Intended for healthcare professionals

General Practice

Primary care: core values Developing primary care: gatekeeping, commissioning, and managed care

BMJ 1998; 317 doi: (Published 11 July 1998) Cite this as: BMJ 1998;317:125
  1. Jennifer Dixon (j.dixon{at}, fellow in policy analysisa,
  2. Peter Holland, associate director, primary careb,
  3. Nicholas Mays, director of health services researcha
  1. aKing's Fund, London W1M 0AN
  2. bLambeth, Southwark, and Lewisham Health Authority, London SE1 7NT
  1. Correspondence to: Dr Dixon

    This is the fourth in a series of six articles reflecting on the core values that will underpin the development of primary care

    Series editor: Mike Pringle

    If Nye Bevan were around today, he might be surprised to find that the basic features of British general practice, not least its administrative separation from hospital care, are still in place half a century after the genesis of the NHS. But primary care has not stood still over that period—both its structure and role have developed continuously.

    This development has not been part of an orchestrated grand plan. Rather, it has been characterised by incremental change in response to wider pressures. In this article we examine briefly how some of these pressures have recently influenced the shape and direction of primary care in the UK, and reflect upon the direction of further change in future.

    Summary points

    • Primary care is being shaped incrementally by external pressures, especially the need to contain costs and demonstrate improved quality

    • As a result, primary care professionals, particularly general practitioners, have been encouraged to take more responsibilty to influence health services, rather than just their own professional practice

    • In recent years general practitioners' influence has increased through being involved in commissioning, or being directly responsible for purchasing care, from providers

    • In future, primary care will be required to take a bigger role in managing resources for primary and secondary care

    • Scrutiny of quality and cost of care will become more intense

    Pressures influencing the shape of primary care

    Of the pressures outlined above, two of the greatest at present are the imperative to control the rising costs of health care and improve quality. Consequently, some of the prime movers shaping the development of health systems in the United Kingdom and other countries in recent years have been funders of health care, whether public or private.

    Three related changes have resulted. Firstly, there has been greater investment in, and expansion of the role of, primary care, and more emphasis on its gatekeeping role. Secondly, general practitioners, and to a lesser extent other primary care staff, have been given more opportunity to shape services that are provided in secondary care, particularly through directly managing a budget. Thirdly, incentives and rules have been applied to providers in secondary and primary care to encourage cost conscious behaviour, reduce inappropriate or ineffective care, and promote good quality care. Each of these aims is an essential element of managed care 1 2 and is referred to in the recent white paper, The New NHS.3

    Greater investment in primary care and the gatekeeping role

    Unlike many other countries, the United Kingdom has developed a strong system of primary care. Firm central direction has ensured universal access to a general practitioner, a healthy balance of general practitioners to hospital doctors, and greater average annual real growth of expenditure on family health services compared to hospital and community health services—3.7% compared with 2.9% over the past 20 years. The solo general practitioner working out of two rooms has been replaced largely by group practice, multidisciplinary teams and multipurpose health centres. The roles of primary care staff, especially nurses, have expanded and teamwork is encouraged.4 The two recent primary care white papers emphasise both the development of primary care organisations to replace the independent general practitioner, and primary care as the main locus for healthcare activity. 5 6 In the 1990s there has been some limited attempt to influence the services provided in general practice—for example, through the national general practice contract—and this is likely to continue.

    Other countries are belatedly learning the value of these types of arrangement, particularly in terms of efficiency, and are rapidly reshaping their healthcare systems. For example, in the United States there are new incentives for doctors to train as primary care physicians and for hospitals not to train more specialists.7 Payment scales have been adjusted to favour primary care physicians over specialists, 8 9 reimbursement for providers has shifted from fee for service to capitation, and payers are increasingly insisting that patients seeking care make first contact with a primary care gatekeeper rather than a specialist. There is thus a worldwide push to promote investment in primary care above specialist care. 10 11

    Greater opportunity to shape services provided in secondary care

    The underlying aim of initiatives in this area is not simply to give primary care providers greater influence over secondary care. Increasingly, the government wants to encourage greater cost control and efficiency at the point where many key decisions relating to subsequent expenditure are made—in primary care. The NHS has done this through increasing the influence of the general practitioner, rather than of other members of the primary care team or patients.

    Three overlapping developments are increasingly being pursued in Britain3: greater contact between general practitioners, health authority purchasers, and secondary care providers; giving general practitioners and primary care organisations direct purchasing power; and, most recently, encouraging vertical and “virtual” integration of providers in primary and secondary care.

    Greater contact between general practitioners, purchasers, and secondary care providers

    General practitioners and other primary care staff have always had opportunities to influence care provided by other providers. They have been able to do this informally through professional networks and formally through representation on the boards of health authorities and hospitals.

