Equitablerationing specialised children: South Africa

Journal ofMedical Ethics 1999;25:224-229

Equitable rationing ofhighly specialisedhealth care services for children: aperspective from South AfricaWillem A Landman and Lesley D Henley East Carolina University, Greenville, North Carolina, USA, and RedCross War Memorial Children's Hospital, Cape Town, South Africa, respectively

AbstractThe principles of equality and equity, respectively inthe Bill of Rights and the white paper on health,provide the moral and legalfoundations forfuturehealth care for children in South Africa. However,given extreme health care need and scarce resources,

the government faces formidable obstacles if it hopesto achieve a just allocation ofpublic health care

resources, especially among children in need of highlyspecialised health care. In this regard, there is adearth of moral analysis which is practically usefulin the South African situation. We offer a set ofmoral considerations to guide the macro-allocation ofhighly specialised public health care services amongSouth Africa's children. We also mention moralconsiderations which should inform micro-allocation.(J7ournal ofMedical Ethics 1999;25:224-229)Keywords: Children; health care priority setting; justice inhealth care; macro-allocation; micro-allocation; rationing;South Africa

In South Africa, the focus of public medicine hasshifted to provision of primary health care

services, leaving the funding of highly specialisedhealth care in a state of uncertainty. How shouldthe South African government discharge itsconstitutional and moral obligations in respect ofadvanced public medical care for children?

Moral and legal commitmentsIn its white paper, the Department of Health pro-

vides the moral underpinnings of future healthcare for children. In addition to equity, defined as

"(T)he universal provision of services on the basisof need rather than any other criterion",' the gov-ernment believes the "...moral and ethical basisfor the provision ofMCWH (maternal, child andwomen's health)" should be in keeping with inter-nationally agreed upon principles in the UnitedNations Convention on the Rights of the Child.2Services must also be "... efficient, cost-effectiveand of a good quality".3

Correspondingly, the Bill of Rights in the con-stitution entitles every child under 18 to basicnutrition, shelter, basic health care services andsocial services.4 The government has yet to define"basic health care services" to which each child islegally (and morally) entitled. "Basic health careservices" could, for example, refer only to a fixedpackage of primary health care services' availablefor all children, or it could also include healthcare for children with special needs which may berelatively rare and costly to treat but arenevertheless basic in the sense of being necessaryfor survival and adequate functioning. Such care,tailored to the needs of identified specialties,could be viewed as essential highly specialisedcare, or the tertiary equivalent of essentialprimary health care. The scope of the definitionof "basic health care services" will have crucialimplications for children with special needs.Children's socio-economic rights enjoy specialprotection in the constitution since they are con-sidered "clear, near-absolute core entitlements"6which could receive priority were the judiciarycalled upon to allocate economic resources toguarantee these rights.7

In practice, the government has taken steps toimprove children's health care. For example, theintroduction of free health care for all childrenunder six and pregnant women,' improved vacci-nation coverage and extensive clinic-upgradingprogrammes should lead to considerable reduc-tions in morbidity and mortality from acute, gen-erally preventable, diseases. In turn, the govern-ment is proposing that unique and highlyspecialised health care services for children withcomplex conditions be limited to ten centrallyfunded hospitals which will be considered na-tional resources. Access to these services will beon the basis of "need and non-discrimination"and, to this end, the white paper singles out "... thepoorest patients" and "... those outside the imme-diate geographical location of the facilities".9

on June 29, 2022 by guest. Protected by copyright.

http://jme.bm

j.com/

J Med E

thics: first published as 10.1136/jme.25.3.224 on 1 June 1999. D

ownloaded from

http://jme.bmj.com/

Landman, Henley 225

Agreement on definitions of highly specialisedhealth care services, and a formula for their fund-ing, is being negotiated between the national andregional departments of health. These services arelikely to include the following characteristics: lowpatient volume, high cost per patient, significanteconomies of scale, better results with large units,and requiring a complex team ofpeople to providelong term follow-up care (Department of Health,unpublished communication, 1996). Clearly, itwill not be possible to provide these services toeveryone who might benefit from them, makingrationing inevitable.

