266 BRITISH MEDICAL JOURNAL VOLUME 291 27 JULY 1985 Vegans Must take vitamin B12 supplements Other May need to improve egetarians their intakes of iron and protein - for example byeating plenty of legumes and nuts Cravings and aversions-At some stage in pregnancy most women experience a distortion of their usual range oflikes and dislikes offoods. Women may develop a nine month aversion to foods they usually like-for example, meat, fried foods, coffee, tea. Contrariwise and at the same time they may experience a craving for certain foods. These are often sweet foods, such as fruits and chocolate ice cream, but some remarkable non- foods-coal, soap, soil-have been recorded. Vegetarians who are pregnant may need extra dietary advice. There are several types of vegetarian. Those most at risk are vegans. It is essential for them to take a supplement ofvitamin B 12 for normal cerebral development ofthe fetus. Other lacto-ovo vegetarians, especially if they are prosperous and belong to a traditional vegetarian group, usually manage well enough but may want or need advice to optimise their protein and iron intakes. Legumes and nuts are an important part ofa balanced vegetarian diet. Professor A Stewart Truswell, MD, FRCP, is Boden professor of nutrition, University of Sydney. The second illustration is reproduced, by permission, from Taggart N, et al. BnrtishJournal ofNutrntion 1967;21:439-5 1; the third from Hytten F and Chamberlain G. ClinicalPhysiology in Obstetrics, published by Blackwell Scientific Publications; the sixth from Papoz L, etal, and the seventh from Naeye RL, both from Maternal Nutnrtion in Pregnancy: Eatingfor Two? edited by J Dobbing and published by Academic Press. Philosophical Medical Ethics Justice and allocation of medical resources RAANAN GILLON In my last article I indicated the wide range of issues concerning justice that are relevant to medical ethics. Even within the sphere of distributive justice the range is dauntingly broad. At one end of the range are what economists call microallocation decisions, of which the most dramatic deal with the allocation of scarce lifesaving resources such as haemodialysis between competing claimants. At the other end are macroallocation decisions taken at a governmental level on the division of the national "cake" between, for instance, health, other welfare, education, arts, and defence budgets. In between are varieties of what one might call mesoallocation decisions. These include decisions on how to distribute the allocated national health budget-the subject matter of the Black report,' which showed so clearly and so shockingly the statistical cor- relations throughout the nation between poverty and low social status on the one hand and adverse health outcomes on the other. (In doing so it also showed the inadequacy of assuming that overtly "health care" decisions are the only or even the most important ones in determining the nation's health.) They also include decisions on how to allocate medical resources at health authority level between the competing medical and other health care claims and decisions within a hospital on how to allocate between competing specialties and firms. More specific still are decisions for allocation among the different members of a hospital firm or health centre; and then come the microallocation decisions of each doctor or health worker distributing his or her available resources among particular patients. Although this range of decisions is exceedingly broad and disparate, all are based on some moral assessment ofhow competing Imperial Coliege of Science and Technology, London SW7 INA RAANAN GILLON, MB, MRCP, director, Imperial College health service, editor, Journal ofMedical Ethics, associate director, Institute of Medical Ethics claims can be fairly adjudicated. They are thus all explicitly or implicitly based on some theory of justice. Preliminary distinctions In the application of such a theory to problems of resource allocation it is worth making some preliminary distinctions. The first is between the formal and substantive contents of the theory. As I indicated in my last article, Aristotle's formal principle (equals should be treated equally, unequals unequally in proportion to the relevant inequality) and the impartiality and fairness it entails are widely accepted in different theories of justice whose substantive contents vary considerably. Among the substantive claims of a theory of justice (the function of which is fair adjudication between competing claims) it is important to distinguish its method for justifying itselfand dealing with competing claims for other theories of justice (for we know that such conflicting claims are likely to occur). That method itself should meet Aristotle's formal require- ments. A democratic voting structure, for example, affords a method of justly choosing between, among other things, the theories of justice preferred by different members of that democratic society. Finally, it is important to distinguish between the theory of justice itself and the equally important practicalities of applying it. Justice is not achieved simply by basing a scheme for resource allocation on a gcod theory of justice. Its decisions must be implemented. Given the wide agreement about Aristotle's formal theory it is worth noting that its acceptance, even before any substantive aspects of a theory of justice are agreed, has important practical implications