 1990s ‘fundholding’ was introduced; this voluntary scheme gave practices a notional budget with which to purchase a range of services for their patients. Whilst few clear benefits resulted from this initiative (Glennerster et al. 1994, Harrison and Choudhry 1996), it was popular with GPs, largely because of the flexibility that allowed any savings made to be reinvested in the practice (Petchey 1996). Many fundholding practices employed managers from outside the health service to negotiate their contracts with hospitals and other provider units (Westland et al. 1996), and this influx of professional managers (who were seen as having a higher status than colleagues who had been promoted from within general practice) resulted in the development of more sophisticated management structures within the practices (Laing et al. 1998). In 1998 the new Labour government reorganised the NHS. Key changes were the abolition of fundholding and the setting up of primary care organisations (PCOs) including Primary Care Trusts (PCTs) (Department of Health 1997). These organisations are charged with commissioning services from both primary and secondary care providers for the populations they serve (each PCT covers a population of approximately 100,000 to 300,000). Their boards are made up of representatives of GPs, nurses, pharmacists and other stakeholders, including users. Currently most GPs remain self-employed, with anything from 50 to more than 100 GPs contracting with each PCT to provide primary care services.

Parallel with these organisational changes have come profound changes in the nature of medical work as performed by GPs and practice nurses. This started in a small way with the 1990 ‘new contract’, which required GPs to perform screening measures on groups of patients in order to earn a proportion of their income. Initially this involved payment for performing screening medicals on new patients and elders, and payments for the provision of so-called ‘health promotion clinics’. It was envisaged that patients with chronic diseases such as hypertension and diabetes would receive care in these clinics. Whilst the health promotion clinic idea cannot be judged a success (Glendinning et al. 1994), this initiative might be said to mark the beginning of official encouragement to deliver proactive care to patients according to their disease rather than in response to the patient's own decision to attend the practice. Thus, whilst 20 years ago the majority of activity within a practice consisted of patient-initiated contacts in response to their own perception of their need, most modern general practices, as well as providing appointments for those who ask for them, provide a parallel service of ongoing and preventative care to patients according to a programme. Much of this care is doctor (or nurse) initiated and tends to be referred to by GPs and practice nurses as ‘chronic disease management’. This change is significant because the demands associated with providing these two kinds of service are very different, and the need for what is sometimes called ‘acute’ or ‘emergency’ care has not diminished as the provision of chronic disease management has increased. Without increases in resources it is difficult for these conflicting demands to be met. In 2003 another ‘new contract’ for UK GPs was agreed that might address some of these issues (Shekelle 2003).

This move towards increased systematic care of patients with chronic diseases has been partially driven by more general changes in the nature of medical work. The work of Cochrane (1999) advocating the application of evidence from randomised controlled trials to the practice of medicine has been extended into the area of chronic disease management (Benech et al. 1996). Harrison (2002) has called this translation of evidence from randomised controlled trials into sets of normative guidelines that rapidly become the accepted ‘gold standard’ of care ‘scientific-bureaucratic’ medicine. National Service Frameworks were introduced by the new Labour government in 1997, and go further than simply providing guidelines about appropriate clinical care for individual patients; in addition to clinical guidelines, they also supply detailed preferred ‘service models’ (Department of Health 1998b) that specify how and by whom that care should be delivered. In the absence of good evidence about appropriate models of service, these recommendations are based upon expert opinion, and are directed at hospitals as well as primary care and other health care providers such as Ambulance Trusts. Four NSFs have been published so far, covering mental health, heart disease, diabetes and the care of elderly people, with (at the time of writing) two more expected for children's services and renal disease. Whilst NSF implementation is not yet clearly a compulsory requirement for GPs, it is an explicit duty on Primary Care Trusts, and it is clearly within the terms of reference of the inspection body, the Commission for Health Improvement (CHI), to report upon the extent to which NSFs are being implemented (Department of Health 1998a). Thus, whilst a decade ago the movement towards the care of patients with chronic diseases in an organised fashion was patchy, with some practices much further ahead than others, there is now a clear presumption from central government that this will become the norm. The new contract for GPs agreed in 2003 institutionalises this change further, with a significant proportion of GP income being dependent upon the attainment of a large number of detailed quality indicators that embody the concept of organised care of those with chronic disease according to normative guidelines.

These changes embody a further departure for general practice. GPs often define themselves and their role (at least rhetorically) in terms of holistic care of the individual, taking into account social and psychological factors as well as physical ones (Baker 1997). The current movement towards programmed care of patients according to their disease category cuts across this commitment to the individual. The ‘scientific-bureaucratic’ approach embodied in NSFs and guidelines is more closely identified with a clinical epidemiology or public health approach than traditional general practice. Thus, it can be seen that within general practice there is a move away from the traditional approach of treating patients who walk through the door, towards a clinic-based approach to groups of patients. This change mirrors current changes in the wider NHS, with the rise of scientific-bureaucratic medicine and a public health or population-based concept of health and disease.

Much of the published analysis of the changes mentioned above has taken as its starting point the concept of GPs and other categories of physician as ‘professionals’, seeking to mitigate the negative effects on their professional status and clinical freedom (Sheaff et al. 2002, Mahmood 2001, North and Peckham 2001). Whilst these pressures and concerns undoubtedly exist, there are, as I have argued elsewhere (Checkland 2003), other ways of conceptualising the situation. One such alternative approach is that advocated by Hoff (2001), who argues that as well as being professionals, doctors are workers who, like other workers, are ‘active agents, negotiating the terms of their daily existence in unique ways across a variety of work situations’. The advantage of this approach is that it focuses attention on the way that physicians struggle to infuse their work with meaning whilst constructing understandings that ‘normalise’ their lives as workers and allow them to cope with uncertainty in their surroundings (Hoff 2001: 54). This in turn shifts attention away from profession-wide concerns towards individuals and their unique work situations. Research questions can then be constructed and explored that may have the potential to explain observed phenomena that do not fit closely with the broad-brush concept of relentless deprofessionalisation resisted by the profession (Dowswell et al. 2002). As Armstrong (2002) concludes in a study of the reality of decision-making by GPs in the treatment of depression:

whilst the imposition of a ‘rational’ basis for clinical autonomy might satisfy the requirements of the profession as a collective, it must struggle against a patient-centred rhetoric and a process of cognitive transformation that underpins everyday activity and the claimed autonomy of individual clinicians. This latter form of clinical autonomy is less a volitional political stance to defend occupational privileges and more the practical mechanism through which medicine is delivered to patients (2002: 1776).

