TheAfrica Malaria Report 2003 - WHO archivesarchives.who.int/prioritymeds/report/append/610pub_africa_malaria... · The Africa Malaria Report – 2003 6 Resource mobilization and

TheAfrica MalariaReport 2003

© World Health Organization/UNICEF 2003 WHO/CDS/MAL/2003.1093

All rights reserved.

The designations employed and the presentation of the material in this publication do not imply the expression of anyopinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory,city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on mapsrepresent approximate border lines for which there may not yet be full agreement.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed orrecommended by the World Health Organization in preference to others of a similar nature that are not mentioned.Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

The World Health Organization does not warrant that the information contained in this publication is complete andcorrect and shall not be liable for any damages incurred as a result of its use.

2

World Health Organization

3

Acronyms and abbreviations 5

Foreword 7

Executive summary 9

Introduction 13

1 The burden of malaria in Africa 17

1.1 Burden of malaria on health in Africa 18

1.2 Burden of malaria on African health systems 19

1.3 Burden of malaria on the poor 19

1.4 Recent trends in the burden of malaria 20

1.5 Future prospects 22

2 Insecticide-treated nets 24

2.1 Evidence 24

2.2 Progress: ITN strategy plans 25

2.3 Progress: taxes and tariffs 25

2.4 Progress: long-lasting insecticidal nets 26

2.5 Progress: coverage 26

2.6. Challenges: increasing coverage 27

2.7. Challenges: overcoming disparities in net coverage 27

2.8. Scaling up 27

3 Prompt and effective treatment 31

3.1 Evidence 31

3.2 Drug resistance 32

3.3 Progress: drug policies 32

3.4 Progress: treatment coverage 33

3.5 Challenges: increasing coverage 34

3.6 Challenges: disparities in use 35

3.7 Scaling up 36

4 Malaria during pregnancy 38

4.1 Evidence 39

4.2 Progress 41

4.3 Challenges 41

4.4 Opportunities 43

5 Response to epidemics and complex emergencies 46

5.1 Magnitude of the problem 46

5.2 Prevention and control of epidemics 48

5.3 Progress and challenges 48

5.4 Strategies for the future 51

Contents

4

The Africa Malaria Report – 2003

6 Resource mobilization and financing 53

6.1 Resources required to roll back malaria 53

6.2 Domestic resources 54

6.3 International resources 58

6.4 Next steps for financing malaria control in Africa 60

Notes on data sources 63

• Table 1: Data availability by source 65

• Table 2: Sources of data for monitoring malaria burden and control 66

Country profile overview 67

Country profiles 70-112

• Table 3: Percentage of children under 5 years who slept under a mosquito net (any) during the night preceding the survey, by background characteristics 114

• Table 4: Percentage of children under 5 years who slept under an insecticide-treated mosquito net during the night preceding the survey, by background characteristics 115

• Table 5: Percentage of febrile children under 5 years who received treatment with an antimalarial, by background characteristics 116

• Table 6: Estimates of percentage of population living in areas at risk for malaria transmission 117

• Table 7: Millennium Development Goals 118

5

Acronyms and abbreviations

ACT artemisinin-based combination therapy

ANC antenatal clinic

AQ amodiaquine

AS artemisinin

CFR case-fatality rate

CQ chloroquine

DHS demographic and health surveys

DSS demographic surveillance systems

ETF early treatment failure

GFATM Global Fund to Fight AIDS, Tuberculosisand Malaria

HBM home-based management of fever

HIPC highly indebted resource-poorcountries

HIS health information system

HIV/AIDS human immunodeficiency virus/acquired immunodeficiencysyndrome

IMCI Integrated Management of Childhood Illness

INDEPTH International Network of field siteswith continuous Demographic Evaluation of Populations and their Health in developing countries

IPT intermittent preventive treatment

IRS indoor residual spraying

ITN insecticide-treated net

LBW low birth weight

LCF late clinical failure

LLIN long-lasting insecticidal net

LPF late parasitological failure

MARA Mapping Malaria Risk In Africa

MEWS malaria early warning systems

MICS multiple indicator cluster surveys

MoH Ministry of Health

n.a. not applicable

NGO nongovernmental organization

NMCP national malaria control programme

PRSC poverty reduction support credits


6


PRSP poverty reduction strategy paper

PTD preterm delivery

RBM Roll Back Malaria

SAMC Southern Africa Malaria Control Programme

SGA small for gestational age

SP sulfadoxine–pyrimethamine

SWAPs sector-wide approaches

TB tuberculosis

UNICEF United Nations Children’s Fund

VAT value-added tax

WHO World Health Organization

■ Where space is limited in some figures and tables, the following abbreviations have beenused for country names:

CAR Central African Republic

DRC Democratic Republic of the Congo

Tanzania United Republic of Tanzania

7

Foreword

In October 1998, together with the Presidentof the World Bank and the Administrator ofthe United Nations Development Programme,we launched Roll Back Malaria as a catalystfor a renewed global commitment to tackle adisease that has been ignored by the worldfor far too long – a single disease that puts abrake on development, particularly in Africa.

This report from UNICEF and WHO suggeststhat, in 2003, malaria remains the singlebiggest cause of death of young children inAfrica and one of the most important threatsto the health of pregnant women and theirnewborns. However, there are clear signs thatthe movement to Roll Back Malaria is havingan impact. The combined strategies suggestedin 1998 for reducing the burden of malaria(insecticide-treated nets, prompt access totreatment, and prevention of malaria inpregnancy) are now widely accepted; theirapplication on a large scale throughoutAfrica is under way.

Although coverage of individualinterventions, such as insecticide-treatednets, is still far too low, the good news isthat there is a clear trend towards increasingcoverage, and other encouraging moves – achange in government taxation policies on nets and netting materials, for example,

and the development of Africa-basedindustrial production of nets – that willhelp sustain this trend.

Parasite resistance to previously effectivelow-cost drugs is an enormous and growingproblem, but governments are now fullyengaged in this challenge, monitoring thedevelopment of resistance and energeticallypursuing the most promising options formore effective treatment.

The financial resources for fighting malariaare increasing. The establishment of the GlobalFund to Fight AIDS, Tuberculosis and Malaria isproviding significant new grants to helpcountries accelerate implementation of theirplans to Roll Back Malaria. In addition, fundsmade available to improve health under debt-relief initiatives are being used to financemalaria interventions in some countries.

Our two organizations remain firmlycommitted to working together, with ourother partners and with Africa, to achieve theambitious goals for Roll Back Malaria set inAbuja on 25 April 2000 and agreed to byAfrican heads of state. We intend that thisreport should be the first of a regular series,tracking progress towards achievement ofthese goals and of the MillenniumDevelopment Goal for malaria.

Foreword

Carol Bellamy Dr Gro Harlem BrundtlandExecutive Director Director-GeneralUnited Nations Children’s Fund World Health Organization

This report – the first of its kind – takes stockof the malaria situation and of continuingefforts to tackle the disease in Africa; it isbased on a review of the best informationavailable to WHO and UNICEF, from samplesurveys and routine reports, at the end of 2002.

Malaria continues to be a majorimpediment to health in Africa south of theSahara, where it frequently takes its greatesttoll on very young children and pregnantwomen. Because malaria is such a commondisease and well known to the people itaffects most, and because many of those whobecome sick with malaria do not visit healthcare facilities, assessing the size of theproblem, and how it is changing over time, isan enormous challenge.

New analyses confirm that malaria is aprincipal cause of at least one-fifth of allyoung child deaths in Africa. The latestavailable data on outpatient visits and onhospital admissions and deaths due to malariaconfirm that this disease makes substantialdemands on Africa’s fragile healthinfrastructure. In endemic countries, as manyas one-third of all clinic visits and at least aquarter of all hospital admissions are formalaria. In some countries, these data suggestthat illness due to malaria has increased overthe past decade; in others, the size of theproblem has remained constant. No country inAfrica south of the Sahara for which data areavailable shows a substantial decline.

Additional information on trends in malariamortality is available for an increasingnumber of countries with “demographicsurveillance systems”. These sources indicatethat the number of children dying of malariarose substantially in eastern and southernAfrica during the first half of the past decadecompared with the 1980s. In west Africa overthe same period there was little change inthe overall malaria mortality rate in children.

In summary, the burden of sickness anddeath due to malaria remained high in Africasouth of the Sahara during the 1990s andincreased in most countries in the eastern andsouthern part of the continent. Monitoringsystems cannot yet reliably track changes in

indicators of the burden of malaria, particularlymalaria mortality, on a yearly basis.

The high burden of malaria in Africa, andthe increasing burden in some parts of thecontinent during the 1990s, is not anindication that the intensified efforts tocontrol the disease over the past few yearshave had no impact. The full impact onmalaria sickness and death of the recentefforts to accelerate malaria controldescribed in this report will be measurableonly some years after high coverage ofinterventions is achieved. It is possible thatthe start of intensified control effortscoincided with increasing malaria mortality,meaning that – without them – the situationmight have been substantially worse than isnow reported. The strengthening of malariasurveillance and monitoring needs to begiven priority in parallel with efforts tocontrol malaria.

The 2000 Summit on Roll Back Malaria, heldin Abuja, endorsed a “shortlist” of relativelyinexpensive malaria control interventionsalready available and known to be effective.Partners in the Roll Back Malaria effort, whichinclude governments of malaria-endemiccountries, donor governments, internationalorganizations, the private sector, and civilsociety bodies, have supported theintroduction of these interventions.

9


Executive summary

The 2000 Abuja Summit established controlpriorities and goals.

Pic

ture

: W

HO

/RB

M

Insecticide-treated nets (ITNs) are a low-cost and highly effective way of reducing theincidence of malaria in people who sleepunder them, and they have been conclusivelyshown in a series of trials to substantiallyreduce child mortality in malaria-endemicareas of Africa. By preventing malaria, ITNsreduce the need for treatment and thepressure on health services, which isparticularly important in view of the increasein drug-resistant falciparum malariaparasites. Although accurate data from the1980s are not generally available forcomparison, it is certain that there are nowmore children sleeping under nets and agreater use of ITNs in Africa than ever before.Recent survey data showed thatapproximately 15% of young children sleptunder a net, but that only about 2% usednets that were treated with insecticide.Untreated nets provide some protectionagainst malaria, but their full protectivebenefits can be realized only if they areregularly retreated with insecticide.

The price of nets has fallen substantially asa result of greater demand, increasedcompetition between producers, andreductions in taxes and tariffs and otherobstacles to trade that many Africancountries instituted after the Abuja Summit.In many countries, both nets and the

insecticide to treat them can now bepurchased in small shops and markets andeven on street corners; only a few years agothey would have been available only in a fewspecialist shops in capital cities. At least fivelarge factories in Africa are now producingnets. Almost all malaria-endemic Africancountries now have active programmes underway to encourage ITN use, and most of thesesupport a variety of different mechanisms toincrease net coverage. Nevertheless, thecommercial price of nets and insecticide –though falling – still puts this life-savingtechnology beyond the reach of the poorestincome groups of the population. Majorefforts are now being made in at least fiveAfrican countries to provide subsidized ITNsto the most vulnerable groups – youngchildren and pregnant women. Newtechnological developments promise netsthat will retain insecticidal activity for manyyears, and novel ways of encouraging regularnet treatment with insecticide should make itpossible to increase the proportion of netsthat are effectively treated.

Treated nets and other means of reducingmosquito bites will not totally preventmalaria. People who become ill with thedisease need prompt and effective treatmentto prevent the development of severemanifestations and death. Since the 1980s,parasite resistance to chloroquine, the mostcommonly available antimalarial drug, hasemerged as a major challenge. In mostcountries in eastern, central, and southernAfrica, chloroquine has lost its clinicaleffectiveness as a malaria treatment. A similarevolution is taking place, though some yearslater, in west Africa, and there is indirect, butcompelling, evidence that this is giving rise toincreasing mortality. Unfortunately, resistanceto the most common replacement drug,sulfadoxine–pyrimethamine, has also emerged,especially in eastern and southern Africa.

Over the past few years, 13 countries inAfrica have changed their national policies torequire the use of more effective antimalarialtreatments. Where current monotherapies arefailing, WHO recommends artemisinin-basedcombination therapy (ACT), which is highlyefficacious and promises to delay theemergence of resistance. So far however, itsuse is constrained by high costs and limitedoperational experience in Africa. To date, four African countries have adopted ACTs as first-line treatment.

10


Insecticide-treated nets are highly effective in reducing malaria mortality in young children.

Pic

ture

: W

HO

AF

RO

Improved management of malaria casesmay be undertaken as part of a generalstrengthening of public health services, forexample as part of the strategy for IntegratedManagement of Childhood Illness (IMCI).However, in many malaria-endemic countriesthe first treatment for malaria is oftenpurchased from a shop. Data fromrepresentative sample surveys indicate thatalmost half of all children under 5 years ofage with fever are treated with anantimalarial drug. Although this isencouraging, some of these treatments mayhave been with failing drugs or been giventoo late or in the wrong dosage. Recentstudies indicate that home treatment,supported by public information and pre-packaging (as an aid, to ensure that patientstake the full treatment course at the righttime), can help to reduce malaria mortality inchildren. Many countries now concentrate onmaking effective malaria treatment availableclose to the home, through support tocommunity initiatives and engagement ofdrug sellers and the pharmaceutical industry.Realizing the full potential of effectivetreatment as a tool for reducing mortalitywill require a systems approach, ensuring thateffective drugs are affordable (which willoften require subsidization) and that they aresupported by appropriate education of formaland informal providers as well as mothers,and by quality assurance and regulation.

The impact of malaria on pregnant womenand their newborns can be substantiallyreduced by the recently recommended use of“intermittent preventive treatment” (IPT). Thisstrategy provides at least two treatment dosesof an effective antimalarial at routineantenatal clinics to all pregnant women livingin areas at risk of endemic falciparum malariain Africa (irrespective of whether they areactually infected with malaria or not). Abouttwo-thirds of pregnant women in Africa southof the Sahara attend clinics for antenatalcare, and incorporating IPT for malaria intotheir routine care should be straightforward.Now an integral part of the “MakingPregnancy Safer” strategy, IPT has beenadopted as policy by six countries to replacechemoprophylaxis; most other countries inthe region are reviewing their policies in thelight of the new recommendation. Thebeneficial effects of IPT will probably beadditive to the proven benefits of ITN use bypregnant women. A comprehensive approach

to the prevention and management of malariaduring pregnancy therefore calls for acombination of IPT, support for ITN use, andprompt access to effective treatment. Fivecountries in eastern and southern Africa haverecently formed a coalition to reduce theimpact of malaria in pregnancy through thiscombined approach.

Areas on the northern and southern fringesof the malaria-endemic belt of Africa, as wellas highland areas in many countries, are at

11

Executive summary

Pre-packaged antimalarial drugs can increase compliance and facilitate home treatment.

Young children are at the greatest risk of malaria in Africa.

Pic

ture

: W

HO

/MA

L

Pic

ture

: W

HO

/RB

M/L

.Tay

lor

12


risk of epidemic malaria. Unlike the endemicdisease, epidemic malaria typically affectspeople of all ages and can have high case-fatality. Roll Back Malaria has beensupporting efforts to improve the earlyrecognition of, and effective and timelyresponse to, malaria epidemics. Indoorresidual spraying can play an important rolein malaria vector control, especially in thecontrol of epidemics. Malaria early warningsystems have been established in southernAfrica to improve outbreak detection andresponse and are being developed in otherepidemic-prone parts of Africa. Fifteenepidemic-prone countries have developed a preparedness plan of action; data on the timeliness and effectiveness of epidemicresponse in these countries are presented in this report.

Tackling malaria effectively requiressubstantial resources. At the Abuja Summit itwas estimated that at least US$ 1 billion isneeded from a combination of increaseddomestic spending and internationalassistance; the report provides informationon resource flows. Since the launch of RollBack Malaria in 1998, international spendingon malaria has more than doubled toapproximately US$ 200 million per year.Further untapped resources for malariacontrol may become available through debtrelief initiatives. Government spending on allhealth care is low in most African countries– typically less than US$ 15 per person per

year – and the costs of malaria control arehigh: artemisinin-based combination drugsto treat resistant malaria are likely to costUS$ 1–3 per treatment for the drug alone,and ITNs cost around US$ 5. Most of thecosts of preventing and treating malaria inAfrica today are in fact borne by peoplethemselves. For example, people buy nets,insecticide sprays, and coils, and spend aconsiderable amount of money on malariatreatment, which may contribute to poverty.Increasing the efficiency of domestic ”out ofpocket” spending is a priority, and this canbe achieved through government support forthe most effective interventions and theappropriate regulation to ensure that onlysafe, effective malaria interventions are soldand that the public is fully informed abouttheir use and effectiveness.

The recently established Global Fund toFight AIDS, Tuberculosis and Malaria (GFATM)is a major new source of grant funding fortackling malaria in Africa. Twenty-fivecountries and one multi-country group havesubmitted successful proposals to the GFATM.Almost all of these proposals build on thenational malaria control plans developed bythese countries with the support of the RollBack Malaria Partnership during the period1999–2001. The countries have been awardeda total of US$ 256 million for an initial twoyears to scale up malaria control activities.Depending on success, it is expected thatadditional funds will be made available for atotal period of five years.

Prevention of malaria during pregnancyresults in healthier mothers and infants.

Pic

ture

: W

HO

/TD

R


This report has been drafted in response to aseemingly very simple question: “What do weknow about malaria in Africa today?” In thepast, the answer to this question would havebeen, “It depends on whom you ask.”Although most experts would have agreed onthe fundamental facts, including the relativeimportance of the disease, its geographicaldistribution, and the key strategies forprevention and treatment, opinions wouldhave begun to diverge at the next level ofdetail – dictated by personal experience inthe absence of routinely collected andauthoritative information on the globalmalaria situation.

Accurate statistics on malaria in Africahave been difficult to collect and reportbecause of the enormity of the diseaseproblem, the weakness of health informationsystems, and the fact that treatment of mostmalaria cases, as well as many deaths fromthe disease, occurs outside the formal healthsystem. Following the period of internationalindifference to malaria, there was also littleinternational agreement on what informationwas needed for monitoring malaria controland how it should be collected. This situationis changing, and there is now a strongconsensus on priority indicators and the bestway of collecting representative information.

During the 1950s and 1960s, the malariaeradication campaign successfully eliminatedor controlled the disease in countries withtemperate climates and in some countrieswhere malaria transmission was low ormoderate. However, the emergence of drugand insecticide resistance, coupled withconcerns about the feasibility andsustainability of tackling malaria in areaswith weak infrastructure and hightransmission, brought an end to theeradication era, as well as to the bulk ofinternational funding for malaria control andinvestment in malaria research. Despiteinternational indifference in subsequentyears, progress continued to be made inunderstanding the problem of malaria andstrategies for its control. By the early 1990sthe international community began toappreciate that the malaria burden wasunacceptably high and worsening,

particularly in Africa, and that real reductionsin malaria mortality and morbidity werepossible with existing but under-used toolsand strategies.

In 1992, malaria control was re-establishedas a global health priority by a Conference ofMinisters of Health held in Amsterdam.Scientific interest in the disease and itscontrol, political commitment to reducing theburden of malaria, and the financial resourcesfor malaria research and control began toincrease rapidly. The project for AcceleratedImplementation of Malaria Control (1997-1998) represented an unprecedentedcontribution to the fight against malaria inAfrica south of the Sahara, in terms of bothtechnical support and funds. The fundingprovided for the project over the two yearswas estimated to have been more than 12times the contributions made by WHO duringthe previous decade.

By the year 2000 a sequence of criticalmilestones had been achieved and an ambitiousglobal commitment had been realized:

■ 1991-1998: malaria control expertiseand capacity were expanded andstrengthened, particularly in Africa,especially through the project forAccelerated Implementation of MalariaControl (1997-1998);

■ 1997: new research collaborations,notably the Multilateral Initiative onMalaria (http://mim.nih.gov/), wereformed.

■ 1998: the Roll Back Malaria Partnership(www.rbm.who.int/) was launched andconsensus on the core technicalstrategies for tacking malariaestablished.

■ 2000: the United Nations declared2001–2010 the Decade to Roll BackMalaria in developing countries,particularly in Africa (United NationsGeneral Assembly, Resolution 55/284).

■ 2000: malaria figured prominently in theUnited Nations’ Millennium DevelopmentGoals (General Assembly official records:27th Special Session: Supplement 3.Document A/S-27/19/Rev.1).

13

Introduction

■ 2000: African heads of state met in ahistoric summit in Abuja, Nigeria, toexpress their personal commitment totackling malaria and to establish targetsfor implementing the technicalstrategies to Roll Back Malaria.

■ 2001: resources for controlling malariawere significantly boosted with theestablishment of the Global Fund toFight AIDS, Tuberculosis and Malaria.

With the renewed internationalcommitment to fighting malaria, the need forregular and reliable information on the globalmalaria situation is greater than it has everbeen. The general indifference of the past hasgiven way to an urgent demand forinformation that can be used to define andanalyse the malaria situation and measureprogress towards the goals established by theinternational community and by nationalcontrol programmes. The World HealthOrganization (WHO) and the United NationsChildren’s Fund (UNICEF) are committed tomeeting this demand.

This report is an initial effort to collect,analyse, and present information on themalaria situation. The report focuses on Africa

and specifically on those African countries with the highest burden of the disease. These countries bear more than 90% of theglobal malaria burden. Emphasis is also givento the technical strategies for malaria control established by the Roll Back MalariaPartnership and the targets set at the Abuja Summit. In addition, with due regard to the importance of understanding the resource requirements of malaria control, a chapter on resource mobilization andfinancing is included.

The data contained in this report havebeen drawn from a variety of sources in

order to provide the most complete pictureof the malaria situation in Africa. The UNICEF Multiple Indicator Cluster Surveysand the and Health Surveys, in particular, are national surveys that represent a majoradvance in collection of baseline data toprovide benchmarks against which progresscan be measured. It is fully expected that the recent consensus on core data needs, well coordinated efforts to collect data, and progress in solving methodological and other data collection problems willtogether fulfil the new demands for malaria information.