for resource allocation. Firstly, it requires resource allocation decisions to be made on moral grounds and it rules out partiality and other arbitrary methods of allocation. For instance, neither doctors nor governments can decide that they prefer blondes or whites and justly allocate their resources accordingly, because o n 5 A p ril 2 0 2 1 b y g u e st. P ro te cte d b y co p yrig h t. h ttp ://w w w .b m j.co m / B r M e d J (C lin R e s E d ): first p u b lish e d a s 1 0 .1 1 3 6 /b m j.2 9 1 .6 4 9 0 .2 6 6 o n 2 7 Ju ly 1 9 8 5 . D o w n lo a d e d fro m http://www.bmj.com/ BRITISH MEDICAL JOURNAL VOLUME 291 whiteness and blondness are not morally relevant characteristics (for a way of justifying this conclusion which purports not to depend on prior intuitive determination of moral principles see Hare's paper on relevance2). Secondly, in requiring fairness and thus an element of mutual agreement about the principles for settling conflicts Aristotle's formal principle seems (and this is an empirical not a logical claim) to require implementation of that legal adage that not only must justice be done it must also be seen to be done. People being as they are, only thus is such agreement likely to be obtained and maintained, and only thus are the agreed principles likely to be implemented consistently and impartially as formally required. Accepting this would have important practical impli- cations for the way such decisions are taken. Which moral principles should take precedence? Once one turns to substantive theories of medical resource allocation-answers to the question "What are the relevant in- equalities that justify giving more to some and less to others?"-one meets the same sort of disagreement and complexity about which moral principles should take precedence (see bibliography) as one does with theories of justice generally. The main alternatives, however, are straightforward enough, as my 8 year old daughter briskly reminded me when I was getting into my usual tangle over these impossible questions. How should I choose one out of three dying people to have the only available lifesaving machine? "Well," she told me, sparing a minute or two from her television programme, "you could give it to the youngest because she'd live longer (welfare maximisation), or you could give it to the illest because she needs it most (medical need), or you could give it to the kindest because kind people deserve to be treated nicely (merit). No, you couldn't give it to the one you liked best (partiality), that wouldn't be fair." Nor, she decided, would "eenie meenie minee mo" (lottery) be fair because the one who needed it most, or the youngest, or the kindest might not get it. Nor did she (much to my surprise) think that the Queen should get it in preference to the poor man (social worth)-"because she's got so much already and the poor man hasn't." Of all the methods, her preferred one was to choose the illest because he needed it most-but, not surprisingly, she could not say why that was a better option than the others. Her list of options, however, is remarkably standard, and she joins many doctors in preferring medical need as the criterion of choice. Perhaps unexpectedly medical need correlates most obviously with the Marxist criterion for justice-"to each according to his need." (I should add that this criterion is not exclusively Marxist, that few doctors are Marxists, and that few would accept the first half of the Marxist slogan-"From each according to his ability.") Unfortunately, the concept of a need-as distinct from a desire, for example-is not at all clear.3 Furthermore, it is at least plausible to argue that assertion of needs entails assertion of implied value or values, in which case what are the implied values of the criterion of medical need? Prolongation of life, elimination of disease and attainment of health, and improved quality of life, in the sense of both reduction of suffering and enhancement of flourishing, are all candidates as values correlating to medical need but how are they to be chosen or ranked and what precisely do we mean by these terms? (Their complexities recall the World Health Organisation's definition of health as a state of complete physical, mental, and social wellbeing or the controversies over "sanctity of life," which I outlined in earlier articles.) Thus the apparently straightforward criterion of medical need, while it is undoubtedly a necessary criterion for just distribution of medical resources,4 in no way evades the need to make explicit the moral criteria it encompasses. Nor does it make any easier the choice between competing candidates agreed to be in medical need. Medical success as a criterion A related but by no means identical medical criterion is that of medical success. Medical resources should, it is often claimed, be allocated according to probability of medical success. This adds to the criterion of medical need one of efficiency and, like my daughter's criterion of maximal prolongation of life, corresponds roughly to the welfare maximising objective of utilitarian theories of justice. There are, of course, straightforward cases when the criterion is unproblematic: it would be absurd and wrong to give the only three available pints of a rare blood group to the patient with an incompatible group rather than to the patient with a compatible group. But the criterion of medical success is plagued by all