implementing comprehensive and decentralised health systems

1This article is published in a peer reviewed section of the International Journal of Integrated Care

International Journal of Integrated Care – Vol. 1, 1 March 2001 – ISSN 1568-4156 – http://www.ijic.org/

Implementing comprehensive and decentralised healthsystemsThe quest for integrated care in post-apartheid South Africa

David Sanders, Director, School of Public Health, University of Western CapeMickey Chopra, Lecturer, School of Public Health, University of Western Cape

Correspondence to: David Sanders, E-mail: [email protected], Mickey Chopra, E-mail: [email protected]

Introduction

Before 1994, South Africa’s public health serviceswere racially segregated, highly unequally distributedbetween the rural and urban areas and rich and poorcommunities, overwhelmingly hospital based andcurative in their emphasis. Management responsibilitywas also fragmented between different authorities—national, provincial, local and the former homelands.Since 1994 policies and structures have been put inplace to reduce the previous fragmentation by creatingan integrated national health system, based upon pri-mary health care, and decentralized in terms of man-agement to geographically defined districts. Theprivate health care sector, which caters for approxi-mately 20% of the population, yet consumes almost60% of the resources (financial and human) devotedto health care, still remains largely unaffected by thesechanges in policy and organization.

The challenge of transforming the public health caresector has raised questions about the optimalapproach to conceptualizing and providing compre-hensive and integrated health care in the present inter-national and national policy climate. This articleinterrogates some of the debates, which have markedthe transformation process in South Africa. In partic-ular it explores the tension between a comprehensiveprimary health care approach (CPHC) as elaboratedin the Alma Ata Declaration in 1978 and more recentapproaches, variously termed ‘selective primary healthcare’ w1x and the ‘new universalism’ w2x. An attemptto develop comprehensive care and an integrated,decentralized system is illustrated through the useof a case study, which centers on implementationof nutrition policy in South Africa’s Eastern CapeProvince.

Selective primary health care andthe primary health care package

The concept of PHC, codified in the Alma Ata Decla-ration explicitly outlined a strategy, which would

respond more equitably, appropriately and effectivelyto basic health care needs and also address theunderlying social, economic and political causes ofpoor health. Certain principles were to underpin thePHC approach (PHCA), namely, universal accessibil-ity and coverage on the basis of need; comprehensivecare with an emphasis on disease prevention andhealth promotion; community and individual involve-ment and self-reliance; inter-sectoral action for health;and appropriate technology and cost-effectiveness inrelation to available resources.

An important response to the challenge of transform-ing the South African health system has been arecently proposed ‘Core Package of Primary HealthCare Services’ w3x. Through the identification of ‘criti-cal’ services it is hoped that a set menu of servicescan be offered at each level of care—community, pri-mary (clinic and health centers) and hospital. Forinstance, for malnutrition the core package consists ofproviding nutrition messages in the community, growthmonitoring in the clinic and treatment protocols in thehospital. This approach is very attractive to healthmanagers with its neat outline of tasks, timetables andcosts for each level of care.

This approach builds upon the legacy of ‘selective pri-mary health care’ w1x, which was a response to theCPHC outlined at Alma Ata. The adoption of certainselected interventions, such as growth monitoring, oralrehydration therapy (ORT), breastfeeding and immu-nization (GOBI), it was argued, would be the ‘leadingedge’ of PHC ushering in a more comprehensiveapproach at a later stage. This approach was nurturedby the prevailing conservative political ideology of the1980s which encouraged a focus away from thebroader determinants of health—such as income in-equalities, the environment, community develop-ment—towards an emphasis on health care tech-nologies. The result was the enthusiastic promotion ofselected interventions, which received generous fund-ing to the detriment of more comprehensive app-roaches.