14


■ 60% of those suffering with malaria should have access to andbe able to use correct, affordable, and appropriate treatmentwithin 24 hours of the onset of symptoms

■ 60% of those at risk for malaria, particularly children under 5years of age and pregnant women, will benefit from a suitablecombination of personal and community protective measures,such as ITNs

■ 60% of pregnant women at risk of malaria will be covered withsuitable combinations of personal and community protectivemeasures, such as ITNs

■ 60% of pregnant women at risk of malaria will have access tointermittent preventive treatmenta

■ 60% of epidemics are detected within 2 weeks of onset

■ 60% of epidemics are responded to within 2 weeks of detection

The goal of Roll Back Malaria is to halve the burden of malaria by 2010. The following targetsfor specific intervention strategies were established at the Abuja Malaria Summit, April 2000

RBM strategy Abuja target (by 2005)

Prompt access to effectivetreatment

Insecticide-treated nets (ITNs)

Prevention and control of malariain pregnant women

Malaria epidemic and emergencyresponse

a: The original Abuja declaration included the recommendation for chemoprophylaxis as well, but present WHO and RBMpolicy strongly recommends IPT — and not chemoprophylaxis — for prevention of malaria during pregnancy.

15


On the left is a normal, healthy, red blood cell with a smooth surface.The flexible and deformable disc shape allows it to flow easily throughnarrow blood capilliaries.

On the right is a similar red blood cell infected for one day withPlasmodium falciparum parasites. It has many knob-like protrusions. The cell’s rapid transformation to a more rigid spherical shape impedesflow through narrow blood capillaries. Additionally the protrusions actlike Velcro, causing the infected blood cell to bind to specific receptorssuch as those on the lining of blood vessels. These adhesions in the brainand the placenta are part of the cause of cerebral and placental malaria.

■ Picture from scanning electron microscope: Lirong Shi, MichaelDelannoy, David Sullivan, Johns Hopkins Bloomberg School of PublicHealth, Malaria Research Institute.

17

Chapter 1: The burden of malaria in Africa

About 90% of all malaria deaths in the worldtoday occur in Africa south of the Sahara.This is because the majority of infections inAfrica are caused by Plasmodium falciparum,the most dangerous of the four humanmalaria parasites. It is also because the mosteffective malaria vector – the mosquitoAnopheles gambiae – is the most widespreadin Africa and the most difficult to control. Anestimated one million people in Africa diefrom malaria each year and most of these arechildren under 5 years old (1).

Malaria affects the lives of almost allpeople living in the area of Africa defined bythe southern fringes of the Sahara Desert in the north, and a latitude of about 28°in the south. Most people at risk of thedisease live in areas of relatively stablemalaria transmission – infection is commonand occurs with sufficient frequency thatsome level of immunity develops. A smallerproportion of people live in areas where risk of malaria is more seasonal and lesspredictable, because of either altitude orrainfall patterns. People living in theperipheral areas north or south of the main endemic area (Figure 1.1) or bordering highland areas are vulnerable to highly seasonal transmission and to malaria epidemics.

In areas of stable malaria transmission, veryyoung children and pregnant women are thepopulation groups at highest risk for malariamorbidity and mortality. Most childrenexperience their first malaria infectionsduring the first year or two of life, when theyhave not yet acquired adequate clinicalimmunity – which makes these early yearsparticularly dangerous. Ninety percent of allmalaria deaths in Africa occur in youngchildren. Adult women in areas of stabletransmission have a high level of immunity,but this is impaired especially in the firstpregnancy, with the result that risk ofinfection increases.

Malaria has been well controlled oreliminated in the five northernmost Africancountries, Algeria, Egypt, Libyan Arab

Jamahiriya, Morocco, and Tunisia. In thesecountries the disease was causedpredominantly by Plasmodium vivax andtransmitted by mosquitoes that were mucheasier to control than those in Africa southof the Sahara. Surveillance efforts continuein most of these countries in order to preventboth a reintroduction of malaria parasites tolocal mosquito populations, and theintroduction of other mosquito species thatcould transmit malaria more efficiently (aparticular risk in southern Egypt). The malariasituation in these countries is not consideredfurther in this report.

Malaria is endemic in some of the offshoreislands to the west of mainland Africa – SaoTome and Principe and São Tiago Island ofCape Verde. In the east, malaria is endemic inMadagascar, in the Comoro islands (both theIslamic Federal Republic of the Comoros andthe French Territorial Collectivity of Mayotte),

1. The burden of malaria in Africa

Distribution of endemic malaria

Source: reference 2

Endemic malariaMalaria marginal/epidemic prone

Roll BackMalaria target

The global target of Roll Back Malaria

is to halve malaria-associated

morbidity andmortality by 2010

compared with levels in year 2000.

Figure 1.1

and on Pemba and Zanzibar, but has beeneliminated from the island of Reunion. InMauritius, malaria has been well controlledsince the 1950s, but occasional outbreaks ofvivax malaria occur, the last in associationwith a cyclone in 1982. Since that year there has been a steady decrease in cases and risk is now extremely low. Seychelles has been free of malaria since 1930, and malaria vectors are believed to no longer exist there.

1.1

Burden of malaria on health in AfricaMortality

There are three principal ways in whichmalaria can contribute to death in youngchildren (Figure 1.2). First, an overwhelmingacute infection, which frequently presents asseizures or coma (cerebral malaria), may kill achild directly and quickly. Second, repeatedmalaria infections contribute to thedevelopment of severe anaemia, whichsubstantially increases the risk of death.Third, low birth weight – frequently theconsequence of malaria infection in pregnant women – is the major risk factor for death in the first month of life (3). In addition, repeated malariainfections make young children moresusceptible to other common childhoodillnesses, such as diarrhoea and respiratoryinfections, and thus contribute indirectly to mortality (4).

The consensus view of recent studies andreviews is that malaria causes at least 20% ofall deaths in children under 5 years of age inAfrica (Figures 1.3 and 1.4). Althoughrespiratory disease caused by a variety ofinfectious agents results in a similarproportion of deaths, P. falciparum is themost important single infectious agentcausing death among young children.

Morbidity and long-term disability

Children who survive malaria may sufferlong-term consequences of the infection.Repeated episodes of fever and illness reduce appetite and restrict play, socialinteraction, and educational opportunities,thereby contributing to poor development.An estimated 2% of children who recover

18

The Africa Malaria Report–2003

Most of the malaria burden isfrom deaths in young children

Although adults also become infected with malaria,the illness is usually less severe thanks to their acquired immunity. Infections in young children are serious and may kill if not treated promptly.

Rest of the world Africa

0–4 5–14 15+Age (years)

1000

800

600

400

200

Deat

hs (t

hous

ands

)

Source: WHO Global Burden of Disease project, estimates for 2000, reference 17

0

Malaria kills children inthree different ways

Low birth weightPreterm delivery

Cerebral malariaRespiratory distressHypoglycaemia

Severeanaemia

Death

Acute febrileillness

Infection inpregnancy

Chronic,repeatedinfection

Figure 1.2

Figure 1.3

from malaria infections affecting the brain (cerebral malaria) suffer from learningimpairments and disabilities due to braindamage, including epilepsy and spasticity (5).

1.2

Burden of malaria on Africanhealth systemsIn all malaria-endemic countries in Africa, 25–40% (average 30%) of alloutpatient clinic visits are for malaria (withmost diagnosis made clinically). In these same countries, between 20% and 50% of allhospital admissions are a consequence ofmalaria (see country profiles for details). With high case-fatality rates due to latepresentation, inadequate management, and unavailability or stock-outs of effectivedrugs, malaria is also a major contributor to deaths among hospital inpatients (Figure 1.5).

This high burden may in fact be partly aresult of misdiagnoses, since many facilitieslack laboratory capacity and it is oftendifficult clinically to distinguish malaria fromother infectious diseases. Nonetheless, malariais responsible for a high proportion of publichealth expenditure on curative treatment, andsubstantial reductions in malaria incidencewould free up available health resources andfacilities and health workers’ time, to tackleother health problems.

1.3

Burden of malaria on the poorPoor people are at increased risk both ofbecoming infected with malaria and ofbecoming infected more frequently. Childmortality rates are known to be higher inpoorer households and malaria is responsiblefor a substantial proportion of these deaths.In a demographic surveillance system in rural areas of the United Republic of Tanzania,under-5 mortality following acute fever(much of which would be expected to be due to malaria) was 39% higher in thepoorest socioeconomic group than in the richest (6).

A survey in Zambia also found asubstantially higher prevalence of malariainfection among the poorest population

19


Under-five all cause mortality

Per 1000 children under-five

19–7980–126

127–165166–211212–316No data

Source: UNICEF, State of the World's Children, 2003

% of outpatient visits due to malaria

Western Africa Central/Eastern Africa Southern Africa

50

40

30

20

10

0

% of hospital admissions due to malaria


50

40

30

20

10

0

Source: AFRO routine Health Information System data. Averages 1998–2001. Error bars give the standard errors.

% of hospital deaths due to malaria


50

40

30

20

10

0

Malaria burden on health facilities

Under-five All ages

Figure 1.4

Figure 1.5

groups (7) (Figure 1.6). Poor families live indwellings that offer little protection againstmosquitoes and are less able to affordinsecticide-treated nets. Poor people are alsoless likely to be able to pay either foreffective malaria treatment or fortransportation to a health facility capable oftreating the disease.

Both direct and indirect costs associatedwith a malaria episode represent a substantialburden on the poorer households. A study innorthern Ghana found that, while the cost ofmalaria care was just 1% of the income of the rich, it was 34% of the income of poor households (8).

1.4

Recent trends in the burden of malariaRoutine case detection and reporting

Data from health facilities are potentiallyuseful for monitoring time trends in thenumber of malaria cases and deaths but havesevere limitations (Figure 1.7). In Africa, mostcases of malaria are diagnosed on the basisof clinical symptoms and treatment ispresumptive, rather than based on laboratoryconfirmation. Moreover, malaria parasitaemiais common among clinic attendees in manyendemic areas, so that a positive laboratoryresult does not necessarily mean that thepatient is ill with malaria. The main clinicalsymptoms of malaria – fever and general

weakness – are nonspecific and may well bedue to other common infections.

Reporting from facilities to districts andfrom districts to the ministry of health variesin its completeness and timeliness fromcountry to country and often does notinclude nongovernment facilities. Thus,routine reports of the number of malariacases and deaths have limited value forcomparisons of the malaria burden betweencountries. Demographic and health surveys(DHS) and other sources (9) indicate that lessthan 40% of malaria morbidity and mortalityis seen in formal health facilities – a smallfraction of the total burden. However,routinely collected data are often the onlyinformation available over a prolonged periodand over a wide geographical area. Whilethese data are of use for local programmeplanning, major investment in improving boththe quality of health information systems andaccess to health services would be requiredbefore their utility for monitoring changes inmalaria disease trends could be assessed.

At present, the most reliable data availableon trends in malaria deaths in children under5 years of age is obtained from demographicsurveillance systems (DSS), which measuredeaths and possible causes prospectively overtime in populations of known size andcomposition. The number of DSS sites isincreasing: 24 sites in 13 African countriesare collaborating under the INDEPTH network(International Network of field sites withcontinuous Demographic Evaluation of

20


Parasite prevalence is higher in poor children

The prevalence of malaria infection was higher in under-fives from poorer families in 10 districts suveyed in Zambia.

Source: reference 7

Poorest Second Middle Fourth Richest0

20

40

60

80

% u

nder

-five

s in

fect

ed

Wealth quintile

Figure 1.6

Populations and Their Health) (10). Most ofthese sites are in eastern and southern Africa;there are a few sites in the west of thecontinent but none in central Africa.

Recently, data from 1982–1998 wereanalysed across 28 DSS sites, adjusting forthe specificity and sensitivity of verbalautopsies that were used to attribute deathsto malaria (11). Malaria mortality in under-5salmost doubled in eastern and southernAfrica over the period 1990–1998 comparedwith 1982–1989. It is known that theprevalence of malaria infections caused bychloroquine-resistant parasites increasedsubstantially from the late 1980s in thesesame areas (Figure 1.8). Thus, although themethodology cannot prove cause and effect,it is very likely that some of this increase inchild mortality was related to some extent tothe spread of chloroquine-resistant malaria.In west Africa the mortality rate remainedthe same; here too, however, malaria becameproportionally more important (11). Analysisof mortality data being collected fromINDEPTH using standardized verbal autopsyquestionnaires since 2000 should soonprovide further insight into more recentdisease trends.

Throughout Africa south of the Sahara, thedecrease in all-cause under-5 mortality thatwas apparent during the 1970s and 1980slevelled off in the 1990s (Figure 1.9), perhapspartially as a result of increased malariamortality. Some of the important factors thatmay have contributed to the increasing malariaburden in these African settings include:

■ drug resistance (12)

21


Source: Routine health information system data,AFRO, for under-fives

UR Tanzania

Trends in outpatient visits and

hospital admissions for malaria

Malawi ZambiaUganda

80

60

40

20

019901985 1995 2000

% hospital admissions due to malaria

80

60

40

20

019901985 1995 2000

% outpatient visits due to malaria

Malaria mortality in DSS sitesEstimates for average prevalences of Plasmodium falciparum of 63% in western Africa and 39% in eastern

and southern Africa, adjusted for variations in the sensitivity and specificity of verbal autopsy.

1982–1989 1990–1998Source: reference 11

0

10

20

30

40

0

4

8

12

Mal

aria

mor

talit

y/10

00 u

nder

-5 y

ears

% u

nder

-five

dea

ths

due

to m

alar

ia

Western Africa Eastern/southern Africa Western Africa Eastern/southern Africa

Figure 1.7

Figure 1.8

■ more frequent exposure of non-immunepopulations

■ emergence of HIV/AIDS (13, 14)

■ climate and environmental change (15)

■ breakdown of control programmes (16).

1.5

Future prospectsFrom the time trends shown, it appears thatRBM is acting against a background ofincreasing malaria burden. With the typical2–3-year delay in national-level databecoming available, it is still too early toevaluate the extent to which RBM hasachieved a levelling-off or reversal of therising trend in the malaria burden. The verylow level of coverage with ITNs and untreatednets documented in 2000 and 2001 falls farbelow the coverage levels in the ITN trialsthat demonstrated substantial healthbenefits. It should therefore come as nosurprise that significant reductions in childmortality have yet to be observed. The impactof treatment coverage levels is more difficultto estimate, given both a lack of informationon promptness and dosage, and varying levels

of drug effectiveness. Coverage levelsapproaching the Abuja target of 60% willprobably be required before the full effect ofITNs and effective treatment on child healthwill become apparent.

References

1. The World Health Report 2002: reducing risks,promoting healthy life. Geneva, World HealthOrganization, 2002.

2. MARA/ARMA collaboration (Mapping Malaria Risk inAfrica), July 2002. www.mara.org.za.

3. Steketee RW et al. The burden of malaria in pregnancyin malaria-endemic areas. American Journal of TropicalMedicine and Hygiene, 2001, 64(1,2 S):28–35.

4. Molineaux L. Malaria and mortality: someepidemiological considerations. Annals of TropicalMedicine and Parasitology, 1997, 91(7):811–825.

5. Murphy SC, Breman JG. Gaps in the childhood malariaburden in Africa: cerebral malaria, neurological sequelae,anemia, respiratory distress, hypoglycemia, andcomplications of pregnancy. American Journal of TropicalMedicine and Hygiene, 2001, 64(1,2 S):57–67.

6. Mwageni E et al. Household wealth ranking and risksof malaria mortality in rural Tanzania. In: Third MIMPan-African Conference on Malaria, Arusha, Tanzania,17–22 November 2002. Bethesda, MD, MultilateralInitiative on Malaria: abstract 12.

22


Trends in all-cause, under-five mortality in Africa

Deat

hs p

er 1

000

unde

r-fiv

es

0

100

200

300

400

Western Eastern CentralSouthern

1960 1970 1980 1995 20001990

Source: DHSFigure 1.9

7. Report on the Zambia Roll Back Malaria baselinestudy undertaken in 10 sentinel districts, July to August2001. Zambia, RBM National Secretariat, 2001.

8. Akazili J. Costs to households of seeking malaria carein the Kassena-Nankana District of Northern Ghana. In:Third MIM Pan-African Conference on Malaria, Arusha,Tanzania, 17–22 November 2002. Bethesda, MD,Multilateral Initiative on Malaria: abstract 473.

9. Breman JG. The ears of the hippopotamus:manifestations, determinants, and estimates of themalaria burden. American Journal of Tropical Medicineand Hygiene, 2001, 64(1,2 S):1–11.

10. Population and health in developing countries. Vol. 1.Population, health and survival at INDEPTH sites.Ottawa, International Development Research Centre,2002.

11. Korenromp EL et al. Measuring trends in childhoodmalaria mortality in Africa: a new assessment of progresstoward targets based on verbal autopsy. [LancetInfectious Diseases, conditionally accepted, March 2003].

12. Trape J-F. The public health impact of chloroquineresistance in Africa. American Journal of TropicalMedicine and Hygiene, 2001, 64(1,2 S):12–17.

13. Grimwade K et al. HIV-infection in adults increasesrates of severe and fatal falciparum malaria in regions ofunstable transmission. In: XIVth International AIDSconference 2002, Barcelona, Spain: abstract ThPeC7604.

14. Nwanyanwu OC et al. Malaria and humanimmunodeficiency virus infection among male employeesof a sugar estate in Malawi. Transactions of the RoyalSociety of Tropical Medicine and Hygiene, 1997,91(5):567–569.

15. Mouchet J et al. Evolution of malaria in Africa forthe past 40 years: impact of climatic and human factors.Journal of the American Mosquito Control Association,1998, 14(2):121–130.

16. Sharp B et al. Malaria control by residual insecticidespraying in Chingola and Chililabombwe, CopperbeltProvince, Zambia. Tropical Medicine and InternationalHealth, 2002, 7(9):732–736.

17. The World Health Report 2001. Mental health: newunderstanding, new hope. Geneva, World HealthOrganization, 2001.

23


24


Before the development of insecticide-treated nets (ITNs) as a new technology in themid-1980s, people in many countries werealready using nets, mainly to protectthemselves against biting insects and forcultural reasons (1–3). It was only recentlyappreciated that a net treated withinsecticide offers much greater protectionagainst malaria: not only does the net act asa barrier to prevent mosquitoes biting, butalso the insecticide repels, inhibits, or killsany mosquitoes attracted to feed. Thus ITNsprovide protection both to individualssleeping under them and to other communitymembers. The effect is so significant that useof ITNs is considered to be one of the mosteffective prevention measures for malaria.

2.1

EvidenceRandomized controlled trials in Africansettings of different transmission intensitieshave shown that ITNs can reduce the numberof under-5 deaths by around one-fifth (5),saving about 6 lives for every 1000 childrenaged 1–59 months protected each year(Figure 2.1). The incidence of clinical episodesof Plasmodium falciparum infection isreduced by 50% on average. When used bypregnant women, ITNs are also efficacious inreducing maternal anaemia, placentalinfection, and low birth weight (6).

This may even be an underestimate of the

Abuja targetIn April 2000, African

heads of stateparticipating in the

Abuja Summit agreedthat at least 60% of

those at risk formalaria, particularly

children under 5 yearsof age and pregnant

women, are to benefitfrom the most suitable

combination ofpersonal and

community protectivemeasures such as ITNs

by 2005 (4).

2. Insecticide-treated nets

ITNs reduce under-5 mortality

Randomized controlled trials showed an overall under-5 mortality reductionof 17% in communities provided with ITNs compared with communities not provided with ITNs. The impact was similar across a range of malaria endemicities. Impact derives not only from a reduction in malaria deaths,but also from reductions in child deaths due to other causes that are associated with, or exacerbated by, malaria, such as acute respiratory infection, low birth weight, and malnutrition.

Source: reference 5, 24

Burkina Faso 1997

Ghana 1996

Kilifi, Kenya 1996

Gambia 1995

Asembo Bay, Kenya 2001

0 10 20 30 40 50

% reduction in under-5 mortality

Figure 2.1

25

Chapter 2: Insecticide-treated nets

efficacy of ITNs because the impact ofreduced mosquito burden extends tohouseholds and communities without nets,which reduces the apparent differencebetween study areas with nets and studyareas without nets. The protection affordedto non-users in the vicinity is difficult toquantify, but it appears to extend overseveral hundred metres. From observedreductions in parasite prevalences, it hasrecently been estimated that, in the longterm, widespread use of ITNs – if regularlyretreated – will massively reduce malariatransmission (7), but this effect will becomefully apparent only after the usual 2-yearduration of a trial.

The ITN trials achieved their impact withclose to 100% of households possessing netsand 50–75% of under-5s sleeping underthem, a level of use similar to the Abujatarget of 60%. Where lower coverage and userates are achieved, the impact on mortalitywill be less.

Subsequent programmes havedemonstrated the effectiveness of ITNs underfield conditions. In a large-scale socialmarketing programme in two rural districts inthe south of the United Republic of Tanzaniawith high perennial malaria transmission, ITNcoverage of infants rose from less than 10%at baseline to more than 50% 3 years later.ITN use was associated with a 27% increasein survival of children aged 1 month to 4years and a 63% reduction of anaemia in thissame age group (evaluated by case–controldesign) (8).

In the Gambia, the National ImpregnatedBednet Programme achieved an 83% nettreatment rate and reported 77% of under-5sand 78% of women of childbearing agesleeping under ITNs (9). Overall under-5mortality fell by 25%, and case–controlstudies suggested that there were 59% fewer episodes of uncomplicated malaria in ITN users (10, 11).

2.2

Progress: ITN strategy plansEighteen of the 40 malaria-endemic countriesin Africa with country strategy plans forrolling back malaria have developed strategicplans which include increasing access to ITNs.Twenty-five African countries have

successfully applied for funding in the secondround of Global Fund applications.

2.3

Progress: taxes and tariffsThe cost of ITNs is a barrier to theirwidespread use. As one element in reducingprices, the Abuja Declaration committedgovernments to “reduce or waive taxes andtariffs for nets and materials, insecticides,antimalarial drugs and other recommendedgoods and services that are needed formalaria control strategies”. Eighteen countrieshave now reduced or eliminated taxes andtariffs (Figure 2.2). Time-limited changes intax or tariff regimes can be introducedthrough informal agreements between healthand finance ministries, but more permanentarrangements normally require nationallegislation.

Most countries apply the “HarmonizedCommodity Description and Coding System”to classify products introduced by the WorldCustoms Office (12). Under this system, eachproduct is assigned a six-digit code for thepurposes of levying tariffs and collectingtrade statistics. Nets are currently classified astextiles and customs offices can be reluctant

Reduction of taxes and tariffs

Source: RBM data from countries, 2003

Before Abuja, April 2000By Africa Malaria Day, 2001By end of 2002

Figure 2.2

26


to give exemption for the whole range ofproducts covered by the code. Somecountries also subscribe to regionalagreements on tariffs and taxation rates,which can influence the adoption of policychange. For example, the West AfricanEconomic and Monetary Union requires all of its eight member states to adhere to the Common External Tariff Resolution,which stipulates fixed rates for import duty of 20% and for value-added tax (VAT) of 18%. Clearly, changes in nationalpolicy would be greatly facilitated by changes to international agreements.

