Primary Care

Managing change in the culture of general practice: qualitative case studies in primary care trusts

BMJ 2003; 327 doi: (Published 11 September 2003) Cite this as: BMJ 2003;327:599
  1. Martin N Marshall, professor of general practice (martin.marshall{at},
  2. Russell Mannion, senior research fellow2,
  3. Elizabeth Nelson, research associate1,
  4. Huw T O Davies, professor of health care policy and management3
  1. 1 National Primary Care Research and Development Centre, University of Manchester, Manchester M13 9PL,
  2. 2 Centre for Health Economics, University of York, York YO10 5DD
  3. 3 Centre for Public Policy and Management, Department of Management, University of St Andrews, St Andrews KY16 9AL
  1. Correspondence to: M N Marshall


    Objectives To explore the potential tension between the need for managers to produce measurable change and the skills required to produce cultural change, and to investigate how managers of primary care trusts are attempting to deal with this tension.

    Design Qualitative case studies using data derived from semistructured interviews and a review of published documents. An established cultural framework was to used to help interpret the findings.

    Setting Six primary care trusts in England purposefully sampled to represent a range of cultural, structural, geographical, and demographic characteristics.

    Participants 42 interviews with 39 different senior and middle primary care trust managers conducted over an 18 month period.

    Results We found two distinct and polarised styles of management. One group of managers adopts a directive style and challenges the prevailing norms and values of clinicians, an approach characteristically seen in organisations with hierarchical cultures. This group is made up mostly of senior managers who are driven principally by the imperative to deliver a political agenda. Managers in the second group are more inclined to work with the prevailing cultures found in general practice, attempting to facilitate change from within rather than forcing change from outside. This management style is characteristically seen in organisations with a clan-type culture. The approach was manifest mostly by middle managers, who seem to act as buffers between the demands of senior managers and their own perception of the ability and willingness of health professionals to cope with change. The different management approaches can lead to tension and dysfunction between tiers of management.

    Conclusions The development of primary care depends on high quality managers who are able to draw on a range of different management skills and styles. Managers are most likely to be effective if they appreciate the merits and drawbacks of their different styles and are willing to work in partnership.


    Managers are under increasing pressure to implement major structural reform in the NHS as well as achieving demanding national performance targets. At the same time they are expected to deliver changes in the culture of NHS organisations—changes which some regard as fundamental if real improvements are to be sustained.13

    The political imperative to deliver rapid objective change can lead managers to adopt an increasingly authoritarian style.4 Such an approach may be effective in producing short term measurable change, but it is likely to bring managers into direct conflict with doctors who value their professional autonomy and resist current attempts to “manage” their performance.5 Management styles that make doctors unhappy may adversely affect both the quality of care and patients' satisfaction.6

    Tension exists between the performance management role of managers and their responsibilities to produce changes in the culture of the NHS. This tension is becoming increasingly important for managers who work in primary care trusts—relatively new organisations that are expected to lead the reform of the NHS. With the help of a model called the competing values framework, we explored this tension and investigated how managers are attempting to deal with it. The study represents one part of a larger investigation of organisational culture and its relation to performance.3 7 8

    The competing values framework

    Organisational cultures are difficult to assess because their shared beliefs, values, and assumptions are not always explicit.9 Qualitative approaches help researchers to move beyond superficial explanations of culture and are greatly aided by conceptual frameworks that seek to explain important cultural dimensions. The competing values framework10 11 has been applied in both healthcare and non-healthcare organisations.12 13 Using two main dimensions—the first describing how processes are carried out within the organisation, and the second describing the orientation of the organisation to the outside world—the competing values framework describes four basic organisational cultural types (figure). Organisations (or subgroups within them) may possess more than one of these types, but one of them is usually dominant. The framework thus provides a way to describe and explain qualitative information about organisational culture.14 15

    Fig 1
    Fig 1

    Competing values model of culture types for organisations (adapted from Cameron and Freeman


    We conducted qualitative case studies, deriving data from open ended interviews with managers, supplemented with documentary evidence produced by the primary care trusts.

    We selected a sample of six primary care trusts from a purposeful sample of 12 that were participating in a longitudinal investigation of clinical governance.16 The subsample was chosen to reflect the emphasis that we knew the organisations placed on cultural change,2 different stages of maturity (range 12-24 months since formation of the trust), number of practices, and geographic and demographic characteristics. Three of the primary care trusts had recently merged with adjacent organisations.

    We conducted a total of 42 interviews in the six organisations. Twenty three of these took place between August and December 2000 and 19 between January and May 2002. The two phases of data collection enabled us to examine the stability of our findings over time. Three people were interviewed in both periods of study, resulting in 39 different interviewees. Nineteen of the participants were senior managers (10 executive and nine non-executive board members) and 20 were middle managers (non-board members, including clinical governance leads, locality managers, and project managers). Interviewees were selected to represent different roles and responsibilities within the organisation. Ethical approval was obtained and anonymity was assured. The interviews were guided by a schedule which addressed the participants' understanding of the term “culture,” their perceptions of cultural traits established in the primary care trust and in practices, the perceived relation between culture and performance, and the facilitators, barriers, and management strategies for change. The interviews were conducted in an open ended fashion, and any references to the style or impact of different management approaches were investigated in detail. Each interview lasted between 25 and 90 minutes and was audiotaped. In addition, we reviewed relevant documents, such as annual and clinical governance reports.

