Intended for healthcare professionals

Education And Debate

Rationing and the health authority

BMJ 1998; 317 doi: (Published 17 October 1998) Cite this as: BMJ 1998;317:1067
  1. Tony Hope, readera,
  2. Nicholas Hicks, consultantb,
  3. D J M Reynolds, consultantc,
  4. Roger Crisp, fellowd,
  5. Sian Griffiths, directorb
  1. aDivision of Public Health and Primary Care, Institute of Health Sciences, Oxford OX3 7LF
  2. bDirectorate of Public Health and Health Policy, Oxfordshire Health Authority, Oxford OX3 7LG
  3. cDepartment of Clinical Pharmacology, John Radcliffe Hospital, Oxford OX3 9DU
  4. dSt Anne's College, University of Oxford, Oxford OX2 6HS
  1. Correspondence to: Dr Hope
  • Accepted 11 June 1998

Most governments would like to pretend that health care rationing does not exist. Those working in the NHS know that it does. There is simply not enough money for all patients to receive the best treatment.1 A few governments—the Netherlands, New Zealand, and Sweden, in particular—have grasped the nettle.24 In the United Kingdom, major decisions about what to fund or, more critically, what not to fund have been made at local level by health authorities, fundholding general practices, and, under the provision of the new white paper, primary care groups. There is very little central guidance on how these decisions are to be made. Health authorities are continually faced with reasonable requests for extra funding. To fund something new, something else of value has generally to be sacrificed, and this happens despite all efforts to eliminate ineffective health care. This paper describes one health authority's approach to this problem.

Summary points

Oxfordshire Health Authority has set up a priorities forum to provide advice on rationing decisions in health care

The forum aims to ensure a reasonable and evidence based process for decision making

The decision process focuses on three key areas: evidence of effectiveness, equity, and patient choice

A wide range of policy issues and individual cases has now been considered

Key issues that remain include determining relative funding for each area of health care, achieving consistency in spending for treatments with broadly similar effects, and involving the public

The priorities forum

Health authorities and primary care groups will increasingly need to be able to justify their decisions to the wider community. There are two key questions:

  • What was the process by which the decision was made?

  • What were the grounds for making the decision?

Oxfordshire Health Authority has set up a priorities forum to provide advice on rationing decisions. One aim of the forum is to provide a reasonable “due process” for decision making. It is a subcommittee of the health authority, and meets once a month to discuss individual cases and policy. The forum brings together general practitioners, medical directors of the local NHS trusts, health authority staff, hospital doctors, and non-executive members of the health authority. Members of the local community health council attend as observers. The other aim of the forum is to make decisions on the right grounds. To help it to do this the forum set up a subcommittee to advise on an ethical framework.

Ethical framework

This framework focuses on three key areas. These are evidence of effectiveness, equity, and patient choice.


Value and efficiency are also considered within effectiveness. “Effectiveness” means the extent to which the treatment (or other healthcare intervention) achieves the desired effect (notably the proportion of patients who would be expected to show the effect). “Value” means a judgement on how valuable that effect is in the relevant individual(s) relative to the value of other treatments. The impact of a treatment is its value, weighted for effectiveness. Efficiency is the impact per unit cost.

In making judgments about the value of a treatment, three factors are of particular relevance: the additional length of life that the treatment brings, the contribution that the intervention makes to the patient's wellbeing, and the level of need of those who benefit from the treatment.5 The third factor might have the effect, for example, of giving a higher priority to immediate treatment after a heart attack than to treatments of less urgent importance to patients' wellbeing, even if the latter lead to more lives saved overall.

Clearly, ineffective treatments should not be purchased. The forum is often faced with making decisions where evidence is insufficient, one way or the other. In these cases, the question of whether or not to purchase will depend on the likely efficiency taking into account the quality of evidence.


The basic principle of equity is that equals should be treated equally. In making its purchasing decisions, the health authority tries to aim for this ideal. For example, the decisions concerning extracontractual referrals should be consistent with those concerning the main contracts. The forum has decided that there should be no discrimination on grounds of employment status, family circumstances, lifestyle, learning disability, age, race, sex, social position, financial status, religion, or place of abode.