2.4

Progress: long-lastinginsecticidal nets

In response to low re-treatment rates ofconventional insecticide-treated nets,especially in Africa, WHO prompted industry to develop long-lasting insecticidal nets (LLINs) – ready-to-use,factory-pretreated nets that require nofurther treatment during their expectedlifespan of 4–5 years. This technologyobviates the need for re-treatment (unlikeconventional ITNs, LLINs resist washing) and reduces both human exposure (at anygiven time, most of the insecticide is hiddenand not bioavailable) and the risk ofenvironmental contamination.

Using the most recent fibre technologies,LLINs are regarded as a major breakthroughin malaria prevention. One LLIN is alreadycommercially available and is recommendedby WHO. At a current price of around US$ 5per net, LLINs are already more cost-effectivethan conventionally treated nets. Efforts arebeing made to scale up production capacityto meet demand, which is already high. TheRBM partnership is facilitating technologytransfer and stimulating local production of LLINs in Africa (13).

2.5

Progress: coverage In nine countries surveyed between 1997 and2001, a median 13% of households possessone or more nets (range 1.1–54%). A median1.3% (range 0.2–4.9%) of householdssurveyed in three countries own at least oneITN (14). The proportion of under-5s sleepingunder nets is also low – about 15% across 28countries surveyed. Even fewer children (lessthan 2%) sleep under ITNs. Only twocountries, the Gambia and Sao Tome and Principe, reported ITN use rates of more than 10% (Figure 2.3).

While current rates of coverage aregenerally low, the availability and use of netshave increased appreciably over the past 10years, particularly in countries where netswere not normally used. In the UnitedRepublic of Tanzania, for example, nets wererare in the 1980s, especially in rural areas,but ownership has increased to 63% in towns

Use of nets in Africa

Note: Dotted line indicates Abuja target. For Mali and Zimbabwe no data on ITNs available.

% under-fives who slept under a net the night before the survey

0 20 40 60 80

Any netITN

Guinea-BissauSao Tome & Principe

GambiaMali

ComorosBenin

CARMadagascar

ChadSudan, northern

TanzaniaNigerKenya

SomaliaEquatorial Guinea

Sierra LeoneSenegal

TogoDRC

CameroonAngola

Côte d'IvoireMalawiUgandaZambia

RwandaZimbabwe

BurundiSwaziland

Source: MICS and DHS, 1998–2001Figure 2.3

27


and to 29% in rural areas (14). Such trendsare encouraging and highlight the progressthat is being made.

2.6

Challenges: increasing coverageMost African households in malaria risk areasdo not possess any net, whether treated withinsecticide or not. To achieve adequatecoverage most countries will require manymore nets; to cover all Africans at risk (16),an estimated total of 260 million nets wouldbe needed.

Increasing ITN availability will requirelarge-scale expansion of supply anddistribution. Barriers to increasing the supplyand distribution of nets and insecticidesinclude taxes and tariffs, regulatory issues,and inadequate distribution systems. Barriersto increasing the demand for nets andinsecticides relate to the price, to theiraffordability for households, and topromotion and marketing.

There is also scope to increase the use ofITNs by providing insecticide treatment forany untreated nets already in houses. Basedon the comparative coverage with untreatedand treated nets, this could double thepercentage of households with ITNs.

Low insecticide re-treatment rates areanother challenge. Insecticide for nettreatment is still an unfamiliar commodity in Africa. Moreover, people’s motivation for using nets is often to reduce mosquitonuisance, not to repel or kill malaria-transmitting mosquitoes. The increasingavailability of attractive branded formulations in Africa should stimulatedemand for insecticides, and the development of LLINs is another potentialsolution to the problem of low re-treatment rates.

2.7

Challenges: overcomingdisparities in net coverageA major barrier to net ownership is poverty.The most common reason cited for notpossessing a net is lack of money: the priceof a net represents a large proportion of theincome of a poor household.

2.8

Scaling upNet possession and use have to increaseconsiderably if the gap between the numberof under-5s who would benefit from a netand those who currently sleep under one is tobe reduced. The challenge is to find thebalance between covering the costs ofincreasing ITN coverage and stimulating thegrowth of commercial markets, whileensuring that the poorest and mostvulnerable are protected (23).

In most malaria-endemic African countriesthe public sector does not have the financialor logistic capacity to extend net use to thescale required. Most countries spend onlyUS$ 4 per capita a year on health – theequivalent of the average cost of anuntreated net. The Abuja target forexpanding ITN use in Africa will thereforerequire synergy between public and privatesector activities.

In providing an enabling environment forscaling-up actions, governments need tofocus on the following priorities:

■ Creating demand for ITNs through healthinformation channels and mass media.

■ Providing sustained subsidies targeted tothe most vulnerable groups, preferablythrough a system that uses public channels(e.g. at antenatal clinics) for delivery of

ITN coverage in Zambia

A community programme in rural Zambia achieved netcoverage of more than 60% of individuals at risk. Withsupport from UNICEF, ITNs have been sold since 1995 by theprovincial community-based malaria control programme. In asurvey, a third of the population reported sleeping under anet the previous night, more than 80% had re-treated theirnets at least once, and pregnant women and under-5s weregiven priority for net use. The incidence of malaria wassignificantly lower among net users than non-users (18).

Eritrea’s national impregnation campaign

The current estimate of the number of nets in use in Eritreais 533 200, with some 226 000 having been procured anddistributed in 2002. Net treatment rate increased from 20%in 2000 to 58.5% in 2002. Most households in endemicareas of Eritrea possess two nets (17).

28


Households are willingto pay for nets

Willingness to pay Current cost

Source: references 19, 20 and 21

0

1

2

3

4

5

US$ 6

UR Tanzania Nigeria Kenya

Studies in Kenya, Nigeria and the United Republic of Tanzania indicate thathousehold members are willing to pay for ITNs, but typically less than the currentcost. One approach to reducing inequitiesis social marketing – subsidizing, to some extent, the cost of ITNs on sale. In the United Republic of Tanzania, a social marketing project that started in 1998 increased coverage while reducing socioeconomic inequities.

Net ownership and use islowest in poor households

Poor households are less likely to have ITNs or any net. The disparity contrasts with a probably higher burden of malaria in poorer households, who live in houses and areas that typically allow for more exposure to mosquitoes, and who are less able to afford treatment once infection has occurred.

Source: MICS, 1998–2001, average across 22 countries

0

10

20

30

40

% u

nder

-five

s sl

eepi

ng u

nder

net

s

Poorest Second Middle Fourth Richest

Untreated nets ITNs

Wealth quintile

Children in rural households are less likely to use nets

In rural areas, fewer children under-5 sleep under a net than in urban areas, even though malaria is often more prevalent in rural areas. The magnitude of the urban–rural difference was larger for ITNs than for any net. Pooled over all surveys, the ratio of urban to rural coverage was 1.8 for any net and 2.8 for ITN. From these same surveys, use of any nets or ITNs is similar in boys and girls.

Source: MICS and DHS, 1998–2001, average across 22 countries

Untreated nets ITNs

0

10

20

30

40

Boys Girls

% u

nder

-five

s sl

eepi

ng u

nder

net

s

% u

nder

-five

s sl

eepi

ng u

nder

net

s

0

10

20

30

40

Urban Rural

Figure 2.4 Figure 2.5

Figures 2.6 and 2.7

29


subsidies but commercial distributionchannels for delivery of the goods.

■ Stimulating and facilitating thedevelopment of commercial markets, throughtax and tariff reduction and by streamliningthe regulation of new insecticide products.Competition between manufacturers anddistributors must be promoted to ensure thatnets are available to the general populationat the lowest possible price.

■ Monitoring insecticide resistance.

■ Possibly, market priming (i.e. thetemporary procurement and distribution ofITNs, aimed at strengthening commercialdistribution channels) in areas where thedemand for nets is too low formanufacturers to make an economicalreturn.

To overcome the challenge of low re-treatment rates, there should be a strongerrole for subsidy of insecticide distributionthrough publicly funded channels. This is thesystem followed in the world’s largest andlongest-sustained ITN programmes, namelythose in China and Viet Nam (23).

References

1. MacCormack CP, Snow RW. Gambian culturalpreferences in the use of insecticide-impregnated bednets. Journal of Tropical Medicine and Hygiene, 1986,89(6):295–302.

2. Robert V, Carnevale P. Influence of deltamethrintreatment of bed nets on malaria transmission in theKou valley, Burkina Faso. Bulletin of the World HealthOrganization, 1991, 69(6):735–740.

3. Aikins MK, Pickering H, Greenwood BM. Attitudes tomalaria, traditional practices and bednets (mosquitonets) as vector control measures: a comparative study infive west African countries. Journal of Tropical Medicineand Hygiene, 1994, 97(2):81–86.

4. The African summit on Roll Back Malaria. Abuja,Nigeria, 25 April 2000. Geneva, World HealthOrganization, 2000 (document WHO/CDS/RBM/2000.17).

5. Lengeler C. Insecticide-treated bednets and curtainsfor preventing malaria (Cochrane Review). In: TheCochrane Library, Issue 4. Oxford, Update Software,2001.

6. Garner P, Gulmezoglu AM. Prevention versustreatment for malaria in pregnant women. In: TheCochrane Library, Issue 2. Oxford, Update Software,2000.

7. Smith T et al. Effects of insecticide-treated mosquitonets on malaria transmission. In: Third EuropeanCongress on Tropical Medicine and International Health,Lisbon, Portugal, 8–11 September 2002.

8. Armstrong Schellenberg JRM et al. Effect of large-scale social marketing of insecticide-treated nets onchild survival in rural Tanzania. Lancet, 2001,357:1241–1247.

9. Cham MK et al. Implementing a nationwideinsecticide-impregnated bednet programme in TheGambia. Health Policy Plan, 1996, 11(3):292–298.

10. D'Alessandro U et al. The Gambian NationalImpregnated Bed Net Programme: evaluation ofeffectiveness by means of case-control studies.Transactions of the Royal Society of Tropical Medicineand Hygiene, 1997, 91(6):638–642.

11. D'Alessandro U et al. Mortality and morbidity frommalaria in Gambian children after introduction of animpregnated bednet programme. Lancet, 1995,345(8948):479–483.

12. Harmonized Commodity Description And CodingSystem. http://www.com-law.net/findlaw/customs/hs.htm.

13. Guillet P et al. Long-lasting treated mosquito nets: abreakthrough in malaria prevention. Bulletin of theWorld Health Organization, 2001, 79(10):998.

14. Demographic and Health Surveys (DHS). Calverton,MD, ORC Macro. http://www.measuredhs.com.

15. Progress Report Regional Procurement Centre,Pretoria, South Africa. New York. United NationsChildren's Fund, 2002.


30


17. Mission Report on Mass Mosquito Net ImpregnationCampaign, Eritrea, 2002. Brazzaville, WHO RegionalOffice for Africa (document WHO/AFRO/CDS/VBC/2002).

18. Chimumbwa J. A community-based programme inZambia. 1999. Luapula Community-based malariaprevention and control programme. Presentation atSecond International Conference on Insecticide TreatedNets, Dar es Salaam, United Republic of Tanzania, 11-14October 1999.

19. Guyatt HL, Ochola SA, Snow RW. Too poor to pay:charging for insecticide-treated bednets in highlandKenya. Tropical Medicine and International Health, 2002,7(10):846–850.

20. Onwujekwe O et al. Hypothetical and actualwillingness to pay for insecticide-treated nets in fiveNigerian communities. Tropical Medicine andInternational Health, 2001, 6(7):545–553.

21. Simon JL et al. How will the reduction of tariffs andtaxes on insecticide-treated bednets affect householdpurchases? Bulletin of the World Health Organization,2002, 80(11):892–899.

22. Hanson K et al. Equity and ITNs in Tanzania: evidencefrom a social marketing project. Third MIM Pan-AfricanMalaria Conference 2002, Arusha, Tanzania: abstractno.446.

23. Global Partnership to Roll Back Malaria. Scaling-upinsecticide-treated netting programmes in Africa. Astrategic framework for coordinated national action.Geneva, World Health Organization, 2002 (documentWHO/CDS/RBM/2002.43.)

24. Phillips-Howard PA et al. The efficacy of permethrin-treated bednets on child mortality and morbidity inwestern Kenya. American Journal of Tropical Medicineand Hygiene, 2003 (in press).

31

Chapter 3: Prompt and effective treatment

Prompt and effective treatment of malaria isa critical element of malaria control (1). InAfrica south of the Sahara, where mostmalaria is due to Plasmodium falciparum andpotentially fatal, early and effective treatmentcould save many lives. It is vital that sufferers,especially children aged under 5 years, starttreatment within 24 hours of the onset ofsymptoms, to prevent progression – oftenrapid – to severe malaria and death (2).

A strong health system would provide forreliable diagnosis as the basis for optimaltreatment. However, in most malaria-endemicareas, access to curative and diagnosticservices is limited and drugs are purchasedthrough the private, informal sector (3, 4).Moreover, diagnosis is complicated by the lackof a specific clinical presentation, frequentoccurrence of several diseases simultaneously,and – in areas of intense transmission –asymptomatic malaria infections. In high-transmission malaria-endemic areas, WHOtherefore recommends that, as part of thestrategy of Integrated Management ofChildhood Illnesses (IMCI), all under-5s withfever be presumptively treated withantimalarials (5). Community-levelinterventions to strengthen homemanagement of children with fever aregaining importance as part of efforts toimprove access to prompt treatment,particularly in isolated rural areas.

3.1

EvidenceThe global consensus that access to prompt,effective treatment should be a key element ofthe RBM strategy is based on the widespreadrecognition that untreated falciparum malariacontributes both directly and indirectly to thedeath of non-immune individuals, sometimeswithin hours of the onset of symptoms (2).Prompt, effective treatment of malaria andappropriate management of clinicalcomplications will be life-saving.

Uncontrolled studies in Madagascar (7) andthe United Republic of Tanzania (8) revealedsignificant reductions in mortality when

research teams provided prompt access toantimalarial treatment. However, these studiestook place in circumstances where the obstaclesto access that characterize most health systemsin endemic countries had been eliminated.

Randomized, controlled trials of treatmentof febrile illness with reduction of mortalityas the end-point are fraught withmethodological and ethical problems andhave produced conflicting results. In a widelyquoted community-randomized trial in anarea of low, seasonal malaria transmission inEthiopia, under-5 mortality was reduced by40% as a result of teaching mothers toprovide prompt chloroquine treatment forfevers at home (9) (Figure 3.1). However, ageneral improvement in child care may havecontributed to this high level of impact.

Other indirect evidence attributed the lowmalaria-specific mortality in Brazzaville to thewidespread use of chloroquine as self-treatment (10). Conversely, an increase in childmortality following the spread of chloroquineresistance was observed at a demographicsurveillance site in rural Senegal (11) (Figure3.2). Demographic and Health Surveysdocumented a 15% reduction in infant and

3. Prompt and effective treatment

Abuja target

In April 2000, Africanheads of state

participating in theAbuja Summit agreedthat by the year 2005at least 60% of thosesuffering from malaria

should have promptaccess to and be able

to use correct,affordable, and

appropriate treatmentwithin 24 hours of theonset of symptoms (6).

With spreading chloroquine resistance, malaria mortality in children increased in three rural sites in Senegal

Source: reference 11

Malaria mortality/1000 child deaths

Niakar

0 4 8 10

Bandafassi

Mlomp 1984–1989

1990–1995

Training of mothers in chloroquine treatmentreduced under-five mortality in Ethiopia

Source: reference 9

All-cause mortality/1000 child-years

No intervention

0 20 40 60

Intervention

Figure 3.2

Figure 3.1

child mortality in Malawi during the 1990s ata time of increasing or stable rates of infantand child mortality in Kenya, Rwanda,Uganda, the United Republic of Tanzania,Zambia and Zimbabwe. The exceptionalmortality reduction in Malawi is likely to be due partly to the 1993 change in drug policy from chloroquine to sulfadoxine–pyrimethamine (SP).

3.2

Drug resistanceAntimalarial drug resistance has become oneof the greatest challenges in malariatreatment. Chloroquine, the cheapest andmost widely available antimalarial drug, haslost its clinical effectiveness in most parts ofAfrica (Figure 3.3). Resistance of Plasmodiumfalciparum to the most affordable alternativedrugs, notably SP, is also an emerging problemin eastern and southern Africa (Figure 3.4).

Several newly developed drugs could replacethose that are no longer effective. Inparticular, artemisinin-based combinationtherapies (ACTs) have enormous potential inmalaria therapy. The combination of multipledrugs enhances clinical efficacy and may delaythe development of resistance of parasites(12). However, these drugs are not yet widelyavailable and not always affordable.

3.3

Progress: drug policiesDrug resistance has led many countries ineastern, southern and central Africa to revisetheir treatment guidelines (Figure 3.5).

The varying levels of drug resistance withincountries make the changing of nationalpolicies difficult. In addition to clinical

32


SP treatment failures in Africa (1996–2001)

West

Source: WHO Regional Office for Africa and WHO headquarters

Central East0

20

40

60

All failures including LPFClinical failures (ETF + LCF)

% fa

ilure

Chloroquine treatment failure in Africa

WHO has established 126 sentinel surveillance sites in 36 African countries that monitor the efficacy of locally used antimalarial drugs by following up patients in clinics. According to standard protocol (13, 14), results are expresssed as i) early treatment failure (ETF); ii) late clinical failure (LCF); in the future, late parasitological failure (LPF) will beconsidered as well. Treatment failure for policy change as shown here consists of the sum of ETF+LCF.

Source: WHO Regional Office for Africa, 1997–2002

Eastern, southern, Great Lakes block Central block Western block

60

80

100

40

20

0

Ethi

opia

Buru

ndi

Eritr

eaRw

anda

Keny

aTa

nzan

iaM

ozam

biqu

eZa

mbi

aU

gand

aBo

tsw

ana

Zim

babw

eG

abon

Cong

oCh

adCA

RCa

mer

oon

DR C

ongo

Gha

naLi

beria

Mau

ritan

iaN

iger

Gui

nea

Beni

nCô

te d

'Ivoi

reTo

goBu

rkin

a Fa

soSe

nega

lM

ali

Sier

ra L

eone

Gam

bia

Nig

eria

% fa

ilure

Note: The box indicates the 25th/75th percentile, the line limits lower/upper values, and where the lines cross, the median.

Figure 3.3

Figure 3.4

efficacy and safety, factors such as adherenceto treatment regimens, cost, and drugmanagement issues must be taken intoaccount in deciding on a policy revision.Implementing a policy change is an expensive venture that requires significantinjection of funds both to finance the change process and for procurement of the required medications.

On average, a period of 18 months has been needed from consensus to completeimplementation of policy in most Africancountries. Factors accounting for this delayincluded obtaining political and financialsupport, training of health care providers, and sensitization of the general population,which is crucial for successful implementationof the policy decision.

3.4

Progress: treatment coverageRecent national household surveys in 28African countries have shown that an averageof 42% of children under 5 years with feverwere treated with an antimalarial (Figure 3.6).However, more than 80% of these reportedtreatments were with chloroquine, so thecoverage with effective treatment is likely tohave been much lower. In addition, manytreatments may not have been within 24hours of onset of symptoms, and dosages may have been inadequate (15–17). Thesecoverage estimates therefore represent anupper limit of the coverage with prompt,effective treatment, and the true value isprobably much lower.

33


Drugs nationally recommended for first-line treatment of uncomplicated malaria, January 2003

Source: RBM data

ACT

Non-ACTSPCQ

Not malariaendemic

Non-ACT: CQ+SP or AQ+SP combination.ACT: artemisinin (AS)+AQ, AS+SP, or Coartem.In Cameroon, amodiaquine is recommended.

Countries that changed first-line policy from CQ to SP: Malawi (1993), Kenya (1996), Botswana (1997), DRC and Tanzania (2001).

Selected provinces in South Africa where change to ACTs has been fully implemented: Coartem in KawZulu-Natal, and artesunate+SP in Mpumalanga (2001).

Countries that have adopted ACTs but have yet to implement this policy fully: Tanzania (Zanzibar) (2001), Zambia (2001), Burundi (2002).

Source: MICS and DHS, 1998–2002

0 20 40

CARCameroon

KenyaAngola

CoromosSao Tome & Principe

Sierra LeoneMadagascar

GhanaBeninTogo

Guinea BissauZambia

Côte d'IvoireGambia

TanzaniaSudan, Northern

Equatorial GuineaNigerDRC

SenegalChad

BurundiMalawi

SwazilandSomaliaRwanda

EritreaEthiopia

Almost half of febrile under-fives are treated with antimalarials. Most treatments involve chloroquine

against which resistance is increasing.

% under-fives with fever

These data do not indicate source of treatment, i.e. formal or informal private sector.

60 80

Chloroquine Any antimalarial

Note: Dotted line indicates Abuja target. No chloroquine data available for Ghana (DHS, 1998) and Eritrea (preliminary data DHS, 2002).

Figure 3.5

Figure 3.6

34


In nine countries where a number ofnational surveys have been conducted overthe past 15 years, there has been anincreasing trend towards use of antimalarialsfor treatment of febrile under-5s (Figure 3.7).These national surveys provide further proofthat use of antimalarials is widespread andcommon. The apparent increasingresponsiveness of caretakers of young childrenin accessing antimalarial treatment highlights

the opportunity for achieving furtherimprovements in their access to moreeffective antimalarials and their compliancewith treatment regimens.

3.5

Challenges: increasing coverageAt least half of the population in the poorestparts of Africa lacks access to essential drugs, including antimalarials (18), for reasonsthat include inadequate financing, poorhealth care delivery systems, and weak drug regulation (19).

