Intended for healthcare professionals

Sacred Cows

Clinical governance

BMJ 1998; 317 doi: (Published 19 December 1998) Cite this as: BMJ 1998;317:1725
  1. Neville W Goodman (Nev.W.Goodman{at}, consultant anaesthetist
  1. Department of Anaesthesia, Southmead Hospital, Bristol BS10 5NB

    Every so often, and seemingly increasingly often, a new “Big Idea” reaches the NHS. The BMJ issue that celebrated the NHS's 50th anniversary contained an essay by Gabriel Scally and Liam Donaldson about the latest: clinical governance.1 This Big Idea is different: unlike previous ones it has no intuitive meaning. Ironically, the word governance in the sense intended (“The manner in which something is governed or regulated; method of management, system of regulations”) is marked as obsolete in the Oxford English Dictionary,2 which gives a quote from 1660: “To enquire of the Foundation, Erection, and Governance of Hospitals.”

    The editor, in his editor's choice of that issue, remarked, perhaps pointedly, that Scally and Donaldson “try to spell out the meaning” of clinical governance. I have read “Clinical governance and the drive for quality improvement in the new NHS in England” carefully, word by word, and some parts several times. I have tried to understand why they needed over four pages to impart the commonsense message that we must all strive after quality in practising medicine; I have retained little beyond that it is our statutory duty now to provide quality in our medical care. The essay is all thought and no action, an epitome of hope over expectation, a high sounding clarion call of wonderful things just over the horizon. Most depressing of all, the authors seem to recognise the real difficulties but ignore just how obdurate these difficulties are. The result is an essay full of the “what” but short on the “how.”

    “Rigorous” rhetoric

    The rhetoric starts with the title—the drive for quality in the new NHS. The NHS is not new; it is 50 years old. It could even be argued that the Labour government—by turning away from “dominat[ion] by financial issues and activity targets”—is trying to return the NHS to how it was before the Conservatives fractured it. Perhaps Scally and Donaldson, invoking industry, take their words directly from descriptions of corporate governance, but “rigorous in its application, organisation-wide in its emphasis, developmental in its thrust, and positive in its connotations” is meaningless. If readers doubt this, try “rigorous in its thrust, positive in its emphasis, developmental in its application, and organisation-wide in its connotations” or any other combination.

    Two of these words reappear when they describe the origins of clinical governance: “Although clinical governance can be viewed generally as positive and developmental, it will also be seen as a way of addressing concerns about the quality of health care.” “Although” at the start of the opening clause suggests that the following clause will be in logical contrast, but surely “addressing concerns about … quality” is positive. Surely, concern about quality is the whole point of clinical governance.

    Serious failings and geographical variations are mentioned. Certainly, if serious failings do not become less likely (though how will we know?) clinical governance will have failed. But Scally and Donaldson use failings in screening programmes as their example of “serious clinical failures.” They should have chosen a clearer unarguably clinical failure: there are some who believe screening failures were, and are, the inevitable result of not thinking through the repercussions of screening before the programmes were introduced.3 On the other hand, it is certain that geographical variations in care will remain. The demonstration of variations is not the necessary condition for change; the condition is the demonstration that some variations are unsatisfactory. But, yes, for some time now medical care has been “subservient to price and quantity in a competitive ethos.”

    Empty phrases

    A box then appears in their text, with the simple heading: “What is clinical governance?”—an occasion for a clear definition. Instead we get a mission statement, a rolling unpunctuated tangle of prepositional and adverbial phrases similar to the many that have appeared in the past few years on the walls of our hospital wards and clinics: “Clinical governance is a system through which NHS organisations are accountable for continuously improving the quality of their services and safeguarding high standards of care by creating an environment in which excellence in clinical care will flourish.” It is disappointing that people with the standing and experience of Scally and Donaldson could not foresee the inevitable response of weary healthcare workers to these empty phrases. While we are accountable for continuous improvement, who is accountable for the continuously rising expectations against which that quality is measured?


    Disintegration of clinical governance

    Scally and Donaldson admit that defining clinical quality, of which professional performance is one aspect, has been difficult. This indeed is central: in many specialties it is extraordinarily difficult to define quality at all, let alone measure it. But they go on blithely to say that clinical governance “is designed to consolidate, codify, and universalise often fragmented and far from clear policies and approaches.” Why “consolidate, codify, and universalise”? Why not just “coordinate”? How does one “codify” a policy and approach? Can this be done in any way other than by producing reams of guidelines and standards and by having hordes of people ticking boxes on other people's work?

    Scally and Donaldson represent quality in health care as a bell shaped frequency histogram—problems to the left, exemplars to the right. They are correct that we need better ways of learning from these extremes, but they seem not to understand their own mathematical analogy: “a major shift towards improved quality will occur only if health organisations in the middle range of performance are transformed—that is, if the mean of the quality curve is shifted.” Simple mathematics dictates that all improvements shift the mean; there is no need to invoke transformation. Shifting the mean does not transform, that is, change the shape, of a distribution in the way that, for example, taking logarithms would. Perhaps they intended to say that it is not enough just to learn from the outliers.