This paper takes the idea of the ‘practical mechanism[s] through which medicine is delivered to patients’ and seeks to use it to understand the impact of National Service Frameworks and normative guidelines on practice. In a book examining what he calls the ‘back-regions’ of medical institutions, Atkinson (1995: 31) argues that this focus on the micro-level of medicine as work is one that has been little pursued by sociologists. The work of Lipsky (1980), identifying public service workers as ‘street-level bureaucrats’, is one theoretical approach that has potential to illuminate some of these micro-level concerns.

Street-level bureaucrats

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

Lipsky defines street-level bureaucrats as ‘public service workers who interact directly with citizens in the course of their jobs, and who have substantial discretion in the execution of their work’ (Lipsky 1980: 3). Although in his list of ‘typical’ street-level bureaucrats in the US Lipsky does include health workers, and some of his work and that of his collaborators focuses on doctors within Emergency Rooms and Veterans’ Administration Hospitals, much of his writing concentrates on those working in social services, the police and the education services. In Lipsky's analysis, the characteristics shared by street-level bureaucrats include: a focus on the need to ‘process workloads expeditiously’ (1980: 18); substantial autonomy in their individual interactions with clients, and an interest in maintaining and maximising that autonomy; conditions of work that include inadequate resources (both monetary and in terms of personnel and time), demand that will always exceed supply, ambiguous and multiple objectives, difficulties in defining or measuring good performance, a requirement that decisions should be taken rapidly and clients who are what Lipsky calls ‘non-voluntary’ (1980: 56) –i.e. they have limited (or non-existent) choice over whether, where or how they present to the service involved. From this list it can be seen how GPs might fit into this definition. With only 10 minutes allocated per patient (and in some practices less than this), GPs will always need to ‘process’ their clients quickly and efficiently. This need will be strengthened by the increase in general workload implied by the rise in clinic-based care of patients with chronic disease alongside the traditional care of patients who walk through the door. Although both individual and practice-level outside scrutiny of GPs’ work is increasing, they retain considerable autonomy in their one-to-one interactions with patients in consultations. Resources in the NHS are always limited, and the demands of modern information management have placed a considerable further strain on practices’ administrative staff. Demand for appointments has been rising inexorably over many years and this has resulted in the development of innovative ways of responding (Gillam and Pencheon 1998). GPs ‘objectives’ will always be ambiguous in a system where cash-limitations result in the need for intrinsic rationing by those with the most direct contact with clients (Harrison and Moran 2000, Heginbotham 1997), and although there is a move towards explicit performance measurement (Department of Health 1998c), definitions of ‘quality’ remain in dispute (Loughlin 2000). Finally, GP clients are ‘non-voluntary’ in the sense that, although there are isolated areas (mainly in London) where private general practice exists, in most areas there are no alternatives to NHS GPs. Patients do have a choice between GPs, but this is not a choice that is often exercised after initial registration. In addition, GPs act as ‘gatekeepers’ for the secondary care sector, with access to consultants, both NHS and private, available mainly by referral from GPs.

Whilst arguing that GPs do fulfil many of Lipsky's criteria for street-level bureaucracy, it must be acknowledged that there are some limitations to this analysis; particularly the fact that, rather than being directly employed, GPs at present remain, for the most part, independent contractors to the NHS. However, Lipsky's use of the word bureaucrat does not imply a Weberian definition of bureaucracy as a hierarchy of formal authority governed by rules, and it is not self-evident that the distinction between bureaucrats and non-bureaucrats should rest simply on formal employment status. Although as ‘independent contractors’ GPs do have more freedom to set the terms and conditions of their work than their employed colleagues, there are practical limitations to this freedom. For example, whilst it is technically possible to lengthen appointments from 10 minutes, levels of demand make this very difficult to achieve in practice. Lipsky himself defines street-level bureaucrats in terms of the characteristics of their work situations – particularly the autonomy they enjoy in face-to-face contacts with the public and the limitations inherent in working in a situation of limited resources – rather than their employment status. It is in the pressures they face as workers that GPs are defined here as meeting Lipsky's definition of street-level bureaucrats.

What are the consequences of these pressures on the work of street-level bureaucrats? Lipsky argues first, that ‘the decisions of street-level bureaucrats, the routines they establish, and the devices they invent to cope with uncertainties and work pressures effectively become the public policies they carry out’ (1980: xii). In her study of decision-making in a UK Health Authority, McDonald (2002) identifies this process occurring as GPs make practical decisions about the management of patients with coronary heart disease that may be at odds with stated local policy. Lipsky also argues that street-level bureaucrats respond to the pressures on them by processing people in a routine and stereotyped way, and they ‘develop conceptions of their work and of their clients that narrow the gap between their personal and work limitations and the service ideal’ (1980: xii). Finally, he argues that in order to cope with the demands of their work situations, street-level bureaucrats will inevitably ration their services, either explicitly or by using such devices as waiting times, gate keepers (receptionists?) and psychological costs associated with seeking help (1980: 87).

The results reported here form part of a larger study investigating the impact of the changes in medical work represented by National Service Frameworks and the new GP contract in general practice. In this paper, the response to the guidelines and service models in NSFs is compared to the response to purely clinical guidelines such as those published by the British Hypertension Society. The use of Lipsky's ideas in interpreting these findings will be discussed.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

Qualitative case studies were undertaken in three practices in an area covered by a single Primary Care Trust in England. Semi-structured interviews of doctors, nurses, practice managers and other staff with managerial or clinical responsibilities (for example health care assistants, data recording clerks and office managers) were undertaken, alongside observation of both set-piece meetings and other incidental practice activity. Documentation relating to practice organisation, audit data and papers from meetings were collected. In keeping with the overall theoretical approach, practice staff were viewed as ‘workers’ in an organisation. A framework for understanding and interrogating the nature of an organisation was derived from the organisational studies literature1. In this context interview questions were directed at understanding how the staff viewed their and their colleagues’ roles (Katz and Kahn 1978), what underlying norms and values could be detected against which judgements about new initiatives were made (Vickers 1995) and how the staff collectively ‘made sense’ (Weick 1979) of what they were doing, as well as more straightforward questions about their responses to the National Service Frameworks (see Box 1). The author undertook all of the interviews and observations, and as a fellow-GP (albeit in a different area), attained a degree of ‘insider’ status (Adler and Adler 1998: 85) that may not have been available to a non-medically qualified researcher. The study received Local Research Ethics Committee approval.