Against this background of extreme health careneed, scarce resources, and a commitment toseemingly conflicting moral demands of equalconstitutional rights, equity, efficiency, cost-effectiveness and quality, how might the govern-ment achieve a just allocation ofpublic health careresources for children? We offer a set of moralconsiderations that, if acted upon, would bringthese commitments together coherently and givepractical expression to the constitutional right ofchildren to highly specialised health care servicesand to the government's commitment to equity.The considerations we propose are not theoutcome of a philosophical thought-experimentabout what rational choosers might decide behinda veil of ignorance in some original position,'0even though such intellectual exercises yieldimportant normative insights. Instead, the frame-work for our discussion is set by the factual egali-tarian commitments in the Bill of Rights and thewhite paper on health, and the realities of greatneed and scarcity.We confine ourselves to provision of highly spe-

cialised health care services, examples of whichmight include organ transplantation, high tech-nology neonatal care, and treatment of childhoodcancers, cystic fibrosis (CF), and chronic renalfailure. We assume there is a dedicated health carebudget for children, a portion of which will be setaside for highly specialised services even thoughresources are at present being directed away fromtertiary care. We further assume public policyseeks to give content to children's constitutionalright to "basic health care services" by setting andimplementing priorities for a social minimumwhich goes beyond essential primary health care.We believe ours is a more rational approach todistribution than the status quo rationing bydefault, which is uneven and reflects arbitrary his-torical and policy choices. Although our mainfocus is macro-allocation, we mention moral con-siderations that should inform micro-allocation.

Macro-allocationPREVENTrIoNThe current, public policy emphasis on an essen-tial primary health care package is the first rationalstep towards achieving equity (as defined), or, in awider sense, equality of basic opportunities. It is apolicy that expresses a basic moral commitment toprimary prevention, amounting to a form oftriage" (a system of priority treatment to maxim-ise the survival rate) on a national scale in the pre-vailing socio-economic conditions, and given thateverything cannot be done at once.

Essential primary health care, which includesmaternal and child health services, offers the mostcost-efficient way of reducing much of the diseaseburden among children without drastically in-creasing public expenditure,'2 which, in turn,safeguards children's basic opportunities to de-velop their potential. Apart from costs incurred,and initially these may be high, there would becosts avoided, such as the future cost of treatinguntreated disease and disability, as well as lostproductivity over the long term.3

NON-ABANDONMENTSince not all disease is amenable to preventionand acute management, on what rational basiscould one set priorities beyond primary healthcare? Conversely, how do we value equally the livesof children with acute, preventable conditions andthose with complex or long term, possiblyincurable, conditions?'5We propose a basic moral and public policy

commitment to non-abandonment. As far as possi-ble, no identified area of vital health care needshould be excluded. Health is a basic need andhighly specialised health care is an intermediateneed necessary for meeting that basic need.'6 Interms of such a commitment, on a macro levelevery child would have an equal chance to get thehealth care services necessary to develop his or herpotential.Non-abandonment takes seriously one of the

basic elements of a just health care system,namely, fair distribution of the burdens ofrationing."7 18 As tertiary health care services arecut, those with conditions that are expensive andcomplex to treat carry an excessive burden. With-out highly specialised health care services in thepublic sector, these children will be denied accessto life-preserving and life-enhancing health care asability to pay becomes the sole criterion forreceiving health care only available in the privatesector. Since this would systematically disadvan-tage the poorest children with special needs, itwould be a further breach of government policy.


http://jme.bm

j.com/

J Med E


ownloaded from

http://jme.bmj.com/

226 Equitable rationing of highly specialised health care services for children: a perspective from South Africa

In addition to these underlying reasons of egali-tarian justice (relating to equal individual rights,equity, and fairness), there are good utilitarianreasons for a commitment to non-abandonmentof highly specialised health care. It would promoteor benefit other elements of a just health care sys-tem, such as effective health care education andtraining, and the pursuit of high quality biomedi-cal research.'8 As experience is gained in treatingexpensive diseases and conditions, cost may wellbe reduced, and much might be learned in theprocess. '

Rationing formulaWe propose the following, rationing or priority-setting formula which might achieve non-abandonment of highly specialised health care ona macro level in a way that maintains a balancebetween the different areas of vital health careneed and their distribution in the population.20Ascertain the number of children in an identifiedarea of highly specialised health care, for exampleCF, and multiply it by a unit cost (the cost ofstandard medical treatment per child per year) todetermine the overall cost of treating children withCF per year. Having done the same for other areasof highly specialised health care, sum the costs forall identified areas of care, express the total avail-able health care budget for highly specialisedhealth care for children for a particular year as apercentage of that sum, and fund each specialtyaccording to that percentage. In these circum-stances non-abandonment serves the end of egali-tarian justice by making provision for numbers in aspecific area of highly specialised health care needin proportion to the incidence and prevalence ofthat need among the population.For various reasons, the methods of calculating

the cost of treatment would vary from one condi-tion to another. We use a biomedical, specialty-based model to estimate need as it most accuratelyreflects de facto practice, although other ap-proaches to estimation of need are possible.2"However, without accurate national data on theextent of met and unmet need for highlyspecialised health care services, reliance on inter-national incidence and prevalence rates would beunavoidable in the short term.Assume, for argument's sake, that only 55% of

overall required funding is available. Each area ofhighly specialised health care would receive 55%of funds needed. These funds would be chan-nelled to the ten central hospitals offering highlyspecialised health care, with allowances foruneven geographical spread of patients withinspecialties. Shortfalls will have to be accommo-dated on an institution-based specialty or micro

level. However, these could be reduced indifferent ways by macro-level measures, such asreapportionment of the national budget to in-crease the overall health care budget for children,or using some of the funds generated by the intro-duction of a national lottery.The Department of Health's commitment to

good quality care requires that funds made avail-able to specialties in central hospitals be used fortreatment that meets universal standards of highlyspecialised medical care. Such a commitmentwould preclude stretching resources to accommo-date more than the number budgeted for on amacro level on the basis ofnon-abandonment andfunding according to the proposed formula.Would a policy based on non-abandonment

result in the best economic use of resources? InSouth Africa, equality is a constitutionally man-dated ideal, the realisation of which might requireusing health care resources in ways that deliverless rather than more benefit. Equality requiresthat some resources be used to combat even themost severe and costly disease and disability sothat every child has some chance to develop his orher potential, and this means sub-optimal use ofresources in terms of maximising overall beneficialoutcome. This illustrates the fundamental norma-tive tension between two basic moral values thatunderlie distributive choices, namely equality andwellbeing.22

Cost-effectiveness considerationsHowever, according to the government, children'shealth care services must be cost-effective andefficient. If this requires using scarce resources toconfer the most overall benefit or well-being (forexample, in the form of quantifiable extension oflife or improvement of quality of life), on thegreatest number, it will conflict with the govern-ment's own commitments to equity or justice asequality of individual opportunities. By contrast, acommitment to non-abandonment, which sup-ports the equitable provision of highly specialisedhealth care, will preclude allocation of totalresources in the most cost-effective way. In suchan approach, considerations of efficiency (definedas "the attainment of the best outcome or result atthe lowest possible cost"),23 would be limited tochoices between different kinds of treatments forthe same conditions within specialties, and wouldnot underpin choices among services orspecialties.24 Cost-effectiveness gains are possiblethrough the use of marginally less effective, butvery much cheaper, treatments, although suchdecisions may be controversial.25 At an institu-tional level, cost-effectiveness should operatethrough the concentration of specialty services


http://jme.bm

j.com/

J Med E


ownloaded from

http://jme.bmj.com/

Landman, Henley 227

that develop tested, efficient treatment pathwaysand adopt sound managerial principles.26 Like-wise, cost-effectiveness considerations requirethat highly specialised health care, for example,paediatric organ transplantation, be restricted tosome, or only one, central hospital.A policy based on non-abandonment requires

the inclusion ofmost disease categories, regardlessof cost. But is it not more rational to give priorityto cheaper treatments that potentially providesmall benefits to many children rather than focuson expensive treatments that potentially providelarge benefits to a few children, for the same over-all expense? Weighting cost against other morallyrelevant variables means that some health careservices should be channelled towards individualhigh-cost/high-benefit treatments for small num-bers of children.22 For some children with specialneeds, this might be the only way to promoteequality of basic opportunities, given extremescarcity.An example of such a commitment to non-