the moral evaluative problems of medical need as well as by additional problems of how to determine medical success: what criteria are appropriate and how is success to be measured? (In this regard the economists' methods of comparing different techniques in terms of "quality adjusted life years" (QUALYs)5 seem to offer conventional methods of clinical trial considerable additional precision for comparative purposes.) The third plank of my daughter's analysis concerned merit and desert-save the life of the kindest because kind people deserve to be looked after. Other merit related criteria are forward looking rather than backward looking. A consultant physician would select "a man who would be able to continue regular work in suitable employment or a married woman with young children . . . in preference to an unemployed labourer with no fixed abode."6 A consultant in clinical renal physiology, also writing about selection for renal dialysis, believed that "[g]ainful employment in a well chosen occupation is necessary to achieve the best results; only the minority wish to live on charity"7 (even when the alternative is death?). More recently a man described as demented, intermittently violent, uncooperative, dirty, doubly incontinent, and with a tendency to expose himself and masturbate while being examined was taken off dialysis treatment "in the patient's best interests."'' How much were the patient's dementia and discomfort the reason for stopping treatment and how much his objectionable behaviour? How much, in general, should a patient's merits and demerits, personal and social, affect his being selected for livesaving medical treatment? Certainly in the medical triage of wartime return to combat duty has been an established medicomilitary criterion for treatment.' In peacetime, however, allocation of medical resources on the basis of a patient's non-medical merits is widely regarded as repugnant. How are we to account for such differing intuitions? And what about extreme cases such as Shackman's hypothetical choice between Fleming and Hitler, where only one of them could be treated?'0 A possible approach Given the fervent disagreement about which moral values should take priority in allocating medical resources it is hardly surprising that doctors on the whole tend to avoid the issue and try to concentrate on doing their best for their individual patients. Two methods of trying to cut the Gordian knot are notable. One commentator, in a different context, has argued that if not all who need scarce lifesaving resources can have them then none should." Two American theologians, Ramsey and Childress, have argued that once a preliminary assessment on broad medical suitability has been made allocation should be by randomisation, either by a lottery or on a first come first served system (with steps taken to ensure that people could not unfairly "use" the system by having inside knowledge).'2 13 I have not yet discovered an acceptable way to give consistent moral priority to any of these substantive criteria for allocation of scarce medical resources (and do not really expect to do so). Calabresi and Bobbitt plausibly suggest that societies tend to try to "limit the destructive impact of tragic choices between fundamental moral values by choosing to mix approaches over time."'4 Within such temporal cycles first one value and then another is emphasised, but "none can, for long, be abandoned." Be that as it may, it would be a mistake to suppose that either the possibility or the need for justice is undermined by such variability and disagreement about which fundamental moral value to abandon when, in a particular situation, not all can be retained. After all, justice is precisely a method for moral resolution of conflicting claims. Provided one or 27 JULY 1985 267 o n 5 A p ril 2 0 2 1 b y g u e st. P ro te cte d b y co p yrig h t. h ttp ://w w w .b m j.co m / B r M e d J (C lin R e s E d ): first p u b lish e d a s 1 0 .1 1 3 6 /b m j.2 9 1 .6 4 9 0 .2 6 6 o n 2 7 Ju ly 1 9 8 5 . D o w n lo a d e d fro m http://www.bmj.com/ 268 BRITISH MEDICAL JOURNAL VOLUME 291 27 JULY 1985 other fundamental moral value is given priority after due consider- ation of the different claims in the light of all of the agreed moral values and in accordance with the formal principle of justice then justice, it seems to me, is done. Thus if, in the context of allocating scarce medical resources, practical systems were set up for resolving conflicts about which value, in a particular case, should have priority, and if those systems took account of the fundamental moral values of respect for autonomy, beneficence, and non-maleficence, and if their deliber- ative structures incorporated Aristotle's formal principle of justice with its demands of formal equity, impartiality, and fairness then they would be just systems and their deliberations could be expected to yield just results despite (perhaps because of) the conflict within them. I doubt if better than that is achievable. Is less acceptable? References 1 Black DAK, Morris JN, Smith C, Townsend P, Davidson N. Inequalities in health: the Black report. Harmondsworth: Penguin, 1982. 2 Hare RM. Relevance. In: Goldman AI, Kim J, eds. Values and morals. Dordrecht: Reidel, 1978:73-90. 3 Daniels N. Health-care needs and distributive justice. Philosophy and Public Affairs 1981;10(2): 146-79. 