Use of health facilities and suboptimaltreatment at home

Data from MICS in 21 countries indicate thatabout 46% of febrile children who receivedantimalarial treatment were treated at a healthfacility, 44% at home, and 10% both at homeand at a health facility (Figure 3.8). There isconsiderable variation between countries. InBurundi, Gambia, and Guinea Bissau, childrentreated with antimalarials were at least fourtimes as likely to be treated in a health facilityas at home, whereas children in Cameroon,

Increases over the past decade in the proportionof under-fives treated with antimalarials

Source: MICS and DHS

Ghana(DHS 1988, DHS 1998)

Zambia(DHS 1992, MICS 1999)

Tanzania (DHS 1991/2, DHS 1999)

Niger(DHS 1992, MICS 2000)

Kenya (DHS 1993, DHS 1998)

Senegal(DHS 1992, MICS 2000)

Burundi(DHS 1987, MICS 2000)

0

20

40

60

% u

nder

-five

s w

ith fe

ver t

reat

edw

ith a

ntim

alar

ials

1987–1993 1998–2000

Madagascar(DHS 1992, MICS 2000)Malawi(DHS 1992, DHS 2000)

Antimalarial drug stock-outs are a commonreason for not using health facilities

Source: RBM Baseline Surveys 2000–2001

BotswanaSwaziland

EritreaTogo

GhanaGuineaMalawi

Burkina FasoChad

MozambiqueNigeriaZambia

BeninTanzania

KenyaUganda

0 20 40 60 80 100

% of health facilities with no stock-out of nationally recommended antimalarial drugs continuously for 1 week during the past 3 months.

% health facilities

Where do young children with fever taking antimalarials receive treatment?

Source: MICS, 2000–2001

Note: Based on data from 21 countries

Both at health facility and at home 10%

At home 44%

At health facility46%

Figure 3.7

Figure 3.8 Figure 3.9

Chad, and Niger were at least three times aslikely to be treated at home.

Distance to the health facility, inadequatedrug stocks, and lack of money to pay forservices are the most common reasons for notusing public health facilities (Figure 3.9). The cost of antimalarial treatment imposes a significant burden on households in manyaffected areas. In the United Republic ofTanzania, for example, the cost of treatment has been shown to be by far the largest component of householdexpenditures on malaria (20), exceeding that on preventive measures such as ITNs.Only in five countries (Botswana, Djibouti,Namibia, South Africa, and Swaziland) is treatment free of charge.

A study of treatment of childhood malariain Zambia found that, in most cases, drugswere bought at pharmacies or local shops. The“informal private sector” is thus a main sourceof antimalarial drugs (21). However, thesetreatments are often inconsistent withnational treatment guidelines: they mayinclude counterfeit drugs, drugs of poorquality, and incorrect dosing and irrationalprescription practices (22, 23).

3.6

Challenges: disparities in useNational MICS have shown that children frompoorer households are less likely than othersto receive antimalarial treatment. In addition,although coverage is equally high for boysand girls, it is somewhat lower in rural areas,where the malaria burden is higher, than inurban areas (Figures 3.10, 3.11, and 3.12).

Household wealth also affects the quality of antimalarial drugs administered. Acommunity study in Ghana, for example,suggests that leftover drugs were used moreoften by the poor (82%) than by the less poor(53%) to treat fever episodes. Drug purchaseswithout prescriptions were also morecommon, and visits to health clinics lessfrequent, among the poor (24).

The seriousness of high fever will promptthe majority of caretakers to seek treatmentfor life-threatening illness in young children.For most surveyed populations, however, thedistribution of prompt effective treatmentamong communities remains unknown andrequires further investigation.

35


Antimalarial treatment is lower for children living in poorer households

Source: MICS, 2000–2001

0

20

40

60

80

% u

nder

-five

s w

ith fe

ver

rece

ivin

g an

timal

aria

lsPoorest Second Middle Fourth Richest

Wealth quintile

Boys and girls with fever are equally likely to receive antimalarial drugs

Source: MICS and DHS, 1998–2001

% u

nder

-five

s w

ith fe

ver

rece

ivin

g an

timal

aria

ls

0

20

40

60

Boys Girls

80

Despite a higher malarial burden, children in rural areas are less likely

to receive antimalarial drugs when they have fever

Source: MICS and DHS, 1998-2001

% u

nder

-five

s w

ith fe

ver

rece

ivin

g an

timal

aria

ls

0

20

40

60

Urban Rural

Figure 3.10

Figure 3.11

Figure 3.12

36


Cost-effectiveness versus affordability

Cost-effectiveness analyses indicate thatantimalarial treatment is generally highlycost-effective, even in the most resource-poorcountries (25). In practice, however, the costsof treating malaria patients with the mosteffective antimalarials may well not beaffordable for communities or households incountries with widespread resistance tocommonly available, inexpensive drugs.

Average cost of a full course of adultoutpatient treatment:

Chloroquine US$ 0.13

SP US$ 0.14

Amodiaquine US$ 0.20

Artemisinin-basedcombinations US$ 1–3

3.7

Scaling upThe achievement of the Abuja target of 60%coverage with prompt and effectiveantimalarial treatment will require moreeffective methods to improve delivery andcompliance with recommended regimens.Measures will include full integration ofmalaria treatment into national health

systems, improving access to effective drugsfor treatment as close to the home as possible,and engaging the private sector (26).However, currently allocated financialresources for health care in most of the low-income malaria-endemic countries will not be sufficient to respond to malariatreatment needs (25).

Although financial support for antimalarialtreatment is increasing, this has not kept pace with the costs incurred by the need tobegin replacing with newer drugs (includingACTs) those that are no longer effectivebecause of parasite resistance. Africangovernments and the global community areasked to address urgently the need to allocatesubstantial resources for the delivery of moreeffective treatment regimens to those most at risk of malaria.

Additional strategies, such as costcontainment by pooled procurement,negotiation of more favourable prices,removal of charges, tariffs and taxes, and theintroduction of subsidies, are key to improvingthe affordability of newer and more expensivetreatment regimens and to the widespreadavailability of these treatment to at-riskpopulations in Africa.

References

1. A global strategy for malaria control. Geneva, WorldHealth Organization, 1993.

2. Greenwood BM et al. Mortality and morbidity frommalaria among children in a rural area of The Gambia,West Africa. Transactions of the Royal Society of TropicalMedicine and Hygiene, 1987, 81(3):478–486.

3. Mwenesi H, Harpham T, Snow RW. Child malariatreatment practices among mothers in Kenya. SocialScience and Medicine, 1995, 40(9):1271–1277.

4. Théra MA et al. Child malaria treatment practicesamong mothers in the district of Yanfolila, Sikasso region,Mali. Tropical Medicine and International Health, 2000,5(12):876–881.

5. Nicoll A. Integrated management of childhood illness inresource-poor countries: an initiative from the WorldHealth Organization. Transations of the Royal Society ofTropical Medicine and Hygiene, 2000, 94(1):9–11.

6. The African summit on Roll Back Malaria. Abuja,Nigeria, 25 April 2000. Geneva, World HealthOrganization, 2000 (document WHO/CDS/RBM/2000.17).

7. Lepers JP et al. Malaria in 1988 in a village of theMalagasy Highland Plateaux. Epidemiological findings [inFrench]. Archives of the Institut Pasteur Madagascar,1989, 56(1):97–130.

8. Rooth I, Björkman A. Fever episodes in a holoendemicmalaria area of Tanzania: parasitological and clinical

Home-based management

of feverHome-based management of fever (HBMF) is a promisingstrategy for improving the coverage of prompt effectivetreatment. Community health workers and mothers ofyoung children are trained in the recognition of symptomsand the benefits of prompt antimalarial treatment.Prepackaged kits of full-course treatments, with appropriatedrawn and written instructions, allow mothers to treatchildren as soon as fever is detected. Programmes have beenlaunched in Ghana, Nigeria, and Uganda in June 2002.

Uganda has gone to scale with the HBMF approach in more than 10 districts and is rapidly expanding coverage.The impact of HBMF is being evaluated in three districts;interim results suggest that, among children under 5 years,the number of outpatient malaria cases has declined sinceprogramme implementation.

findings and diagnostic aspects related to malaria.Transactions of the Royal Society of Tropical Medicineand Hygiene, 1992, 86(5):479–482.

9. Kidane G, Morrow RH. Teaching mothers to providehome treatment of malaria in Tigray, Ethiopia: arandomised trial. Lancet, 2000, 356:550–555.

10. Trape JF et al. Malaria and urbanization in CentralAfrica: the example of Brazzaville. Part V: Perniciousattacks and mortality. Transactions of the Royal Society ofTropical Medicine and Hygiene, 1987, 81(Suppl.2):34–42.

11. Trape J-F et al. Impact of chloroquine resistance onmalaria mortality. Comptes Rendus de l'Académie desSciences, Paris, 1998, 321:689–697.

12. Antimalarial drug combination therapy. Report of aWHO Technical Consultation, April 2001. Geneva, WorldHealth Organization, 2001 (documentWHO/CDS/RBM/2001.35).

13. Assessment of therapeutic efficacy of antimalariadrugs: for uncomplicated falciparum malaria in areaswith intense transmission. Geneva, World HealthOrganization, 1996 (document WHO/MAL/96.1077).

14. Monitoring antimalarial drug resistance. Report of aWHO consultation, Geneva, Switzerland, 3–5 December2001. Geneva, World Health Organization, 2002(document WHO/CDS/RBM/2002.39).

15. Slutsker L et al. Treatment of malaria fever episodesamong children in Malawi: results of a KAP survey.Tropical Medicine and Parasitology, 1994, 45(1):61–64.

16. Breman JG, Campbell CC. Combating severe malaria inAfrican children. Bulletin of the World HealthOrganization, 1988, 66(5):611–620.

17. Holtz TH et al. Health care seeking behavior and hometreatment of febrile illness in Blantyre District, Malawi,2000. American Society of Tropical Medicine andHygiene, 50th Annual Meeting November 11-15, 2001,Atlanta, GA: abstract 536.

18. WHO medicines strategy 2000–2003. Geneva, WorldHealth Organization, 2000 (document WHO/EDM/2000.4).

19. Haak H. Access to antimalarial medicines: improvingthe affordability and financing of artemisinin-basedcombination therapies (in preparation).

20. Jowett M, Miller N. Malaria expenditure analysis:Tanzania case study. York, University of York, 2000.

21. Baume C, Helitzer D, Kachur SP. Patterns of care forchildhood malaria in Zambia. Social Science andMedicine, 2000, 51(10):1491–1503.

22. Dabis F et al. Monitoring selective components ofprimary health care: methodology and communityassessment of vaccination, diarrhoea, and malariapractices in Conakry, Guinea. ACSI-CCCD team. Bulletin ofthe World Health Organization, 1989, 67(6):675–684.

23. Deming MS et al. Home treatment of febrile childrenwith antimalarial drugs in Togo. Bulletin of the WorldHealth Organization, 1989, 67(6):695–700.

24. Biritwum RB, Welbeck J, Barnish G. Incidence andmanagement of malaria in two communities of differentsocio-economic level, in Accra, Ghana. Annals of TropicalMedicine and Parasitology, 2000, 94(8):771–778.

25. Goodman CA, Mills AJ. The evidence base on the cost-effectiveness of malaria control measures in Africa.Health Policy and Planning, 1999, 14(4):301–312.

26. Mendis K et al. Effective delivery methods for malariatreatment (in preparation).

37


38


Malaria infection during pregnancy is a majorpublic health problem in tropical andsubtropical regions throughout the world. Inmost endemic areas of Africa, pregnantwomen are the main adult risk group formalaria. The main burden of malaria infectionduring pregnancy results from infection withPlasmodium falciparum. The impact of theother three human malaria parasites (P. vivax,P. malariae, and P. ovale) is less clear. Everyyear at least 30 million women in malariousareas of Africa become pregnant; most ofthese women live in areas of relatively stablemalaria transmission.

The symptoms and complications ofmalaria during pregnancy differ with theintensity of malaria transmission and thuswith the level of immunity acquired by thepregnant woman (2). Since malariatransmission intensity may vary within thesame country from areas of relatively stabletransmission to areas of unstable or epidemictransmission, the clinical picture of malariainfection during pregnancy may likewiserange from asymptomatic to severe, life-threatening illness.

In areas of epidemic or low (unstable)malaria transmission, adult women have notacquired any significant level of immunity andusually become ill when infected with P. falciparum. For pregnant women in theseareas the risk of developing severe malaria is2–3 times higher than that for non-pregnantwomen living in the same area. Maternal

death may result either directly from severemalaria or indirectly from malaria-relatedsevere anaemia. In addition, malaria may resultin a range of adverse pregnancy outcomes,including low birth weight, spontaneousabortion, and neonatal death (Figure 4.1).

In areas of high and moderate (stable)malaria transmission, most adult women havedeveloped sufficient immunity that, evenduring pregnancy, P. falciparum infectiondoes not usually result in fever or otherclinical symptoms. In these areas, theprincipal impact of malaria infection ismalaria-related anaemia in the mother andthe presence of parasites in the placenta. Theresulting impairment of fetal nutritioncontributes to low birth weight and is aleading cause of poorer infant survival anddevelopment (Figure 4.1). In areas of Africawith stable malaria transmission,

4. Malaria during pregnancy

Abuja target

In the AbujaDeclaration African

heads of states agreedto achieve by 2005:

■ at least 60%coverage of pregnant

women at risk ofmalaria with the mostsuitable combination

of personal andcommunity protective

measures

■ at least 60% of allpregnant women at

risk of malaria,especially those in

their first pregnancies,shall have access to

intermittent preventivetreatment (1).

Malaria during pregnancyIn areas of low or

epidemic transmissionIn areas of high or moderate

(stable) transmission

Acquired clinical immunity—low or none

Clinical illness

Maternal anaemia Placental infection

Higher infant mortalityImpaired child development

Severe disease

Death from severe malaria Spontaneous abortionPremature deliveryStillbirth

Risk to mother Risk to fetus Maternal morbidity Low birth weight

Acquired clinical immunity—high

Asymptomatic infection

Strategic framework for malaria controlduring pregnancy in WHO Africa Region

To reach the Abuja Summit goala, a three-pronged approach is recommended:

■ Intermittent preventive treatment

■ Insecticide-treated nets

■ Effective case management of malarialillness.

a: The original Abuja declaration also recommendedchemoprophylaxis, but present WHO and RBM policystrongly recommends IPT – and not chemoprophylaxis –for prevention of malaria during pregnancy.

Figure 4.1

39

Chapter 4: Malaria during pregnancy

P. falciparum infection during pregnancy isestimated to cause an estimated 75 000 to200 000 infant deaths each year (3).

Despite the toll that malaria exacts onpregnant women and their infants, this was –until recently – a relatively neglected problem,with less than 5% of pregnant women havingaccess to effective interventions. During thepast decade, however, potentially moreeffective strategies for prevention and controlof malaria in pregnancy have been developedand shown to have a remarkable impact onthe health of mothers and infants.

4.1

EvidenceIntermittent preventive treatment (IPT)

For many years WHO recommended thatpregnant women in malaria endemic areasshould receive an initial antimalarialtreatment dose on their first contact withantenatal services, followed by weeklychemoprophylaxis (given at less thantherapeutic dose) with an effective and safeantimalarial drug (4). In most countries inAfrica, chloroquine (CQ) has been the drug ofchoice. However, the emergence and spreadof CQ-resistant falciparum malaria, poorpatient compliance with multiple doses, anda high incidence of CQ-induced pruritus havelimited the effectiveness and hence theimplementation of this policy.

In 2000, the WHO Expert Committee onMalaria recommended that intermittenttreatment with an effective, preferably one-dose, antimalarial drug, should be madeavailable as a routine part of antenatal careto women in their first and secondpregnancies in highly endemic areas (5). Atpresent, sulfadoxine–pyrimethamine (SP) –given at a therapeutic dose – is the single-dose antimalarial with the best overalleffectiveness for prevention of malaria inpregnancy in areas with high transmission,and low resistance to SP. Other antimalarialsare being evaluated for potential use in IPT.

Studies in Kenya (6,7) and Malawi (8) haveshown that IPT with at least two treatmentdoses of SP is highly effective in reducing theproportion of women with anaemia andplacental malaria infection at delivery. Thebenefits of IPT for both maternal and infanthealth have been seen in a range of differentmalaria transmission settings (Figures 4.2–4.4).

Low seasonal transmissionA trial in Kilifi, Kenya, showed that the use of IPT with SP reduced the risk of severe anaemia and peripheral parasitaemia among women in their first pregnancy living in malarious areas.

Source: reference 6

Placebo Two-dose SP

Peripheralparasitaemia

Anaemia

% prevalence0 10 20 30 40

High seasonal transmissionTwo treatment doses with SP were highly effective inreducing the proportion of women in their first or secondpregnancy with placental malaria infection at delivery during peak transmission season in Malawi.

Source: reference 8

SP/SP: two-dose SPSP/CQ: one-dose SP followed by weekly CQCQ/CQ: CQ treatment followed by weekly CQ

SP/SP

SP/CQ

CQ/CQ

0 10 20 30 40 50% placental infection

High perennial transmissionA trial in Kisumu, an area with high prevalence of malaria and HIV infection in Kenya, found that a two-doseregimen with SP was safe and efficacious for the preventionof placental malariaa in areas with low HIV seroprevalence.

Source: reference 7

a: In women in their first and second pregnancy.Comparison of SP-based fever case management with IPT with SP.

SP/SP: two-dose SP. Monthly SP: treatment doses at enrolment and then monthly through 34 weeks of gestation.

SP/SP

Monthly SP

Case/fevermanagement

0 10 20 30% prevalence

Placental malaria Low birth weight

Efficacy of IPT in different transmission settings

Figure 4.2

Figure 4.3

Figure 4.4

40


Insecticide-treated nets

If used during pregnancy in areas of stablemalaria transmission, ITNs reduce the overallrisk of morbidity and mortality amongpregnant women and their infants. A trial inthe Gambia found that, during the rainyseason in villages where ITNs were used, theprevalence of malaria infection amongpregnant women was lower and fewerbabies were classified as premature (9).Further evidence comes from a recent studyin a highly malarious area of Kenya. During the first four pregnancies, womenwho were protected by ITNs at night gave birth to 25% fewer premature or small-for-gestational-age babies than women who did not sleep under ITNs (10) (Figure 4.5).

Case management of malaria illness

Malaria in pregnant women requiresimmediate treatment, focusing on completecure of the infection. Each country in malaria-endemic areas of Africa needs a policy thatguides effective management of malaria inpregnant women. Collaboration betweenmalaria control programme and reproductivehealth programme staff can facilitate thedevelopment of systematic managementprotocols and drug supply strategies (11).

Cost-effectiveness

Weekly CQ chemoprophylaxis was comparedwith two-dose IPT using SP in primigravidaein an area of moderate to high malariatransmission: IPT-SP proved to be more cost-effective than CQ chemoprophylaxis, largelybecause of lower costs and highercompliance with SP (12). In a further cost-effectiveness analysis, three different SPregimens were compared with febrile casemanagement using SP for a hypotheticalcohort of 10 000 pregnant women in Kenya.The results suggested that the two-dose SPregimen (when the costs only of antenatalcare are considered and at HIVseroprevalence below 10%) would be theleast expensive strategy for preventing lowbirth weight (13).

ITN use by children has been shown to becost-effective in several settings (14,15). Thesubstantial benefit of ITNs in reducing theburden of malaria during pregnancy makesit likely that the cost-effectiveness of ITNuse by pregnant women will be of similarattractiveness to that for children.

ITN use among women in their first four pregnancies reduced LBW and premature birth by 25%


Stillbirth/abortion

Preterm delivery (PTD)

Small for gestational age (SGA)

Low birth weight (LBW)

LBW or PTD or SGA

Any adverse outcome

0 0.4 0.80.2 0.6 1.0 1.2 1.4

0 5 10 15 20 25 30 35

Control ITN Prevalence ratio (95% Cl)

% prevalence

Prevalence ratio

Source: DHS, 1998–2001

0 20 40 60 80 100

% attendance at ANC

EthiopiaNiger

MauritaniaBurkina Faso

NigeriaGuinea

CameroonTanzania

GabonTogo

Côte d'IvoireBenin

GhanaKenya

UgandaMalawi

Zimbabwe

Most women in Africa south of the Saharaattend an antenatal clinic at least once

Attending ANC at least twiceb

Attending ANC at least oncea

a: Attended by skilled providerb: Does not reflect skill level of care provider

Figure 4.5

Figure 4.6

41


4.2

ProgressAntenatal care

WHO recommends that IPT be administeredto pregnant women during routine visits toantenatal clinics. As recent surveys confirm,at least two-thirds of pregnant women inmost countries do have access to, and use,antenatal care, and most of them attendantenatal clinics at least twice (Figure 4.6).The high level of antenatal care coverage anduse provides a unique opportunity to deliverprevention packages to pregnant women inthe Africa region.

Prevention with IPT and ITNs

Coverage of pregnant women with IPT andITNs is a fundamental part of prevention ofmalaria during pregnancy. Although there areexamples of successful delivery of IPT andITNs through antenatal clinics, large-scaleprogrammes to deliver these prevention toolsto pregnant women are only now beingdeveloped. Thus, data on IPT and ITNcoverage are limited at present.

Surveys conducted on a national scaleindicate that net use among women ofreproductive age (15–49 years) – anindication of use by pregnant women –remains very low. Among pregnant women,coverage with any net (treated or untreated)was less than 10% in three of the fourcountries for which recent data wereavailable. Coverage with ITNs in these threecountries was even lower, at 3% or less(Figure 4.7). As countries accelerate efforts tocontrol malaria during pregnancy, ITNcoverage is expected to increase.

4.3

ChallengesBottlenecks

Timely antenatal clinic attendance is key for delivering the prevention package topregnant women. Some 40% of pregnantAfrican women present for the first time toantenatal clinics in the second trimester ofpregnancy, and about 25% come for the firsttime in the third trimester (16) (Figure 4.8).This means that at least the first dose of IPT could be given in time to most pregnantwomen. The ITN part of the preventionpackage delivered during the first antenatal

Timing of intermittent preventive treatment


BeninBurkina Faso

CameroonCAR

ChadComoros

Côte d'IvoireEthiopia

GhanaGuineaKenya

MadagascarMalawi

MaliMozambique

NamibiaNiger

NigeriaRwandaSenegal

TanzaniaTogo

UgandaZambia

Zimbabwe

Conception Quickeninga Birth2010 30

IPT1 IPT2

Weeks of pregnancy

a: Quickening is the first noted movement of the fetus

0–3 months 4–5 months 6–7 months

8+ months Unknown/missing

% attendance of pregnant women0 20 40 60 80 100

When do pregnant women attend antenatal clinics?