Selective PHC was given further credence by theWorld Bank’s 1993 World Development Report,‘‘Investing in Health’’ w4x, which recognized the impor-tance of health to development. Based on calculationsof burden of disease globally and in different regions,it specified the most cost-effective health interventions,and culminating in the formulation of a core packageof health services to be provided at the different levelsof care. The identification of core packages became amechanism to ration the cost of health services pro-vided by the state, as other activities were to be takenup by non-government organisations. This fitted inneatly with the Bank’s wider economic and fiscal pol-icies, encouraging the privatisation of health caredelivery and the cutting back of State services.

Proponents of this approach have pointed to thesuccesses in increasing immunization rates, reducinginfant mortality and the eradication of polio. However,closer examination of these achievements reveals theshortcomings of an approach, which does not inte-grate the different levels of care or situate health withina wider context. There is growing evidence that immu-nisation coverage is stagnating and even declining inmany countries w5x, infant mortality rates are rising inmany Sub-Saharan countries w6, 7x and the health ofmany children is now deteriorating w8x. Evaluationshave raised questions about the sustainability of massimmunisation campaigns w9x, the effectiveness ofhealth facility based growth monitoring w10x and theappropriateness of ORT when promoted as sachets orpackets and without a corresponding emphasis onnutrition, water and sanitation w11x. A recent reviewhas even pointed out the lack of evidence for theeffectiveness of directly observed therapy for TB(DOTS) in the absence of a well functioning healthservices and community engagement w12x. Evalua-tions at both national and provincial levels have foundthat it is only when these core service activities areembedded in a more comprehensive approach (whichincludes paying attention to health systems andhuman capacity development) that real and sustaina-ble improvements in the health status of populationsare seen w13–15x.

A comprehensive strategy forhealth development1

CPHC, however, is based on the understanding thathealth improvement results from a reduction in boththe effects of disease (morbidity and mortality) and its

1 This section draws heavily upon Sanders D, ‘‘PHC 21 – Everybody’sBusiness’’, Main background paper for an international meeting to celebratetwenty years after Alma Ata, Almaty, Kazakhstan, 27-28 November 1998,World Health Organisation WHOyEIPyOSDy00.7

incidence as well as from a general increase in socialwell-being. The effects of disease may be modified bysuccessful treatment and rehabilitation and its inci-dence may be reduced by preventive measures. Well-being may be promoted by improved social environ-ments created by the harnessing of popular and polit-ical will and effective intersectoral action.

Of particular relevance to the development of compre-hensive health systems is the clause in the Alma Atadeclaration stating that PHC ‘‘addresses the mainhealth problems in the community, providing promo-tive, preventive, curative and rehabilitative servicesaccordingly’’ w16x.

Comprehensive health systems include, therefore,curative and rehabilitative components to address theeffects of health problems, a preventive component toaddress the immediate and underlying causative fac-tors which operate at the level of the individual, and apromotive component which addresses the more basic(intersectoral) causes which operate usually at thelevel of society.

Table 1 below illustrates, using some common healthproblems, the complementary role of the differentcomponents in holistically tackling them. Such amatrix, which starts from a disease focus, is useful forhealth professionals in providing them with an under-standing of both the health care as well as broader,developmental interventions required to comprehen-sively address them.

Strategies for comprehensively tackling such healthproblems can be grouped essentially under two com-plementary headings: promoting healthy policies andplans and implementing comprehensive and decen-tralised health systems. Success of these strategiesdepends upon the creation of a facilitatory environ-ment through such actions as advocacy, communitymobilisation, capacity-building, organisational change,financing and legislation.

Below we outline the principles of implementation ofcomprehensive and decentralized health systems. Wefirst refer briefly to the strategy of health promotionwhose operationalisation is usually explicitly inter-sectoral from the outset. We then focus upon the keysteps in developing comprehensive, community-basedprogrammes structured around common health prob-lems, and on the integration of such programmes intodecentralized health systems.