abandonment would be to regard some, non-experimental, transplant surgery no differentlyfrom other forms of expensive therapy. Indeed, athoughtful comparison of high treatment costs ofpaediatric heart transplantation, CF, childhoodacute lymphoblastic leukaemia, and preterm carefor infants could find no moral justification forarbitrarily excluding any particular treatment cat-egory, simply because it was too expensive."5"Why," the authors ask, "... should a child withincurable heart disease be denied the chance of ahealthy childhood granted to others with similarlydevastating afflictions?", the cost of all being pro-hibitively high.27 Moreover, given good survivalrates, heart, kidney and liver transplantation, inthe South African context, may be cost-effectivesince it is cheaper to transplant these organs thanto treat patients repeatedly for complications ofchronic heart, kidney or liver disease.2" Treatmentof a relatively small number of children withchronic diseases, although expensive, is justifiedwhen cost is weighted against other morallyrelevant considerations, such as equality of oppor-tunity and non-discrimination, and individualbenefit.2

Medical futilityA commitment to non-abandonment also requirestreatment of severe disease and extreme disability.Severity of disease is not, however, an absolutemoral consideration and, given scarcity and greatneed, other morally relevant variables, such as costand beneficial outcome, must be balanced in thepursuit of equality of basic opportunities.22Therefore, we would not support treatment,

beyond comfort care, of a child whose clinicalcondition is so severe that no beneficial outcomewhatsoever is likely. For example, aggressive treat-ment, beyond comfort care, of an anencephalicinfant is futile by all generally accepted goals ofmedicine, such as caring, healing or saving life.Some might hold the extreme position that a goalof medicine is to preserve life at all cost, evenphysiological life with rudimentary or no con-sciousness. We reject this view. However, all judg-ments about medical futility are value-laden andshould not be used as a method of rationing care.29Even a judgment about medical futility thatappeals to purely physiological outcomes is notvalue-neutral but contains embedded normativechoices, for example, about the value of a life andthe goals of medicine.30

In sum, we believe our commitment tonon-abandonment is the best way to balance allmorally relevant considerations in a situation ofextreme health care need and resource scarcity. Itaccommodates the formidable tensions that existbetween equality of basic opportunities andachieving the most beneficial overall outcome, aswell as other tensions created by factors such ascost, and prevalence and severity of disease.

Micro-allocationOnce a procedure is in place, on a macro level, toprevent arbitrary exclusion of categories of highlyspecialised care, it seems rational to use availablehealth care resources to generate the mostpossible good, and this suggests a moral commit-ment to maximise benefit or promote the mostbeneficial outcome.

If we assume 45% underfunding for all special-ties, and if specialties are funded according to theproposed formula, then only 55% of identifiedneed in a specialty can be treated. Health careprofessionals will have to decide which childrenwill receive highly specialised care that cannot beprovided to everyone who needs it. On what justi-fiable moral basis could such exclusion be done ona micro level, since the public system is designedprecisely to assist those without the ability to pay,and since children present a special case andshould not be judged in terms of standards such asprevious contribution to society, effort, or prudentplanning?We propose, within specialties, children be

selected, in the first instance, according to a "firstcome, first served" (queuing) rule."1 This kind ofrandom selection aimed at addressing equity on amicro level, in respect of children with equal needfor survival or adequate functioning, could bemade fairer by adopting corrective measures sen-sitive to considerations such as access to care and


http://jme.bm

j.com/

J Med E


ownloaded from

http://jme.bmj.com/

228 Equitable rationing of highly specialised health care services for children: a perspective from South Africa

geographical location. Does this mean thatpatients already receiving treatment have absolutepriority over those who arrive later with moreurgent needs or better chances of success? Webelieve not. For example, in a paediatric intensivecare unit (PICU), there may be times when achild, with a more urgent need or a greater likeli-hood of benefit, might need to displace a childwith a poor prognosis. Although admission to aPICU establishes a presumption in favour of con-tinued treatment, requirements of medical utilitysometimes justify early discharge to make roomfor others with more urgent need or higher prob-ability of benefit. This does not constituteabandonment if alternative, more appropriate,care is provided.3'Queuing would provide a pool of potential

patients who, depending on the nature of theircondition, would need to meet a set of selectioncriteria32 based on medical utility, defined as themaximisation of the welfare of patients in need oftreatment. Where rationing is unavoidable, medi-cal utility is justified because scarce resourcesshould only be distributed to individual childrenwith a reasonable chance of benefiting from them.Failure to do so would result in a waste of scarceresources.3' The development of explicit criteriawould also be in line with a recent constitutionalcourt ruling on access to life-prolonging re-sources. According to the judgment:

