Intended for healthcare professionals


Will prescriptions for cultural change improve the NHS?

BMJ 2013; 346 doi: (Published 01 March 2013) Cite this as: BMJ 2013;346:f1305
  1. Huw Talfryn Oakley Davies, professor1,
  2. Russell Mannion, professor2
  1. 1School of Management, University of St Andrews, Fife KY16 9RJ, UK
  2. 2Health Services Management Centre, University of Birmingham, Birmingham B15 2RT, UK
  1. Correspondence to: H T O Davies hd{at}

The recent Francis report diagnoses serious cultural deficiencies in the NHS and recommends fundamental cultural change. Huw Davies and Russell Mannion examine what research tells us about the likelihood of success

It is hard to escape the conclusion from the Francis report into care at Mid Staffordshire NHS Foundation Trust that the primary culprit at the heart of this latest NHS scandal is “the culture” of our healthcare organisations.1 Francis suggests that “a fundamental culture change is needed” and is clear that he is seeking a move to something overarching and comprehensive for the whole NHS.

The Francis inquiry, like the Kennedy inquiry into paediatric cardiac surgery at Bristol more than a decade earlier,2 has gone to considerable trouble to try to understand the meaning of culture in a healthcare context. Yet the subtlety of some of the supporting evidence to the inquiry has not been matched by the same degree of nuance in the inquiry’s recommendations about culture, which are somewhat aspirational and broad brush (box 1). This is a pity, since the research literature has much to say about the nature of culture and the possibilities for shaping cultural change to produce benefits.3 4 5

Box 1: Core recommendations relating to culture in the Francis report 1

Recommendation 2: The NHS and all who work for it must adopt and demonstrate a shared culture in which the patient is the priority in everything done. This requires:

  • A common set of core values and standards shared throughout the system

  • Leadership at all levels from ward to the top of the Department of Health, committed to and capable of involving all staff with those values and standards

  • A system that recognises and applies the values of transparency, honesty, and candour

  • Freely available, useful, reliable, and full information on attainment of the values and standards

  • A tool or methodology such as a cultural barometer to measure the cultural health of all parts of the system.

Recommendation 185 (and others): Addresses the need for a “focus on culture of caring,” mostly within nursing, with an emphasis on “an increased focus in nurse training, education and professional development on the practical requirements of delivering compassionate care in addition to the theory.”

Recommendation 198: Healthcare providers should be encouraged by incentives to develop and deploy reliable and transparent measures of the cultural health of front-line nursing workplaces and teams, which build on the experience and feedback of nursing staff using a robust methodology, such as the cultural barometer.

Recommendation 290: The Department of Health should promote a shared positive culture by setting an example in its statements by being open about deficiencies, ensuring those harmed have a remedy, and making information publicly available about performance at the most detailed level possible.

Unpacking culture

By finding fault with culture and providing prescriptions for change, Francis is making several assumptions that require examination. Firstly, he presupposes that we can identify and assess common aspects of culture,6 as well as identify which aspects are supportive of or inimical to high quality care.3 4 Secondly, he assumes that these aspects of culture can be purposely changed, that any changes will lead to improvements, and that the costs and dysfunctions from such prescriptive changes will be outweighed by the benefits.4 7 Finally, while acknowledging that culture may vary “from organisation to organisation and from department to department,” Francis emphasises the need for “a positive and common culture throughout.” This presumes that common cultures are possible and desirable, even in systems as large and distributed as the NHS. However, research shows more complex and nuanced relations between cultures, practices, and outcomes than Francis implies.