Implementing comprehensive anddecentralised health systems

The implementation of PHC has too often focused onlyon the (often facility-based) curative and personal pre-



Table 1. Comprehensive primary health care for some common diseases: a summary framework of priority interventions w17x

Intervention

Disease Rehabilitative Curative Preventive Promotive

Diarrhoea Nutrition Oral Education for Waterrehabilitation rehydration personal and

food SanitationNutrition hygienesupport Household

Breastfeeding FoodSecurity

Measlesimmunisation Improved child

care

Pneumonia Nutrition Chemotherapy Immunisation Nutritionrehabilitation

Housing

Clean Air

Measles Nutrition Chemotherapy Immunisation Nutritionrehabilitation

Nutrition HousingSupport

Tuberculosis Nutrition Chemotherapy Immunization Nutritionrehabilitation

Nutrition Contacts HousingSocial support (tracing)rehabilitation Ventilation

Cardiovascular Weight loss Drug treatment Nutrition Nutrition policydisease education

Graded exercise Supportive Tobacco controltherapy Increased exercise

Stress control Recreationalfacilities

Treat hypertension

Smoking cessation

ventive components of comprehensive care, while thehealth promotion initiative has stressed the broadersocial components. This gap needs urgently to bebridged since they are clearly indivisible in the processof health development and providing integrated care.

Health promotion in practice

The implementation of healthy public policies, whichemphasise the role of intersectoral activity, has beensignificantly enhanced since the late 1980s by thegrowth of the health promotion movement. TheHealthy Cities initiative w18x, and subsequently a focuson other settings such as schools, markets, work plac-es and hospitals, demonstrate an approach whichtakes forward the policy development activitiesdescribed above. These initiatives can garner politicalsupport and encourage local agencies and sectors toreassess their policies and practices in influencing

health. By 1997 there were over 1000 communitiesparticipating worldwide in the Healthy Cities initiative.

Facilitating organisational change and encouraging(particularly government) staff to be more flexible,innovative and responsive to local communities arekey actions in achieving success w17x. In the pastmany of the initiatives to promote community partici-pation in health have concentrated on inviting com-munity people to participate in activities established(and largely controlled) by the health services. Arecent WHO study uncovered a wide range of com-munity groups or organisations—which have beentermed ‘health development structures’ (HDSs)—thatplay some role in promoting health. A report noted that‘‘the majority of HDSs owe their origins to age-oldcommunity traditions of mutual support and coopera-tion and have a long history of community action’’ w19x.They include, in addition to representative healthcouncils, women’s groups, youth groups, social clubs,



cooperative societies, mutual aid societies and sport-ing clubs. There are many roles, often invisible, playedby such groups, that contribute to improving health.This could be achieved more systematically in part-nership with the health sector, but has hitherto beenlargely unexploited. Settings-based health promotioninitiatives offer a perspective and mechanism for thiskind of relationship. The concept of health promotingdistricts holds much promise and should be developedas a means of extending health services towards amore intersectoral and developmental role.

The development of comprehensive,community-based programmes

Whereas health promotion activities, recognising thefundamental contribution to health of equitable socialand economic development, commence with a multi-sectoral focus, programmes originating around dis-eases or health problems start from a health care orservice response. While curative, preventive and car-ing actions are very important and still constitute thecore of medical care, comprehensive PHC demandsthat they be accompanied by rehabilitative and pro-motive actions. By addressing priority health problemscomprehensively as indicated in the table above, aset of activities common to a number of health pro-grammes will be developed as well as a horizontalisedinfrastructure. The promotive activities will necess-arily involve other sectors and, if successful andwidespread, create pressure for supportive policyresponses.

The principles of programme development apply toall health problems, whether specific communicable(e.g. diarrhoea) or non-communicable diseases (e.g.ischaemic heart disease) or health-related problems(e.g. domestic violence).

Much experience has been gained internationally inthe development of comprehensive and integratedprogrammes to combat undernutrition: these ex-periences can provide useful lessons for otherprogrammes.

