Intended for healthcare professionals

Education And Debate Modernising the NHS

A promising start, but fundamental reform is needed

BMJ 2000; 320 doi: (Published 13 May 2000) Cite this as: BMJ 2000;320:1329
  1. Alain C Enthoven, Marriner S Eccles professor of public and private management (enthoven_alain{at}
  1. Graduate School of Business, Stanford, CA 94305-5015, USA

    This is the first of seven articles

    The NHS has just received its largest sustained increase in resources since the service was started in 1948. To ensure that the money is spent wisely, Britain's prime minister has set about producing a national plan for health, with the help of six action teams—on prevention and inequalities, partnership, performance and productivity, the professions and the wider NHS workforce, patient care (empowerment), and patient care (speed of access). The BMJ has asked six commentators to state what their priorities would be in each of these areas, and we will be publishing these articles over the next few weeks. This week, however, we start with a commentary from Alain Enthoven, professor of public and private management at the graduate business school, Stanford University. Enthoven's ideas were a driving force behind the creation of the internal market in the NHS in the 1990s; he recently revisited the subject in last year's Rock Carling lecture, and we invited him to give his impressions on the prime minister's current plans for the NHS.

    The prime minister was right to promise more money for the NHS. I recently wrote: “The NHS is obviously very short of resources needed to achieve its objectives. One sees it in the buildings, the pay, the staff and equipment shortages, the very short times doctors spend with patients, and the headlines about crises.” 1

    The prime minister was also right to say “that the NHS needs fundamental reform if it is to provide the standard of care that people deserve in the 21st century.” 2 Money alone will not get us there.

    And the prime minister was right to call attention to the wide variations in performance in the NHS. Among his many examples: “Why is there a twofold difference in the cost of care between the best and the least efficient hospitals? … The top 25 per cent of trusts use their consultants twice as productively as the bottom 25 per cent.” These variations mean that large amounts of resources are being wasted. That is, opportunities to get a great deal more care out of the existing or expanded resources are being lost because inefficient practices are unchecked. Open, competitive markets work to drive out such variations as competition forces the least efficient to adopt the methods of the more efficient, or be forced out of business. What will do that in the NHS?

    What is fundamental reform? Apparently the prime minister does not have a plan. He has appointed six action teams to produce one by July. What is missing from the speeches and the press releases is a coherent strategy for effectively identifying and motivating systematic improvement. Here are some suggestions.

    Summary points

    Quick fixes will fail: changing cultures and processes will take time.

    Information is fundamental, and the NHS currently has poor information and makes poor use of what it has.

    The government's centralising approach will fail: much better would be to rely on incentives in a decentralised approach.

    Quality improvement is a sound philosophy that takes time and resources.

    Quick fixes will fail

    The fundamental problems of the NHS cannot be fixed in time to make a noticeable difference by the next election. It will take years to change NHS culture, train and retrain the people, create the necessary information systems, analyse and change the care management processes, repair and build the needed buildings, and procure modern reliable equipment. Though there is a political imperative to find quick fixes, they will inevitably fail. Resources should be directed to reforms that can sustain large improvements over the long run.

    Information is fundamental

    The government must give top priority to creating high quality clinical and management information systems that measure the quality and economy of all NHS operations. The NHS today lacks high quality clinical and financial databases and makes poor use of the information it has. Last year, Black wrote, “Irrespective of the uses to which they wish to put the data, clinicians, managers, consumers, and researchers all need data from consecutive cases, that are complete and accurate, that are based on standard definitions of clinical disorders, interventions, and outcomes, and that include information on those characteristics of patients that affect outcome.” 3

    Embedded Image

    (Credit: PHILIP WOLMUTH)

    It was a serious mistake to leave information systems off the list of challenges and modernisation action teams—unless information is seen as an integral part of performance improvement or adequately addressed by other initiatives. Good data are needed to measure and evaluate outcomes and efficiency, to plan improvements, and to measure progress. Without vastly improved information, and people trained and motivated to use it, nothing else will do much good.

    Creating the necessary information systems will require some fundamental changes in organisation and payment. A crucial problem for information systems is motivating staff to report accurately, promptly, and completely. Some real “out of the box” thinking is needed here. For example, the new plan might propose that a substantial part of each hospital's revenue should be based on fixed payments for each completed case weighted by healthcare resource group (HRG), with payment to follow submission of a completed discharge abstract signed by the attending doctor to attest to its accuracy. Perhaps the same should be done for consultants. Such a payment system could add a much needed incentive for productivity, an issue about which the prime minister expressed concern.2 Such data must be audited for accuracy and completeness.

    Not just a problem of technology

    The creation of information systems is not merely an information technology problem (though information technology can help a great deal). Accurate information must be made an integral part of everyone's work. Information must have consequences. It must not be seen as a mere academic exercise. Information systems must be good enough that management (including doctors making clinical management decisions) will act on what these systems produce.

    There is a crisis of confidence in the quality of NHS care. A high quality clinical database is an essential part of the cure. Without it, the NHS lacks the data systems needed to tell it what is happening. For example, clinical governance is impossible or meaningless if hospital executives have no way of knowing how their outcomes compare with those of other hospitals. Good data would make it possibleto analyse variations in treatment patterns and methods and to compare risk adjusted outcomes in different hospitals.

    The high quality clinical database should include all inpatient cases and all outpatient encounters. All parts of the healthcare system should be linked. Encounter data can be used to address questions such as, “Have asthmatic children been instructed in use of peak flow meters?” A link to prescribing data linked to patients can address, “Are patients receiving beta blockers after a heart attack?” Capitation payments to primary care trusts should be “risk adjusted” on the basis of diagnostic information obtained from encounter or prescribing data, so that primary care trusts that have a disproportionate share of chronically ill patients can be paid appropriately.

    The database should be linked to measured outcomes, including mortality. Well designed computer applications could make that convenient and relatively cheap.

    Data on costs

    The NHS needs and does not yet have good information on comparative costs in hospitals. As James Raftery recently pointed out, The New NHS—1998 Reference Costs suffers from several serious deficiencies, including that the costs accounted for in the report cover only about 40% of total hospital costs, the costing methods are not standardised, and data are reported at hospital, not patient, level.4 In other industries, management considers it essential to measure costs per unit of output with reasonable accuracy. It may be harder in health care, but the NHS cannot achieve efficiency without it—because without it, nobody knows who or what is efficient.

    Incentives are fundamental

    This brings us to the fundamental strategic choice of how to translate information into action. The prime minister is clearly heading for a centralised approach: create league tables and set the Commission for Health Improvement to work pursuing and changing the poor performers. I doubt that prime ministerial exhortation of NHS doctors to adopt his latest ideas on how to practice medicine will be effective in the long run.

    Experience teaches that the centralised approach will fail. It assumes that the centre knows best. Centralised economic management has failed everywhere it has been tried. The former Soviet Union and other centrally planned economies failed. The bureaucrats at the centre consider variation and innovation to be a threat and they resist it. Moreover, the centralised approach is likely to be seen as punitive and coercive. It disempowers people in the field. And it is likely to do little or nothing to motivate the best performers to improve. Large corporations decentralise into smaller operating units.


    The alternative is to rely on incentives in a decentralised approach. For all its limitations, the internal market was the beginning of a decentralised, incentives oriented approach. It was wrong for the prime minister to make so much of abolishing it. It would have made more sense to seek ways to make it work better. The “abolition” conflicts with what is said in the government's white paper The New NHS: “New primary care groups will be established in all parts of the country to commission services for local patients … devolved commissioning will go hand in hand with greater equity.” Commissioning by primary care groups or trusts is an internal market-like idea that could be seen as generalisation of general practitioner fundholding.

    Continuing a policy that worked to prevent the internal market from achieving its objectives, the government could, in effect, tell primary care trusts not to make any decisions that might “destabilise hospitals” or cause other political problems, not to move their purchasing from one hospital department that provides poor quality at a high cost to another that provides better quality at a lower cost. This would be a mockery of the trusts' commissioning role and surely lead to disillusionment.

    The alternative would be to encourage primary care trusts to purchase aggressively on behalf of their patients, to support them with data and other resources, and to back them when they want to move patient flows in pursuit of better quality and service. This approach would focus on incentives and performance measurement. Hospitals and their doctors would have to provide high quality and good service economically or risk losing patients and revenue. The prime minister would have to let go and trust the trusts to do the right thing, even if what they did caused pain to some favoured MPs.

    The decentralised approach is far more likely to succeed. It unleashes and draws on innovation and creativity throughout the NHS, as did the establishment of fundholding. Decentralisation empowers the front line professionals.

    Decentralisation recognises that not all good ideas come from the centre. Indeed, a great danger of overcentralisation is that bad decisions are reached—for lack of information or for partisan, ideological, or other political reasons—and then forced on the whole NHS. Better to let the local people evaluate the evidence in their own context, while holding them accountable for quality and cost. The decentralised approach recognises that a lot of important information is local, may be qualitative, and not available to the centre—especially today with such weak NHS information systems.

    Quality improvement

    The answers to the present crisis of confidence in the quality of care in the NHS must be found in basic organisational and cultural changes that lead to much greater transparency and to a total commitment to continuous quality improvement (CQI). This seems to be what clinical governance is about. I had hoped the internal market would motivate and reward it. Continuous quality improvement is a comprehensive and integrated management philosophy that has been adopted with impressive success in some industrial companies.1 Government cannot force this from the top. Real continuous quality improvement is a “grass roots” movement. But government can state expectations that a profound change in NHS culture and method of operation must take place.5 And it can create a framework of appropriate incentives.

    Yet the government must prioritise: it can't have continuous quality improvement and everything else it is demanding every week. It must provide resources to support the extensive training that will be needed. It was an excellent idea to include the quality improvement leader Donald Berwick of Harvard on the recently appointed performance and productivity team.6 He is leading the campaign to bring continuous quality improvement to American health care (which is definitely not an easy task).

    Embedded Image

    (Credit: ULRIKE PREUSS)

    Consumer choice and competition are modern

    This government is committed to the objectives set out in Modernising Government, including modernising the NHS.7 The March 1999 white paper emphasised responsive public services focused on the needs of users, efficient high quality services, “information age” government using new technology, and valuing public service. It noted that “the British public has grown accustomed to consumer choice and competition in the private sector. If our public service is to survive and thrive, it must match the best in its ability to innovate, to share good ideas and to control costs.” The government intends to modernise “wherever practical by giving the public the right to choose … The incentives to modernise have been weak.” The author of that white paper should be included in the drafting of the new plan.

    I doubt that it is possible to create and sustain a culture of innovation, efficiency, and good customer service in a public sector monopoly whose services are in excess demand and whose units do not get more resources for caring for more patients. That is asking too much. Money has to follow patients. I doubt that the NHS can achieve modernity without consumer choice and competition, and substantially more resources. The prime minister has promised additional resources. Let us hope that his July plan includes some consumer choice and competition and strong incentives to modernise.


    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.