" ... while each claimant seeking access to publicmedical resources is entitled to individualisedconsideration, the lack of principled criteria forregulating such access could be more open tochallenge than the existence and application ofsuch criteria".33

Where medical utility is roughly equal betweenchildren eligible for the same resource (for exam-ple, a donor heart), we propose queuing, randomi-sation or a lottery, whichever procedure is mostappropriate and feasible under thecircumstances.3' Here, where there are no obviousdifferences among patients in terms of medicalneed and likelihood of success, random selectionis justified on grounds of equity and equal evalua-tion of lives.On the micro level, because rationing decisions

are value laden, determination of medical utility(including choice and weighting of patient selec-tion criteria) will require constant medical andpublic scrutiny. Significantly, the governmentproposes periodic national health summits andthe formation of hospital committees to encour-age public participation in the planning andprovision of health care services.34 However,

democratic debate about normative questionssuch as fair rationing procedures will likely createits own in-built tensions, as society is forced tomake tragic choices between individual lives andthe overall good, in other words to put a price onlife.22 Still, such debate would form part of anhonest, open and rational attempt to achieveequitable rationing or egalitarian justice consist-ent with the demands of a democratic constitu-tion, and given the intractable combination ofextreme need, scarce resources, and the inevitabil-ity of rationing exclusions.Does micro-allocation mean many children

needing highly specialised health care will beabandoned after all? At least on a macro level, apolicy based on non-abandonment should giveevery child an equal chance to get the health carenecessary for developing basic opportunities, noone being discriminated against on account offactors beyond his or her control such as cost oftreatment, or incidence or severity of disease.Importantly, non-abandonment, as proposed, willmake a material difference to where health careresources are channelled. Therefore, a distribu-tion policy premised on non-abandonment willyield outcomes entirely different from otherpossible approaches. For example, a straightfor-ward overall cost-benefit approach would likelyresult in the exclusion of whole categories of chil-dren with special needs (for instance, those need-ing organ transplantation) because resourcescould be used more cost-effectively on childrenelsewhere in the health care system. However, ona micro level, given extreme need and scarcity,rationing exclusions (and therefore tragic choices)are inevitable. Still, we believe a policy based onnon-abandonment is the fairest way to giveconcrete expression to the government's moraland legal commitment to greater equality andnon-discrimination in the distribution of basichealth care services among children.

ConclusionThe principles of equality and equity, respectivelyin the Bill of Rights and the white paper on health,provide the moral and legal foundations for futurehealth care for children in South Africa. In reality,extreme need and scarcity will seriously limit thegovernment's ability to fulfil its commitment toequal health care rights for children, especiallytheir equal right to highly specialised health care.To achieve equality of basic opportunities on

the macro level, we propose a national healthpolicy for children premised on the principles ofprevention and non-abandonment. We argue thatall effective highly specialised health care servicesfor children should be accommodated in the


http://jme.bm

j.com/

J Med E


ownloaded from

http://jme.bmj.com/

Landman, Henley 229

health care system. Apart from an egalitarian jus-tification, it could be argued that abandoningexisting specialties would have adverse conse-quences throughout the health care system andthat, once abandoned, they would be very costly toresurrect. We reject grounding the macro-allocation of health care resources on purely cost-effectiveness calculations because they are indif-ferent to equality of opportunity and equity.Indeed, attaining equality on the macro level,through non-abandonment, may require usingresources that deliver less rather than more overallbenefit. Non-abandonment also addresses theformidable tensions in health care resourceallocation generated by considerations of equalityand utility, treatment costs, and severity andprevalence of disease and disability. Whereas apolicy grounded in cost-benefit and cost-effectiveness principles would likely result inwhole categories of highly specialised health carebeing excluded because better outcomes for chil-dren could be obtained elsewhere in the healthcare system, non-abandonment on a macro levelwould give most children with special needs anequal chance of realising their potential.Given extreme need and scarcity, it would not

be possible to provide highly specialised healthcare services to all children who might benefitfrom them, making rationing inevitable. To thisend, we propose that within each specialty,children be selected on a "first come, first served"basis. Thereafter, to promote the best use ofscarce resources with the least waste, childrenshould meet selection criteria based on medicalutility. Where medical need and probability ofsuccess are roughly equal among eligible childrenwith similar conditions, equal evaluation of livesrequires selection according to an appropriateform of random selection.