After the priority health problems in a district or locallevel have been identified, the first step in programmedevelopment is the conducting of a situation analysis.This should identify the prevalence and distribution ofthe problem, its causes, and potential resources,including community capacities and strengths, whichcan be mobilised and actions which can be undertak-en to address the problem. The more effective pro-grammes have taken the above approach, involvinghealth workers, other sectors’ workers and the com-munity in the three phases of programme develop-

ment, namely, assessment of the nature and extent ofthe problem, analysis of its multilevel causation andaction to address the linked causes (Figure 1).

This approach to implementation has thus beentermed the ‘‘Triple A cycle’’ (Figure 2).

Similar minimum or core service components can alsobe identified for other health programmes e.g. activi-ties in the Safe Motherhood Initiative, the IntegratedManagement of Childhood Illness, DOTS, technicalguidelines for the management of common non-com-municable diseases etc. There is an advantage instandardising and replicating these core activities inhealth facilities at different levels, thus reinforcing theirpractice throughout the health system.

The development of a decentralisedhealth system

There are at local level in most countries a number ofprogrammes, often vertically organised and centrallyadministered, with specialised staff who performsonly programme functions. The development of com-prehensive programmes which are integrated into adecentralised district service inevitably requires trans-formation of both management systems and practice.Making the transition from a centralised bureaucraticsystem to a decentralised, client-oriented organisa-tional culture calls for a significant investment in devel-oping both management systems and structures andthe management capacity of health personnel. Districtlevel staff must be able to support decentralised devel-opment of comprehensive programmes with clearroles, goals and procedures.

The case study which follows illustrates a number ofthe principles described above of the process of devel-opment of a comprehensive programme and the im-pact of such a programme on the creation of an inte-grated district health system (Photo 1).

Nutrition programme and healthsystems development in Mount Frere

Mount Frere health district is one of the poorest dis-tricts in the country. It lies 100 km north of Umtata inthe old homeland region of Transkei and is home toapproximately 300,000 people. The whole area is abeautiful part of the country full of magnificent scenery.However, the terrain is also very rugged and inacces-sible. There are only two tarred roads in the wholedistrict. At times of heavy rain, or after heavy snowsome roads can become totally impassable. In 1999 itwas declared a national emergency area following



Figure 1. Causes of undernutrition.

Figure 2.

devastation by a tornado and, more recently, by veldfires.

A recent local survey found the crude birth and deathrates to be significantly higher than those for the restof the Province. More than two thirds (71%) of house-holds use water from unprotected sources for drinkingpurposes. More than two fifths of births (45%)occurred at home and were therefore not attended byhealth professionals. Regional surveys have foundmore than one in three children to be stunted in theirgrowth and a majority of children and women probablysuffer from undernutrition.

Being a health worker in Mount Frere is not easy.Years of neglect and poor resources have resulted ina health system which is barely functional. If the poorof Mount Frere are to become empowered they needa health service which delivers adequate quality serv-ices and also seeks to involve them in tackling theirhealth and development challenges. This in turn re-

quires public sector workers to gain the confidenceand skills to optimise the quality of service they offerand to understand and reach out to the people they



Photo 1.

Photo 2.

serve. Transforming the public sector to deliver highquality services is one of the major challenges of thenew South Africa.

A local NGO—Health Systems Trust—contracted thePublic Health Programme, University of the WesternCape to work with the local health services to devisea comprehensive programme to tackle malnutri-tion which had been prioritised by the local healthmanagement.

The aims of this project are:

● To increase the capacity of provincial, regional anddistrict health workers and the district health systemto improve the quality of care and service that theyprovide.

● To bring together a multi-sectoral team andempower them to initiate community-based nutri-tion programmes.

● To help develop appropriate systems, structuresand policies for the smooth and integrated function-ing of different sectors and programmes, at the dif-ferent levels (provincial, regional and district) of thepublic sector.

● To assist in the development of an efficient districthealth system.

● To develop processes and tools to assist in theimplementation of the Integrated Nutrition Pro-gramme in other districts

Forming partnerships, teambuildingand advocacy

For innovation in the public sector to be successful,sustainable and replicable it must have the supportof all role-players at all levels of the service. The man-agement of the project is through a partnershipbetween the Eastern Cape Department of Health, theHealth Systems Trust and the Public Health Pro-gramme, University of the Western Cape.