Decision processisotretinoin treatment

  • The issue—Dermatologists asked the health authority for extra funding to pay for isotretinoin in the treatment of severe acne

  • Evidence on effectiveness—Isotretinoin improves severe acne in most patients, often when other treatments have failed. It improves the patient's appearance in the short to medium term at a time of life (adolescence and early adulthood) when many are very conscious of their appearance. It may also reduce long term scarring

  • Value and equity—Isotretinoin costs considerably more than alternative treatments, at least in the first year.6 Acne is not life threatening. With adequate funding, all patients who might benefit could have an equal chance of receiving treatment

  • Patient choice—Many patients would choose isotretinoin if it were available. Outcome measures include issues of great importance to patients, such as appearance

  • Priority—Dermatologists considered that isotretinoin should have a higher priority than some other services currently provided

  • Decision—The forum agreed that isotretinoin was a high priority, but did not agree that resources should be taken from other services to increase the size of the dermatology resource envelope. Dermatologists were asked to identify other services they provided that were of lower priority

  • Conclusion—Dermatologists returned to the priority forum with suggestions of lower priority services. It was agreed that they should stop providing electrolysis

Patient choice

The authority will not make an exception to a decision simply because a patient chooses an ineffective intervention, since this would deny another patient access to a more effective treatment. However, patients can choose between treatments of similar efficiency. The forum sees it as important, however, that trials testing the effectiveness of treatments should include outcome measures important to patients.

Decision processriluzole treatment

  • The issue—Riluzole is a new drug licensed for use in motor neuron disease. Should the health authority pay for its prescription?

  • Effectiveness—Riluzole slows progression in this rare and universally fatal disease, prolonging life by an average of 3 months. It does not improve function and quality of life but may delay deterioration and the need for interventions such as tracheostomy

  • Value and equity—It costs about =A335 000-=A345 000 to buy one year of additional life with riluzole.7 By contrast, it costs about =A320 000 to buy a year of life with renal dialysis, and about =A35000 to buy a year of life with statins in high risk patients. Many of these treatments are not yet delivered to all those who might benefit

  • Patient choice—Many patients with motor neuron disease want to receive riluzole

  • Priority within the neurosciences resource envelope—Neurologists and neurosurgeons did not regard riluzole as having a higher priority than other services that competed for the same resources such as βinterferon and increased surgical capacity to treat young adults with subarachnoid haemorrhage

  • Decision—The forum agreed with the neurologists' assessment. It also wanted to increase the size of the neurosciences resource envelope, but recommended that increased neuroscience resource be used to ensure that young adults with treatable subarachnoid haemorrhage were offered appropriate treatment

  • Conclusion—Except in unusual circumstances, the authority does not purchase riluzole

Putting the principles to work

“Envelopes” of money

The health authority provides each clinical area with an “envelope” of money. It leaves many of the decisions about how best to use that money to the doctors working within these areas who are often in the best position to decide about priorities.

When doctors ask the health authority for extra funding to carry out a new treatment or service the first question asked is, why shouldn't this new treatment be funded within the current “envelope”? Is this new treatment more valuable than some of the other activities that are being carried out within this clinical area? If it is, then perhaps these less valuable treatments should not be funded. This is illustrated in the box on isotretinoin treatment.

Marginal health care activities within each envelope

However, perhaps more money should be given to this clinical area at the expense of some other clinical area—assuming that overall resources are fixed. The forum, therefore, needs to consider whether the relative sizes of the envelopes are right. To do this it needs to identify the care at the margins of being resourced within each envelope of money. Through comparing these “marginal” activities between different clinical areas, decisions can be made as to whether the relative sizes of each of the envelopes should be changed.

Comparing different effects

Choices sometimes need to be made between treatments with different effects—for example, one prolonging life, the other enhancing life. The need for such comparisons can be lessened by improving consistency across treatments with similar overall effects. Thus, many treatments used in quite different clinical areas extend life without curing the underlying condition. For example, the following treatments palliate symptoms and prolong life for up to one year: riluzole for motor neuron disease, chemotherapy for non-small cell carcinoma of the lung, and paclitaxel for breast cancer. Are there gross differences in the amount of money spent to extend life by a month across these three areas? By comparing treatments with broadly similar outcomes, we are no longer comparing chalk with cheese, though perhaps we are comparing stilton cheese with cheddar.

Priorities forum in action

The forum has considered a wide range of policy issues as well as individual cases. In the decision process box we give two specific examples of its work.

The future

The forum's work is now at the stage where it needs to tackle three key issues:

  • What should the relative sizes of the health care envelopes be?

  • Is the authority consistent in how much it spends, in different clinical areas, for treatments with broadly similar effects?

  • How can the public be involved?


We thank Paul Brankin, Penny Thewlis, Tom Jones, and Vivyenne Rubinstein, who are all members of the ethics subgroup as well as the priorities forum, for their input to the ideas expressed in this paper.


  • Funding None.

  • Conflict of interest None.


View Abstract