Intended for healthcare professionals

Analysis And Comment Health service reform

Seven years of feast, seven years of famine: boom to bust in the NHS?

BMJ 2006; 332 doi: (Published 13 April 2006) Cite this as: BMJ 2006;332:906
  1. Alan Maynard, professor of health economics (akm3{at},
  2. Andrew Street, senior research fellow2
  1. 1 Department of Health Sciences, University of York, YO10 5DD
  2. 2 Centre for Health Economics, University of York
  1. Correspondence to: A Maynard
  • Accepted 3 April 2006

Service improvements in the NHS have not kept pace with the dramatic increase in expenditure. As expenditure growth slows, attention must focus on reforming the incentives faced by NHS staff and institutions rather than on indulging in yet another reorganisation

The Labour government was elected in 1997, trading on the popular belief that the National Health Service was “safe in their hands.”1 In office the government initially adhered to expenditure levels inherited from the Conservatives and chose to focus on structural and process reforms, including reorganisation of purchasing bodies, the introduction of national service frameworks, and the creation of regulatory bodies such as the National Institute for Clinical Excellence (NICE) and Commission for Health Improvement (CHI).

In 2000, breaking from the Conservative's expenditure plans, the government decided to initiate a programme of NHS investment at a level (annual budget increases of 7%) and for a period (seven years) unequalled in any other healthcare system.2 The Wanless report was used to provide some intellectual support for this bold policy.3

Enjoying the feast

Almost seven years on, it is clear that the spending has improved services. More patients are being treated and cared for—in acute and community hospitals and primary care, and by newly established service providers such as walk-in centres and NHS Direct.4 Waiting times for elective surgery have been driven down, by targeting people who have been waiting a long time, so that no one waits more than six months for an inpatient admission.57 The national service frameworks have significantly improved the quality and timeliness of service delivery and there has been investment in new facilities and some reduction in the awesome maintenance backlogs inherited in 1997.7 8 But these improvements have not kept pace with the spending increases. From 1998-9 to 2003-4 NHS productivity seems to have declined.4 9

The government's funding increase of 2000 was a historic opportunity to change the culture of NHS reform. It could have been used to fundamentally restructure the incentives faced by NHS staff and organisations, and this was the hope. Negotiations were opened with consultants and general practitioners about their contracts, and hospitals were to be paid on the basis of their activity, under arrangements termed “payment by results.”10 But by early 2006, the medical profession had secured large pay increases subject to limited additional contractual conditions, payment by results had been suspended because of problems in setting tariffs, and the chief executive of the NHS had resigned in the face of criticism of the organisation's mounting deficit, which was predicted to increase from £250m (€358m, $433m) in 2004-5 to £790m in 2005-6.

What went wrong?

The fundamental problem was that, as funding flowed into the NHS, many spending controls were overridden as politicians and managers concentrated on targets for activity and service delivery. The lack of control is most evident with respect to the role of NICE, in pay negotiations, and in contracting arrangements.

Embedded Image

The NHS has already had its years of plenty, so any famine will be quite a problem (Joseph Interprets Pharoah's Dream by Julius Schnorr von Carolsfeld, 1860)


Problems with NICE

There have been four problems with NICE. Firstly, after appraising new healthcare technologies, it has usually concluded that some subgroup of patients will benefit even at a high cost per quality adjusted life year (QALY). By approving technologies of marginal cost effectiveness, NICE is inflating NHS spending with small health gains for the population.11 Secondly, it has failed to appraise, sanction, and expedite the removal of redundant technologies. Thirdly, much of its advice has been implemented unevenly12 because of local financial constraints, creating both a new form of post-code rationing and pressure to increase NHS spending. Finally, it has to accept the prices set by pharmaceutical companies. NICE cannot be relied on to control expenditure if it fails to exercise a rationing function and has no influence over prices.

Recruitment and retention

A substantial proportion of the funding increase has been swallowed up with pay awards. Although labour costs consume 70% of NHS expenditure, during the 1990s it was increasingly apparent that the NHS employed relatively low numbers of doctors, nurses, and other staff: staffing levels in Germany, for instance, are almost twice those of the NHS.13 To address the situation, efforts were made to improve recruitment and retention. Recruitment of staff focused on removing barriers to entry, mainly by expanding the intake to medical and nursing schools by over 50%.14 This was complemented by aggressive international recruitment of doctors and nurses. However, the requirements for more staff may have been overestimated, with emerging evidence of unemployment in some professions.15 It has not helped that the price of labour increased dramatically.

The medical profession has benefited from substantial pay awards, with the 2004 pay settlements for general practitioners and consultants increasing annual costs by several billion pounds.16 The generosity of these awards may have stemmed from a belief that they were necessary to ensure staff retention. The medical profession has long voiced discontent with pay, but rarely do doctors vote with their feet. And with good reason: NHS doctors are now the best paid in Europe.17 Pay increases should have made conditional on productivity gains but were not. A decade long decline in consultant activity seems to have been unaffected by the 2004 pay awards.18

The pay award for general practitioners increased costs in two ways. Firstly, practitioners could opt out of out of hours cover by foregoing £6000 of salary. Many took up this option, and the rate paid for general practitioners who opted back in rose to over £90 per hour and £1000 for an overnight locum session. Secondly, general practitioners were given fees per service for targeted care of chronically ill patients. This “quality and outcomes framework” increased general practitioner pay considerably, but it is unclear whether these extra payments were made for work previously done or for improved service provision. The pay rise has been so generous that hospital doctors and general practitioners can work fewer hours and still be better off financially—so, in effect, the NHS is paying more for less.