A cultural mosaic

What is organisational culture? The answers are many, complex, and contested.8 9 But at the heart of many definitions is that culture consists of the values, beliefs, and assumptions shared by occupational groups (box 2). These shared ways of thinking are then translated into common and repeated patterns of behaviour: patterns of behaviour that are in turn maintained and reinforced by the rituals, ceremonies, and rewards of everyday organisational life.3 8 In everyday language, culture is “the way things are done around here,” together with the shared ways of thinking that support these norms of practice.3

Box 2: Defining culture910

Culture can be defined as “the pattern of shared basic assumptions—invented, discovered or developed by a given group as it learns to cope with its problems of external adaptation and internal integration—that has worked well enough to be considered valid and, therefore, to be taught to new members as the correct way to perceive, think and feel in relation to those problems.”10

Many authors talk of culture as operating at three levels:

  • Level 1: artefacts—The most visible manifestations of culture, including the physical layout of services, established processes of care, staff rotas and reporting arrangements, dress codes, rituals, reward structures, and ceremonies. This would include, for example, normal working patterns, the agenda and processes of board meetings, and the arrangements for handling patient complaints and staff concerns

  • Level 2: beliefs and values—Used to justify particular behaviours, provide a rationale for choosing between alternate courses of action, and distinguish “right” from “wrong.” Examples include respect for patient autonomy and dignity and the prevailing views on current individual and collective performance

  • Level 3: assumptions—The unspoken, largely unconscious, expectations and presuppositions that underpin day to day work. For example, assumptions about the nature of the caring role, the knowledge and perspectives of patients and relatives, and the relative role and power of doctors, nurses, and managers.

Many years of organisational research, some of it in healthcare, have shed light on what we can hope for from cultural analysis and change.3 4 5 6 7 8 9 10 It is clear that culture in large complex organisations is rarely uniform.11 12 13 One study suggested that NHS staff were united on only two main issues: the need for care to be based on individual need rather than funding and a dislike of “constant interference into healthcare provision by successive UK governments.”13

In reality, cultural divergence of basic beliefs and assumptions is the norm. Staff may work for the same organisation (a hospital, say) and employ the same language (talking of care, quality, evidence, or performance) but the meaning of these terms can vary between different staff groups. Doctors, for example, tend to interpret good quality care as treatment following best evidence, whereas nurses tend to take a more holistic view, emphasising the alleviation of symptoms and restoration of health. When even the meaning ascribed to basic language is not shared, it is hard to see how cultural alignment is possible.14

Moreover, research has consistently found differences between staff groups in their fundamental beliefs and assumptions. Doctors, nurses, and managers, for example, take different views on the balance between a focus on individual patients and the need for more corporate approaches.15 16 Team based care is also conceived of differently, with doctors often assuming a more hierarchical approach, and nurses tending to be more open and inclusive.13

While it is tempting to focus on subcultures as being aligned to professional groups, this too is an oversimplification. Those in hybrid roles (such as clinician-managers) take on some of the orientations associated with their new roles but never quite shed all of their original assumptions.16 Cultural divergences are also seen within professional groups—for example, tribal loyalties and professional identities associated with new subspecialisms undermined attempts at quality improvement in acute pain services.17 Given such cultural complexity, simply urging people to think differently is unlikely to over-ride the complex personal and social forces that shape organisational behaviour.

Our healthcare organisations, then, are better viewed as multiple (often competing) subcultures, stratified by hierarchy, hospital, service, ward, team, and, most obviously, occupational group. Subcultures may, at different times, be driving forces for change, overt defenders of the status quo, or covert counter cultures quietly undermining new initiatives. Understanding this cultural mosaic is essential for any effective cultural diagnosis. Moreover, it can be difficult to decide whether counter cultures are stubborn resisters to necessary change, simply defenders of professional traditions, or perceptive opponents of damaging new directions. In that sense, conflicts between cultural subgroups reflect differences in power and legitimacy as well as outlook, and they can be as much about a struggle for expression of identity, meaning, and purpose as they are about competition for resources, autonomy, and control.5 9 17

Striving for cultural uniformity in such complex, dynamic, and (in all likelihood) conflict ridden organisational arrangements may be overoptimistic. Exhortations that “patients be at the centre of the NHS” have been a regular feature of reports, inquiries, and General Medical Council recommendations for well over a decade but remain “stubbornly resistant to adoption.”18