Table Box 1 . Areas covered in interview questions
• The organisation:
 • The formal and informal roles that are being filled in this organisation.
 • The organisational expectations of behaviour in those roles.
 • The values that underpin both behaviour in role and expectations about that behaviour. Questions such as ‘What is the best/worst thing about your practice? What is the most important part of being a GP to you? What metaphor would you use to describe your practice?’ were used to elicit information about this.
• National Service Frameworks
 • Attitudes to the idea of central models of service
 • Attitudes to the reality of NSFs
 • Activity within the practice directed towards their implementation

Practices were theoretically sampled (Strauss and Corbin 1990). First, practices that had a documented track record of being actively involved in developments in the NHS (for example fundholding, local quality improvement schemes and groups undertaking medical audit work) were chosen to allow comparison of reaction to this new type of innovation (involving significant top-down pressure to conform) with responses to previous changes in the NHS. Secondly, practices within a single English PCT area were chosen for this part of the study in order to eliminate any differences due to differences in approach to NSF implementation between PCTs. Finally, practices with differing degrees of involvement with the current ‘reform’ agenda were chosen. Thus, in one practice a GP was currently an active participant in the local PCT. In a second practice there had been some PCT involvement, but this had recently ceased. In the third practice there was no engagement. None of the managers or nurses were actively involved in any PCT initiatives or committees. Fieldwork took place between March 2002 and March 2003. Data collection took place over approximately a 6–8 week period in each practice and an average of 15 contacts made with each practice. In each practice at least one clinical session was undertaken as a locum. As well as recompensing the practice for the time taken in interviews etc, this allowed the researcher to gain a more in-depth understanding of the practice organisation, patient population and approach. The interviews were transcribed and analysed thematically (Miles and Huberman 1994) using the computerised qualitative data analysis package ATLAS.ti (© 1997–2003 Scientific Software Development, Berlin). Using the framework derived from the organisational studies literature, these data were then further analysed alongside detailed field notes from observation in the practice and any available documentation to develop an internally-coherent set of explanations for the observed behaviour relating to National Service Framework implementation in each practice. These explanations were expressed in terms of the observed roles occupied within the practice, the inferable underlying standards against which practice activity and external events were judged, any identifiable important ‘relationships’ that practice activity tended to work to maintain (a special case of which would be expressed practice goals) and the collective sense that participants made of their worlds (Vickers 1995, Weick 1995). In one practice, for example, being and remaining a ‘small practice’ was one such important relationship. This was closely related to an expressed need to be closely in control of practice activity, and was manifest in the appointment of an administrator rather than a manager; the doctors believed they should do the management of the practice themselves. This need to be in control was also expressed in their attitudes to centralised initiatives such as NSFs, with negative comparisons being made with previous local, voluntary schemes to improve quality of care. The three case studies were then compared, and recurring issues identified. Data from three practices relating to their responses to NSFs and guidelines are presented here. These practices are all inner-city practices, covering ethnically and socially-diverse populations. They are medium sized, with patient numbers between 5,000 and 6,700, and are in an area covered by a single PCT. Approximately 30 per cent of their patients attract deprivation payments.

Principal findings

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

In keeping with case study methodology (Yin 2003: 50), the cases were initially analysed separately. Cross-case analysis was then performed, comparing responses across the cases. It was found that, whilst the three practices differed in many organisational aspects, a similar pattern of response to National Service Frameworks and guidelines was present. In particular, all three practices had reacted positively to the general move towards the incorporation of normative guidelines in clinical practice, whilst failing to respond proactively to National Service Frameworks. No differences in reactions to NSFs and guidelines were found between the practice with an active involvement with the local PCT and the others, suggesting that these findings were not a product of more general attitudes to the overall agenda of change.

Attitudes to and use of guidelines

In all three practices, participants spoke positively about the idea of clinical guidelines. The practical help in managing patients afforded by good guidelines was emphasised by both doctors and nurses. These quotes illustrate this:

Because it might be that some of these things are fairly lower order in terms of activities so that [if] it's a system that works we’ll do it, like the computer, if I press this I know this comes up on the screen, I don't have to think about it, or maybe seeing things in that way is better and it releases me to do something else more exciting . . . (GP2).

It's good to have a guideline to work from, or like the British Hypertension Guideline or something like that, you know exactly, you've got the flow diagram and you know what you should be dealing with . . . (GP4).

So I thought, no, I didn't have any problems, actually you know at the end of the day our clinical decisions are very hard, and the more things we have that help us make a clinical decision easily, such as the guidelines, actually make it easier, the consultation (GP8).

And [a guideline] makes it easier because it's sort of more tangible, you know what road to go down with somebody rather than not knowing what to do next . . . (PN4).

So I think it's good that we've actually got, you know, some guidelines, you know, on what to do and how to implement it and things like that (PN2).

In addition to this evidence of an expressed positive stance towards evidence-based guidelines that were regarded as being of practical value, all three practices demonstrated evidence of behaviour indicating a positive attitude to guidelines (see Table 1):

Table 1. Practical use of guidelines in study practices
PracticeEvidence
1Use of British Hypertension Society guidelines as baseline for a recent audit.
Use of British Thoracic Society asthma guidelines in nurse-run clinic.
2One partner previously worked with the local Medical Audit Advisory Group developing guidelines
Involvement in the development of a local quality mark scheme, one strand of which required demonstration of the use of guidelines in practice. The practice was one of the first to achieve this quality mark.
3One partner previously worked with the local Medical Audit Advisory Group developing guidelines.
Use of local diabetes-centre guidelines as baseline for recent audit.

Participants in two practices did sound a note of caution. Whilst happy to be working with guidelines they drew attention to the fact that some of the guidelines that reach primary care were not practically useful:

Ninety nine percent of guidelines passed down to primary care are never used, because they are unworkable or impractical, or they only deal with the first time a patient comes in (GP3).

Another GP pointed out that guidelines sometimes failed to take into account the needs of deprived populations:

But obviously, one of the difficulties is [guidelines] doesn't take into account the area or the population that we have. And that sort of rigid arrangement doesn't always necessarily work particularly well for the sort of clients that we have (GP9).

Attitudes to National Service Frameworks

Participants were asked about both their attitude to the concept of National Service Frameworks and their reaction to the reality of them when they first arrived in practice. In keeping with the generally positive attitude of these health professionals to the idea of clinical guidelines, NSFs as an idea were generally accepted. These quotes from members of all three practices illustrate this:

The idea is excellent, because it means that we know what we should be doing instead of reading the literature and saying ‘Well, which of these bits of information do we as an individual practice take on board and what don't we?’ (GP3).

Well I think it is good to have an idea of what people, who have had time to think about it, think we should be doing, a model of care (GP7).

I, well I suppose, you know, I think in essence you know, it's good to try and standardise care of patients with coronary heart disease or with diabetes. So you know, I'm in favour of it from that point of view. And ultimately I think we are, we do need to have quality measurements and prove that we are actually looking after people in a positive way. And a beneficial way that can be measured (GP10).

However, asked about reactions to the actual NSFs when they appeared in practice, attitudes were more negative, and concentrated on the failure of NSFs to offer practical help:

. . . but I mean when you get into the subtext, I mean some of it sounds like these very glorious. erm, edict's, doesn't it? You know what's our mission statement? You know we’ll do a better job for everybody in the universe. Fine, okay, right, how are you going to do that? Give me the details? (GP1).

I think we, we, our approach to audit has tended to be, to do simple but easy to repeat audits, like say blood pressure, because then they’re a marker for other things, aren't they? If you’re getting the blood pressure right, it shows that you’re thinking about the patient. That gives you an even chance of thinking about other things to do with their health. The NSF requires this, you know, 20-page audit, I mean have you seen it? You've seen, well you must have the same ischaemic heart disease template that we've got, it's got six pages on it, you know, it's a complete nightmare (GP6).

Against that, I mean it doesn't take into account how we, you know the other side of it which is consultation[s] and you know the things that you can't measure in a sense (GP8).

Thus, whilst having no ideological objections to the concept of national models of practice (contrary to what might be expected if they were taking a stance based upon the importance of maintaining professional autonomy), these GPs felt that the actual documents were not helpful in the practical business of seeing patients – and in some cases, were actually making life more difficult.

Practical responses to National Service Frameworks

The most striking aspect of the practices’ practical responses to NSFs was the fact that, whatever their overall attitudes to the concept of National Service Frameworks, none of these practices had a plan for their systematic evaluation and implementation. Whether or not any part of the documents had been read was generally a matter of personal choice. In one practice, there was an informal agreement that doctors would look at those areas of clinical practice for which they took the lead and ‘cascade’ this information to their colleagues. In spite of this informal agreement, there was no evidence of any systematic attempts to follow it up. Table 2 shows how many people in each practice reported that they had read any part of the NSFs, expressed as a proportion of the practice personnel including doctors, nurses and managers. Managers are included here because whilst the adoption of purely clinical guidelines might be said not to be the responsibility of managers, the implementation of NSFs requires considerable co-ordination across the practice, and has significant implications for administration. Thus, in a practice formally implementing NSFs, the manager would need to be involved at an early stage.

Table 2. Proportion of staff in each practice who had read all or part of the NSFs
PracticeMental healthCHDOlder peopleDiabetes
SumryFull doc.SumryFull doc.SumryFull doc.SumryFull doc.
Pr 11/71/7 5/70/70/70/72/71/7
Pr 22/50/5 3/51/50/50/51/50/5
Pr 30/60/6 3/61/62/60/64/60/6
Total3/181/1811/182/182/180/187/181/18

Thus in three practices with a track record of both positive attitudes to the idea of guidelines and practical evidence of their use, and a generally positive approach to the idea of National Service Frameworks, very few people had read even the summary versions. The only two documents that had been looked at by an appreciable number of people across all three practices were the summary versions of the NSFs for CHD and diabetes. The NSFs for mental health and older people come out particularly badly, with even the summaries receiving little attention. Whilst in theory it might be desirable for one staff member to read each document and cascade the information to their colleagues, there was no evidence of this happening, with no formal meetings to discuss NSFs having taken place at the time of the fieldwork.

Explanations for these findings were fairly consistent across all three practices (Table 3).

Table 3. Explanations for failure to read NSFs
ExplanationNumber of citations
NSFs too large/staff too busy8
NSFs too detailed and complex1
NSFs coming too rapidly1
No sense of imperative, and no clear idea of who is responsible5
Flow of information within practice confused – where are the NSF documents?7
‘We are doing it already and so don't need to look at it in detail’5
Particular NSF not relevant1
Unaware of their publication dates1
Not interested1
NSFs not primary care focused1

From this it can be seen that there are three broad groups of explanations offered. The first relates to the characteristics of the NSFs themselves. The size of the documents was mentioned frequently and was a significant factor influencing responses. In addition, the complexity of the documents was felt to be a disincentive to reading them, and it was commented that they were being published too fast. These quotes illustrate these concerns:

I don't think anyone reads the full documents. I think we all have good intentions and have a look at it, but I think basically we look at summaries of it, because it's just impossible to get through everything (PN1).

They’re too big, they’re too detailed, they’re too prescriptive, and they don't pay attention to the causes of the problems (GP6).

There's too many, too fast and too complicated. No way of fitting it in or just using it, nothing (GP8).

The second group of explanations relates to the way the practices were organised. Although the members of one practice informally took responsibility for different clinical areas, even in this practice there was no clear mechanism whereby new initiatives such as NSFs would be dealt with when they arrived in the practice. Thus in all three practices, staff commented that they didn't know where all the documents were, or whose responsibility it was to look at them. In none of these practices was there a strong manager who was able to co-ordinate and direct practice activity. This confusion is illustrated by this GP:

Umm I've only, sort of, flicked through the . . . the Ischaemic Heart Disease one, because I was doing an assignment for my course, it involved discussing something like that – so I actually got it off the Internet – so I'm not sure whether I was actually ever sent a copy and chucked it in the bin, or did something with it, but the diabetes one, I think, is out, but I don't recall ever touching it – I think from . . . [GP1] was talking to somebody about it recently, I think it's in the practice library, but I don't know how it got there, so I don't know if it came to him as the lead partner, and then he put it in the library or if it comes to the practice manager, I really don't know (GP4).

As one doctor neatly summed it up:

But if anyone thinks that things arrive here, somebody with the ability to look at it properly has the time to look at it and then spread it as useful information that everyone else thinks sensibly about, they've got another think coming (GP8).

What seems to be lacking here is any kind of clear pathway whereby information or initiatives that arrive at the practice are dealt with in a systematic way. This has been referred to as the ‘information process’ and has been identified as a significant problem by other authors (Dowswell et al. 2001).

Thirdly, as practices that saw themselves as being conscientious and already striving to provide a high standard of care, there was a feeling that ‘NSFs are not directed at us’ and therefore did not need to be looked at in detail. In the practice where only the odd person had looked at more than just the NSF for CHD, the following comment was made in answer to a question about the respondent's initial reaction to NSFs following their publication:

But then you look at it a bit more closely and think, ‘Well, we kind of do those, and that one's not relevant to me. . . . and, oh, you know we’ll have to beef the things up a little bit and do, and move towards that’ (GP1).

It is the first group of explanations that will be concentrated upon here.

Changes in response to NSFs

All participants were asked to name any aspect of their practice that they felt had changed as a result of National Service Frameworks (see Table 4).

Table 4. Changes attributed to NSFs in practice
PracticeChangeNumber of citations
Practice 1Update of CHD register3
Delegation of CHD secondary prevention to nurse1
Nothing2
Practice 2Better coronary heart disease audits2
Involvement in PCT heart failure project1
Nothing1
Practice 3Review of diabetic protocol1
Development of call and recall for patients with CHD2
Development of CHD register2
Development of diabetes register1
Redirection of practice priorities towards the care of patients with chronic disease1

Staff were by no means in agreement as to the cause of these changes. For example, whilst in one practice two staff members commented that they thought the recent updating of the coronary heart disease register had occurred as a result of the NSF, others disagreed, saying that this work had predated the NSF and had been influenced by a local pre-existing project co-ordinated by the local hospital. It is interesting that no one in any of the practices identified anything to do with mental health or the care of elderly people as having changed as a result of the NSFs. This differential implementation between NSFs has been commented upon by others (Rogers et al. 2002). One study practice was in the process of developing a proposal for a Personal Medical Services (PMS) pilot. This scheme was introduced by the Conservative government in 1996 and involves the contracting by practices with the local primary care organisation to provide a defined set of services over and above those normally provided by GPs. These might include services for drug addicts, older people or the provision of investigations more usually undertaken in hospitals, such as ultrasound scanning or endoscopies. The doctors in this practice intended to use their pilot to improve the care of elderly patients. When asked, the doctors specifically denied that the NSF for older people had influenced their work on this, and they were keen to identify the origins of the scheme in their own concerns about the quality of care these patients received.

There was one exception to this picture of limited and rather haphazard implementation of National Service Frameworks. One of the practices studied was, at the time of the fieldwork, suffering a workload crisis. They described themselves as being ‘close to breaking point’ following a large influx of new patients (a 25 per cent rise in their list size in the last year) and a series of personnel crises over several years resulting in high staff turnover and insufficient administrative staff to deal with the workload. One of the results of this was a significant increase in the demand for appointments that they were struggling to meet. The doctors’ felt extremely stretched and as a result their morale was low. During the fieldwork period it was decided to take on a new partner to help with this, but it would be some time before the impact of this on demand was felt. At interview, the doctors all expressed negative attitudes to NSFs and identified a very limited impact of them on their practice. Subsequent observation of a practice clinical meeting revealed that a young, newly qualified and newly appointed nurse was in fact systematically working through and implementing the NSF for CHD. She presented hand-outs of the major elements of the framework, had spent much of her free time developing and refining a computerised register of patients with heart disease, was running regular audits and had a plan for developing a call and recall system for these patients. The significance of this work will be discussed.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

That change in organisations is complex is a truism. Whilst there is some evidence to support the adoption of a multifactorial, context-dependent approach to facilitating change (Effective Health Care 1999), studies that pay attention to the organisational context of the situation in which change in behaviour is desired are uncommon (Solberg 2000). Allery et al. (1997) found an average of three reasons cited for even simple changes in practice in both primary and secondary care; it is likely that changes involving the activities of more than one staff member and a variety of routine processes will be even more complex. As part of a wider investigation of the impact of National Service Frameworks in UK general practice, in-depth case studies in three general practices reveal that even in practices with positive attitudes to clinical guidelines and a track record of their practical use, NSFs, whilst being generally welcomed as a concept, have had little practical impact. By taking this qualitative approach, detailed data have been collected about the minutiae of practice activity in response to NSFs. Whilst macro-level concerns about things such as the perceived ‘secondary care focus’ of NSFs were expressed, many of the factors advanced to explain what was happening in practice were to do with what Armstrong (2002: 1776) calls ‘the practical mechanisms through which medicine is delivered to patients’ rather than higher-level, profession-wide concerns with status identified by those taking a view of doctors as archetypal professionals. Examples of these factors include the size and complexity of NSFs, their failure to address the things that cannot be measured, such as the conduct of the consultation, and their perceived lack of practical focus. Other issues also arose, such as a lack of any clear structure within these practices to deal with incoming information, and attitudes that identified NSFs as being directed at ‘them not us’. In the following analysis, attention will be focused on the first group of explanations, to do with the characteristics of NSFs themselves. The organisational and other factors will be addressed in subsequent papers.

The work of Lipsky, identifying public sector workers as street-level bureaucrats, provides a possible framework within which to analyse some of these observed effects. Assuming GPs to act as street-level bureaucrats according to Lipsky's definition, what interpretations of the behaviour observed in this study does this suggest? Lipsky (1980: 82) discusses the development of ‘routines and simplifications’ to help deal with the pressures associated with demand that outstrips supply. A simple, well-designed and easily internalised clinical guideline can be identified as a device that will aid this process. Thus, some guidelines, such as those published by the British Thoracic Society for the care of asthma, were welcomed by GPs and nurses in the study as ‘making the job easier’, whereas others were dismissed as ‘too complicated’. Applying Lipsky's framework to NSFs might explain why GPs in three practices that had a track record of welcoming innovations (indeed, for one practice this was identified as an integral part of their collective sense of ‘self’) showed little enthusiasm for NSFs in practice. Rather than aiding the processing of clients, NSFs were perceived as too complicated, and this played a part in preventing their explicit implementation. One doctor nicely summed this up, saying that only if NSFs ‘scratched where it itches’ would they be considered and adopted. In this study the clinical aspects of the guidelines for the care of patients with coronary heart disease had generally been adopted early (before their publication in the NSF) because they were seen as ‘making the job easier’. By contrast, clinical guidelines for the management of conditions such as strokes and dementia included in the NSF for Older People had not been adopted. It is possible that the inclusion of detailed service models, the implementation of which involved action by more than one actor (Harrison 1994), contributed to an overall perception of NSFs as part of the problem of excessive workload rather than as part of the solution. As a result, little explicit attention was paid to them, and those clinical guidelines within the NSFs that might have helped ‘make the job easier’ were not noticed or adopted. Whilst it is possible that some guidelines may be more easily accepted than others because they do not challenge accepted practice, this cannot fully explain the observed empirical findings. For example, the British Thoracic Society asthma guidelines have achieved almost universal acceptance in the UK; it is unlikely that before their introduction, practise in this field was entirely uniform. It seems likely that the acceptance of guidelines such as these results from a combination of factors, including the perceived ‘practical’ value of the guidelines, their ease of use and their relationship to current practice. Further qualitative empirical work will be required to tease out how far each of these factors contributes to the observed reality with regard to both normative clinical guidelines and the more complex normative models of service embodied in NSFs.

