Intended for healthcare professionals


New Labour, new NHS?

BMJ 1997; 315 doi: (Published 20 December 1997) Cite this as: BMJ 1997;315:1639

The white paper spells evolution not revolution

  1. Jennife Dixon, Fellow in policy analysisa,
  2. Nicholas Mays, Director of health services researcha
  1. a King's Fund, London W1M 0AN

Since May the new Labour government has repeatedly pledged to abolish the reforms of the NHS carried out by the previous Conservative government in 1991. The internal market, competition, the business ethic, and general practitioner fundholding would be swept away and bureaucracy and inequities in access to care reduced. But with no new big idea to hand akin to the radical changes of 1991, and with an awareness that not all the reforms were bad after all,1 2 this government had a problem. Should it believe its own rhetoric and reverse most of the changes, as promised, without anything new to replace them? Or should it swallow its pride, concede that some features of the internal market were worth keeping, and build on the best?

The result, published last week in the white paper, The New NHS,3 is, of course, a compromise. The rhetoric is that the internal market, which supposedly resulted in damaging competition, has been abolished. In reality, competition was weak, the purchaser-provider split will remain, and purchasers will still have some choice between providers. So what's new?

Quite a lot. The main change concerns primary care. The chief responsibility for purchasing health care will move from the current 100 health authorities, 3600 fundholders, and 90 total purchasing pilots to 500 primary care groups each covering “natural communities” of roughly 100 000 people. Primary care groups are to consist of groups of general practitioners (around 50) and community nurses which will eventually hold a budget for virtually all hospital and community health services for the area plus the cash limited part of the general medical services budget—for example, for prescriptions and practice staffing. Health authorities will continue to purchase only selected specialist services, and fundholding will be scrapped from April 1999. The plan is for primary care groups to develop in four stages over the next five years: at a minimum they could leave all purchasing to the health authority and have an advisory role only; at a maximum they could purchase almost all services and merge with community trusts to form primary care trusts providing all primary and community health care. The overall budget for patient care will be cash limited, and the primary care groups will be able to keep any savings made. The current management costs of the health authority and fundholders will be pooled, capped, and shared out between the health authority and primary care groups through a process yet to be defined.

At first glance, this seems like a sensible evolution from the current plethora of purchasing models.4 The trend has been for single practices to team up into groups and take on greater responsibility for either commissioning (through general practitioner to locality commissioning5) or purchasing (through fundholding to total purchasing6). What is new is that all practices in a natural community will be required to take part in primary care groups to reduce “two tierism.” And there's the rub: what is the incentive for reluctant general practitioners to participate? On the one hand, since the primary care group will control the cash limited general medical services budget for all practices in the area, general practitioners will have an incentive to get involved to influence the flow of funds to their practice. But on the other hand, general practitioners, particularly those least experienced in purchasing, will need significant management and information technology support to participate. And here's the next rub: the implication in the white paper is that funds available to cover the management costs of the primary care groups and the health authority will be cut to pay for other goodies in the white paper. Without adequate management support, primary care groups will be a damp squib.

The second main change relates to health authorities. Eventually shorn of most purchasing responsibilities, their role of shaping, monitoring, and regulating local services will be strengthened. For example, as shapers they will be responsible for developing a health improvement programme—a strategy for meeting population health and healthcare needs. They also have important new reserve powers to influence decisions on capital development and new consultant posts in NHS trusts. They have new statutory responsibilities to collaborate with other local bodies, including local authorities, primary care groups, and NHS trusts, for example in developing a health improvement programme. As regulators they will support and monitor primary care groups and have powers to intervene should they fail. These are all logical developments but raise questions about the capacity of health authorities to take on these new roles, especially with a reduction in management resources.7

The third main change relates to NHS trusts. While remaining semiautonomous, they will now have a statutory duty to collaborate with other NHS organisations, for example, in developing a local health improvement programme. Annual contracts with purchasers will be replaced by three year service agreements and payment for what were known as extracontractual referrals will be made by an undefined method of “retrospective” reimbursement (again to cut management costs). New measures of performance will emphasise outcomes—patients' experience of care, and access to care— rather than productivity. Importantly, no information will be classified as “commercial in confidence,” so, for example, trusts will have to publish the costs of the treatment they offer and a national list of reference costs will be available for comparison.

These changes should bring NHS trusts in from the cold and encourage more openness and collaboration. Better information on comparative costs—which the 1991 reforms largely failed to produce—will strengthen the purchasers' hands to tackle less efficient trusts. The emphasis on measuring outcomes, rather than the number of patients treated, is also welcome. However, the white paper is silent on how payment for trusts will reflect this: rewarding trusts exclusively for higher productivity, as at present, will run against the efforts of primary care groups to treat more patients outside hospital. Retrospective reimbursement for extracontractual referrals could result in abuse, requiring arbitration. Five year agreements may be too cosy, slow down needed change, and prevent frustrated purchasers from seeking alternative providers.

The last main change relates to quality control. A new National Institute of Clinical Excellence is to draw up national evidence based guidelines on the costs and effectiveness of treatments. These will be used to help develop national service frameworks spelling out standards of quality and access to care in specific services, similar to the Calman-Hine recommendations for cancer care.8 Using this information, a national inspectorate—the Commission for Health Improvement—will be responsible for developing and overseeing the quality of clinical care and tackling shortcomings. Chief executives of NHS trusts will be held to account for the first time for clinical quality of care, and health authorities and primary care groups will be able to call in the NHS Executive regional offices or the Commission for Health Improvement if a trust is failing to deliver.

The proposals amount to three main things: softening the harsher edges of the internal market by increasing collaboration and openness; involving all general practitioners in commissioning/purchasing; and strengthening central control over the quality of, and access to, clinical care. They rest on several beliefs, which, as in all policymaking, are the messy product of political values, aspiration, practical judgment, and evidence: that competition in the NHS has generated bureaucracy and inequity; that the most promising way to manage scarce NHS resources is through devolving budgets to clinicians; and that existing systems to monitor the quality of clinical care (Royal Colleges and General Medical Council take note) are poor.

But are they the right way forward? The content of the proposals looks sound. The overall way they will be introduced—bottom up evolution over 5–10 years rather than top down revolution overnight—is sensible and welcome. But their success rests on significant assumptions: that enough general practitioners can, or will, participate; that primary care groups will have enough bite to improve services and will manage budgets effectively; and that health authorities will be able to develop the new primary care groups into robust and cohesive organisations. There are also notable omissions. For example, there is nothing new on overall funding of the NHS except that the changes in themselves will save £1bn in bureaucracy over five years—a fiction since developing the primary care groups will need high start up costs. At best these reforms could give the service a real chance to manage scarcity better—through effective managed care. At worst they could just be the internal market with its motor removed, while perennial problems which undermine support for the NHS— haphazard rationing, financial deficits, the “winter crises,” and lengthening waiting times—go unaddressed.

Finally, it is worth remembering that there are four separate white papers: The New NHS refers to England only. In Scotland's version, Designed to Care,9 health boards, not general practitioners, will be the main purchasers, and instead general practitioners are encouraged to group into primary care trusts and form closer bonds with hospitals. This difference provides a useful opportunity for a natural experiment but undermines the notion of a one nation NHS emphasised so heavily by the current government. Then again, disgruntled general practitioners in England could always move north of the border.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
View Abstract