    Embedded Image

    The NHS reforms of 1991 channelled general practitioners' influence into the purchasing process instead.12 General practitioners have been encouraged to influence providers indirectly through the health authority via locality commissioning and variants such as general practitioner led commissioning, or through the new primary care groups.3 The existing initiatives have had some impact, particularly in developing services at the interface between primary and secondary care.1315 General practitioners who purchase care (for example, through fundholding or total purchasing16) can influence providers directly through purchasing services.

    The reforms in 1991 offered little to encourage greater direct links between providers in primary and secondary care, other than through purchasing, possibly because efficiency was a higher concern than quality of care. Yet these links remained and have grown, despite the incentives of the internal market and other policies such as the requirement to increase hospital productivity.17 Hospital at home, shared care, and outreach schemes are widespread, and some trusts are making efforts to work jointly with general practitioners on a wide range of issues.18 The 1997 Primary Care Act and the recent white papers for England, The New NHS,3 and Scotland, Designed to Care,19 mark a break with the recent past because they explicitly encourage links of this kind.

    Giving general practitioners and primary care organisations direct purchasing power

    The general practitioner fundholding scheme, introduced in 1991, and its subsequent variants—community fundholding, extended fundholding, and total purchasing—gave general practitioners the opportunity to influence secondary care providers directly and provided modest incentives to shift costly hospital care to community settings. Currently around 55% of people in Britain are registered with practices operating some kind of fundholding scheme.19

    If hard outcome measures of efficiency, equity, effectiveness, and choice for patients are used as a measure, the impact of fundholding has been uncertain.2023 There may be at least five reasons why the impact on curbing costs or demands, where appropriate, has been modest.

    • Fundholding practices, at least in the early days, may have had relatively generous budgets that provided weak incentives to scrutinise expenditure.24

    • Peer review of clinical behaviour is undeveloped, and adequate information to support it is often lacking.

    • The scope for reducing hospitalisation for elective surgery may be limited, since there is little opportunity to shift it into primary care; in any case fundholding offers no significant remuneration for taking on extra work.

    • NHS trusts may obstruct change because they see nothing positive in greater general practitioner power for general practitioners, share no mutual sense of mission, and have incentives to increase hospital activity while general practitioners try to reduce it. On soft outcomes such as increasing general practitioners' sense of empowerment and ability to influence other providers, fundholding has had more obvious success.20 25

    • Finally, fundholders and total purchasers as organisations may be too undeveloped and weak to have had much impact.20

    This apparent lack of impact so far, plus the higher administrative costs of devolved purchasing, raise important questions about the future impact of different forms of purchasing or commissioning. The new primary care groups, covering a population of around 100 000 (set out in The New NHS), which will largely replace existing forms of general practitioner purchasing and commissioning, will need considerable support and help from health authorities to develop into robust and cohesive organisations. Will they be strong enough to manage demands effectively and appropriately and persuade providers to make necessary changes? Other, more fundamental, questions also need urgent answers, such as the accountability and purchasing competence of primary care groups and the future role of health authorities,26 only hinted at in The New NHS.

    Regardless of the pros and cons of existing models, greater incentives to use resources for NHS care more efficiently and to manage demand must be here to stay. The current proposals seek to draw all general practitioners into the mainstream task of managing NHS resources. No one model will suit all areas, however, and the umbrella term primary care groups will probably cover a range of organisations.

    Encouraging vertical and virtual integration

    Since 1991 the NHS has tried to separate purchasers and providers and, to some degree, push purchasing into primary care. While primary and secondary care have worked together there was no push to merge them into one “vertically integrated” organisation—until the 1997 Primary Care Act and the recent white paper, The New NHS.

    Vertically and “virtually” integrated organisations linking primary and secondary care (box) are most strongly developed in the United States (particularly in California). They have developed largely in response to the pressure to control costs and to reduce cost shifting between different providers.27

    Linking primary and secondary care

    • Vertical integration usually comprises large networks of primary care physicians and their teams working with secondary care providers in one single organisation. The organisation receives capitated payment for patient care, bears all the financial risk, and shares the benefits of any reduction in use of resources (such as fewer admissions to hospital) among employees, who are thus encouraged to work towards the same broad mission.

    • “Virtual” integration is where primary care organisations (often large networks of primary care physicians) receive capitated payment for patient care, bear the financial risk of that care, and contract with preferred secondary care providers (often entering into long term relationships) without being part of the same organisation.27

    In the United Kingdom, local vertical partnerships between hospitals and community services and primary care have developed at the interface between primary and secondary care. Examples include hospital at home schemes, outreach, shared care, general practitioners working in accident and emergency departments, and community staff attached to general practices as part of the primary care team. These have developed mostly to improve the quality and seamlessness of services provided, and in response to new technologies that allow more treatment at home and easier communication with hospital. Recently, the potential of such partnerships to contain costs by reducing unnecessary hospital use has become important.28

    The 1997 Primary Care Act provided the opportunity for further vertical integration. The act allowed NHS trusts (acute or community) to employ the primary team directly, including the general practitioners, and allowed the merger of budgets for general medical services and hospital and community health services.2931 But the underlying aim of this legislation is not clear—is it to promote more seamless care and teamwork,32 facilitate a shift of care from hospital into the community, ease recruitment of general practitioners and practice staff, or protect the income of NHS trusts? If a main aim is to contain costs by shifting care into the community, then there may be insufficient incentives for secondary care providers to change spots and become more primary care led. But strong and stable partnerships could develop between providers in different settings under these arrangements.