Culture and performance

Cultural diagnoses aside, any relations between organisational culture and organisational outcomes such as healthcare quality or patient safety are unlikely to be simple.19 It seems that organisations do better on those aspects of performance that are most valued, affirmed, and celebrated within the organisation.20 This means that trade-offs are inevitable and that there can be no one right culture.9 20

Moreover, it is becoming clear that much as we might like to think that the direction of travel is from culture to performance (however that is assessed), judgments on organisational performance also affect local cultures.9 21 This means that organisations castigated for “failing” are likely to see a spiral of decline, demoralisation, and cultural drift.9 It also means that organisations feted for their success on specific aspects of performance are likely to reinforce further developments in that performance, perhaps to the detriment of other areas.9 21 Thus the pursuit of highly visible success measures (such as foundation trust status) can lead to the neglect of basic care processes, as happened in Mid Staffordshire.1

If culture is a shared narrative that staff use to make sense of their environment—a narrative that also shapes behaviour—then any narrative about organisational performance is an intimate part of that cultural backdrop. And it is here that the organisation’s leaders have a key role, because it is leaders who interpret quality and performance data and reflect this in their accounts of achievements and future directions.4 9 22

Crucially, the policy environment and overarching regulatory regime provide a context within which senior managers assess performance and orient their priorities and practices. Longitudinal data on the NHS suggest that during 2001-08 board cultures shifted away from an emphasis on organisational cohesion to prioritising rules based achievement and external competitiveness.23 In that sense, some of the preoccupations at Mid Staffordshire revealed at the Francis inquiry were predictable.9 The Mid Staffordshire board did not pursue foundation trust status on a whim: it was strongly guided to do so by prevailing policy imperatives.

Finally, all attempts at cultural manipulation and performance management are likely to have some unexpected and perhaps dysfunctional consequences.24 It is disappointing therefore that while the first Francis report25 identified target driven priorities as part of the problem, this latest report advocates that healthcare providers be given incentives to deploy transparent measures of culture in the workplace.1 Thus culture itself becomes a target driven priority. When attached to sanctions and rewards, what Francis termed a “cultural barometer” for healthcare providers 1 seems destined to reinforce rather than reform a culture of compliance.

Taking on board the evidence

So where does this leave Francis’s cultural prescriptions? We would do well to tone down our rhetoric around culture, be cautious about the idea of cultural uniformity, and be sceptical that top-down prescriptions will bring about the desired changes. Instead the emphasis needs to be on careful local nurturing, reaching for gardening metaphors in place of those rooted in ideas of engineering. Local contexts provide for organic, home grown approaches that are sensitive to local histories and preoccupations, and real change requires detailed and sustained work on the ground.7 We also need better understanding of how the overarching policy arrangements and dynamics either facilitate locally focused change or frustrate local initiatives by being too rigid.

Getting the cultural diagnosis right is an essential first step. But we need to recognise the variety and depth of cultural diversity in the NHS and make it more visible to the players. Such recognition needs to be explicit about not just cultural diversity in terms of language, values, and identity but divergence in terms of power, authority, status, and reward. Any cultural prescription for the NHS must deal with this diversity, paying attention to change within cultural subgroups and attending to the interactions between cultural subgroups.

Moreover, while recognising the power of cultures to constrain, individuals still have choices to make, choices for which they should be held accountable.26 Here ideas of “behavioural justice” might prove helpful.27 Such notions recognise that individuals have responsibility for their behaviour but can reasonably be held to account only when they are empowered and have adequate resources that facilitate appropriate courses of action. Thus nurses and healthcare assistants might reasonably be supposed to have less room for self directed culture change than, say, doctors and managers. Senior managers too are located differently from policy makers, potentially challenged from below and oppressed from above.9

Francis pulled back from laying formal blame for organisational failure at the feet of individuals, although senior trust managers do receive substantial criticism.28 Moreover, he calls for “all who work in it [the NHS] to take personal and collective responsibility to root out poor practice wherever it is found.” Such general exhortations neglect the varying degrees of autonomy and empowerment of different players.