Use of ITNs by pregnant women is a fundamental part of

the Strategic Framework for Africa


Note: refers to use the night before the survey

Benin Malawi Rwanda Uganda

40

30

20

19

0

% u

se o

f net

% women aged 15–49using any net

% women aged 15–49using ITN

% pregnant women using any net

Figure 4.7

Figure 4.8

42


clinic visit would provide additional protection for the mother during theremainder of the time of pregnancy and into the post-partum period, as well as protection for the newborn through at least the first year of life.

Overcoming challenges

High coverage with antenatal care, i.e. a minimum of two visits to an antenatal clinic, does not necessarily translate into full coverage with IPT. However, a multidisciplinary team in theBlantyre district in Malawi (population about950 000, with approximately 35 000 birthsannually) focused from mid-2001 through2002 on resolving barriers to completecoverage with IPT. Improved education about the benefit of IPT and modifiedrecommendations for the scheduledantenatal clinic visits after quickeningresulted in a rapid increase to 75% coverage, with two doses of SP in pregnant women (Figure 4.9).

Disparities

A striking determinant of attendance atantenatal clinics is household wealth.According to a recent study, poor women areless likely to use antenatal services than arewomen from the richest households (16)(Figure 4.10). In seven African countries southof the Sahara for which recent data areavailable, the percentage of rich womenattending antenatal clinics was at least twicethat of poor women. The same study indicatedthat antenatal care coverage was significantlyhigher in urban than in rural areas.

Does household wealth affect antenatal care?

Across 22 countries in Africa south of the Sahara, rich women were about 1.5 times more likely to attend antenatal clinics than poor women.


100

80

60

40

20

0Poorest Second Middle Fourth Richest%

pre

gnan

t wom

en a

tten

ding

AN

C

Wealth quintile

Coverage with chemoprophylaxis/IPT inpregnant women is lower in rural areas


100

80

60

40

20

0Benin

CPRwanda

CPMalawi

IPT1Malawi

IPT2Ethiopia

CPUganda

CP

Note: CP= chemoprophylaxis

Urban Rural

Malawi demonstrates the feasibility of achieving high coverage with two doses of SP

In 2000, less than half of pregnant women attending ANC received the full course of IPT

Follow-up surveys in the Blantyre district in late 2002 indicate high coverage withtwo doses of SP

ANC 1 and 2: Proportion of women attending ANC at least once or twice Sources: DHS 2000 and Malawi Ministry of Health, 2003

ANC 1

ANC 2

IPT 1

IPT 2

0 20 40 60 80 100% pregnant women receiving intervention

IPT 1

IPT 2

0 20 40 60 80 100% pregnant women receiving IPT

Figure 4.9

Figure 4.10

Figure 4.11

43


Nationally representative surveys confirmsimilar difference in coverage of pregnantwith chemoprophylaxis, IPT, or nets (Figures4.11 and 4.12). Women living in rural areaswere less likely to receive chemoprophylaxisor, in Malawi, IPT with SP.

4.4

OpportunitiesAt present, strategies to ensure safe pregnancy in malaria-endemic areas areplanned with a focus initially on strengtheningmalaria preventive services and correct case management for pregnant womenattending antenatal care. A number ofopportunities will facilitate the acceleratedimplementation of malaria control during pregnancy:

■ High levels of coverage with antenatalcare visits make a clinic-based approach feasible

■ Partnerships have been establishedbetween RBM and programmes such asMaking Pregnancy Safer and nationalreproductive health services committedto strengthening entry-level antenatalclinic services

■ The Strategic Framework for MalariaControl During Pregnancy in the WHOAfrica Region (2) has been developed

Policy summary: Best practices for malaria

control during pregnancies

■ Effective case management of malaria illness for all women of reproductive age in malarious areasmust be ensured.

■ The policy for malaria control during pregnancyshould emphasize a preventive package ofintermittent preventive treatment (IPT) andinsecticide-treated nets (ITNs), particularly in areas of stable transmission.

■ All pregnant women should receive at least 2 doses of IPT after quickening, during routinelyscheduled antenatal clinic visits as recommended by WHO.

■ Programmes should seek the highest possible coverage of pregnant women with these interventions— at least 60% (Abuja RBM goals) and preferablyhigher — and document this accomplishment. Given current high rates of antenatal clinicattendance in most African countries south of theSahara, this should be achievable.

Source: adapted from reference 2

Coverage with any net and/or ITN among women aged 15–49 and/or pregnant women (PW)


PW ITN

40

30

20

10

0

% n

et c

over

age

PWAny net

ITN Any net ITN Any net ITN Any net PWAny net

Any net

Uganda Rwanda Malawi Benin

Urban Rural

Figure 4.12

■ Several countries are in the process ofupdating policies that are consistentwith new evidence and with WHOrecommendations for prevention andcontrol of malaria during pregnancy

References

1. The African summit on Roll Back Malaria, Abuja,Nigeria, 25 April 2000. Geneva, World HealthOrganization, 2000 (document WHO/CDS/RBM/2000.17).

2. Strategic Framework for Malaria Control DuringPregnancy in the WHO African Region. Geneva, WorldHealth Organization, 2003 [final draft, in press].

3. Steketee RW et al. The burden of malaria in pregnancyin malaria-endemic areas. American Journal of TropicalMedicine and Hygiene, 2001, 64(1,2 S):28–35.

4. WHO Expert Committee on Malaria. Eighteenth report.Geneva, World Health Organization, 1986 (WHOTechnical Report Series, No. 735).

5. WHO Expert Committee on Malaria. Twentieth report.Geneva, World Health Organization, 2000 (WHOTechnical Report Series, No. 892).

6. Shulman CE et al. Intermittent sulphadoxine-pyrimethamine to prevent severe anaemia secondary tomalaria in pregnancy: a randomised placebo-controlledtrial. Lancet, 1999, 353:632–636.

7. Parise ME et al. Efficacy of sulfadoxine-pyrimethaminefor prevention of placental malaria in an area of Kenyawith a high prevalence of malaria and humanimmunodeficiency virus infection. American Journal ofTropical Medicine and Hygiene, 1998, 59(5):813–822.

8. Schultz LJ et al. The efficacy of antimalarial regimenscontaining sulfadoxine-pyrimethamine and/orchloroquine in preventing peripheral and placentalPlasmodium falciparum infection among pregnantwomen in Malawi. American Journal of Tropical Medicineand Hygiene, 1994, 51(5):515–522.

9. D'Alessandro U et al. The impact of a nationalimpregnated bed net programme on the outcome ofpregnancy in primigravidae in The Gambia. Transactions

44


Making Pregnancy Safer:access to essential care for pregnant

women and their newbornsIn support of the Safe Motherhood Initiative, the WHO Making Pregnancy Safer focuses oneffective evidence-based interventions that target the major causes of maternal andnewborn morbidity and mortality. Making Pregnancy Safer aims to strengthen healthsystems and to identify actions atcommunity level needed to ensureaccess to essential care for pregnantwomen and their newborns.

Skilled attendance at delivery andprovision of an appropriate andeffective continuum of antenatal and perinatal care are particularlyimportant in this initiative.

The evidence-based interventions of Making Pregnancy Safer focus on six areas:

■ Technical and policy support

■ Advocacy

■ Partnership building

■ Development of norms and tools

■ Research and dissemination

■ Monitoring and evaluation

Pic

ture

: W

HO

/TD

R

of the Royal Society of Tropical Medicine and Hygiene,1996, 90(5):487–492.

10. Kuile T et al. Permethrin-treated bednets reducemalaria in pregnancy in an area of intense perennialmalaria transmission in western Kenya. American Journalof Tropical Medicine and Hygiene, 2003 [in press].

11. Managing complications in pregnancy and childbirth:a guide for midwives and doctors. Geneva, World HealthOrganization, 2000.

12. Goodman CA, Coleman PG, Mills AJ. The cost-effectiveness of antenatal malaria prevention in sub-Saharan Africa. American Journal of Tropical Medicineand Hygiene, 2001, 64(1,2 S):45–56.

13. Wolfe E et al. Cost-effectiveness of sulfadoxine-pyrimethamine for the prevention of malaria-associatedlow birth weight? American Journal of Tropical Medicineand Hygiene, 2001, 64(3,4):178–186.

14. Wiseman V et al. The cost-effectiveness ofpermethrin-treated bednets in an area of intense malariatransmission in western Kenya. American Journal ofTropical Medicine and Hygiene, 2003 [in press].

15. Goodman CA, Coleman PG, Mills AJ. Cost-effectiveness of malaria control in sub-Saharan Africa.Lancet, 1999, 354:378-385.

16. UNICEF, WHO. Antenatal care in developingcountries: promises, achievements, and missedopportunities. An analysis of levels, trends anddifferentials, 1990-2001, 2003 [in preparation].

45


Commitment to accelerating prevention and control of malaria during pregnancy

Countries are developing national policies on the prevention of malaria in pregnancyaccording to WHO recommendations and have started to document experiences fromcontrol efforts.

In March 2002, a technical meeting on malaria prevention and control during pregnancywas held in Malawi where reproductive health and malaria staff from five countries, Kenya,Malawi, Uganda, the United Republic of Tanzania, and Zambia, came together to discuss acoordinated approach to accelerating prevention and control of malaria in pregnancy. Theparticipants and RBM partners have agreed to forge the East and Southern Africa Coalitionfor malaria prevention and control during pregnancy, MIPESA. The coalition wasinaugurated in June 2002.

This coalition is a unique concerted effort in which national reproductive healthprogrammes, Making Pregnancy Safer and malaria control programmes are coming together

■ to support inter-country programme collaboration in accelerating prevention andcontrol of malaria in pregnancy

■ to provide a forum of information exchange between and among countries, partnersand involved institutions

■ to strengthen models for programme collaboration between national reproductivehealth, Making Pregnancy Safer and Malaria Control Programmes for quality evidence-based antenatal care

■ to develop mechanisms that facilitate consultation among member programmes

■ to support documentation of best practices

■ to develop a collaborative monitoring and evaluation system

■ to develop collaborative approaches to information, education and communicationthat improve knowledge of antenatal care providers and pregnant women

■ to facilitate the development of a sub-regional operational research agenda andconduct of key research protocols.

46


Some 110 million Africans live in areas at riskfor epidemic malaria (Figure 5.1). Theincreasing frequency of epidemics in bothlow-risk areas and areas of moderatetransmission make imperative the institutionof special responses to epidemics, in additionto regular malaria control activities. Theimpact of epidemics can be greatly reducedby timely detection or, ideally, prediction andprevention. Timely response may include thedeployment of additional drug stocks, use ofdifferent drugs, and vector control.

5.1

Magnitude of the problemDevastating malaria epidemics have recentlybeen reported from Botswana, Mozambique,South Africa, Swaziland and Zimbabwe(Figure 5.2) and have been attributed mainly

to heavy rainfall following a drought.Epidemics also affected several Saheliancountries in the 1990s.

In East Africa and the Great Lakes region, Burundi, Ethiopia, Kenya, Uganda, and the United Republic of Tanzania, aresubject to frequent and recurrent malariaepidemics that often affect large numbers of people (Table 5.1). Anomalies of rainfalland/or temperature are thought to be the most important causes of theepidemics here.

Although the extent of suffering caused by malaria epidemics is notadequately documented, it is generallybelieved that morbidity, mortality, and the overall economic impact of theseepidemics are enormous. For example, the number of deaths during the malariaepidemic that hit Ethiopia in 1958

5. Response to epidemicsand complex emergencies

Abuja target

The target of Roll BackMalaria in the AbujaDeclaration is that

60% of epidemics aredetected within 2

weeks of onset and60% of epidemics areresponded to within 2weeks of detection (3).

What is a malariaepidemic?A sharp increase in malaria incidenceamong populations in whom the diseaseis infrequent, or an increase in clinicalmalaria in areas of moderatetransmission constitutes an epidemic.

Malaria epidemics occur principally inareas of low transmission, where nosingle age group in the population isimmune. The introduction of malaria,particularly if exacerbated by changes inrainfall and temperature, can triggerexplosive epidemics that affect bothadults and children (1). However,epidemics can also occur in areas ofhigher transmission as a result of theabandonment of control programmes,immigration of non-immune people, andreduced access to treatment.

Areas at risk ofepidemic malaria


Figure 5.1

was estimated at more than 150 000 out of 3 million clinical cases (a case-fatalityrate of 5%) (1). Adults account for arelatively large proportion of epidemic cases and deaths.

On the basis of the following assumptions(4) it is estimated that there are 110 000deaths from epidemics each year:

■ 110 million Africans are at epidemic risk

■ Epidemics occur on average every 5years (cyclical) (5)

■ Assuming 0.5 malaria episodes perperson during epidemics:

— 0.1 episodes per person at risk peryear

— 11 million malaria episodes inepidemics each year

■ Up to 5% of malaria episodes are severe,with case-fatality rate for severeepisodes in epidemics possibly up to20%:

— possible 110 000 malaria deaths fromepidemics each year.

47

Chapter 5: Epidemics and complex emergencies

Historic examples of severe epidemicsPlace, year Population Episodes Deaths Causes

exposed

Madagascar 2.5 million Over 200 000 15 000-30 000 Abandonment ofhighlands, each year; at peak, each year IRS and shortage 1987–1988 (1) 27% of outpatient of antimalarials

attendances

Ethiopia highlands/ 8-10 million 3 million Over 150 000 High rainfall andDumbia plain, (case fatality temperature1958 (1) rate >5%)

Ethiopia, Dec. 1997 45 million >1 million 3271 officially High rainfall andFeb.1998 (1) reported temperature

NE Burundi, 2 million during 1287 reported, Abandonment of Oct-2000- 6 months — a 4-fold true number control (in relationMay 2001 (9) increase in estimated to be to complex

confirmed cases (10) 10-15x higher. emergency),chloroquine resistance and expandedrice cultivation.

Malaria epidemics in Africa1997–2002

Source: RBM data

Epidemic-prone countriesLocation of documented epidemic within past five years

Figure 5.2

Table 5.1

5.2

Prevention and control of epidemicsThe main functions of a malaria controlprogramme in combating epidemics areillustrated in Figure 5.3 and described ingreater detail in several WHO documents (1).

The burden of an epidemic will be much lessif the number of cases is contained by earlydetection and rapid response (1) (Figure 5.4).

Antimalarial treatment during epidemics

It may be necessary to formulate drugpolicies for treatment during epidemics,because malaria epidemics usually involvenon-immune patients who have lowerspontaneous parasite clearance than partiallyimmune patients from endemic areas. It isessential that treatments selected for use inthese situations are highly effective. Thetrend is toward artemisinin-basedcombinations (ACT) which have the addedadvantage of reducing gametocyte carriage.Some countries, e.g. Burundi, Ethiopia, andMozambique, have already adopted specialdrug policies for epidemics.

5.3

Progress and challengesIdentification of risk areas

Table 5.2 shows 15 of the 25 Africancountries which have identified their areas of

risk and have prepared plans to respond to epidemics.

Actual detection and response

It is difficult to say with certainty whethercountries are close to the Abuja target of60% adequate detection and response.Evaluation is hampered by the diverse natureof epidemics and associated definition issues.

In evaluations of control programmes,timely detection has been found to be abigger challenge than adequate and timelyresponse once an epidemic has beenidentified (7). (Table 5.3) However, there isindication that response, especially withmeasures like indoor residual spraying, isoften too slow (more than 2 weeks afterdetection of the epidemic) as well.

Integrated Disease Surveillance

A recently initiated programme on IntegratedDisease Surveillance and Response, promotedby WHO/AFRO, is helping epidemiologists toselect and use accurate indicators – initiallyon a monthly basis. In defined epidemic-prone districts, the second step is to collectweekly malaria morbidity and mortality datain order to detect any unusual increaseoccurring within 2 weeks and to takeimmediate action. This programme hasconsiderable potential for rapidly improvingthe capacity for early detection of epidemicsof malaria and other diseases; itsimplementation is making good progressacross the continent (Table 5.4).

48


Epidemic

CONTROL(Investigation and response)

POST-MORTEM(Review and

report)

FORECASTING

PREVENTION(Mitigation)

PREPAREDNESS

EARLY WARNING SYSTEMSEarly

resp

onse

Source: reference 6Figure 5.3

49


Country with epidemic risk Risk areas

and preparedness plan of action

Angola All risk areasBotswana All risk areasBurundi HighlandsEthiopia HighlandsKenya HighlandsMozambique All risk areasNamibia All risk areasRwanda HighlandsSenegal Close to Senegal riverSouth Africa All risk areasSwaziland All risk areasSudan Sahelian areasUganda HighlandsZambia HighlandsZimbabwe Highlands

Sources: references 6, 12, 13

Malaria Early Warning Systems (MEWS) areintended to facilitate the timely responses thatwill prevent and contain epidemics. MEWScomprise forecasting, early warning, and earlydetection. Forecasting usually refers to seasonalclimate forecasts; early warning refers to themonitoring of meteorological conditions such asrainfall and temperature; and early detection isbased on routine clinical surveillance.

RBM has supported the development of asimple tool to monitor epidemic risks in marginaltransmission areas based on anomalies in rainfalland temperature, i.e. rainfall in excess of or lessthan the expected amount, as identified fromsatellite observations. The predictions point toepidemics that might occur around 6 weeks after the detected meterological change. The risks are displayedon maps which are updated every 10 days and can be freely accessed via the RBM web site or athttp://edcsnw4.cr.usgs.gov/adds/.

MEWS are generally performing well in southern Africa, and studies have started in Ethiopia, Kenya,Uganda, the United Republic of Tanzania, and Sudan. Decision-tree models are being constructed andvalidated. Work remains to be done to better involve staff from meteorological services in joint ventures withministries of health to regularly compile and evaluate data and disseminate warnings to communities.

Outbreak detectionand response

Source: reference 1

Impact of theepidemic whenepidemic is notcontained byearly detectionand response.

Earlyresponse

Containment of epidemic when detected and responded to rapidly.

0

10

20

30

40

50

60

70

80

90

Earlydetection

Time

Case

s

Figure 5.4

Malaria Early Warning Systems

Table 5.2

Complex emergencies

A large proportion of malaria deaths in Africa occur among populations affected by conflicts. Populationdisplacement, malnutrition, and breakdownof health systems characterize these conflicts,making victims highly vulnerable to disease.Given the large numbers of cases occurringwithin short periods of time, and theconsequent heavy burden on the healthsystem, malaria epidemics in complexemergencies might be considered as “man-made epidemics”.

International organizations are the mainproviders of relief assistance – in the firstinstance, emergency shelters, clean water,and food – to displaced populations, thenbasic health care.

A main requirement in complexemergencies is coordination of the activitiesof different NGOs to streamline the technicalapproaches for malaria-specific interventions(cost-effective drugs, ITNs, and residualspraying) and to ensure that existingresources are used cost-effectively, within acontext of basic health care.

50


Epidemics Percentage Percentage Sourceidentified detected within responded to within 1998–2000 2 weeks 2 weeks of detection

Angola 3 0 0 SAMC

Botswana 2 100 100 SAMC

Malawi – – _ SAMC

Mozambique 3 67 67 SAMC

Namibia 3 0 33 SAMC

South Africa 4 100 100 SAMC

Swaziland 2 100 100 SAMC

Tanzania – – – SAMC

Zambia _ - - SAMC

Zimbabwe 3 33 67 SAMC

Eritrea 4 (districts) 75 100 RBM baseline survey

Ethiopia 102 55 100 RBM baseline survey

Uganda ? 0 100 RBM baseline survey

Note: SAMC – reference 7

Table 5.4

Table 5.3

Status of Integrated Disease Surveillance implementation, 2002

Guidelines adapted Botswana, Eritrea, Ethiopia,Ghana, Kenya, Malawi, Mali,Nigeria, Swaziland, Tanzania,Zambia.

Guideline Burkina Faso, Cameroon, Chad,adaptation Congo, DRC, Equatorial Guinea,initiated Gabon, Gambia, Guinea-

Conakry, Mozambique, Namibia, Rwanda, Seychelles,Uganda, Zimbabwe.

Plan of action Lesotho, Madagascar.completed

Assessment Benin, CAR, Mozambique, completed

51


Other activities supported by WHO andother RBM partners include:

■ producing and distributing anemergencies handbook;

■ operational research into emergencymanagement, including suitable drugregimens;

■ increasing capacity on the ground,including training;

■ stimulating the development of tools,such as insecticide-impregnated sheetshelters and blankets.

5.4

Strategies for the futureA key challenge for district and nationalhealth managers is to improve the use ofdata routinely generated at the peripherallevel, using Integrated Disease Surveillancewhere appropriate. In areas affected by

epidemics related to anomalies in seasonalrainfall and temperature, jointMoH/meteorological services teams should beset up to promote the use of local epidemicpredictions. Similar coordination mechanismsmay need to be established amongappropriate government services in areas thatare affected by epidemics linked to, forinstance, migration or agricultural practices.

Control options are as important asprediction and early detection; they must bewell defined and agreed upon in advance bythe MoH and partners of malaria controlprogrammes as part of a preparedness plan ofaction allowing for rapid and cost-effectivedeployment. Treatment options may need toinclude new and more effective drugs or drugcombinations (such as ACT) to be suppliedfree of charge to all malaria patients in theearly stage of an epidemic. As regardsefficient vector control approaches, thecontroversy surrounding the choice ofinsecticides for IRS needs to be resolved.Logistic support capacity should be

Epidemic control in southern Africa

Systematic epidemic control is standardized to the greatest extent in southern Africa, where it is beingcoordinated under the Southern African Malaria Control programme (SAMC). Since 1998, a regional approach to reducing the impact of malaria epidemics includes strategic planning, forecasting, prevention,preparedness and early warning to facilitate an improved early response (6, 14).

Several SAMC countries use residual house spraying and distribution of ITNs for epidemic prevention, andmost use residual house spraying as early response to reduce epidemic impact.

Capacity for detection – southern Africa

Country Forecasting Weekly Evaluation Epidemic Manual Epidemic Contingency Epidemicsurveillance short-term plan guidelines control (drug) funds

epidemic-prone weather committee stocksdistricts

Angola No No No Yes No No Yes No

Botswana Yes Yes Yes Yes Yes Yes Yes No

Malawi No No No No No No No No

Mozambique Yes Yes No Yes No No No Yes

Namibia Yes Yes No Yes Yes Yes No Yes

South Africa No Yes No Yes Yes Yes No No

Swaziland Yes Yes Yes Yes No Yes Yes No

Tanzania No No No No Yes No No No

Zambia No No No Yes No No No No

Zimbabwe Yes Yes No Yes No No No Yes

Table 5.5

strengthened in countries considering the useof IRS for early response or prevention. Inepidemic-prone areas, all strategies toincrease the coverage with ITNs to a levelthat reduces transmission are relevant toepidemic prevention.

Broad agreement on the choice of efficientprevention and control options willcontribute to securing funds from RBMpartners in advance of the epidemic season,particularly for the establishment of essentialemergency stocks. In this context,collaboration in epidemic risk assessment andresponse with other epidemic-prone countriesby managing subregional/cross-bordersurveillance systems and emergency stocks isvital. The participation of NGOs in nationaland international partnerships is particularlyimportant, because these groups can rapidlydeploy assistance at field level.

References

1. A practical guideline for prevention and control ofmalaria epidemics in Africa. Geneva, World HealthOrganization, 2003 (draft DIP/MAL/03.06).



4. Worrall E. The burden of malaria epidemics and cost-effectiveness in epidemic situations. In: Third MIM Pan-African Conference on Malaria, Arusha, Tanzania, 17–22November 2002, pre-MIM meeting/workshop, “Theintolerable burden of malaria: what's new, what'sneeded”. Bethesda, MD, Multilateral Initiative on Malaria:Annex II, p. 367, number 12.

5. Hay SI et al. Etiology of interepidemic periods ofmosquito-borne disease. Proceedings of the NationalAcademy of Sciences of the USA, 2000,97(16):9335–9339.

6. Malaria epidemics and their control in SouthernAfrica. Epidemic post-mortem for the 2000/2001 malariaseason. Harare, Southern African Malaria ControlProgramme (SAMC), 2002.

7. A baseline for RBM in Southern Africa - WHOSouthern Africa Malaria Control Programme. Harare,Southern Africa Malaria Control Programme, 2002.

8. Hay SI et al. The utility of forecasting, early warningand early detection during the 2002 malaria emergencyof western Kenya. Lancet, 2003 [in press].

9. Prevention and control of malaria epidemics. Thirdmeeting of the Technical Support Network. Geneva,World Health Organization, 2002 (documentWHO/CDS/RBM/2002.40).

10. Bonora S et al. Rising temperature and the malariaepidemic in Burundi. Trends in Parasitology, 2001,17(12):572–573.

11. Mouchet J et al. [The reconquest of the Madagascarhighlands by malaria.] Bulletin de la Société dePathologie Exotique, 1997, 90(3):162–168.

12. Cox et al. Mapping malaria risk in the highlands ofAfrica. London, MARA/London School of Hygiene andTropical Medicine, 1999.

13. Jambou R et al. Malaria in the highlands ofMadagascar after five years of indoor house spraying ofDDT. Transactions of the Royal Society of TropicalMedicine and Hygiene, 2001, 95(1):14–18.

14. Malaria control progress in Southern Africa 1997 to2000 – Review. Harare, Southern Africa Malaria ControlProgramme, 2002.

15. Connor S. Map of epidemic risk areas in Africa. NewYork, International Research Institute for ClimatePrediction, The Earth Institute at Columbia University,2003 (available at http://edcsnw4.cr.usgs.gov/adds/).

52


Epidemic detection in KenyaEarly warning based on district-level rainfall estimates hadhigh potential for detecting two epidemics in summer 2002in four highland districts in south-west Kenya. The district-specific warnings could have given 4 weeks’ notice ofpossible emergency conditions, as identified in aretrospective analysis.

The rainfall estimates are freely available and their userequires minimal training and technical facilities. In thissetting, seasonal forecasts proved to be less specific andless accurate. Early detection based on surveillance ofoutpatient cases provided sentinel information too late,because the reporting – although fairly complete – wasmonthly (weekly reporting under Integrated DiseaseSurveillance was not yet active at this time) (8).

53

Chapter 6: Resource mobilization and financing

Since the inception of Roll Back Malaria in1998, and particularly since the AbujaSummit in 2000, malaria prevention andcontrol have once again become domesticand international priorities. Internationalspending for malaria has increased at leasttwofold since 1998, and this does not eveninclude the complementary financing of arange of primary health care services (such as reproductive health and IMCI) thatalso have an impact on malaria (1).

In 2002, approximately US$ 200 millionwas earmarked for malaria control worldwide,compared with an estimated US$ 60 millionin 1998. Of the total in 2002, about US$ 70–80 million can be attributed todomestic commitments (1). Through the workof national and international RBM partners,the renewed importance of malaria isreflected in the development of countrystrategic plans, the recognition of malariacontrol actions as a global public good, andthe prioritization of malaria in developmentinitiatives such as the Highly Indebted PoorCountries initiative and the MillenniumDevelopment Goals.

6.1

Resources required to roll back malariaA broad resource base is essential to reducethe burden of malaria in Africa. Criticalresources include financial and humanresources, institutional know-how, politicalcommitment, and community support. Theseresources are located in a diverse set ofinstitutions including governments,nongovernmental organizations, theinternational community, the private sector,and civil society.

Harnessing and coordinating theseresources require the development ofconsensus among potential partners and theformation of effective working relationships.The establishment of partnerships workingtoward a common goal is an integral part ofthe strategy to Roll Back Malaria.

The Abuja Declaration in 2000 (2) called forthe allocation of new resources – at leastUS$ 1 billion per year – from Africancountries and their development partners.Although more resources are available formalaria control today than 3 years ago, asignificantly greater investment is needed tosupport those fighting malaria on theground. In 2000 and 2001, African heads ofstate pledged to allocate the resourcesrequired for sustained implementation ofplanned Roll Back Malaria actions, includingsignificant increases to country healthbudgets. However, in order for additionalinvestments to have an impact on malaria,

6. Resource mobilization and financing

Source: World Health Report 2002Statistical Annex 5

Private – primarily household – expenditures, combined with donor support, make up the bulk of health and malaria control financing in Africa south of the Sahara. The available data suggest that expenditure on health care and malaria control by African governments is low. Successful efforts to reduce the financial burden of malaria on African families must concentrateon reducing out-of-pocket expenditures on malaria commodities,such as drugs, while making these expenditures more efficient, for instance, through improving the quality of drugs and encouraging correct drug use. Governments are highly dependent on external support for health, and increased governmental health expenditures would enable greater support for malaria control.

Malaria control is financedthrough mainly three channels

0

10

20

30

Government Donors Private/Out-of-pocket

40

% to

tal

heal

th e

xpen

ditu

res

50

Contribution of different sources in African health expenditure

Figure 6.1

54


countries must have mechanisms by whichresources can be translated into effectiveprogrammes that reach those most in need.This requires functioning health systems thatallocate appropriate resources to malariaprogrammes for recurrent costs and essentialcommodities (such as antimalarial drugs),administrative systems that ensure timeliness,regularity and accountability in the flow offunds, infrastructure and supplies fordelivering services, a skilled and motivatedwork force, and sufficient organizationalcapacity to manage the system. In addition,an enabling environment is required tostimulate private sector growth to providecommodities for malaria prevention andtreatment, such as insecticide-treated netsand antimalarial drugs.

Rationalization of taxes and tariffs ondrugs and nets, for example, has alreadycontributed to this environment in somecountries. While private sector growth isstimulated, governments will have anincreased responsibility to implementsufficient regulatory controls to ensure thesafety, quality, and effectiveness of malariacontrol commodities.

6.2

Domestic resources It is not possible to ascertain accurately theamount of government resources supportingmalaria control. Malaria control today isfinanced through a range of complementaryprogrammes that all have an impact onmalaria. Public health activities such as IMCI,reproductive health programmes, and otherprimary health care services will all have directpositive impacts on malaria but are notrecorded as expenditures attributed to malariacontrol programmes. In addition, many malariacosts are covered by general budgets forhealth facility staff or drugs. In countries witha heavy malaria burden, the disease mayaccount for as much as 25–40% of outpatientvisits and up to 50% of inpatient admissions,generating costs that may amount to as muchas 40% of public health expenditure. Figure 6.2 illustrates the expenditures onhealth in 44 African countries.

Continued fiscal shortfalls faced by thoseworking on malaria suggest that, despiteprogress, government resource allocations formalaria are still insufficient. Governments in

Per capita governmentexpenditures on health (US$)

BotswanaSouth Africa

GabonNamibia

SwazilandEquatorial Guinea

DijboutiZimbabwe

CongoSomaliaAngola

SenegalZambia

MauritaniaGambia

ComorosGuinea

TanzaniaRwanda

MozambiqueMadagascar

KenyaGuinea-Bissau

GhanaEritrea

DRCCôte d'Ivoire

CameroonBurkina Faso

BeninSao Tome & Principe

MaliMalawiUganda

TogoSierra Leone

ChadCAR

SudanNigeria

NigerEthiopiaBurundiLiberia

Source: WHO Statistics Office

0 50 100 150

120108

8280

4037

2018151313121111119976666666666666

55544444322221

Figure 6.2

55


African countries generally support publicsector salaries without providing resourcesfor the other recurrent expenditures, such asantimalarial drugs, needed to deliver services.In addition, although the “cost” of achievingthe scale of action required may besignificant, human resources are required forexpenditures to have an impact on reducingthe burden of malaria. In many Africancountries today, however, shortages oftrained personnel impede malaria controlmore than the lack of financial resources.

Most people in Africa south of the Saharapay out-of-pocket for malaria control

Overwhelmingly, people in Africa south ofthe Sahara pay for malaria prevention andtreatment through out-of-pocketexpenditure. The lack of governmentresources for malaria prevention andtreatment contributes to shifting the burdenof expenditure to households. Evidencesuggests that most of this expenditure is onpharmaceuticals, the costs of which willfurther rise with the introduction of newantimalarials made necessary by increasingresistance to current drugs. Measures shouldbe taken to reduce the costs of these drugs,for instance through targeted subsidies. It isimportant to note, however, that high out-of-pocket expenditure indicates a willingnessto pay for malaria treatment, which mayhave positive implications for thesustainability of malaria control financing.

Inevitably the poorest households will spenda greater proportion of their income on

The prioritization of malaria control in debt relief: A reflection of renewed importance

The Heavily Indebted Poor Countries (HIPC) Initiative is the first international response to providecomprehensive debt relief to the world’s poorest, most heavily indebted countries. The 26 countries thatentered the HIPC programme, of which most are in Africa south of the Sahara, are saving an average US$ 1.3billion per year compared with debt payments in 1998. Resources freed from debt are being used to supportcountry-driven poverty reduction strategies – in the form of Poverty Reduction Strategy Papers (PRSPs),which are developed by national governments in consultation with civil society.

As of January 2003, PRSPs (and Interim PRSPs) in every country in Africa south of the Sahara note malariaas a significant contributor to poverty and stress the importance of malaria control in their approaches tosustainable development. The prioritization of malaria in these documents will have tangible implications forfinancing malaria control. For instance, as of 2002, all World Bank activities in countries that have preparedpoverty reduction strategies must respond to the priorities detailed in the PRSP. In addition, the Bank hasdeveloped Poverty Reduction Support Credits (PRSCs) to directly support implementation of a country’spoverty reduction strategy. Although the process is still relatively new, these resources have already beenmade available for malaria control in some countries, such as Cameroon.

Government expenditures on health are low

Health expenditure in Africa is generally less than US$ 15 perperson per year – and less than US$ 5 per year in severalcountries. Some of the variation between countries can beattributed to their relative wealth but there is also aconsiderable range in the percentage of governmentexpenditure allocated to health –from 3.0% to 16%. In 2001,the year after the summit on malaria, leaders of Africancountries assembled once again in Abuja, Nigeria, and pledgedto earmark 15% of their national budgets to health spending.Fulfilling this pledge would no doubt have a significantimpact on the financial resources available for malaria.

The cost of malaria control is high

Original data source: references 4, 9, 10

The average direct and indirect costs (opportunity cost of time, for instance) of malaria can exceed per capita total spending on health, highlighting the importance of increased government health expenditures to mitigate the health and economic impact of malaria.

0

5

10

15

Burkina Faso Ghana Rwanda

20

Average costs of malaria treatment (US$)

Average costs (direct and indirect) associated with treating a malaria episode

Total expenditure on health, public and private (per capita)

Figure 6.3

56


malaria prevention and treatment. Strategiesto mitigate the financial burden of malaria –particularly on the poorest households, leastlikely to afford the necessary expenditures –should be combined with appropriatecommunication and behaviour changeactivities both to reduce the costs of malaria-related commodities and to supportappropriate use of those commodities.

Household spending should be directedtoward the most effective interventions

Although household spending on malariacontrol is very high, it is not necessarilyefficient. Most purchasing of malaria controlcommodities occurs in the private sector –through local markets and drug vendors, forinstance – with the result that individualsreceive little guidance and may purchase anduse poor-quality, inappropriate, or incompleteinterventions. Significant expenditures onmosquito coils and insecticide aerosol sprayshave been reported, but while thesecommodities reduce nuisance mosquitoesthere is little evidence that their use reducesthe burden of malaria. Drug vendors mayprovide little guidance on treatment choiceand usually will not encourage the completionof a full course of treatment, contributing tothe development of parasite resistance.

Communication programmes designed tobring about behaviour change will play anessential role in guiding consumers towardsthe purchase of malaria control interventionsof demonstrated effectiveness in reducingmalaria transmission, such as insecticide-treated materials and antimalarials to whichresistance is less common. There is evidencethat preventive spending directed at ITNs mayreduce the annual cost per case of all-causeclinic visits by sick children from US$ 49 toUS$ 38 (3), suggesting a significant costsaving in addition to the reduction in painand suffering from disease.

Planning for impact

Several countries are well advanced informing coalitions between partners that areworking towards an agreed plan of action.The process differs from one country toanother but some common themes areevident. The process usually starts with asituation analysis of the state of healthsector development, as well as of theepidemiology of malaria. The analysis alsoidentifies the resources available to roll backmalaria, including the diverse range of

Source: reference 5

Households are spending the most for malaria

In the United Republic of Tanzania, more than 70% of total expenses formalaria control come from households and are spent in the private sector. A similar financial burden is felt by households across Africa south of theSahara, suggesting a need for improved targeting of interventions –particularly preventive interventions – at the household level. In addition, better understanding of domestic expenditure flows could lead to improvements in the distribution of resources for malaria control.

Total malaria expenditures by source in the United Republic of Tanzania, fiscal year 1998

Government20%

Donors9%

Households71%

The financial burden of malaria is greatest on the poor

Source: reference 6

A study from Malawi demonstrated that both poor and less poor households spent approximately US$ 19 for malaria treatment annually. This represents a relatively greater burden on poorer households. Strategies must be devised to reduce costs for the poor while maximizing the health impact of expenditures.

0

5

10

15

Very-low-incomehouseholds

20

Low-to-high-incomehouseholds

25

30

Mal

aria

exp

endi

ture

as

% o

f to

tal a

nnua

l hou

seho

ld in

com

e

Figure 6.4

Figure 6.5

57


stakeholders already involved in malariacontrol. Formal partnerships are created anda statement of intent issued, indicating whatwill be achieved in 5 years, and how.Resources are mobilized from partners andsystems are set up to monitor achievementsin rolling back malaria. Forty Africancountries have drawn up National StrategicPlans, and at least 25 countries are ready toimplement. Many of these plans have servedas the basis for successful proposals to theGlobal Fund to Fight AIDS, Tuberculosis andMalaria – and some are already receivingfunds from this source to supportimplementation of the plans (Table 6.1).

Partnerships with industry are increasingresources for malaria

The business community is increasinglyplaying a role in the Roll Back Malariapartnership and providing vital resources andexpertise to support malaria controlprogrammes. Partnerships have been forgedwith the drug industry, companies with majorinvestments in malaria-endemic countrieswhose workers are at risk from malaria, andmanufacturers of nets and insecticides. Inaddition, companies have provided funds fortraining and collaborating with youngscientists and medical officers to allow themto gain valuable experience in the field.

Understanding resource flows to improveeffectiveness of expenditures

Financial resources are only one part ofresource mobilization. African countries andthe international community need to placegreater emphasis on understanding the flowof resources to actual malaria prevention andcontrol activities. Expenditure analyses thattrack spending on malaria can elucidate thepathways that allow financing for malaria toovercome administrative bottlenecks andsupport essential services and commoditiesrequired for achieving impact.

Efforts to roll back malaria in Ethiopia, forinstance, have been hampered byinefficiencies in the flow of scarce resourcesto malaria prevention and control activities.Total expenditure for malaria control isapproximately US$ 10.25 million, or only US$ 0.35 per person at risk, but even theselimited resources could have a greater impacton malaria control if they were reachingthose most in need.

Through a recent malaria expenditure

Angola ✓ ✓ ✓ ✓ ✓

Benin ✓ ✓ ✓ ✓ ✓ ✓

Botswana ✓ ✓ ✓ ✓ ✓ ✓

Burkina Faso ✓ ✓ ✓ ✓ ✓ ✓

BurundiCameroon ✓ ✓ ✓ ✓ ✓ ✓

CAR ✓ ✓ ✓ ✓ ✓

Chad ✓

Comoros ✓ ✓ ✓

Congo ✓ ✓ ✓

Côte d’Ivoire ✓ ✓ ✓ ✓ ✓ ✓

DRC ✓ ✓ ✓ ✓ ✓ ✓

Djibouti ✓ ✓ ✓

Equitorial Guinea ✓ ✓ ✓ ✓ ✓ ✓

Egypt ✓ ✓ ✓ ✓ ✓

Eritrea ✓ ✓ ✓ ✓ ✓ ✓

Ethiopia ✓ ✓ ✓ ✓ ✓ ✓

GabonGambia ✓ ✓ ✓ ✓ ✓ ✓

Ghana ✓ ✓ ✓ ✓ ✓ ✓

Guinea ✓ ✓ ✓ ✓

Guinea-Bissau ✓ ✓ ✓ ✓

Kenya ✓ ✓ ✓ ✓ ✓ ✓

Liberia ✓ ✓ ✓ ✓ ✓

Madagascar ✓ ✓ ✓ ✓ ✓ ✓

Malawi ✓ ✓ ✓ ✓ ✓ ✓

Mali ✓ ✓ ✓ ✓ ✓ ✓

Mauritania ✓ ✓ ✓ ✓ ✓ ✓

Mozambique ✓ ✓ ✓ ✓ ✓ ✓

Namibia ✓ ✓ ✓ ✓ ✓ ✓

Niger ✓ ✓ ✓ ✓ ✓

Nigeria ✓ ✓ ✓ ✓ ✓ ✓

RwandaSao Tome and Principe ✓ ✓ ✓ ✓

Senegal ✓ ✓ ✓ ✓ ✓ ✓

Sierra Leone ✓

Somalia ✓ ✓ ✓ ✓ ✓

South Africa ✓ ✓ ✓ ✓ ✓ ✓

Sudan ✓ ✓ ✓ ✓ ✓ ✓

Swaziland ✓ ✓ ✓ ✓ ✓ ✓

Tanzania ✓ ✓ ✓ ✓ ✓ ✓

Togo ✓ ✓ ✓ ✓

Uganda ✓ ✓ ✓ ✓ ✓ ✓

Zambia ✓ ✓ ✓ ✓ ✓ ✓

Zimbabwe ✓ ✓ ✓ ✓ ✓ ✓

Source: RBM data

RBM policy planning and implementation

Situ

atio

nan

alys

is

Co

ord

inat

ion

mec

han

ism

Nat

ion

alst

rate

gic

pla

n

Co

sted

pla

n

Mo

nit

ori

ng

&ev

alu

atio

n p

lan

Imle

men

tati

on

pla

n

Table 6.1

58


review in Ethiopia, commissioned by Roll BackMalaria, it is clear that different parts of themalaria control programme are competing forthe same scarce resources, leaving little roomfor transportation, antimalarials, and otherrecurrent operational costs. For example,public facility-based pharmacies, wherepatients can obtain drugs free of charge, areoften out of stock of essential drugs, forcingpatients to purchase drugs at private facilities(7). In addition, the allocation of funds acrossthe different levels of the health sector thathave contact with malaria patients may notbe proportional to need.

Ethiopia has recently decentralized thedelivery of health services and has not yetestablished sustainable management andfinancial structures within all levels of thehealth sector. However, the bottlenecks thatimpede effective resource flows for malariahave now been identified and malaria controlis back on the government agenda. Withincreases in management and

implementation capacity, Ethiopia will beable to use effectively the approximately US$38 million in new financing from the GlobalFund to Fight AIDS, Tuberculosis., andMalaria. Further organizational andinstitutional strengthening, accompanied bypolitical commitment and increasedinternational support, will assist Ethiopia touse new and existing resources to bettersupport effective malaria control measures.

Similar reviews of resource flows to malariacontrol should be undertaken in othercountries in Africa south of the Sahara tomake better use of existing resources andimprove targeting of new resources, so thatthe impact of both domestic and donorresources can be maximized.

6.3

International resourcesDespite the increased level of internationalfinancing, the amount of money available inmost countries to control malaria is stillinadequate. In some cases, the necessaryadditional resources can be made availableinternally through greater governmentallocations for increased preventive andcontrol activities. If this alone is insufficient,government spending needs to be supportedby increased international funding.

Although continued financial support fromthe international community for malariacontrol is essential, it is equally important forthe international community to supportcountries in developing implementationcapacity. For instance, donors should workwith countries and the private sector todevise and implement innovative financingschemes – the use of targeting mechanisms,such as vouchers for ITNs for instance – toimprove financial flows to malaria control on the ground.

A health sector-wide approach for malariasupport in the United Republic of Tanzania:a reduction in malaria will be a mark ofsuccess

The World Bank-funded Health SectorDevelopment Programme Project in theUnited Republic of Tanzania broadly supportsthe first phase of the Health Sector ReformProgramme of the country’s government.Although project financing is not specificallydesignated as “malaria” financing, the

Source: reference 8

Expenditure for malaria can be attributed to many different sources

Ethiopia: Malaria expenditure by source (Total: US$ 10.25 million)

Understanding resource flows to malaria control and the essential interventions that are part of it is important to improving efficiency and utilizing both existing and new resources.

Other 1%

Net NGO expenditure 5%

Net governmentmalaria budget 68%

Household expenditure 14%

Donor assistance 7%

Community in-kind support 3%

Industry/Companies 2%

Figure 6.6

59


Countries and donors still need to increase spending for malaria

Recent estimates of what is financially necessary for malaria control with available tools suggest an approximate US$ 0.6–0.9 (2002 US$) per capita by 2007 and 2015, respectively, a significant increase from current levels of financing.