    Working again from developments in “the industrial sector,” Scally and Donaldson write about “an … approach to quality improvement with emphasis on preventing adverse outcomes through simplifying and improving the process of care.” This is inane: deleting the middle section leaves the tautology of “an approach to quality improvement … through … improving the process of care.” Why, anyway, should simplifying care necessarily prevent adverse outcomes? How, exactly, are we to simplify our care given the increased complexity of modern medicine, patients' rights, informed consent, and the greater need, contingent on measuring quality, for collecting data? Yes, we need “Leadership and commitment … team work, consumer focus, and good data.” We know that: what we want is a clear description of how to get it.

    A concrete example?

    They recognise that “clinical audit in the NHS is not a complete success” and introduce NICE (National Institute for Clinical Excellence) and CHIMP (Commission for Health Improvement) for external support by “inspecting, investigating, advising, supplying expertise, facilitating, accrediting.” In this section of their article I felt that I had at last reached something concrete. I turned the page to read their example, a case study of the ailing “Gridstone Royal Infirmary NHS Trust.”

    Their example fails. In an imagined interview for the post of medical director, the panel ask non-specific questions about the applicant's “vision” of clinical governance. The applicant answers with generalities: “mechanisms for effective clinical audit … learn from complaints … clear skills and competencies.” If I had been on the panel I would have asked, “How?” Eventually, a panel member does ask, “Okay, could you be a bit more specific?” But nothing is offered, just more about wrong culture, minimum of hierarchies, environments of learning and evaluation, leadership skills of staff nurses, and so on. The applicant says that patients must be more involved, but, to take one example of consumer choice in the United States, mothers-to-be tend to choose between obstetric services because of free diapers or baby buggies rather than the standard of medical care.

    Vague concepts

    Scally and Donaldson believe clinical governance will thrive if we have good leaders, if everyone is open and participates, and if there is a strong working relation between senior managers and health professionals, but they admit that we know little about how to improve leadership skills. They describe leadership as “a rather vague concept” (which seems odd, coming from the chief medical officer). They think the introduction into undergraduate medical education of problem based learning will improve teamworking skills. Perhaps it will. Perhaps it will also produce doctors who are unable to make decisions on their own.

    Moving on, Scally and Donaldson write that the “evidence based medicine movement has always had a major influence on many healthcare systems of the world.” As the movement is only a few years old, it cannot have “always” had an influence, although the current influence is indeed major, largely because its enthusiasts publish endless books and articles telling everyone how to do it. Scally and Donaldson make no mention of the vast areas of medical practice for which the evidence base is small and likely to remain so.

    They then discuss learning from complaints and critical incidents to prevent their recurrence. This has not proved easy, but we are told that “clinical governance has the opportunity to address this weakness.” How? More complaints, all taking time to investigate,4 may indicate greater familiarity with complaining rather than a worsening service. Critical incidents (incidents that can harm) are easy to define, but it is less certain that reducing their number reduces the number of incidents that actually do harm.

    Uninspiring idea

    Scally and Donaldson conclude that clinical governance is “a big idea that … can inspire and enthuse.” I know people who think that clinical governance is a small idea, a rehash of all sorts of “management speak” and poorly thought through generalisations that can depress and dishearten. Scally and Donaldson represent clinical governance (their figure “Integrating approaches of clinical governance”) as a hexagon, each segment containing some items: for example, in the segment “Coherence” is the item “goals of individual, team, and organisation aligned.” Would that alignment were that simple (see figure).

    Perhaps the article was rushed into print before Scally and Donaldson could sit back and think hard about it. But that is not acceptable from people of their authority. Clinical governance emerges as a mixture of the blindingly obvious (people should lead well and work well in teams) and the unproved (clinical audit). In recent years, clinicians at the coalface have suffered a succession of Big Ideas—achieving a balance, continuing medical education, Calmanisation—based on unimpeachable principles. Each has fallen short because of other factors.57 Now Scally and Donaldson tell us that clinical governance is the Big Idea that will really work, apparently just by making it a statutory duty. I am sure they are sincere in wanting better quality of care in the NHS. We all want that, though I am bound to ask, by what comparisons is overall quality unsatisfactory? For Scally and Donaldson to convince us that clinical governance will work, they must give us realities not generalisations: a few real problems and their solutions. Then leave us to decide if we are inspired and enthused.

    The most important elements in the delivery of quality in health care are contained in the relations between human beings. With good working relations, clinical governance (or whatever it is called) happens naturally; with poor working relations, setting up committees and defining quality on pieces of paper delivers only pieces of paper. If the service providers are measuring their own activity, why should we trust their measurement more than we trust the activity they provide? When all the necessary measuring is done, why should we trust the measurers more than we do the service providers? The inevitable consequence is loss of trust by, whether or not there is loss of trust in, the providers.

    This obsession with measurement and accountability is not unique to medicine8 and fosters “the illusion that life can be reduced to manipulable numbers, the delusion that something which is said to be so therefore is so.”9 The delivery of health care is complicated, so complicated that there are no easy solutions, no curative Big Ideas. There will always be problems and strains within the service. It would have been enough for Scally and Donaldson to have announced the end of the age of competition within the NHS, asked us and allowed us to collaborate for better health care, and given us encouragement: “people work best if they are given a worthwhile job and are allowed to get on with it.”10

    Scally and Donaldson do not convince me that they will spare the emperor's embarrassment.


    I thank Chris Johnson for helpful suggestions for the final draft.

    Competing interests: None declared