Lipsky also states that the decisions of street-level bureaucrats become the policies carried out by their organisations. It is part of the centrally imposed remit of Primary Care Trusts (PCTs) to implement NSFs; part of their assessment by the UK inspection body, the Commission for Health Improvement, revolves around how far they are deemed to have fulfilled this obligation. Different PCTs have interpreted this in different ways and, in the area covering the practices studied, committees have been established with a specific remit to facilitate implementation of each of the frameworks. Whilst most participants in this study mentioned small administrative changes in their practice in response to NSFs, the one participant with an active role in the local trust listed a series of area-wide initiatives, including exercise referral schemes and other programmes intended to facilitate activity at practice level. Discussions with other staff revealed little awareness of these programmes; those schemes that were mentioned tended to be ones that had predated the publication of NSFs, and related only to heart disease. Thus, whilst it is the explicit policy of the government that PCTs will implement NSFs and resources are devoted to this, implementation of those parts of the policy that affect primary care depends upon the practical actions of GPs in their practices. The evidence from this study indicates that these practical activities are only taking place to a limited extent; policy is being ‘enacted’ by GPs on the ground, and official priorities are not being followed.

There is an apparent exception to this Lipskian explanation of the observed behaviour: the systematic efforts of one practice nurse to implement the NSF for CHD in a practice struggling to meet the ordinary demands of their patients. This work is significant in the context of this practice for two reasons. First, it has the potential directly to increase workload. Sending for patients who suffer from chronic diseases without increasing the number of available appointments to react to patient demand will have knock-on effects for the rest of the practice. As the nurses spend their time with these patients, work that has traditionally been done by practice nurses will have to be done by someone else – most likely the doctors. This was seen in one of the study practices, where the development of call and recall systems (unrelated to the publication of NSFs) had resulted in an increase in waiting time to see the practice nurses to almost a month. This resulted in an increase in the doctors’ workloads, as they had to deal with things such as ear syringing that could not wait to see the nurse. Workload for the administrative staff had also increased as they dealt with the letters generated and the phone calls from patients responding to the letters, as had the doctors’ administrative work. Secondly, the time spent on developing call and recall systems and other aspects of NSF implementation is time that cannot be spent on other things. For example, many practices have found that using some of their practice nurse time to ‘triage’ (assess and redirect) requests for same-day appointments has reduced the burden of these requests by up to 50 per cent (Richards et al. 2002). Nurses can also take over some existing work done by the doctors – such as routine blood pressures, or family planning – which would further reduce the pressure on the doctors. Thus, in a practice that is seriously struggling to meet the everyday demands of their population and which sees itself as being in a state of ‘crisis’ as a result, one staff member is unilaterally undertaking a programme of work that will increase the general workload and will not do anything to relieve the pressures on the doctors. It is not being argued here that working to implement the NSF for heart disease is intrinsically wrong; simply that it could be said to be inappropriate in the specific circumstances of this practice, as they struggled to provide appointments for the patients who presented. The nurse stated that the initiative for this work had come from her own concept of her role, shaped by her recent degree course, and that the doctors had not asked her to do it. When the doctors became aware of what was being done during the period of fieldwork, rather than being concerned about the potential increase in workload they were enthusiastic, and encouraged the nurse to continue with the work.

This ‘deviant case’ allows the exploration of the limitations of the identification of GPs with Lipsky's definition of a street-level bureaucrat. As discussed earlier, it is the contention here that Lipsky's use of the word bureaucrat may be misleading. Lipsky himself pays little attention to the employment status of his subjects or the organisational structure within which they work, emphasising instead the conditions of their employment such as their autonomy and the pressure on resources that they face. In these terms, the work of GPs clearly falls within Lipsky's definition. However, the unique status of GPs as independent contractors to the UK NHS offers a possible explanation for the apparent paradox of doctors welcoming work that has the potential to cause problems for the practice's overall delivery of services to patients. Whilst it is misleading to view GPs as clearly self-employed (many of the freedoms implied by their independent status are limited by resource constraints and increasingly by the need to meet external performance targets, such as the requirement for all patients to be seen within 48 hours), as independent contractors GPs are also in the position that Lipsky attributed to public service managers: ‘. . . interested in achieving results consistent with agency objectives’ (Lipsky 1980: 18). From the point of view of GPs as street-level bureaucrats, NSFs were seen as too complicated to implement, and were not felt to be helpful in the day-to-day business of dealing with an increasing workload. When they found, however, that the nurse was actually doing this work, they greeted it with enthusiasm and agreed that it should continue. This can be explained if it is accepted that in spite of the lack of overt pressure from primary care bodies to implement NSFs it was generally accepted in the practice that they ‘should’ be doing so. The nurse's work provided them with the opportunity to ‘meet agency objectives’ without doing the work themselves. The wider implications of this, such as the potential impact on an already overstretched appointments system, were not ‘noticed’ as significant. The organisational explanations for this failure to notice the potential impact of this work on the ability of the practice to cope with their workload are outside the scope of this paper.

It is also possible to use Lipsky's ideas to develop a micro-level understanding of some of the factors that might underlie the nurse's choice of activity. Within general practice, nurses also have considerable autonomy in their face-to-face contacts with patients and are constrained both by the way that demand outstrips supply and by limited resources; thus nurses may also be characterised as street-level bureaucrats in their own right. Lipsky points out that one of the mechanisms used by street-level bureaucrats to cope with these demands is to develop what he calls ‘private goal definitions’. As a newly-qualified practitioner, this nurse commented that at first, in common with most doctors and nurses in her situation, she had found the demands associated with independent practice to be high. Focusing on implementing NSFs might be said to have allowed her, in Lipksy's (1980: 145) words, to ‘modify [her] concept of the job in order to close the psychological gap between capabilities and objectives’ in the face of the constraints she faced. By doing this, her actions had become the practical policy of the organisation. In spite of the potentially negative effect on the overall need for the practice to meet the demand for routine appointments, new systems to send for patients with heart disease were being developed. Using Lipsky in this way has the potential to focus empirical work; for example, the analysis outlined above suggests that the operation of power and authority within a practice is altogether more complex than a simple view of doctors as powerful professionals would suggest.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

In summary, I suggest that GPs can be characterised in part as Lipskian street-level bureaucrats, striving to resolve the conflicts associated with working directly with the consumers of public services. This characterisation leads to the following conclusion. The fact that GPs in the three practices, with a track record of adopting clinical guidelines, did not specifically address centrally imposed NSFs can more easily be explained by the failure of these weighty documents to make the job easier than by invoking profession-wide concerns relating to autonomy and control over working practices. One of the limitations of this way of conceptualising GPs is illustrated by the example of one practice in which work to implement NSFs that will tend to make the overall work of the practice more difficult is not prevented by the partners. It is suggested that in this the partners are simultaneously acting as ‘business owners’ (and managers) interested in achieving results ‘consistent with agency objectives’ (Lipsky 1980: 18), and that the nurse is here acting in her own right as a street-level bureaucrat, enacting policy on the ground.