Source: reference 1

0

0.20

0.40

0.60

0.80

US$ 1.00

2002 2007 2015

Project resource requirement

Current earmarked domestic spending

Current earmarked international spending

Note: Chart refers to malaria control in Africa south of the Sahara

International financing earmarked for malaria has more than doubled

Source: reference 1

International expenditures on malaria increased from US$ 60 million in 1998 to US$ 120 million in 2002. Including spending for reproductive health, IMCI, and primary health care, all of which have an impact on malaria, raises the total amount to significantly more than US$ 120 million. The figure (right) shows estimated earmarked expenditures on malaria worldwide by bilaterals, multilaterals and NGOs.

0

50

100

150

BilateralsUS$ 80 million

MultilateralsUS$ 47.2 million

NGOsUS$ 3.3 million

1998 2000

US$

mill

ions

Figure 6.7

Figures 6.8 and 6.9

project’s objectives have profoundimplications for resource mobilization formalaria. A target of reducing the malariainpatient case-fatality rate for children under5 from 12.8% (1997) to 8% (2011) has beenincluded as a key indicator of success, andthe government has successfully allocatedapproximately US$ 10 million for malaria in2003 (Figure 6.10).

The potential flexibility and countryownership of sector-wide approaches(SWAPs) have dramatically changed theresource envelope for malaria in a number ofAfrican countries. As of February 2003, atleast eight countries in Africa south of theSahara have prioritized malaria throughSWAPs. With detailed district-level planningand coordination with sectoral funding cyclesand planning meetings, additional resourcesfor malaria prevention and control can bemade available while the health sector as awhole is strengthened.

6.4

Next steps for financing malariacontrol in Africa Countries have demonstrated commitment to RBM by driving the Country StrategicPlanning process. International expendituredirectly financing malaria has risendramatically, and additional support formalaria through sectoral financing schemescan be mobilized as required. In addition, the Global Fund for AIDS, Tuberculosis and Malaria is an exciting new source for significant direct support for malaria control.

Countries should improve domestic healthsector investment and management,including improved integration of malariacontrol activities into health sectordevelopment efforts.

Policies that provide a greater resource basefor health in general and malaria inparticular must be pursued, including policiesthat promote equity of access to goods andservices and facilitate the delivery of qualityservices at low cost through the privatecommercial sector. Public sector linkages withthe private sector and NGOs, throughcontracting out of services for instance,would allow countries to expand significantlythe resource envelope for malaria activities.

60


The Global Fund: A new resource to fight malaria

Commitments from the Global Fund to Fight AIDS,Tuberculosis and Malaria have significantly increased thefinancial resources for malaria control in Africa. Over thenext 2 years, the Global Fund is expected to disburse US$256 million for malaria control activities in 25 Africancountries. The challenge now is for countries to develop thecapacity to spend the money efficiently and for theinternational community to support Africa in that effort.

Africa: Global Fund component approvals as ofJanuary 2003

Country Total commitment over two years (US$)

Benin 2 389 185

Burkina Faso 7 144 703

Burundi 13 792 126

Comoros 1 534 631

Eritrea 2 617 633

Ethiopia 37 915 012

Ghana 4 596 111

Guinea 6 893 509

Kenya 10 526 880

Madagascar 2 023 424

Malawi 20 872 000

Mali 2 023 424

Mauritania 824 125

Mozambique 12 273 573

Namibia 3 719 354

Nigeria 17 828 808

Senegal 4 285 714

Somalia 8 890 497

Sudan (North) 14 237 853

Sudan (South) 12 855 490

Swaziland 978 000

Tanzania 11 959 076

Tanzania (Zanzibar) 781 220

Uganda 23 211 300

Zambia 17 892 000

Zimbabwe 6 716 250

Multicountry Africaa 7 424 815

Total 256 206 713

a: Lubombo Spatial Development Initiative Area

Source: The Global Fund Secretariat, January 2003Table 6.2

Communications for behaviour change mustcoincide with resource mobilizationactivities to improve the efficiency ofhousehold spending on malaria preventionand control

A significant portion of the financial burdenfrom malaria falls on households. Thoughmeasures must be taken to mitigate thefinancial impact on the poorest who areunable to afford such high levels of out-of-pocket expenditure, the prominent role ofhouseholds in malaria control marks anopportunity as well. If, for instance,household expenditures were made only onappropriate and effective malaria preventionand treatment commodities, current spendingcould make a significant impact on themalaria burden. Behaviour changecommunications combined with innovativefinancing mechanisms would make the mostof existing household expenditures.

The international community must assistcountries and partners in spendingincreased resources for malaria effectivelyand demonstrating their impact

Donors should work with governments toimprove management and organizationalcapacity for monitoring and evaluatingresource flows, both human and financial.Existing resources do not always flow to

those who need them most. It is essentialthat governments respond by prioritizingmalaria within their own health sectorbudgets to maximize the disbursement ofdomestic and international resources toprevent and control malaria. At the sametime, the international community mustassist countries in the efficient and effectivespending of available resources for malaria,such as those mobilized through healthsector development programmes and the

61


Despite the lack of explicit earmarking, a significant portion of the funding for World Bank financing of health sector development in Tanzania will support malaria control, either for direct malaria activities at national or district level, infrastructure and vehicles for health services, or through improved supplies of pharmaceuticals. Should the need arise, the Tanzanian government could request that additional malaria-specific inputs be financed from this source of funds.

Malaria control is a priority in health sector development

Health sector development in the United Republic of Tanzania50

40

30

20

10

02000–2003 2003–2007 2007–2011

Estim

ated

Wor

ld B

ank

finan

cing

(US$

mill

ions

)

Estimated implementation period

Source: World Bank Country Office for the United Republic of Tanzania

Source: World Bank Country Office forthe United Republic of Tanzania

US$ 10.8 million set aside for malariain fiscal year 2003

“Pooling” and “non-pooling” refer to the different mechanisms used by donors in financing through the sector-wide approach (SWAp).

Non-poolingSWAp partnersUS$ 8 875 000

Pooling SWAp partnersUS$ 1 600 000

GovernmentUS$ 325 000

Figure 6.10

Figure 6.11

Global Fund. Public health success stories,such as the Onchocerciasis ControlProgramme and the polio eradicationcampaign, have depended on thecombination of country commitment andinternational support translating into impacton the ground. There is good reason tobelieve that similar progress can be achievedin malaria. As recent increases in malariafinancing demonstrate, the internationalcommunity stands ready to support countriesin their malaria control efforts, and shouldreadily increase expenditures with furtherprogress in countries.

References

1. Final report of the external evaluation of Roll BackMalaria. London, Report for Department forInternational Development, UK, 2002.


3. Wiseman V et al. The cost-effectiveness of permethrin-treated bednets in an area of intense malariatransmission in western Kenya. American Journal ofTropical Medicine and Hygiene, 2003 (in press).

4. Anenso-Okyere WK, Dzator JA. Household cost ofseeking malaria care. A retrospective study of twodistricts in Ghana. Social Science and Medicine,45(5):659-667.

5. Jowett M, Miller N, Mnzava N. Malaria expenditureanalysis. Tanzania case study. Prepared for DFID-EA(Tanzania) and the Roll Back Malaria Initiative (availableat www.rbm.who.int).

6. Ettling M et al. Economic impact of malaria inmalawian households. Tropical Medicine andParasitology, 1994, 45:74-79.

7. Valdelin J, Walgelin N, Fairbank A. Private expendituretrends in Ethiopia and implications for health systemsfinancing. Cambridge, MA, Essential Services for Healthin Ethiopia (ESHE) Project, Abt Associates Inc., 2001.

8. Study on malaria financing and expenditures inEthiopia. Washington, DC, World Bank, 2000.

9. Ettling MB, Shephard DS. Economic cost of malaria inRwanda. Tropical Medicine and Parasitology, 1991,42:214-218.

10. Guiguemde TR et al. Household expenditure onmalaria prevention and treatement for families in thetown of Bobo-Dioulasso, Burkina Faso. Transactions ofthe Royal Society of Tropical Medicine and Hygiene,1994, 88:285-287.

62


63

Data on the progress in coverage with RBMinterventions and on the burden of malaria atpopulation level in Africa are increasing. Thisreport drew information mainly from thefollowing sources, which are described herewith a focus on their quality, reliability,representativeness, and potential for futureRBM monitoring.

Demographic and Health Surveys

Demographic and Health Surveys (DHS) arenationally representative household surveysthat focus on reproductive and child health.They are organized by Macro International,Calverton, MD, USA and sponsored by theUnited States Agency for InternationalDevelopment (USAID) (1). Data are freelyavailable on the Internet, athttp://www.measuredhs.com/, approximately1 year after completion of the fieldwork.Typically, DHS consist of interviews withbetween 4000 and 12000 women aged 15–49years living in households that are sampled ina multiple-stage cluster design. Between1990 and 2002, 10 African countries havehad 1 DHS, 17 have had 2, and 4 have had 3or more. The average interval between twoDHS is 5–6 years. The questionnaire addresses,inter alia, household living conditions andassets, and child health through birthhistories. Because questionnaire arestandardized and structured and change littlebetween surveys, DHS outcomes arecomparable between countries and over time.

Malaria-relevant indicators include thereported treatment and care (whetherantimalarials were given and facilitiesattended) to under-5s who had fever in the 2 weeks preceding the survey. Since 1998,some DHS have used specific questions onmalaria prevention and treatment, includingthe type of antimalarial drugs given, timingand dosage, possession of mosquito nets and their use for under-5s and pregnantwomen, and the use of by pregnant women.Since 2001, most of these questions havebeen grouped in a standard malaria module,now to be used in all surveys conducted inmalarious countries. The interpretation ofDHS malaria modules must take into account that these surveys are (for logistical

reasons) mostly conducted in the dry, least malarious season.

DHS are additionally a primary source ofinformation on under-5 all-cause mortalityrates, obtained by the direct estimationtechnique, i.e. from birth histories. RecentDHS also measure the prevalence of anaemiaby haemoglobin measurement in childrenunder 5 and (pregnant) women; in future thismay prove to be a very useful indicator ofmalaria burden and a proxy for malaria-related mortality.

Multiple Indicator Cluster Surveys

Between 1999 and 2001, Multiple IndicatorCluster Surveys (MICS) were conducted in 67countries around the world with supportfrom UNICEF. MICS are nationallyrepresentative and sampled through a two-stage cluster design, with an average samplesize of around 6000 households. A householdquestionnaire covers living conditions andhousehold assets, allowing the data to bestratified by such factors as place ofresidence, education of the mother, andwealth quintile of the household.

For 24 countries in Africa, the surveyincluded a malaria module with questionsrelated to the prevalence of fever in theprevious 2 weeks, type of treatment receivedand place of treatment, as well as use of anynets and of insecticide-treated nets by childrenunder 5. Relevant data were also collected onindicators such as all-cause mortality amongchildren under 5 and coverage of antenatalcare. Survey results and questionnaires arefreely available on the Internet athttp://childinfo.org/. As DHS most MICS wereconducted outside of the peak malaria season.

The next round of MICS is planned for2004–2005 and these will again includequestions on malaria prevention andtreatment, and possibly additional questionson household availability of nets, dosage andtiming of antimalarial treatment, and IPTamong pregnant women.

Demographic Surveillance Sites

Demographic surveillance sites (DSS) monitorbirths, deaths, and health in geographically

Notes on data sources


defined populations, continuously over time.They measure mortality due to specificcauses, including malaria, by using verbalautopsies – interviews with surviving relativeson the symptoms that preceded death. Of allDSS sites, 29 from 16 African countriescollaborate under the RBM-supportedINDEPTH network (International Network offield sites with continuous DemographicEvaluation of Populations and Their Health indeveloping countries). Further information onthe INDEPTH network is available athttp://www.indepth-network.net/. INDEPTHaims to strengthen technical and researchcapacity, and to promote the use ofstandardized methodological tools (such as verbal autopsy questionnaires) andthe application of information to policy and practice (3).

Although coverage is limited to about 0.2%(1.3/650 million) of the African population (3,4), DSS are practically the only source ofinformation on malaria-attributed mortalityat a population level. Also, their prospectivenature means that DSS-based estimates ofall-cause under-5 mortality are typicallymore precise and more up-to-date than thosefrom the retrospective DHS or MICS.Nevertheless, there are concerns about therepresentativeness. Large-scale phase III andphase IV community health intervention trialsare often carried out at DSS (5, 6); ifsuccessful, the tested interventions areimplemented on a large scale thereafter.Coverage with interventions, including ITNs,may therefore be higher than in thesurrounding population, and the diseaseburden – in general or of the specifictargeted disease – consequently lower.

RBM baseline surveys

In 2000 and 2001, 17 countries in Africasouth of the Sahara conducted surveys todetermine their baseline situations withrespect to rolling back malaria. The surveysincluded a facility component and acommunity (household) component. Defaultquestionnaires were adapted for local use,and outcomes vary in comparability betweencountries. Most countries followed thesampling guideline, which specified thatthree or four districts of different malariaendemicity were to be chosen. In eachdistricts, the highest-level health facilitiy(usually a hospital), two health centres andthree clinics were randomly selected; for eachsampled facility, the nearest and the farthest

communities in its catchment area weresampled, bringing the total to around 15facilities, 30 communities, and 1000households. Outcomes are not nationallyrepresentative but provide indications ofdisease status of malaria and controlimplementation in the districts most affectedby malaria and most targeted by the nationalmalaria control programme (NMCP).

The household component addresses netpossession and use and treatment-seeking forfever. For some core RBM indicators (drugstock-outs, correct clinical management ofmalaria patients, and availability ofdiagnostic services), the health facilitycomponent of the RBM baseline surveys themain or only current source of data. Whencoverage estimates from the RBM baselinesurveys are compared with national estimatesfrom DHS or MICS, the baseline surveystypically give higher rates, as expected fromtheir focus on targeted areas of the NMCP.

Routine Health Information Systems

Routine Health Information Systems (HIS) arethe source of national data on malaria casesand deaths seen in health facilities. Inprinciple, these data are national, but inpractice not all facilities and districts report.The numbers of cases and deaths reported aretherefore less than the actual clinical burden.More importantly, the clinical burdenrepresents only a fraction of the total burdenin the population, since most malaria patientseither do not seek treatment or are treatedoutside the formal health sector (7). To addto the uncertainty, case loads are basedlargely on clinical, rather than parasitologicaldiagnosis. To reduce the consequent bias, thisreport focuses not on absolute numbers ofrecorded malaria cases and deaths, but onthe proportion of the burden of recordedevents (outpatient clinic visits, hospitaladmissions, deaths in hospitals) accounted for by malaria.

64


65


DHS MICS RBM baseline surveys, DSS Routinecomponent: HIS

community health NMCPfacility

Burden / Impact

All-cause under-5 mortality (in population) ✓ ✓ ✓

Malaria-attributed mortality (in population) ✓

Malaria as % of outpatient visits ✓

Malaria as % of hospital admissions ✓

Malaria as % of hospital deaths ✓

ITNs

Household possession of nets/ITNs ✓ ✓

% of under-5s sleeping under net/ITN ✓ ✓ ✓

Antimalarial treatment

% of febrile under-5s treated with antimalarial ✓ ✓

% of clinical cases correctly managed ✓

Drug stock-outs in clinics ✓

Malaria control in pregnancy

% of pregnant women attending ANC ✓ ✓

% of pregnant women taking IPT ✓

% of pregnant women sleeping under net/ITN ✓

Epidemic response ✓ ✓

Data quality

Nationwide Yes Yes No No n.a. No Yes

Sample size (households) 8000 6000 1000 15 15 000 n.a.(facilities)

Table 1. Data availability by source, for indicators used in this report, as of February 2003

66


Table 2. Sources of data for monitoring malaria burden and control

Dat

a av

aila

bili

ty a

nd

(d

is-)

adva

nta

ges

Sou

rce

Typ

e, in

terv

al, c

ou

ntr

ies

Gen

eral

Mo

rbid

ity

Mo

rtal

ity

Co

ntr

ol i

mp

lem

enta

tio

n

Dem

ogra

phic

Popu

lati

on-b

ased

, lar

ge,

Retr

ospe

ctiv

e U

nder

-5s

and

All-

caus

e U

nder

def

ault

mal

aria

an

d H

ealt

h re

pres

enta

tive

at

nati

onal

/ (r

ecal

l bia

s) c

an

(pre

gnan

t) w

omen

,un

der-

5m

odul

ea , po

sses

sion

and

Su

rvey

s (D

HS)

prov

inci

al b

ut n

ot d

istr

ict

leve

l;be

adj

uste

dm

ainl

y re

port

ed f

ever

sm

orta

lity

use

of (i

nsec

tici

de-t

reat

ed)

incr

easi

ng s

ubse

t of

dev

elop

ing

for

co-c

olle

cted

i.e. n

on-s

peci

fic.

nets

and

of

anti

mal

aria

l co

untr

ies,

expe

nsiv

e, e

very

5—

8so

cioe

cono

mic

trea

tmen

t fo

r un

der-

5s

year

s. D

ata

avai

labl

e on

Inte

rnet

an

d ge

nera

l an

d pr

egna

nt w

omen

, w

ithi

n 1

year

of

end

of f

ield

wor

khe

alth

var

iabl

esIP

T fo

r pr

egna

nt w

omen

Mul

tipl

ePo

pula

tion

-bas

ed, l

arge

,Re

tros

pect

ive,

Re

port

ed f

ever

s in

All-

caus

e U

se o

f (in

sect

icid

e-tr

eate

d)In

dica

tor

Clus

ter

repr

esen

tati

ve a

t na

tion

al/

can

be a

djus

ted

unde

r-5s

unde

r-5

nets

and

ant

imal

aria

l Su

rvey

(MIC

S)pr

ovin

cial

but

not

dis

tric

t le

vel;

for

co-c

olle

cted

i.e

. non

-spe

cifi

cm

orta

lity

trea

tmen

t fo

r un

der-

5sin

crea

sing

num

ber

of

soci

oeco

nom

icde

velo

ping

cou

ntrie

s, ev

ery

5 ye

ars.

gene

ral h

ealt

hM

ain

tabl

es a

vaila

ble

on In

tern

et

varia

bles

wit

hin

1 ye

ar o

f en

d of

fie

ldw

ork.

Dem

ogra

phic

Com

mun

ity-

base

d, lo

ngit

udin

al,

Cont

inuo

usVa

riabl

eVe

rbal

aut

opsy

Varia

ble

Surv

eilla

nce

smal

l and

non

-rep

rese

ntat

ive

on d

eath

s, Si

tes

(DSS

)se

ntin

el p

opul

atio

ns, l

ong

part

icul

arly

for

re

port

ing

dela

yun

der-

5s

Rout

ine

HIS

Clin

ic-b

ased

, gen

eral

ly le

ast

Cont

inuo

us, l

imit

edCl

inic

al a

nd/o

r Pa

rasi

tolo

gica

llyN

o, o

r on

ly a

t le

vel o

f(c

linic

-bas

ed)

func

tion

al in

cou

ntrie

s m

ost

to s

ubgr

oup

ofpa

rasi

tolo

gica

lco

nfirm

ed

clin

ic a

tten

dees

affe

cted

by

mal

aria

(sou

th o

f cl

inic

att

ende

es,

diag

nosi

s, al

ldi

agno

sis,

all a

ges

the

Saha

ra).

whi

ch is

– e

spec

ially

ages

agg

rega

ted

and/

or u

nder

-5s

for

child

ren

– an

d/or

und

er-5

sse

para

tely

a sm

all,

sele

cted

sepa

rate

ly.

and

varia

ble

frac

tion

.

a : As

of A

pril

2003

, the

DH

S m

alar

ia m

odul

e, h

as b

een

cond

ucte

d in

Ben

in (2

001)

, Erit

rea

(200

2), M

alaw

i (20

00),

Mau

ritan

ia (2

000/

01),

Rwan

da (2

000)

, Uga

nda

(200

0/01

), Za

mbi

a (2

001–

2002

) and

ong

oing

in K

enya

(200

3), M

ali (

2003

), an

d N

iger

ia (2

003)

.

67

General indicators

■ Population: estimates are based on UnitedNations Population Division (2000 revisionestimates), available online at:http://www.un.org/popin/.

■ Population growth rate: expressed aspercent per year, is based on estimatespublished annually in the World HealthReport, available online at:http://www.who.int/whr/2002/en/.

■ Infant mortality rate: expressed asnumber of infants per 1000 dying before firstbirthday. See:

- UNICEF, State of the world’s children2003, available online athttp://www.unicef.org/sowc03/

- Hill K. al. Trends in child mortality in thedeveloping world: 1960 to 1996, availableonline at: http://childinfo.org/cmr/revis/kh98meth.html.

■ Under-5 mortality rate: expressed asnumber of under-5s per 1000 dying beforefifth birthday. See:

- UNICEF, State of the World’s Children2003, available online athttp://www.unicef.org/sowc03/

- Hill K et al. Trends in child mortality inthe developing world: 1960 to 1996,available online at:http://childinfo.org/cmr/revis/kh98meth.html.

■ Crude birth rate: the number of births peryear per 1000 population; furtherinformation is available athttp://unstats.un.org/unsd/.

■ Population at risk: expressed aspercentage of population living where theclimate is suitable for malaria transmission.Estimates from Mapping Malaria Risk inAfrica (MARA) (see http://www.mara.org.za./)derived from a theoretical model of thedistribution of endemic and epidemic malariabased on climate conditions (long-term meanrainfall and temperature) suitable for malariatransmission. Classification of risk as endemic,epidemic or overall is based on (8,9).

Malaria morbidity and mortality

■ Hospital deaths due to malaria: thepercentage of recorded inpatient deaths that are attributed to malaria; derived from Ministry of Health reports where data are available.

■ Malaria admissions: the percentage ofrecorded health facility admissions that areattributed to malaria; derived from Ministryof Health reports where data are available.

■ Malaria outpatient attendance: thepercentage of recorded outpatient visits tohealth facilities that are attributed tomalaria; derived from Ministry of Healthreports where data are available.

■ Outpatient malaria cases per 1000persons: the total number of outpatientmalaria cases among all age groups reportedin a country in a year, expressed per 1000total population per year; total population isbased on the United Nations PopulationDivision estimates for the same year as thecase reporting.

■ Inpatient malaria cases per 1000 persons:the total number of inpatient malaria cases(i.e. hospital admissions) among all agegroups reported in a country in a year,expressed per 1000 total population; totalpopulation is based on the United NationsPopulation Division estimates for the sameyear as the case reporting.

Nets and insecticide-treated nets (ITNs)

■ Households with mosquito nets (or ITNs):the percentage of households that possessone or more nets, treated or not (or one ormore ITNs). Estimates are derived fromnationally representative household surveys,such as MICS or DHS.