    The New NHS and Scotland's version, Designed to Care,19 both encourage primary care staff and community trusts to team up to form a single primary care trust. Hinted at in The New NHS, and made more explicit in Designed to Care, is the possibility of primary care organisations linking up more closely with hospitals through innovative local arrangements. Possible developments include vertically integrated disease management packages (for example, for chronic diseases),33 as well as schemes to pool resources and share financial incentives to keep patients out of hospital where appropriate.

    In many ways virtual integration already exists in the NHS. Through fundholding and its variants, purchasers with capitated budgets, who are also primary care providers, have entered into long term contractual relationships with other providers. This has already encouraged greater efforts to provide seamless care and curb costs. For example, many of the new total purchasing pilots have made a priority of attempting to reduce both length of stay and medical admissions where appropriate16 in order to be able to use the resources elsewhere. Some have employed “tracker” nurses to work in provider units to encourage prompter discharge for patients,34 and others have persuaded NHS trusts to employ specialist nurses to help manage patients with chronic disease in the community. It remains to be seen whether these schemes will be effective, or whether the new primary care groups will develop them further. This partly depends on whether hospitals will have strong incentives to increase inpatient activity or whether they will develop wider roles for themselves.

    More incentives and rules to improve efficiency and quality

    Policies to encourage efficiency have mostly been heavily directed by the NHS Executive; for example, the discipline of living within the means of a global budget, and achieving the targets of the purchaser efficiency index35 and cost improvement programmes. The NHS reforms of 1991 aimed to increase the incentives for efficiency at a more local level through introducing the purchaser-provider split and, in particular, by devolving budgets to primary care.

    The incentives operating locally are still weak, however, and this may be one reason why purchasing seems to have had a modest impact on effective management of demands. Although there are early signs that general practitioner fundholders and total purchasers are beginning to think about peer reviewing their colleagues, health authorities have been reluctant to investigate or act even on gross variations in clinical practice. Through the research and development initiative, more information is becoming available on the costs of treatments and on the effectiveness of care, yet there are few direct incentives, as well as inadequate help, to use this knowledge. Proposals in The New NHS are designed to strengthen scrutiny of clinical performance and variations and to make much more information on the costs and effects of treatment available. The proposed Commission for Health Improvement, the nomination of a senior professional in each primary care group who will be responsible for the quality of clinical care, and the publication of a list of reference costs for hospital treatments should all help to improve monitoring of performance. But whether the new primary care groups will act on these initiatives depends on how far they will be supported by health authorities, who are already stretched.

    Even greater scrutiny of clinical behaviour is likely if resource constraints become tighter in future, if the incentives set up by different forms of purchasing through the primary care groups do not result in demands being managed more effectively, and if patients' demands for information increase. Such scrutiny may take a more aggressive form, as seen in the United States: retrospective or prospective authorisation of care before payment, utilisation review and physician profiling, and more direct financial rewards for doctors to provide high quality and cost effective care as well as sanctions for those who do not.36 Sanctions could include exclusion from networks of providers or purchasers. These developments raise many important questions, such as who would set the criteria for, and conduct, utilisation reviews, what will be done about providers who perform poorly, and whether the national GP contract will stand.


    Primary care will develop in response to several key pressures, as it has in the past. The latest developments push the NHS only into the foothills of fully formed managed care. Unless the reforms result in better management of demand and increasing quality, they may curtail the freedom of primary care professionals as providers and purchasers. Direct and powerful tools to scrutinise and control clinical behaviour may become the norm, such as utilisation review with sanctions and rewards. The lesson for doctors may well be “manage or be managed.” In the United States some of these changes have resulted in doctors having greatly diminished control over the healthcare delivery system; these doctors are described as being “still in shock,”37 something that would have surprised Mr Bevan.

    This article has been adapted from Primary Care: Core Values, edited by Mike Pringle, which will by published by the BMJ Publishing Group in July.


    Funding: None.

    Conflict of interest: None.


    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.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    23. 23.
    24. 24.
    25. 25.
    26. 26.
    27. 27.
    28. 28.
    29. 29.
    30. 30.
    31. 31.
    32. 32.
    33. 33.
    34. 34.
    35. 35.
    36. 36.
    37. 37.
    View Abstract