Rather than a single set of prescriptions as advocated by Francis, more balanced strategies that reflect difficult trade-offs and respect local contingencies may be more appropriate for both local and national policy (box 3).

Box 3: How senior management and national policy can facilitate cultural change 9

One size does not fit all. Cultural change requires a judicious blend of approaches that recognises tensions across several important dimensions:

  • Values (retention/ renewal): many existing values are supportive of high quality care and need to be retained; others need to be challenged, changed, or replaced

  • Accountability (checking/trusting): targets have a role in providing direction, focus, and incentive but high trust organisations can more easily learn and adapt to multiple competing demands

  • Targets for change (external/internal): local ownership of targets may be important; data used formatively rather than summatively have different effects

  • Performance data (hard data/soft intelligence): multiple sources of data, and a capacity and willingness to explore contradictions in these, are prerequisites for openness to learning

  • Blame game (high blame/no blame/just blame): blame is an insidious aspect of many cultures that creeps back in despite overt attempts to recognise its chilling effect

  • Motivations (extrinsic drivers/intrinsic drivers): too much focus on extrinsic rewards (salary, status, power) may crowd out intrinsic motivations (professional ethic, care, compassion).

  • Key senior management balances relating to cultural change include:

  • Senior management style may need to vary between the transactional (getting daily tasks done) and the transformational (setting challenging new directions) depending on local performance, resources, and priorities

  • Management orientation may need to blend a corporate focus (to ensure cross-organisation alignment of values and goals) with a pro-professional orientation (that seeks to ensure proper account is taken of professional values around care)

  • Middle management capacity needs to facilitate local change and performance without adding bureaucratic overhead. How much middle management, and of what kind, remains a perennial balancing act.

We need to examine the linkages between national policies and local practices, to see how good policy intentions can become distorted at the local level.29 For example, it can be difficult to nurture more appropriate accountabilities and learning at local level if organisations are named and shamed nationally. The growing cluster of trusts being investigated for high hospital death rates suggests that “the government may already have decided that cultural change will be too hard and take too long.”18

Francis rightly draws attention to the need for “impact assessments before structural change.”1 His report sets out a comprehensive set of questions that should be answered for any new policy directive and draws attention to the importance of the political backdrop against which all NHS cultural change takes place. If properly and comprehensively implemented, such guidance could open up debate on the likely cultural effect of new policies. Regrettably, the recent experience of the passage of the Health and Social Care Bill is not reassuring in this regard.

Finally, the evidence presented at the Francis inquiry makes it clear that we need stronger voices across the system that can speak against dominant or vested narratives. To be effective, such advocates will require full access to quantitative and qualitative data; opportunities to talk with frontline workers, patients, carers, and relatives; and a mandate to focus entirely on quality, safety, and patient experience. We also need to strengthen the support and protection for whistle blowers and challenge the questionable role of gagging orders.30

Culture may indeed lie at the root of many of the service failings of complex organisations. But more sophisticated understandings of cultural dynamics, together with an appreciation of the role of policy in shaping these, are needed if we are to tackle healthcare failings with any hope of success.

Key points

  • Organisational cultures in healthcare are manifold, complex, and dynamic

  • Healthcare organisations are better seen as multiple coexisting subcultures than as uniform cultural monoliths

  • The national policy framework and top down strategic initiatives provide a powerful shaping context for local cultures

  • National policy prescriptions need to tread a delicate balancing act across a wide range of key dimensions and should be rigorously assessed before implementation


Cite this as: BMJ 2013;346:f1305


  • Contributors and sources: HD and RM have been researching healthcare quality, safety, and performance, with a particular interest in the role of organisational culture, for around 20 years. A wide range of academic, policy, and service collaborators have contributed to our work over many years and many projects. Both authors contributed to this article fully in inception, design, analysis, and write up. In addition, we received helpful comments from three referees (Kieran Walshe, Chris Ham, and Naomi Fulop) and useful argument from a long time collaborator (Alison Powell).

  • Competing interests: We have read and understood the BMJ Group policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract