Intended for healthcare professionals

General Practice

Here we go again: repeating implementation errors

BMJ 1999; 319 doi: (Published 04 September 1999) Cite this as: BMJ 1999;319:616
  1. Donald W Light, professor (lightd{at}
  1. Division of Social and Behavioral Medicine, University of Medicine and Dentistry of New Jersey, Stratford, NJ 08084, USA

    The current reforms of the English NHS are being carefully watched in Eastern and Western Europe, parts of Latin America and Asia, and in former Commonwealth nations, because they aim to develop the concept of a “primary care led health service” more fully than has ever been realised before. Building on the transformation of the NHS from an administered to a commissioned service, the Labour reforms reject the costly, inequitable, and demoralising policies of competition and advocate sweeping changes based on cooperation. They call for bringing together formerly autonomous general practitioners into large groups, having them integrate their services with community health and social services, having them develop community-wide plans for health improvement, and devolving nearly the entire budget into their hands.

    While these and other parts of Labour's grand plan are much better thought out than the Conservative reforms, some of the same mistakes of implementation are being repeated. As an excellent analysis of the current reforms warns, “If we have learned anything from the 1991 reforms … it is that implementation is all.”1

    Summary points

    The NHS reforms are promising, but some mistakes of implementation by the last government are being repeated by this one

    Primary care groups, like health authorities in the 1990s, lack the resources, latitude, and incentives to commission effectively

    Primary care groups will be taking on greater proportions of healthcare services but are not in line for greater proportions of the budget

    The reforms embody conflicting models of governance—hierarchical control versus devolved commissioning—with doctors and nurses caught in the crossfire

    Integrating primary, community, and social services and commissioning them effectively is the first order of business

    Health authorities have major new responsibilities on top of the old ones and need to be strengthened

    Too much too fast

    Good ideas and design can be ruined if the agents of implementation are overloaded or overwhelmed. The Conservatives, with their ill considered mandatory efficiency gains from secondary care, produced serious problems of morale and exhaustion that the Nuffield Trust has effectively documented, such as high turnover of staff, wastage, early retirement, sick absences, high stress, and the danger of serious clinical errors.2 If Labour does not slow its pace of implementation, it could create the same exhaustion and low morale in primary care that the Conservatives created in secondary care. Ironically, pressured implementation could crush the very boldness among general practice leaders on which the “new NHS” design depends. Re-engineering clinical care and integrating services take time and a great deal of effort. This is one of the seven lessons from effective commissioning groups,3 and recent reports from the field show how complex a job it is.4 5

    Too few incentives, too little risk

    Implementing the English white paper requires a lot of extra work, but why should nurses and general practitioners bother? They cannot get much (if any) extra pay, and they bear little financial risk for inefficiencies in the current arrangements or unnecessary services Primary care groups, in theory, can keep any money they save, and each practice can keep up to £45 000, but central costing and restrictions will prevent such savings from being made. The same mistakes were made with health authorities as commissioners in the 1990s, with tepid results. While Americans go dangerously overboard with incentives and risks, another of the seven lessons for effective commissioning is that purchasers and providers must have some incentives and bear some risk for the decisions they make.3

    Further, most healthcare costs and potential savings lie in specialty and hospital care. Therefore, cost effective commissioning will devolve more clinical work to the primary and community healthcare teams—more management of serious clinical problems, more post-hospital care, more work integrating care in the community, as well as more work to educate patients and prevent ill health. But it is quite difficult now for money to follow patients and work, even though in theory integrated budgets should enable primary care teams to have more of the budget as they do more of the work.

    Short-changing commissioners

    The new reforms, like the old ones, depend on better commissioning and management of services. Yet both are being underfunded. One of the Conservatives' great mistakes was to strengthen hospitals, by giving them trust status and highly paid executive teams, while they underpaid and underresourced health authorities as purchasers.6 This resulted in top talent migrating from the purchasing side to large hospitals and fiercely defending their grip on the healthcare budget.7 Yet those who hold budgets and purchase or commission are the key to any commissioning model of health care.

    Primary care commissioning is inherently a more costly design than a more centralised approach, such as the approach in the Scottish white paper. The hope is that general practitioners and nurses can save more than the extra cost by being smarter, tougher, and more creative commissioners than can managers in a health authority or health board. But an average allocation of £3 per head of population to do the job short-changes these new commissioners: evidence from the pilot schemes of total purchasing shows those teams needed an average of £7 per person,8 and even then most of the teams were struggling with their commissioning role. It is already evident that most of the primary care groups lack the time or resources to identify sources of waste and address them.

    The government is also perpetuating the Conservatives' mistake of keeping health authorities weak through deeply ambivalent policies. It has assigned them seminal new roles at the same time as it has indicated they are to work themselves out of existence by developing successful primary care trusts. Health authorities have more to do than ever to make these reforms succeed, but not the resources. In what other industry would one meet, as I recently did, a director of finance responsible for a budget of £300m who has only one assistant and a part time secretary? He can do little more than keep up with the blizzard of directives from the centre and sign block contracts. But if the government does not resource commissioners well, why bother to commission at all? Weak or poorly informed commissioning leaves the healthcare system open to all sorts of manipulations and waste.

    Poor information and accountability

    As commissioners of integrated care with an integrated budget, general practitioners and nurses need to know how their patients are treated across all sectors and sites, at what cost, and to what effect Otherwise they will be commissioning with blindfolds. Another great mistake by the Conservatives was to let hospital trusts collect what data they wanted, how they wanted, and to decide what they would share with their purchasers. Current implementation efforts recognise the central need for comparative clinical and financial data across sectors of care, and openly publishing costs is a substantial advance; but the problem seems greater than the efforts. Space precludes detailing the evidence that, even after eight years, many hospitals do not know their true costs, count staff as fixed costs, set marginal prices well above what international experience would estimate they should be, and overcharge. One business manager of a fundholding practice analysed a hospital bill for finished consultant episodes (FCEs) by patient and found 14 attributed to one patient. He also found a £7000 bill from a surgical department which turned out to be for an outpatient visit that the consultant did not attend. Yet health authorities, and now primary care groups, usually have no choice but to take the prices given them. Reference costing may help, but they may also lock in current inefficiencies and self protective accounting of practices.9

    Stifling commissioners with central control

    The core idea of commissioning, especially if devolved to general practitioners and nurses, is to empower those “close to the customer” to attain better service and quality at lower cost than corporate officers or governmental officials can from headquarters.10 This means that the central office sets general goals and provides the necessary resources (see above) but gives the leaders in the field the power and freedom to fulfil those goals. The problem, of course, is that local commissioning groups, or even health authorities, may be careless, make mistakes, or try something that fails. Devolved commissioning is a gamble that promises smarter, more effective solutions but will almost certainly produce its share of mistakes and failures. These in turn generate bad press, so the centre creates restrictions and requirements, performance indicators, and procedural rules that safeguard current institutions and arrangements, just as the Conservatives did.1113 The result is conflicting paradigms of governance, with general practitioners and nurses feeling empowered by one and stifled by the other.

    A principal goal of reforms and commissioning in nearly all countries is to break up and reconfigure the overhospitalised specialty services that built up during the 20th century, and the radical implication of commissioning for evidence based outcomes is that one wants to contract with specialists and let them subcontract for hospital services as they see fit. But the prime object of damage control from the centre is to protect the hospitals, and recent hospital mergers only increase their powers against effective commissioning.

    When it comes to commissioning it is unclear to me what the government wants, though bad weather over the Atlantic may impair my vision. It is perpetuating the implementation error of keeping health authorities underresourced and then concluding that they are inept and clunky That is why primary care commissioning is needed—to juice up all those doctors and nurses to do it right. But then they are given less than half the evidence based resources they need, and three times as many of their chief executives come from health authorities as from general practice fundholding and commissioning groups.14 And to make sure that the juiced up commissioning leaders don't upset currently wasteful arrangements, a raft of rules and requirements are put in place.

    Suggestions for success

    Some of issues I have raised are so basic that they cannot be addressed here and need full discussion. But one of the seven lessons for effective commissioning3 is that primary care itself needs to be commissioned effectively. The capacities, clinical decisions, and inequities of primary care providers determine how equitable and cost effective the rest of the system is. Requiring small primary care teams to come together, overcome their many professional and cultural differences, and develop a collective responsibility for their own practices is an exciting part of the Labour reforms. Integrating them with community health and social services for health gain is a still larger agenda of great importance for nations everywhere. Getting this right seems to me the first order of business, together with developing an integrated information system on clinical performance and costs across all sectors.

    This means that for several years the health authorities will do the major commissioning. They will also develop primary care groups, pull together health improvement plans, plan overall changes in services, nurture networks of diverse local agencies and organisations, and be accountable for evaluation.15 They need talented, well paid leadership teams with good technical resourcing. Beyond these basics lie the real savings through integrated pathways of care between secondary, primary, and community care. One key is to align primary care providers with consultants Win-win contracts have been worked out in the United States that would reward both groups for developing ways to reduce the need for costly services.


    • Competing interests None declared.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.