    The audiotapes were fully transcribed, and we produced individual site summaries before conducting a content analysis of each transcript. All summaries and transcripts were read by EN and MM, who identified any issues describing management styles by iterative examination. We discussed themes as they emerged and explored them using the competing values framework as an interpretive framework where appropriate. We assessed the trustworthiness of our analysis by triangulation between data sources and explored any differences in the researchers' interpretations at project team meetings.


    The managers readily identified a set of common values which they perceived were manifest by the constituent practices in their primary care trusts (box). These values are characteristic of the clan-type culture described by the competing values framework. Some of the participants referred to the benefits of these values to patients and to the NHS. However, most spoke about the negative impact of the values on the ability of managers to improve the “performance” of their primary care trust. In broad terms managers agreed that the term “performance” described the primary care trusts' capacity to improve the health of the local population and reduce inequalities. They thought that this would be difficult to achieve unless the practices changed their culture to one that valued greater collaboration and sharing of expertise, and a willingness to be more flexible in the way that they operated. In the competing values framework, this suggests a desire to change the dominant culture of general practice from “clan” to “developmental,” and this would be achieved by maintaining the relationship based orientation but changing the focus of the practices to one that is more outward looking (figure).

    When questioned in detail, however, managers clearly differed in the ways that they thought this change should be achieved. Some focused primarily on outputs and outcomes, and their discussions were dominated by the need to achieve objective government targets. They made few references to how these targets should be achieved. Others, however, spoke principally about the need to improve processes, and their contributions were dominated by references to “getting the right culture” in order to achieve change. Outcomes, though acknowledged as important, were not the immediate concern of this group of managers. These different management objectives were reflected in contrasting management styles, which can be described as directive or facilitative (table).

    Comparison between directive and facilitative managers

    View this table:

    Managers'perceptions of the current cultural characteristics of general practice

    A sense of history and tradition: Manifest by an awareness of the work that has gone into building a practice and a feeling that “we've always done things this way and don't see any reason to change for change's sake.” A feeling that new initiatives come and go but that general practice goes on unchanged

    A sense of cohesiveness and loyalty to the practice as an organisation: Manifest by a tendency to support the practice in preference to any other organisation

    A strong orientation towards professional autonomy: Manifest by a suspicion of anything that potentially erodes this autonomy

    A tendency to be inward looking: Manifest by a resistance to work with, be compared with, or learn from other practices

    A tendency towards paternalistic leadership styles: Manifest by a lack of strong and radicalleadership at a practice level

    Directive managers

    A directive style was more common among senior managers than middle managers. This was particularly true for executive directors—eight out of 10 executive directors had a predominantly directive style, in comparison with three of the nine non-executive directors and two of the 20 middle managers. Directive managers showed a willingness to challenge the prevailing values manifest by the health professionals, such as their rigid and inappropriate desire… to remain autonomous” (chief executive). They wanted to achieve measurable outcomes within a short period of time, and their narratives were dominated by references to the need to be more responsive to patients. They sought greater uniformity between practices—one senior manager expressed a desire to produce a new “corporate identity” (chief executive) for his primary care trust. The approaches of directive managers largely ran against the grain of the values embedded in a predominantly clan culture—for example, the desire to introduce greater uniformity and the use of directives (hierarchical cultural attributes), and the promotion of competition and target setting (rational cultural attributes). Directive managers described the main obstacles to cultural change as relating to the individuals concerned, rather than relating to organisational or environmental factors.

    Facilitative managers

    A facilitative style was more likely to be used by middle managers than by senior managers—18 of the 20 middle managers had a predominantly facilitative style. Facilitative managers spoke about their desire to work with the prevailing values shown by general practices, and were less inclined to challenge these values:

    “We have one practice which has all along believed that clinical governance was rubbish… I have never challenged them and never had an argument with them that their view of the world is not right because I think that if we can get them by action to change… they will realise that they have changed without me having to rub it in” (clinical governance lead).

    They regarded the facilitation of cultural change as their raison d'etre and expressed a desire to encourage a change in values from within, rather than forcing change from outside. They made few references to government targets, financial incentives, or competition but spoke frequently about “building relationships based on trust and respect” (locality manager) and “encouraging dialogue” and “understanding what makes the practices tick” (clinical governance lead).