An Eastern Cape Provincial Steering Committeeoversees the project. This committee has provincialrepresentatives from the following sectors: agriculture,education, environmental health, maternal, child andwomen’s health, nutrition and welfare and a nationalrepresentative from the nutrition directorate.

At the district level this is mirrored by the Mount FrereNutrition Team which includes representatives fromnutrition, maternal and child health, agriculture, edu-cation, local government, water affairs and environ-mental health. They have fostered teambuildingthrough a series of workshops which have aimed tocome to a common understanding of the causes ofmalnutrition and to apply a planning cycle of assess-ment, analysis and action.

Performing an assessment

The district nutrition team has used the research skillsacquired in the above workshops to conduct rapidassessments of key nutrition services and the nutritionsituation in communities in the district. By developingobservational checklists and interview guidelines theyhave assessed the quality of care of severely mal-nourished children in the paediatric ward, the perform-ance of growth monitoring and promotion (GMP) atthe local clinics, the environmental situation of select-ed communities, and the implementation and opera-tion of the primary school nutrition programme at localschools.

For example, in examining the performance of GMPthe team drew up an observation checklist of whatthey would expect to see. This starts with the nursegreeting the mother and ends with the nurse having adialogue with the carer about the growth of the child.From the observations, it was clear that there weremany things that could be improved. Few of the moth-ers were greeted, there was little or no feedback to themothers and opportunities for group discussions werenot taken up (Photo 2).

The team also assessed the ability of nurses to inter-pret the growth chart and what kind of advice they



give for different growth curves. To do this a series ofgrowth charts were designed which showed normalgrowth, growth faltering, growth failure and catch upgrowth. Nurses were asked to interpret the charts andstate what advice they would give the mothers basedon the chart.

Finally, the possible reasons for the poor performancewhich was detected were explored through a series ofsemi-structured interviews with a selection of nursesfrom the clinics and hospitals. When asked about thereasons why GMP may not be done well in the clinic,most clinic workers mentioned the heavy workload andnumber of tasks (especially administrative). Only aminority mentioned a lack of training or supervision.Most thought the main use of GMP was to find childrenwho were undernourished and would consequentlyqualify for a feeding scheme or require referral to thehospital. Some felt that it was also done in order to‘collect statistics’ for surveillance; only a few men-tioned that it was to initiate dialogue with the motherto promote growth of the child.

Planning and implementing interven-tions

The above assessment was followed by an analysisof the results and the implementation of nutrition-relat-ed interventions in the hospital, clinics and local com-munities. In the hospital practices to improve thein-patient care of severely malnourished children havebeen implemented. In the clinics training workshops toimprove the management of diarrhoea and growthmonitoring and promotion have been conducted. Insome communities the team has assisted a localNGO, the Mvula Trust, to redesign water improve-ments and prioritise hygiene education programmes.

All of these interventions were critically informed bythe situational assessment the team performed. So inthe case of growth monitoring the team stressed theimportance of practitioners being clear about the aimsand objectives of GMP and to emphasise this in train-ing. More specifically, the use of GMP for tailoring indi-vidual nutrition education messages, in the promotionof good growth and in the integration of services wasstressed.

The team then thought about the skills and resour-ces required to implement such a programme. Thisinvolved drawing up a detailed plan of activities asso-ciated with growth monitoring in the local settings.Local teams then planned how they would changetheir work patterns and clinic organisation to optimisegrowth monitoring activities.

To complete the triple A cycle the teams have beencollecting routine data to assess the effectiveness oftheir interventions. In the hospitals there has been fallof nearly fifty percent in deaths from severe malnutri-tion. Supervisors and outside evaluators have con-firmed much improved growth monitoring in many ofthe peripheral clinics. It is important to feed this backto the team in order to encourage them to tackle othernutrition problems as well (Photo 3).