This use of Lipsky's ideas, alongside a focus on doctors as workers who must seek to make sense of their working lives, provides a different way of theorising responses to wider policy changes. Whilst doctors will undoubtedly respond as professionals and as business owners/managers to many things that affect their practices, they will also respond as workers. Lipsky's analysis of street-level bureaucracy predicts many of the features of the NHS at present. These include frequent changes in structure in an attempt to enforce the implementation of central policy; the difficulty of involving clients in a meaningful way in a service in which they are ‘non-voluntary’ (Pickard et al. 2002); and the push towards explicit performance management by targets as embodied in the new GP contract (Shekelle 2003). Lipsky predicts that this latter development will direct activity largely towards meeting the targets, distorting overall activity; this has been seen in the health service with targets for waiting times in Accident & Emergency departments (Carvel 2003).

Following these ideas through leads to interesting new empirical questions. For example, Lipsksy's emphasis on the imperative for street-level bureaucrats to ‘process work’ which will always exceed available resources, and the devices they develop to cope with this, leads to a consideration of what counts as work in general practice. Historically GPs have seen their jobs in terms of sessions spent seeing patients; it is not uncommon for disputes to arise within partnerships about the ‘fair’ distribution of this work. The development of explicit performance targets and an emphasis on audit suggests a future in which sessions of work that do not involve patient contact will have to be explicitly timetabled and rewarded; this is currently rare in the UK. If GPs are behaving like street-level bureaucrats it is possible to predict a conflict between these new responsibilities and the imperative to ‘process’ as many clients as possible. This hypothesis could be tested empirically in the context of the new GP contract: how far will GPs respond as managers, redesigning their practices to meet the new ‘agency targets’, and will this lead to conflict with the street-level bureaucrat role of processing clients who walk through the door? Thinking of GPs as street-level bureaucrats makes the conflict identified in this study between managing acute demand and scheduling elective appointments more obvious; it also suggests that strong management will be needed to deal with this conflict.

It is also possible to link the findings discussed here with work that looks at the impact of recent changes in the health service on the professional role. Lipsky argues that the two best defences against the negative effects of street-level bureaucracy are a strong ethic of professionalism and the intelligent involvement of service users. General practitioners in the UK are protected from the full force of the pressures outlined by Lipsky by their status as independent contractors and by their professional power. Current trends in the health service, however, may limit these protective factors. First, salaried practice is becoming more common, with 43 out of 87 adverts in a recent general practice edition of the British Medical Journal being for a salaried doctor. This will put many GPs in the position of employees, subject to management by their employers. Secondly, it has been argued that the move towards scientific-bureaucratic medicine and performance management by targets will reduce professional autonomy and result in ‘restratification’ within the profession (Charlton 2000, Sheaff et al. 2002, Mahmood 2001, Harrison 1999). Lipsky's analysis provides a theoretical framework within which to think about the potential impacts of these changes on the micro-level of work in general practice, asking questions, with Hoff (2001, 2003), about effects on the meanings that doctors attribute to their work.

In summary, this paper presents evidence from qualitative case studies in general practice, and uses the theoretical model of street-level bureaucracy to explain some of the findings. Data collected in this way are rich and multi-layered; it is not the contention that this approach is the only way of theorising the results, and it is accepted that the narrative offered here must be a simplification of complex and multi-faceted results. In particular, this paper addresses only the similarities in response across three practices and not the individual organisational factors that themselves offer a rich picture of behaviour in response to change. However, shifting the primary focus of research away from profession-wide concerns of autonomy and restratification towards the sense that individuals make of their working lives may, when set alongside these other approaches, contribute to a deeper understanding of the complex responses to centrally-imposed change.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References

Thanks are due to the participants, who gave generously of their limited time. Also thanks to Professor Stephen Harrison and Professor Martin Marshall who made helpful comments throughout the process of writing this paper. I am also grateful to three anonymous reviewers for their comments on an earlier draft; many of their suggestions have been incorporated into this final version.