4 Williams B. The idea of equality. In: Williams B, ed. Problems ofthe self. Cambridge: Cambridge University Press, 1973:230-49. 5 Williams A. The economic role of "health indicators." In: Teeling-Smith G, ed. Measuring the social benefits ofmedicine. London: Office of Health Economics, 1983:63-7. 6 Nabarro JDN. Who best to make the choice? BrMedJ 1%7;i:622. 7 Parsons FM. A true "doctor's dilemma." BrMedJ 1967;i:623. 8 Brahams D. When is discontinuation of dialysis justified? Lancet 1985;i: 176-7. 9 Winslow GR. Triage andjustice. Berkley: University of California Press, 1982:8. 10 Shackman R. Surgeon's point of view. BrMedJ 1%7;i:623-4. 1 I Cahn E, cited by Calabresi G, Bobbitt P. Tragic choices. New York: Norton, 1978:188, 234. 12 Ramsey P. The patient as person. 10th ed. New Haven: Yale University Press, 1979:239-75 passim. 13 Childress J. Who shall live when not all can live? Soundings: An Interdisciplinary Journal 1970;53(4):339-55. (Reprinted in: Gorovitz S, Macklin M, Jameton A, O'Connor J, Sherwin S, eds. Moral problems in medicine. 2nd ed. Englewood Cliffs: Prentice-Hall 1983:640-9. 14 Calabresi G, Bobbitt P. Tragic choices. New York: Norton, 1978:1%. Bibliography (See also cited references) Maxwell RJ. Health and wealth. Lexington: Lexington Books, 1981. Boyd KM, ed. The ethics ofresource allocation. Edinburgh: Edinburgh University Press, 1979. Campbell A. Medicine, health andjustice-the problem ofpriorities. Edinburgh: Churchill Livingstone, 1978. Wolstenholme GEW, O'Connor M, eds. Ethics in medical progress, with special reference to transplantation. London: Churchill, 1%6. Childress J. Priorities in biomedical ethics. Philadelphia: Westminster Press, 1981. Shelp E, ed. Justice and health care. Dordrecht: Reidel, 1980. Engelhardt HT. Shattuck lecture: allocating scarce medical resources and the availability of organ transplantation. N EnglJ Med 1984;311:66-71. Rescher N. The allocation of exotic medical lifesaving therapy. Ethics 1969;79(3):173-86. Parsons V, Lock P. Triage and the patient with renal failure.J MedEthics 1980;6:173-6. Klein R. Rationing health care. BrMedJ 1984;289:143-4. Mooney G. Medical ethics: an excuse for inefficiency? Journal ofMedical Ethics, 1984;10: 183-5. Dr Gillon's articles on philosophical medical ethics will resume later in the summer. Green College Lectures Contemporary challenges in education for the caring professions: education for nursing, midwifery, and health visiting JEAN McFARLANE Diversity of roles in nursing The "greater nursing profession" encompasses many roles as expressed in the Report of the Royai Commission on the National Health Service': "Within nursing there are many levels of skill and different roles.... Nursing is an imunensely varied profession. In hospitals, nurses work in acute, long stay, children's, psychiatric, maternity, and other specialised units. Outside hospital there are health visitors, home nurses, midwives, and nurses working in clinics and in general practice, as part of the primary health care team. Nurses work in administration in the NHS and health departments, in education and research, the armed forces, voluntary organisations such as the Red Cross, occupational health and international agencies. There is a great deal of overlap in the knowledge required in many branches of nursing." This diversity of roles, however, encompasses many different competencies and models of care. The nurse in an intensive care unit needs to combine a high level oftechnological competence with individualised care that supports the patient, who is experiencing Department of Nursing, University of Manchester, Manchester M13 9PT BARONESS McFARLANE OF LLANDAFF, MSC, FRCN, professor and head of department Based on a lecture given at the college on 4 February 1985. not only physical but psychological and spiritual trauma. By contrast, the psychiatric nurse uses the skills of person to person interaction as a therapeutic tool. Much of the work of nurses in the area of mental handicap lies in the use of education strategies tor social skill training. The paediatric nurse needs to be able to integrate therapeutic skills with care that will facilitate and maintain normal human development. Midwifery and health visiting have been identified as separate professional roles. The midwife assists the mother in a normal human function that is only potentially related to pathology (which the midwife must be able to recognise). She is a practitioner in her own right and has considerable decision making autonomy in her area of competence. She spends a high proportion of time using health education skills. The health visitor has a major concern with primary prevention and uses skills of health education, counselling, and social advice. The role of the health visitor is in the domain ofhealth, but she uses knowledge and skills from the disciplines of education and social work. Commonalities and interfaces The nursing curriculum has thus to be designed to prepare for a diversity of roles and models of care. We have to question what commonalities exist in the knowledge and skills needed and to what extent there is a core curriculum or foundation course that, desirably, should be common to all. At what stage should specialisa- o n 5 A p ril 2 0 2 1 b y g u e st. P ro te cte d b y co p yrig h t. h ttp ://w w w .b m j.co m / B r M e d J (C lin R e s E d ): first p u b lish e d a s 1 0 .1 1 3 6 /b m j.2 9 1 .6 4 9 0 .2 6 6 o n 2 7 Ju ly 1 9 8 5 . D o w n lo a d e d fro m http://www.bmj.com/