In his seminal work, A Theory of Justice, Rawls ^{Reference Rawls1} proposed a mechanism that both allows us to formulate a theory of justice as fairness and at the same time provides for a method to test how far our theory meets the desire to treat others fairly. His proposal involved invoking the notion of ‘the original position’ which corresponds to a position in which ‘no one knows his place in society, his class position or social status, nor does any one know his fortune in the distribution of natural assets and abilities, his intelligence, strength, and the like’ (p. 12). This lack of knowledge of one's own original position is termed the ‘veil of ignorance’ and it allows the parties to choose their preferred principles of justice such that no one is advantaged or disadvantaged in the choice of principles by the outcome of natural chance or the contingency of social circumstances. Of course, this is a hypothetical condition that Rawls sets up in order to examine and characterise his preference, a principle of justice, that is not selfishly tailored to one's own circumstances but that genuinely produces fairness in society. Rawls's conceptualisation of justice is an elegant updating of Kant's categorical imperative: act only on that maxim through which you can at the same time will that it should become a universal law. Now, how are these ideas relevant to the paper by Singh et al in this issue? ^{Reference Singh, Khawaja, Pala, Khaja, Krishmanu and Walker2}

Principles of justice

The cost of drugs or services in the National Health Service (NHS) is only relevant in so far as there is a need to consider the distribution of a limited resource. Distribution or allocation of resources by definition requires some guiding principle of justice. But, there are varying terms and concepts of justice including notions of fairness, just desserts, entitlement and need. Distributive justice refers to fair, equitable and appropriate distribution in society determined by justified norms that structure the terms of social cooperation, and problems of distributive justice arise most keenly under conditions of scarcity and competition. ^{Reference Beauchamp and Childress3}

Justice in medicine

In medicine, usually, principles of justice and the processes that guarantee justice operate at policy level and at arm's length from individual patient care. In England, the National Institute for Health and Clinical Excellence through its appraisal mechanisms and guidelines attempts to provide a health economy environment that facilitates or at least maximises fairness. Individual treatment decisions then take place within this ‘fair’ environment. In other words, once a treatment is available within the NHS, individual treatment decisions are made within discrete doctor–patient relationships. These clinical judgements are, or ought to be, solely determined by clinical factors, patient characteristics that are relevant to the treatment under consideration, and patient choice. Non-relevant factors such as age, gender, race and social status that do not have an immediate bearing on the treatment decision do not, or ought not to, influence clinical decisions. Until recently, the cost of drugs too would not have been a consideration for treatment decisions.

Would drug costs considerations endanger the doctor–patient relationship?

The question, therefore, is whether drug costs ought to be a consideration in individual treatment decisions. The doctor–patient relationship is characterised among other things by veracity, privacy, confidentiality and fidelity. Fidelity can be conceived as the obligation to keep faith and promises, fulfil agreements, and maintain relationships and fiduciary responsibilities where fiduciary responsibilities encompass trust and confidence that the doctor will act in the interest of the patient's welfare. Most patients interpret this to mean that the doctor will recommend treatments on their merits not because they are cheap or the cheapest available. If the doctor is aware of the cost of drugs and this awareness influences treatment decisions, it could be argued that ultimately the integrity of the doctor–patient relationship may come under strain from the patient's realisation that other considerations are at play.

Patients' priorities

An unstated assumption underlying the notion that drug cost should guide treatment decisions is that the individual patient is willing to trade their own benefit for someone else's gain. In other words that patient A would be agreeable to the prescription of a cheap drug X so that patient B can have a more expensive drug Y. Harris ^{Reference Harris4} made this case very persuasively in relation to quality-adjusted life-years when he said: ‘It does not follow that where the choice is between three years of discomfort for me or immediate death on the one hand, and one year of health for you, or immediate death on the other, that I am somehow committed to the judgment that you ought to be saved rather than me’ (p. 118). Doctors acting in the interest of the patients directly in front of them cannot at the same time act on behalf of institutions as agents of distributive justice. Neither can doctors assume that patients wish to trade off their own benefits against someone else's.

To return to Rawls and Kant, we ought only to act towards others in such a way that we would enact our actions as a law governing how others act towards us. Somehow, it seems counterintuitive that we would want to be given the cheapest possible drug simply because it is the cheapest possible drug, rather than because it is the most appropriate drug for us whether it is cheap or not. Furthermore, there is very little evidence that it is ever commonly the case that everything being equal, a substantially cheaper drug is both equally effective and harmful except for generic v. branded drugs. If this situation commonly existed, there would be no dilemma for the clinician.

Morally speaking, the best course of action is probably to separate out the policy-making situation that attempts to regulate cost, particularly those policies that aim to ration treatments, from the individual treatment decisions that clinicians are involved in on a daily basis. There are general principles such as the recommendation that generic prescriptions are preferable to branded drugs that can guide clinical decisions. However, the treatment decisions within the doctor–patient relationship ought to remain guided principally by clinical factors and patient choice. This way the insidious erosion of the sanctity of this relationship may at least be ameliorated. Fidelity still ought to count for something.