Note

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Street-level bureaucrats
  5. Methods
  6. Principal findings
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. References
  • Adler, P.A. and Adler, P. (1998) Observational techniques. In Denzin, N.K. and Lincoln, Y.S. (eds) Collecting and Interpreting Qualitative Materials. Thousand Oaks, Ca.: Sage.
  • Allery, L.A., Owen, P.A. and Robling, M.R. (1997) Why general practitioners and consultants change their clinical practice: a critical incident study, British Medical Journal, 314, 8704.
  • Armstrong, D. (2002) Clinical autonomy, individual and collective: the problem of changing doctors’ behaviour, Social Science and Medicine, 55, 17717.
  • Atkinson, P. (1995) Medical Talk and Medical Work. London: Sage.
  • Baker, R. (1997) Will the future GP remain a personal doctor? British Journal of General Practice, 47, 8313.
  • Baker, R. and Thompson, J. (1995) Innovation in general practice: is the gap between training and non-training practices getting wider? British Journal of General Practice, 45, 297300.
  • Benech, I., Wilson, A.E. and Dowell, A.C. (1996) Evidence-based practice in primary care: past, present and future, Journal of Evaluation in Clinical Practice, 2, 24963.
  • British Medical Association. (1966) Third Joint Report of Discussions. London: BMA.
  • Carvel, J. (2003) Hospital targets hit – then missed, The Guardian, 3rd June 2003.
  • Charlton, B. (2000) The new management of scientific knowledge in medicine: a change of direction with profound implications. In Miles, A., Hampton, J.R. and Hurwitz, B. (eds) NICE, CHI and the NHS Reforms: Enabling Excellence or Imposing Control? London: Aesculapius Medical Press.
  • Checkland, K. (2003) Changing the lens: widening the approach to primary care research, Journal of Health Services Research and Policy, 8, 24850.
  • Checkland, P.B. (1998) Information, Systems and Information Systems. Chichester: John Wiley.
  • Cochrane, A. (1999) Effectiveness and Efficiency: Random Reflections on Health Services. London: Royal Society of Medicine Press.
  • Department of Health (1997) The New NHS: Modern, Dependable. London: The Stationery Office.
  • Department of Health (1998a) A First Class Service: Quality in the NHS. London: The Stationery Office.
  • Department of Health (HSC 1998/074) (1998b) National Service Frameworks. London: The Stationery Office.
  • Department of Health (1998c) The New NHS, Modern, Dependable: a National Framework for Assessing Performance. London: The Stationery Office.
  • Dowswell, G., Harrison, S. and Wright, J. (2001) Clinical guidelines: attitudes, information processes and culture in English Primary Care, International Journal of Health Planning and Management, 16, 2, 10724.
  • Dowswell, G., Harrison, S. and Wright, J. (2002) The early days of primary care groups: general practitioners’ perceptions, Health and Social Care in the Community, 10, 4654.
  • Effective Health Care (1999) Getting evidence into practice, University of York Centre for Reviews and Dissemination, 5, 1.
  • Gillam, S. and Pencheon, D. (1998) Managing demand in general practice, British Medical Journal, 316, 18958.
  • Glendinning, C., Chew, C. and Wilkin, D. (1994) Professional power and managerial control: the case of GP assessments of the over-75s, Social Policy and Administration, 28, 31732.
  • Glennerster, H., Matsaganis, M., Owens, P. and Hancock, S. (1994) Implementing GP Fundholding. Wild Card or Winning Hand? Buckingham: Open University Press.
  • Handysides, S. (1994) Building an efficient and healthy practice, British Medical Journal, 308, 17982.
  • Harrison, S. (1994) Knowledge into practice: what's the problem? Journal of Management in Medicine, 8, 916.
  • Harrison, S. (1999) Clinical autonomy and health policy. In Exworthy, M. and Halford, S. (eds) Professionals and the New Managerialism in the Public Sector. Buckingham: Open University Press.
  • Harrison, S. (2002) New labour, modernisation and the medical labour process, Journal of Social Policy, 31, 46585.
  • Harrison, S. and Choudhry, N. (1996) General practice fundholding in the UK National Health Service: evidence to date, Journal of Public Health Policy, 17, 33146.
  • Harrison, S. and Moran, M. (2000) Resources and rationing: managing supply and demand in health care. In Albrecht, G.R., Fitzpatrick, R. and Scrimshaw, S. (eds) The Handbook of Social Studies in Health and Medicine. London: Sage.
  • Heginbotham, C. (1997) Health care priority setting: a survey of doctors, managers and the general public. In New, B. (ed.) Rationing: Talk and Action in Healthcare. London: BMJ Publishing Group.
  • Hoff, T.J. (2001) The physician as worker: what it means and why now? Health Care Management Review, 26, 5370.
  • Hoff, T.J. (2003) How physician-employees experience their work lives in a changing HMO, Journal of Health and Social Behavior, 44, 7596.
  • Katz, D. and Kahn, R.L. (1978) The Social Psychology of Organizations. New York: John Wiley and Sons.
  • Laing, A., Marnoch, G., McKee, L., Joshi, R. and Reid, J. (1997) Administration to innovation: the evolving management challenge in primary care, Journal of Management in Medicine, 11, 7187.
  • Laing, A.W., Cotton, S., Joshi, R., Marnoch, G., McKee, L. and Reid, J. (1998) The buying centre: patterns of structure and interaction in primary health care, The Service Industries Journal, 18, 20.
  • Lewis, R. and Gillam, S. (2002) A fresh new contract for general practitioners, British Medical Journal, 324, 10489.
  • Lipsky, M. (1980) Street-level Bureaucracy: Dilemmas of the Individual in Public Services. New York: Russell Sage Foundation.
  • Loughlin, M. (2000) Quality and ‘excellence’: meaning versus rhetoric. In Miles, A., Hampton, J.R. and Hurwitz, B. (eds) NICE, CHI and the NHS reforms: enabling excellence or imposing control? London: Aesculapius Medical Press.
  • Mahmood, K. (2001) Clinical governance and professional restratification in general practice, Journal of Management in Medicine, 15, 24252.
  • McDonald, R. (2002) Street-level bureaucrats? Heart disease, health economics and policy in a primary care group, Health and Social Care in the Community, 10, 12935.
  • Miles, M. and Huberman, A. (1994) Qualitative Data Analysis: an Expanded Sourcebook. Thousand Oaks, Ca.: Sage.
  • North, N. and Peckham, S. (2001) Analysing structural interests in primary care groups, Social Policy and Administration, 35, 42640.
  • Petchey, R. (1996) From stableboys to jockeys? The prospects for a primary care-led NHS. In May, M., Brundson, E. and Craig, G. (eds) Social Policy Review. London: Social Policy Association.
  • Pickard, S., Marshall, M., Rogers, A., Sheaff, R., Sibbald, B., Campbell, S., Halliwell, S. and Roland, M. (2002) User involvement in clinical governance, Health Expectations, 5, 187198.
  • Richards, D.A., Meakins, J., Tawfik, J., Godfrey, L., Dutton, E., Richardson, G. and Russell, D. (2002) Nurse telephone triage for same day appointments in general practice: multiple interrupted time series trial of effect on workload and costs, British Medical Journal, 325, 1214.
  • Rogers, A., Campbell, S., Gask, L., Sheaff, R., Marshall, M.N., Halliwell, S. and Pickard, S. (2002) Some National Service Frameworks are more equal than others: implementing clinical governance for mental health in primary care groups and trusts, Journal of Mental Health, 11, 199212.
  • Sheaff, R., Smith, K. and Dickson, M. (2002) Is GP restratification beginning in England? Social Policy and Administration, 36, 76579.
  • Shekelle, P. (2003) New contract for general practitioners, British Medical Journal, 326, 4578.
  • Solberg, L.I. (2000) Guideline implementation: what the literature doesn't tell us, Joint Commission Journal on Quality Improvement, 26, 52537.
  • Strauss, A. and Corbin, J. (1990) Basics of Qualitative Research. Newbury Park: Sage.
  • Vickers, G. (1995) The Art of Judgement. Thousand Oaks: Sage.
  • Weick, K.E. (1979) The Social Psychology of Organizing. Reading, Mass.: Addison-Wesley Publishing Company.
  • Weick, K.E. (1995) Sensemaking in Organizations. Thousand Oaks, Ca.: Sage.
  • Westland, M., Grimshaw, J., Maitland, J., Campbell, M., Ledingham, E. and McLeod, E. (1996) Understanding practice management: a qualitative study in general practice, Journal of Management in Medicine, 10, 2937.
  • Yin, R.K. (2003) Case Study Research: Design and Methods. Thousand Oaks: Sage.