Willem A Landman, DPhil, is Professor in theDepartment ofMedical Humanities, School ofMedi-cine, East Carolina University, Greenville, NorthCarolina, USA. Lesley D Henley, PhD, is a SeniorLecturer in the Department of Paediatrics and ChildHealth, Institute of Child Health, Red Cross WarMemorial Children's Hospital, Cape Town, SouthAfrica.

References1 Department of Health. White paper for the transfornmation of the

health system in South Africa. Pretoria: Government Printer,1997:224.

2 See reference 1:103.3 See reference 1:98.4 Constitution of the Republic of South Africa Act, No 108 of 1996.5 See reference 1:36-8.6 De Vos P. Pious wishes or directly enforceable human rights?:

social and economic rights in South Africa's 1996 constitution.South African Journal ofHuman Rights 1997;13:67-107.

7 See reference 6:88.8 McCoy D. Free health carefor pregnant women and children under

six in South Africa: an impact assessment. Cape Town: ChildHealth Unit, University of Cape Town for the Health SystemsTrust, 1996.

9 See reference 1:176-7.10 Rawls J. A theory of justice. Oxford: Oxford University Press,

1972.11 Beauchamp TL, Childress JF. Principles of biomtedical ethics [4th

ed]. New York, Oxford: Oxford University Press, 1994: 385.12 World Bank. Development report: investing in health. Oxford:

Oxford University Press, 1993.13 Hussey GD, Lasser ML, Reekie WD. The costs and benefits of

a vaccination programme for haemophilus influenzae type Bdisease. South African MedicalJournal 1995;85:20-5.

14 Viljoen DL, Buccimaza S, Dunne T, Molteno C. Theprevalence and prevention of neural defects in Cape Town.South African Medical Journal 1995;85:630-2.

15 Hugo-Hamman CT, Vosloo SM, de Moor MMA, Odell JA.Paediatric heart transplants: should we do them? South AfricanMedicalJournal 1991 ;80:434-6.

16 Doyal L, Gough I. A theory ofhunman need. London: Macmillan,1991.

17 Buchanan AE. Privatization and just health care. Bioethics1995;9:220-39.

18 Benatar SR. What makes a just healthcare system? British Med-cial Journal 1996;313: 1567-8.

19 Harris J. More and better justice. In: Bell M, Mendus S, eds.Philosophy and medical welfare. Cambridge: Cambridge Univer-sity Press, 1988:75-96.

20 Doyal L. Needs, rights, and equity: moral quality in healthcarerationing. Quality in Health Care 1995;4:273-83.

21 McPherson M, Arango P, Fox H et al. A new definition of chil-dren with special health care needs. Pediatrics 1998;102:137-40.

22 Landman WA, Henley LD. Tensions in setting health care pri-orities for South Africa's children. Journal of Medical Ethics1998;24:268-73.

23 See reference 1:224.24 Klein R. Priorities and rationing: pragmatism or principles

[editorial]. British Medical Journal 1995;311:761-2.25 Lee C, Sabin C, Miners A. High cost, low volume care: the case

of haemophilia.British MedicalJournal 1997;315:662-3.26 Marais AD, Benatar SR. Planning rational management of

chronic diseases-lessons from a lipid clinic. South AfricanMedical_Journal 1995;85:340-1.

27 See reference 15:435.28 Kahn D. Supraregional funding for transplantation [editorial].

South African Medical J7ournal 1994;84:590- 1.29 Jecker NS. Futility and rationing. American Journal ofMedicine

1992;92: 189-96.30 Halliday R. Medical futility and the social context. Journal of

Medical Ethics 1997;23: 148-53.31 See reference 11:378-86.32 Kilner JF. Microallocation. In: Reich WT, ed. Encyclopedia of

bioethics: vol 2 [rev ed]. New York: Macmillan, 1995:1075-84.33 Soobramoney v Minister of Health (Kwazulu-Natal) CCT 32/97

(27 November 1997).34 See reference 1:33-6.


http://jme.bm

j.com/

J Med E


ownloaded from

http://jme.bmj.com/

Documents

Equitablerationing specialised children: South Africa