Sharing the success

As a result of the success of the Mount Frere projectthe team members are already being used to sharetheir success with other districts in the region. All thehospitals in this region of the Eastern Cape are nowimplementing the programme for improving the hos-pital management of severe malnutrition. All fourdistricts in the region have also initiated training forimproving nutrition services in the clinics. The com-munity participatory hygiene and sanitation toolsdeveloped by the team are also being adapted nation-ally. In all these cases, the Mount Frere team has beencentrally involved in sharing their new-found skills andconfidence.

Capacity development

The heart of this project has been the capacity devel-opment of public sector personnel in some of the poor-est parts of the country. The success of this has beenpartly reflected above. Eastern Cape personnel havetaken the lead in organising and presenting at localworkshops and many have already presented theirexperiences at national meetings and conferences.

Conclusion

Given the historical legacy of wide inequalities inhealth and health care which persist six years after theadvent of democracy, the South African Governmentis facing increasing public pressure to provide re-sources and quality services. In these circumstancesthe allure of currently dominant policies with theirapparently cost-effective packages of care is under-standable. However, the pressure for immediateresponses needs to be reconciled with the necessityto develop human capacity and integrated systems,without which even the most robust and effective oftechnical interventions have been shown to be unsus-tainable. This paper has attempted to argue for andillustrate a more comprehensive approach in whichtested technical interventions are located in a broader



Photo 3.

process of capacity strengthening and systemsdevelopment.

The key issues raised in this article relate to interpre-tation and implementation of the Primary Health CareApproach. A tension which has informed internationaland national policies and programmes has existed,almost from the time PHC was conceived, between amore comprehensive and integrated approach andone which is more selective. The former combinestechnical health care with intersectoral activities, andrests upon and facilitates a long-term social processinvolving capacity development of both communitiesand technicians. The latter selects technical healthcare interventions, primarily on the basis of their cost-effectiveness, and often ‘‘delivers’’ these through ver-tical programmes—with their own infrastructure,administration and personnel. The latter ‘‘selective’’approach, albeit in sophisticated form, is currentlydominant in international health policy and appearsalso to be highly influential in contemporary South Afri-ca, which is facing the enormous challenge of imple-mentation following years of apartheid neglect. Thecase study described in this paper outlines the earlyphases of a programme to combat child undernutritionin a remote, impoverished rural district in the EasternCape Province of South Africa.

The choice between a selective and a more compre-hensive approach is one which is both immediate andstark in present-day South Africa, where new policies

and activities are being engaged with on a daily basis.The authors contend, however, that the issues raisedin this article have broader application and profoundimplications for health systems development. In thecase of other complex, multicausal health problems,for example, HIVyAIDS and T.B., it is already clearthat narrow, vertical approaches are having littleimpact on their spread. A case has been made for amore multi-faceted and integrated approach, central towhich is an action learning process whose outcomesinclude improved understanding of the causes of par-ticular problems and enhanced capacity to addressthem (Photo 4).

Key questions to be answered revolve around sustain-ability and replicability, since it is clear that impact canand does result from such approaches—in commonwith more selective technical approaches. Thus, in thehospital-based component of the Mt Frere nutritionprogramme it has been possible to substantially re-duce the case fatality rate from severe malnutritionamong hospital inpatients. This has been accompa-nied by demonstrable improvements in knowledge andskills amongst both nurses and doctors. Moreover,some of these skills are ‘‘generic’’ and transferable toother aspects of hospital child care. It is, however,important to demonstrate that this improved practiceis sustained without ongoing external support (fromthe UWC team) and that this experience is replicablewithin the South African public health services. Such



Photo 4.

research is not only necessary to justify past and con-tinuing investment in the Mt Frere project, but crucialfor the longer-term development of comprehensiveand integrated health systems in South Africa andbeyond.