■ Under-5s using mosquito nets (or ITNs):the percentage of children under 5 years oldwho slept under nets (or ITNs) the nightbefore the survey. Estimates are derived fromnationally representative household surveyssuch as MICS or DHS.

Antimalarial treatment

■ Under-5s receiving any antimalarial: thepercentage of children under 5 years old,

Country profile overview


with reported fever in the previous 2 weeks,who received any antimalarial for the fever.Estimates are derived from nationallyrepresentative household surveys such asMICS or DHS.

■ Under-5s receiving chloroquine: thepercentage of children under 5 years old,with reported fever in the previous 2 weeks,who received chloroquine for the fever.Estimates are derived from nationallyrepresentative household surveys such asMICS or DHS.

■ Under-5s receiving antimalaria within 24hours: the percentage of children under 5years old, with reported fever in the previous2 weeks, who received any antimalarial forthe fever within 24 hours of onset of thefever symptoms. Estimates are derived fromnationally representative household surveyssuch as MICS or DHS.

Attending health facility

■ Under-5s receiving any antimalarial: thepercentage of children under 5 years old,with reported fever in the previous 2 weeks,who were taken to a health facility andreceived any antimalarial for the fever.Estimates are derived from nationallyrepresentative household surveys such asMICS or DHS.

■ Under-5s receiving chloroquine: thepercentage of children under 5 years old,with reported fever in the previous 2 weeks,who were taken to a health facility andreceived chloroquine for the fever. Estimatesare derived from nationally representativehousehold surveys such as MICS or DHS.

Malaria in pregnancy

■ Pregnant women receiving antenatal careat least once: the percentage of women witha live birth during the 3 years preceding thesurvey who received antenatal care during thepregnancy by a skilled attendant. Skilledattendant is defined as a doctor, trainednurse, or midwife;traditional birth attendantsare not included in this analysis. For womenwith multiple pregnancies during the previous3 years, only the most recent pregnancy isconsidered (DHS; for MICS the interval underconsideration is the past 1 year). Estimates arederived from nationally representativehousehold surveys such as MICS or DHS.

■ Pregnant women attending antenatal careat least twice: the percentage of women with

a live birth during the 3 years preceding thesurvey who reported at least two visits forantenatal care. This does not reflect the skilllevel of the care provider. For women withmultiple pregnancies during the past 3 years,only the most recent pregnancy is considered.Estimates are derived from nationallyrepresentative household surveys such asMICS or DHS.

■ Pregnant women receiving IntermittentPreventive Treatment (IPT) at leastonce/twice: the percentage of womenreportedly pregnant at the time of the surveywho used sulfadoxine–pyrimethamine (SP)for malaria prevention during the pregnancy.The number of data points collected for thisindicator so far is limited, because IPT hasonly recently been adopted nationally andinternationally. Estimates are derived fromnationally representative household surveyssuch as MICS or DHS.

■ Pregnant women using nets/ITNs: thepercentage of women reportedly pregnant atthe time of the survey who slept under anet/ITN the night before the survey. Estimatesare derived from nationally representativehousehold surveys such as MICS or DHS.

Antimalarial treatment policy

Information on the current officialantimalarial treatment policy. Date adoptedreflects the date of recent policy change onthe type of antimalarial drugs used fortreatment. Treatment regimens are given,where information is available, foruncomplicated malaria (differentiated furtheras “probable” and parasitologically confirmedmalaria), treatment failure, severe malaria,and prevention during pregnancy.

Taxes and tariffs on insecticide-treated nets (ITNs)

Information on import tariffs and Value-Added Tax (VAT) on ITNs is based on currentlyavailable information. Data are available onlyfrom countries where a change of policy hasbeen made. Date reflects the last year ofknown update or available information. If thechange occurred before the April 2001 AbujaSummit, date is specified as “before Abuja”.

Proportional rank for selected indicators

Information on selected indicators expressedas the country’s proportional rank among allcountries in Africa south of the Sahara forwhich information is currently available.

68


References

1. Demographic and Health Surveys (DHS). Calverton,MD, ORC Macro. http://www.measuredhs.com.

2. End-decade multiple indicator cluster survey manual.New York, United Nations Children's Fund, 2000(available at http://www.childinfo.org/index2.htm).

3. Population and health in developing countries. Vol. I.Population, health and survival at INDEPTH sites.Ottawa, International Development Research Centre,2002.

4. Snow RW et al. Estimating mortality, morbidity anddisability due to malaria among Africa's non-pregnantpopulation. Bulletin of the World Health Organization,1999, 77(8):624–640.

5. Byass P et al. The role of demographic surveillancesystems (DSS) in assessing the health of communities: anexample from rural Ethiopia. Public Health, 2002,116(3):145–150.

6. Adult Morbidity & Mortality Project - Phase 1 (AMMP-1). The policy implications of adult morbidityand mortality end of phase 1 report. 1997:

7. McCombie SC. Treatment seeking for malaria: a reviewof recent research. Social Science and Medicine, 1996,43(6):933–945.

8. Snow RW et al. A preliminary continental risk map formalaria mortality among African children. ParasitologyToday, 1999, 15:99-104.

9. Craig MH et al. A climate-based distribution model ofmalaria transmission in sub-Saharan Africa. ParasitologyToday, 1999, 15:105–111.

69


General indicators Source 2001

Population (thousands) UNPOP 13527Population growth rate (%) UNPOP 3.2Infant mortality rate (per 1000) UNICEF 154Under-5 mortality rate (per 1000) UNICEF 260Crude birth rate (per 1000) UNPOP 51.3

Pop. at endemic (epidemic) risk (%) MARA1 90 (8)

Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 25Malaria admissions MoH 20Malaria outpatient attendance MoH 50

Outpatient malaria cases (per 1000) MoH 94 62 105 66Inpatient malaria cases (per 1000)

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets Under-5s using mosquito nets MICS 10Households with ITNs Under-5s using ITNs MICS 2

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial MICS 63Under-5s receiving chloroquine MICS 57Under-5s receiving any antimalarial <24 h

Attending health facilityUnder-5s receiving any antimalarial MICS 45Under-5s receiving chloroquine MICS 40

Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once MICS 66PW attending ANC at least twice PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets PW using ITNs

Antimalarial treatment policyDate adopted: ----

Uncomplicated Probable CQ 25 mg/kg Confirmed CQ 25 mg/kg

Eline^s Chart Treatment failure SPSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention CQ weekly

Taxes and tariffs on insecticide-treated nets DateImport tariff - netsVAT - nets

NotesFor an explanation of indicators, ranks, and other information, see the Country Profile Overview. Sources listed reflect most recent value.Case notification rate is based on number of reported outpatient cases.Nationally representative surveys (MICS and DHS) are used for reporting intervention coverage estimates for 1) mosquito net and ITN possession and use and 2) antimalarial treatment. 1 Population at risk is determined from modelling retrospective climate data and population projections. Percentage of population at risk does not vary by year.

Angola

Under-5s receiving any antimalarial

Under-5s using mosquito nets

Hospital deaths due to malaria

Malaria admissions

Under-5 mortality rate

Proportional rank for select indicators

Lowest Second Middle Fourth Highest

0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

350

400

per

100

0

Hospital deaths due to malaria (%) Malaria admissions (%)

Malaria outpatient attendance (%) Under-5 mortality rate (per 1000)

Malaria case notification rate (per 1000)

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 15 6Malaria admissions MoH 26 19Malaria outpatient attendance MoH 35 36 36

Outpatient malaria cases (per 1000) MoH 115 85 Inpatient malaria cases (per 1000) MoH 2

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 40Under-5s using mosquito nets DHS 32Households with ITNs Under-5s using ITNs DHS 7

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial DHS 60Under-5s receiving chloroquine DHS 59Under-5s receiving any antimalarial <24 h

Attending health facilityUnder-5s receiving any antimalarial Under-5s receiving chloroquine

Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once DHS 88PW attending ANC at least twice DHS 83PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets PW using ITNs




Taxes and tariffs on insecticide-treated nets DateImport tariff - nets 18% 2000VAT - nets 18% 2000


Benin




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%



Pop. at risk (%) MARA1 32

Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 2Malaria admissions Malaria outpatient attendance


Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets Under-5s using mosquito netsHouseholds with ITNs Under-5s using ITNs

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarialUnder-5s receiving chloroquineUnder-5s receiving any antimalarial <24 h


Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once MICS 97 PW attending ANC at least twice PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets PW using ITNs

Antimalarial treatment policyDate adopted: 1997

Uncomplicated Probable SPConfirmed SP

Eline^s Chart Treatment failure Quinine 3 times a day, 7 daysSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention SP - IPT



Botswana




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 33 31Malaria admissions MoH 26 27Malaria outpatient attendance MoH 30 31

Outpatient malaria cases (per 1000) MoH 81 78 6 Inpatient malaria cases (per 1000) MoH 3 3

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets Under-5s using mosquito nets Households with ITNs Under-5s using ITNs

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial Under-5s receiving chloroquine Under-5s receiving any antimalarial <24 h


Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once DHS 63 PW attending ANC at least twice DHS 56 PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets PW using ITNs


Uncomplicated Probable CQ 25 mg/kgConfirmed CQ 25 mg/kg




Burkina Faso




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 47Malaria admissions MoH 51Malaria outpatient attendance


Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets Under-5s using mosquito nets MICS 3 Households with ITNs Under-5s using ITNs MICS 1

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial MICS 31 Under-5s receiving chloroquine MICS 23 Under-5s receiving any antimalarial <24 h

Attending health facilityUnder-5s receiving any antimalarial MICS 25 Under-5s receiving chloroquine MICS 19



Uncomplicated Probable SP (AS+AQ)Confirmed SP (AS+AQ)




Burundi




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

100

200

300

400

500

600

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 35Malaria admissions MoH 23Malaria outpatient attendance

Outpatient malaria cases (per 1000) Inpatient malaria cases (per 1000)






Uncomplicated Probable AQConfirmed AQ




Cameroon




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria Malaria admissions Malaria outpatient attendance





Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once PW attending ANC at least twice PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets PW using ITNs






Central African Republic




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%





Outpatient malaria cases (per 1000) MoH 53 2 47 Inpatient malaria cases (per 1000)

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito netsUnder-5s using mosquito nets MICS 27Households with ITNsUnder-5s using ITNs MICS 1









Chad




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%



Pop. at endemic (epidemic) risk (%) MARA1

Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 58 12 Malaria admissions MoH 37 34Malaria outpatient attendance MoH 52 31

Outpatient malaria cases (per 1000) MoH 6 14 14 Inpatient malaria cases (per 1000) MoH 2 3 6










Comoros




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

-10

10

30

50

70

90

110

130

150

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria Malaria admissions Malaria outpatient attendance MoH 18

Outpatient malaria cases (per 1000) MoH 38 Inpatient malaria cases (per 1000) MoH 21

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets Under-5s using mosquito netsHouseholds with ITNs Under-5s using ITNs

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarialUnder-5s receiving chloroquineUnder-5s receiving any antimalarial <24 h








Congo




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

-10

10

30

50

70

90

110

130

150

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%





Outpatient malaria cases (per 1000) MoH 93 24Inpatient malaria cases (per 1000) MoH 28 2




Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once MICS 84 88 PW attending ANC at least twice DHS 73 PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets PW using ITNs






Côte d'Ivoire




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%


Population (thousands) UNPOP 52522Population growth rate (%) UNPOP 3.2Infant mortality rate (per 1000) UNICEF 129Under-5 mortality rate (per 1000) UNICEF 205Crude birth rate (per 1000) UNPOP 48


Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malariaMalaria admissions Malaria outpatient attendance MoH 28











Democratic Republic of Congo




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Equatorial Guinea




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1999 2000 2001 2002Hospital deaths due to malaria MoH 13 6 7 4Malaria admissions MoH 26 17 21 12Malaria outpatient attendance MoH 11 9 8 5

Outpatient malaria cases (per 1000) MoH 40 33 34 20Inpatient malaria cases (per 1000) MoH 4 2 3 2

Mosquito nets and ITNs Source 1999 2000 2001 2002

Households with mosquito nets DHS2 34

Under-5s using mosquito nets DHS2 12Households with ITNs

Under-5s using ITNs DHS2 4

Antimalarial treatment Source 1999 2000 2001 2002

Under-5s receiving any antimalarial DHS2 4Under-5s receiving chloroquine Under-5s receiving any antimalarial <24 h


Malaria in pregnancy Source 1999 2000 2001 2002

PW receiving ANC at least once DHS2 70PW attending ANC at least twice PW receiving IPT at least once PW receiving IPT at least twice

PW using mosquito nets DHS2 7

PW using ITNs DHS2 3


Uncomplicated Probable CQ 25 mg/kg + SPConfirmed CQ 25 mg/kg + SP

Eline^s Chart Treatment failure Quinine 3 times a day, 7 daysSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention CQ + Proguanil weekly


NotesFor an explanation of indicators, ranks, and other information, see the Country Profile Overview. Sources listed reflect most recent value.Case notification rate is based on number of reported outpatient cases.Nationally representative surveys (MICS and DHS) are used for reporting intervention coverage estimates for 1) mosquito net and ITN possession and use and 2) antimalarial treatment. 1 Population at risk is determined from modelling retrospective climate data and population projections. Percentage of population at risk does not vary by year. 2 DHS data for 2002 is preliminary

Eritrea




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%






Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 1 Under-5s using mosquito nets Households with ITNs DHS 0 Under-5s using ITNs

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial DHS 3 Under-5s receiving chloroquine DHS 2 Under-5s receiving any antimalarial <24 h





Eline^s Chart Treatment failure Quinine 3 times a day, 7 daysSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention



Ethiopia




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

40

80

120

160

200

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%





Outpatient malaria cases (per 1000) MoH 23 Inpatient malaria cases (per 1000)










Gabon




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%





Outpatient malaria cases (per 1000) MoH 184 101 Inpatient malaria cases (per 1000)










Gambia




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%






Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito netsUnder-5s using mosquito nets Households with ITNs Under-5s using ITNs

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial DHS 61 Under-5s receiving chloroquine Under-5s receiving any antimalarial <24 h






Taxes and tariffs on insecticide-treated nets DateImport tariff - nets 13% 2001VAT - nets


Ghana




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%



Pop. at endemic (epidemic) risk (%) MARA1 100 (<1)

Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 1 1 26 15Malaria admissions MoH 23 23 23Malaria outpatient attendance MoH 27 27 33 30

Outpatient malaria cases (per 1000) MoH 104 65 109 Inpatient malaria cases (per 1000) MoH 2










Guinea




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Guinea-Bissau




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 26Malaria admissions MoH 19 Malaria outpatient attendance MoH 40








Eline^s Chart Treatment failure AQSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention SP - IPT

Taxes and tariffs on insecticide-treated nets DateImport tariff - nets 18% 2001VAT - nets


Kenya




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Liberia




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

350

400

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Madagascar




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 19 28Malaria admissions MoH 22Malaria outpatient attendance MoH 50 40 26

Outpatient malaria cases (per 1000) MoH 259 201Inpatient malaria cases (per 1000) MoH 3

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 13 Under-5s using mosquito nets DHS 8 Households with ITNs DHS 5 Under-5s using ITNs DHS 3

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial DHS 27 Under-5s receiving chloroquine DHS 1 Under-5s receiving any antimalarial <24 h DHS 22


Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once DHS 92 PW attending ANC at least twice DHS 90 PW receiving IPT at least once DHS 68 PW receiving IPT at least twice DHS 29PW using mosquito nets DHS 7 PW using ITNs






Malawi




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

100

200

300

400

500

600

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 42 40 30Malaria admissions MoH 42 46 38Malaria outpatient attendance MoH 33 40 36 32

Outpatient malaria cases (per 1000) MoH 42 11 9 7Inpatient malaria cases (per 1000) MoH 2 2 1

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 54Under-5s using mosquito nets DHS 37Households with ITNs Under-5s using ITNs

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial Under-5s receiving chloroquine DHS 38Under-5s receiving any antimalarial <24 h








Mali




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malariaMalaria admissions MoH 22Malaria outpatient attendance

Outpatient malaria cases (per 1000) MoH 70 115 97 Inpatient malaria cases (per 1000) MoH 7 4


Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial Under-5s receiving chloroquine DHS 21Under-5s receiving any antimalarial <24 h








Mauritania




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%



Pop. at endemic (epidemic) risk (%) MARA1 100 (<1)

Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 24Malaria admissions Malaria outpatient attendance MoH 28 48 44











Mozambique




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 2 6Malaria admissions MoH 6Malaria outpatient attendance MoH 17











Namibia




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Niger




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

350

400

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 10 7 7Malaria admissions MoH 18 24 9 20Malaria outpatient attendance MoH 25 22 19 27











Nigeria




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 58 56 56Malaria admissions MoH 21 15 11Malaria outpatient attendance MoH 34 25 23


Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 7 Under-5s using mosquito nets MICS 6 Households with ITNs Under-5s using ITNs MICS 5

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial MICS 13 Under-5s receiving chloroquine MICS 7 Under-5s receiving any antimalarial <24 h DHS 6




Uncomplicated Probable AQ + SPConfirmed AQ + SP

Eline^s Chart Treatment failure Quinine 3 times a day, 7 daysSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention CQ weekly



Rwanda




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%


Population (thousands) UNPOP 140Population growth rate (%) UNPOP 1.8Infant mortality rate (per 1000) UNICEF 57Under-5 mortality rate (per 1000) UNICEF 71Crude birth rate (per 1000) UNPOP


Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 46 70 Malaria admissions MoH 56 Malaria outpatient attendance MoH 55











Sao Tome and Principe




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Senegal




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Sierra Leone




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

80%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

350

400

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%











Uncomplicated ProbableConfirmed

Eline^s Chart Treatment failureSevere malariaPregnancy Prevention



Somalia




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 0.2Malaria admissions Malaria outpatient attendance





Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once DHS 94 PW attending ANC at least twice PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets PW using ITNs


Uncomplicated Probable Coartem (Kwazulu Natal); SPConfirmed Coartem (Kwazulu Natal); SP

Eline^s Chart Treatment failure Quinine 3 times a day, 7 daysSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention CQ + PG



South Africa




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

-10

10

30

50

70

90

110

130

150

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 2 5Malaria admissions MoH 6 6Malaria outpatient attendance MoH 3 2 2











Swaziland




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%






Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets

Under-5s using mosquito nets MICS2 23

Households with ITNs

Under-5s using ITNs MICS2 0

Antimalarial treatment Source 1998 1999 2000 2001

Under-5s receiving any antimalarial MICS2 50

Under-5s receiving chloroquine MICS2 49

Under-5s receiving any antimalarial <24 h

Attending health facility

Under-5s receiving any antimalarial MICS2 33

Under-5s receiving chloroquine MICS2 32




Eline^s Chart Treatment failure SP, MQ or QSevere malaria Q or ARTPregnancy Prevention


NotesFor an explanation of indicators, ranks, and other information, see the Country Profile Overview. Sources listed reflect most recent value.Case notification rate is based on number of reported outpatient cases.Nationally representative surveys (MICS and DHS) are used for reporting intervention coverage estimates for 1) mosquito net and ITN possession and use and 2) antimalarial treatment. 1 Population at risk is determined from modelling retrospective climate data and population projections. Percentage of population at risk does not vary by year.2 MICS values represent Northern Sudan

Sudan




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%















Togo




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 23 16 35Malaria admissions MoH 20 38 38Malaria outpatient attendance MoH 33 31 37 39

Outpatient malaria cases (per 1000) MoH 90 136 150 Inpatient malaria cases (per 1000) MoH 0.1

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 13Under-5s using mosquito nets DHS 7Households with ITNs Under-5s using ITNs DHS 0



Malaria in pregnancy Source 1998 1999 2000 2001PW receiving ANC at least once DHS 92PW attending ANC at least twice DHS 84PW receiving IPT at least once PW receiving IPT at least twice PW using mosquito nets DHS 7PW using ITNs DHS 1



Eline^s Chart Treatment failure Quinine 3 times a day, 7 daysSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention CQ weekly

Taxes and tariffs on insecticide-treated nets DateImport tariff - netsVAT - nets 0% 2000


Uganda




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 22 44Malaria admissions MoH 39 35 47Malaria outpatient attendance MoH 33 40 42


Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 30 Under-5s using mosquito nets DHS 21 Households with ITNs DHS 1 Under-5s using ITNs DHS 2

Antimalarial treatment Source 1998 1999 2000 2001Under-5s receiving any antimalarial DHS 53 Under-5s receiving chloroquine DHS 53 Under-5s receiving any antimalarial <24 h





Eline^s Chart Treatment failure AQSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention SP - IPT

Taxes and tariffs on insecticide-treated nets DateImport tariff - nets 0% Before AbujaVAT - nets 0% Before Abuja


United Republic of Tanzania




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

350

400

450

500

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%





Outpatient malaria cases (per 1000) MoH 342 268 342 204Inpatient malaria cases (per 1000) MoH 24 15






Uncomplicated Probable CoartemConfirmed Coartem

Eline^s Chart Treatment failure QuinineSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention CQ weekly (SP-IPT)**



Zambia




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

350

400

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%




Malaria morbidity and mortality Source 1998 1999 2000 2001Hospital deaths due to malaria MoH 8Malaria admissions MoH 15 20Malaria outpatient attendance MoH 12 15

Outpatient malaria cases (per 1000) MoH 141 96 121 Inpatient malaria cases (per 1000) MoH 1

Mosquito nets and ITNs Source 1998 1999 2000 2001Households with mosquito nets DHS 10 Under-5s using mosquito nets DHS 3 Households with ITNs Under-5s using ITNs





Uncomplicated Probable CQ + SPConfirmed CQ + SP

Eline^s Chart Treatment failure Quinine 3 times a day, 7 daysSevere malaria Quinine 3 times a day, 7 daysPregnancy Prevention Maloprim weekly

Taxes and tariffs on insecticide-treated nets DateImport tariff - nets 10% Before AbujaVAT - nets 20% Before Abuja


Zimbabwe




Malaria admissions




0% 20% 40% 60% 80% 100%

0%

20%

40%

60%

1990 1992 1994 1996 1998 2000 2002

0

50

100

150

200

250

300

per

100

0




0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

0% 20% 40% 60% 80% 100%

Documents

TheAfrica Malaria Report 2003 - WHO archivesarchives.who.int/prioritymeds/report/append/610pub_africa_malaria... · The Africa Malaria Report – 2003 6 Resource mobilization and