At the same time, there was a radical reform of all other staff grades in the NHS. “Agenda for change” involved the appraisal of all staff in relation to a new, simplified grading system, with “protected status” accorded to staff grades in short supply, such as clinical coders. This has inflated annual expenditure by about £500m.16 The government now plans to review training and education to facilitate increased flexibility in work patterns. This timing is poor: strategies to improve labour productivity should have been considered in parallel with the grading review.

Payment by results

Having conceded large pay increases, the government introduced payment by results. This was intended to underpin the agenda for improving patient choice, to increase activity both from NHS and independent service providers, and to encourage service providers to compete on the basis of quality not price. Laudable aims, which have met with only partial success in that payment by results increased activity in 2005-6 but contributed to the deficit. Implementing the system for 2006-7, after the withdrawal of the initial set of tariffs in February 2006, remains problematic.

The problem is that early policy design has concentrated more on satisfying accounting rules than on devising appropriate incentives. As in its dealings with the drug industry, the Department of Health has acted as a price taker, setting tariffs on the basis of the average costs reported by providers. And it set no ceiling on the amount of extra activity it was prepared to fund, leaving it to primary care trusts to find ways to exert volume controls.

However, primary care trusts lack the ability to curb activity and, therefore, expenditure.19 They are in a weak bargaining position compared with hospitals because they lack timely information to monitor and hold providers to account for their behaviour. Primary care trusts also have few levers with which to influence referrals of self employed general practitioners, whose decisions are clearly crucial in ensuring that trusts stay within budget. The main influence that the trusts can bring to bear over general practitioners is by introducing practice based commissioning budgets, an updated version of the general practitioner funding scheme that Labour dismissed in its 1997 white paper.20 21 The problem is that, although general practitioners have incentives to underspend their budgets, the penalties for overspending are virtually non-existent. The result is that expenditure control through the contracting process has been all but lost.

Preparing for famine

The end of the annual funding increases is imminent with NHS expenditure growth unlikely to exceed 3% a year after 2008.22 What needs to be done to prepare for the famine years ahead? The government has responded by instigating yet another reorganisation of the NHS. This will create a period of organisational stasis and divert attention from the real task of improving the NHS. Rather than reorganisation, what should be the priorities for the future?

Reclaiming NICE—NICE needs to be re-established as an independent body, particularly after being undermined by political and media pressure for fast-track approval of trastuzumab (Herceptin).23 But NICE needs a wider remit, with the authority to ration services, not merely recommend them; to remove redundant technologies; and to negotiate on the price of drugs. Price negotiation may be forced on the government depending on the outcome of the investigation by the Office of Fair Trading into the Pharmaceutical Price Regulation Scheme.24 Several commentators have argued that, in order to exercise greater cost control, the arbitrary cut-off level for approval of drugs should be reduced from £30 000 per QALY to about £15 000 per QALY.11 25

Summary points

Expenditure controls have been relaxed during seven years' “feast” of increased NHS funding

NICE's limited powers, generous pay awards, and payment by results have driven cost increases

After 2008, the growth in NHS expenditure will be reined in

To cope with this “famine” NICE requires a wider remit, incentives need to be overhauled, and health outcomes data should be collected

Providing incentives—With no immediate possibility of restructuring consultant or general practitioner contracts, other tools to reward clinical performance must be deployed. Clinical excellence awards, for instance, should be awarded explicitly on the basis of activity and outcomes. Practice based budgets for general practitioners must be based on sensible incentives, and the quality and outcomes framework renegotiated so that desirable and affordable objectives are rewarded.

Developing a structured payment system—A properly structured payment regime is needed. The simplistic formulation of payment by results has produced adverse behavioural responses and financial meltdown.26 27 Paying for health care is a complex matter and requires a sophisticated payment system. Other countries have recognised this. The pursuit of simplicity by the English has come at the expense of a coherently incentive regime.28

Collecting outcome data—We need to collect better information and encourage all staff to use it. How do we know what progress the NHS is making if no data are available on health outcomes? BUPA (a private health insurance scheme) has been collecting and analysing such data routinely for elective care since 1998 for 27 private hospitals. The Department of Health has indicated a desire to apply BUPA's example to the more complex mix of NHS activity,29 but the funding crisis will probably undermine this opportunity.


The deficiencies inherent in policy in the recent “feast” years were predictable and foreseen.30 Now what are needed are “hard” budgets and improved incentives that create behaviours that are consistent with efficiently improving the health of the nation. The feast is over. The famine will be harsh for patients, practitioners, and politicians.


  • Contributor and sources AM and AS wrote the paper. The views expressed are those of the authors and not necessarily of the funding body.

  • Funding AS is funded by the Department of Health in England as part of a programme of policy research at the Centre for Health Economics, University of York.

  • Competing interests AS worked as an economic adviser in the Department of Health from February 2005 to January 2006.


  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.
View Abstract