Vitae

Professor David Sanders qualified as a medical doctor inZimbabwe and later specialised in Britain in Paediatrics andTropical Public Health. In 1980 he returned to newly inde-pendent Zimbabwe as Coordinator of a rural health pro-gramme developed by OXFAM in association with theZimbabwe Ministry of Health. He later joined the Universityof Zimbabwe Medical School, initially as a lecturer in Pae-diatrics and Child Health and later became Associate Pro-fessor and Chairperson of the Department of CommunityMedicine. In 1993 he was appointed as Professor and Direc-

tor of a new Public Health Programme at the University ofthe Western Cape, Cape Town, South Africa, which providespractice-oriented education and training in public health andprimary health care to a wide range of health and develop-ment workers. He is the author of two books: ‘The Strugglefor Health: Medicine and the Politics of Underdevelopment’and ‘Questioning the Solution: the Politics of Primary HealthCare and Child Survival’ as well as several booklets andarticles on the political economy of health, structural adjust-ment, child nutrition and health personnel education.

Dr Mickey Chopra is a senior lecturer at the School of Pub-lic Health, University of the Western Cape. His main areasof interest are community child health and nutrition andhealth systems research. These interests grew out of hisexperiences of being a medical officer in rural South Africawhere the previous lack of health systems and communityprogrammes resulted in high levels of childhood mortalityand morbidity.

References

1. Walsh JA, Warren KS. Selective primary health care: an interim strategy for disease control in developing countries. NewEngland Journal of Medicine 1979;301:967–74.

2. World Health Organisation. The world health report 2000: health Systems: Improving Performance Geneva; 2000.3. Daviaud E, Cabral J, Elgoni A. Developing a core package of primary health care services. South African Medical Journal

1998;88:371.4. World Bank. World development report: investing in health Oxford University Press, Oxford, England 1993.5. UNICEF State of the world’s children. Oxford England: Oxford University Press; 1996.6. World Health Organisation World Health Report 1996. Geneva: WHO Press; 1996.



7. UNICEF State of the world’s children. Oxford England: Oxford University Press; 1994.8. Gadomski A, Black, Mosley WH. Constraints to the potential impact of child survival in developing countries. Health Policy

and Planning 1990;5:236–41.9. Hall AJ, Cutts FT. Lessons from measles vaccination in developing countries. British Medical Journal 1993;307:1294–5.

10. Chopra M, Sanders D. Is growth monitoring worthwhile in South Africa? South African Medical Journal 1997;87:875–8.11. Werner D, Sanders D. Questioning the solution: the politics of primary health care and child survival. Palo Alto USA:

Healthwrights; 1997.12. Volmink J, Garner P. Systematic review of randomised controlled trials of strategies to promote adherence to tuberculosis

treatment. British Medical Journal 1997 Nov 29;315(7120):1403–6.13. Halstead SB, Walsh JA, Warren K, editors. Good health at low cost. New York: Rockefeller Foundation; 1985.14. Fitzroy H, Briend A, Fauveau V. Child survival: should the strategy be redesigned? experience from Bangladesh. Health

Policy and Planning 1990;5:226–34.15. Gutierrez G, Tapia-Conyer R, Guiscafre H, Reyes H, Martinez H, Kumate J. Impact of oral rehydration and selected public

health interventions on reduction of mortality from childhood diarrhoeal diseases in Mexico. Bull WHO 1996;74:189–97.16. World Health Organisation and UNICEF Report of the International Conference on primary health care. Alma Ata: USSR;

1978.17. Sanders D, ‘‘PHC 21—Everybody’s Business’’, Main background paper for an international meeting to celebrate twenty

years after Alma Ata, Almaty, Kazakhstan, 27–28 November 1998, World Health Organisation WHOyEIPyOSDy00.7.18. Baum F. The new public health: an Australian perspective. Melbourne: Oxford University Press; 1998.19. World Health Organisation (WHO). Health development structures in district health systems: the hidden resources. Gene-

va: World Health Organisation, Division of Strengthening Health Services; 1994. Available from: URL: http:yywhqlib-doc.who.intyhqy1994yWHO_SHS_DHS_94.9.pdf.

implementing comprehensive and decentralised health systems

Documents