    Facilitative managers adopted strategies which were aligned to their philosophical approach and compatible with the prevailing clan culture. They placed considerable emphasis on evolutionary and developmental change. For example, they highlighted the importance of practice visits by primary care trust managers:

    “Before decisions are made, we consult with practices about how they feel. It's a bit time consuming but it's worthwhile because at the end of the day, the value of any relationship or the way it develops is based on the amount of communication between parties. If managers are just stuck away in offices dictating from the centre, then clearly you are not going to have the same collaboration or co-operation as you would have if people meet and discuss problems on a regular basis” (chair of local health group)

    In addition, they encouraged practices to set aside “protected time” for learning and used respected peers to promote new ideas. They also attempted to influence general practitioners (who they saw as resistant to change) by working with their practice managers and practice nurses (who they saw as more willing to embrace change).

    “We've done a lot of work around CHD registers… and the original work—creating a lot more sharing amongst GPs about CHD, getting enthusiasm going—came from the nurses not from the GPs” (chair of primary care trust)

    Facilitative managers did not completely dismiss demands to produce measurable improvements. They felt, however, that real change would not happen, or at least would not be sustained, without attention to the values of those who deliver health services. They regarded this as a complex and time consuming task and dismissed “quick fixes.” They described the main obstacles to cultural change in environmental or organisational terms, rather than relating to the attitudes of individual health professionals.

    Tensions between managerial approaches

    The different management styles seemed to result in tensions between managers in five of the six primary care trusts. Conflict was particularly marked between the senior managers with a directive style and middle managers with a facilitative style and was more likely to be recognised and voiced by the middle managers. Middle managers thought that they were more in tune with the needs of those working in the front line of the service and criticised their senior colleagues for being out of touch. They saw themselves as the ones who “did the real work” and stated that they often felt excluded from the strategic thinking of the senior managers. Middle managers saw themselves as “buffers” between the demands of senior managers and their perception of the willingness and capacity of the practices to embrace change. Tensions were exacerbated in two of the primary care trusts because middle and senior managers were based in different buildings, and were more commonly expressed in the three primary care trusts that had undergone recent mergers.


    The relationship between managers and clinicians has been a subject of considerable interest in recent years.1719 Much of the literature seems to treat managers as if they are a homogenous group. This study describes how primary care managers think in different ways and use different techniques to promote change. The robustness of these styles is supported by the fact that they emerge strongly both explicitly and implicitly from the narratives and are consistent in the two periods of data collection. The study design does not allow us to comment on the relative effectiveness of the approaches, but use of the competing values framework sheds light on the potential merits and problems of different management styles in achieving different objectives.

    What is already known on this topic

    Doctors value their autonomy, and this can lead to conflict with health service managers

    What this study adds

    Managers' responsibility to demonstrate measurable improvements is in tension with their desire to produce changes in the culture of the NHS

    Middle managers tend to interact with doctors in a facilitative way, working with prevailing professional values; senior managers tend to adopt a more directive style by challenging these values

    Primary care trusts should value the skills of their facilitative managers, and policymakers need to consider the trade offs between short term improvements in measurable performance and long term changes in the culture of NHS organisations

    Why might middle and senior managers adopt different approaches? Management style is likely to be contingent on roles, responsibilities, and the context within which managers work.20 In particular, senior managers may have to reflect the demands of their political masters. They are also in a position to exercise power and authority, whereas middle managers are more likely to have to negotiate to achieve their aims.21 Managers with a directive style may be more likely to be promoted within the NHS. Middle managers have a particularly difficult role to fulfil, acting as buffers between two powerful groups: their senior colleagues and clinicians. To our knowledge this is the first time that this tension between management styles has been reported and explored in the primary care sector.

    Improving the quality of a complex health system is likely to require a range of different management approaches.22 Managers who adopt a directive style, focused on the political imperatives, willing to challenge prevailing norms and drive rapid measurable change, serve an important function in reforming the NHS. However, this style may be more likely to deliver short term measurable results than long term cultural change14 and may induce unwanted dysfunctional consequences.23 Managers with a facilitative style who work with health professionals and maintain their good will are essential to organisations that need to get the best out of their professional workforce. However, this style may be oversensitive to the demands of the more conservative professionals.

    Real and sustained reform of the health system is more likely to be achieved if organisations are able to draw on a repertoire of management styles and skills. To achieve this, primary care trusts need to place greater value on the skills of their facilitative managers. At the same time, policy makers and politicians need to consider the likely trade offs between achieving short term improvements in measurable performance and sustained change in the culture of NHS organisations. These aims are not mutually exclusive in policy terms—but they may feel that way to the managers and clinicians working in the front line of the service.


    • Contributors MM, RM, and HD devised the study; EN conducted the second round interviews. All authors contributed to the analysis and interpretation of the data. MM wrote the first draft of the paper and all authors contributed to subsequent drafts. MM acts as guarantor.

    • Funding UK Department of Health through core support for the Centre for Health Economics, University of York, and the National Primary Care Research and Development Centre, University of Manchester. The views expressed in the paper represent those of the authors and not necessarily those of the funders.

    • Competing interests None declared.

    • Ethical approval This study had multicentre research ethics approval.


    View Abstract