Articles

www.thelancet.com Vol 371 April 12, 2008 1247

Countdown to 2015 for maternal, newborn, and child survival: the 2008 report on tracking coverage of interventionsCountdown Coverage Writing Group,* on behalf of the Countdown to 2015 Core Group

SummaryBackground The Countdown to 2015 for Maternal, Newborn, and Child Survival initiative monitors coverage of priority interventions to achieve the Millennium Development Goals (MDG) for reduction of maternal and child mortality. We aimed to report on 68 countries which have 97% of maternal and child deaths worldwide, and on 22 interventions that have been proven to improve maternal, newborn, and child survival.

Methods We selected countries with high rates of maternal and child deaths, and interventions with the most potential to avert such deaths. We analysed country-specifi c data for maternal and child mortality and coverage of selected interventions. We also tracked cause-of-death profi les; indicators of nutritional status; the presence of supportive policies; fi nancial fl ows to maternal, newborn, and child health; and equity in coverage of interventions.

Findings Of the 68 priority countries, 16 were on track to meet MDG 4. Of these, seven had been on track in 2005 when the Countdown initiative was launched, three (including China) moved into the on-track category in 2008, and six were included in the Countdown process for the fi rst time in 2008. Trends in maternal mortality that would indicate progress towards MDG 5 were not available, but in most (56 of 68) countries, maternal mortality was high or very high. Coverage of diff erent interventions varied widely both between and within countries. Interventions that can be routinely scheduled, such as immunisation and antenatal care, had much higher coverage than those that rely on functional health systems and 24-hour availability of clinical services, such as skilled or emergency care at birth and care of ill newborn babies and children. Data for postnatal care were either unavailable or showed poor coverage in almost all 68 countries. The most rapid increases in coverage were seen for immunisation, which also received signifi cant investment during this period.

Interpretation Rapid progress is possible, but much more can and must be done. Focused eff orts will be needed to improve coverage, especially for priorities such as contraceptive services, care in childbirth, postnatal care, and clinical case management of illnesses in newborn babies and children.

Funding Bill & Melinda Gates Foundation, UK Department for International Development, Norwegian Agency for Development Cooperation, Partnership for Maternal, Newborn and Child Health, Save the Children US and UK, United Nations Children’s Fund, United Nations Population Fund, and World Health Organization.

IntroductionThe Countdown to 2015 collaboration aims to track coverage for interventions that are essential to attainment of Millennium Development Goals (MDG) 4 and 5 and elements of MDGs 1, 6, and 7.1 Countdown to 2015 represents a common evaluation framework1 for tracking coverage of proven interventions and measures of mortality and nutrition in countries with the highest burden of mortality in mothers and children.2 The fi rst round of the Countdown,3,4 in 2005, reported on 60 countries and 17 interventions, and focused on child survival. The Countdown process has since evolved4 from attention to the individual child or mother to focus on the continuum of care from before pregnancy, through to pregnancy, childbirth, and the postnatal period, and on to early childhood.5 Tracking of single, biologically based interventions was complemented by inclusion of broader approaches or packages, such as antenatal or postnatal care, that can serve as platforms

for delivery of multiple interventions. The list of priority countries was also expanded, from those with the highest burden of mortality in children to those with high burdens of either maternal or child mortality.

The aim of Countdown has remained unchanged: to assess every 2 or 3 years until 2015—the target date for the MDGs—whether proven interventions and approaches are reaching more women, newborn babies, and children, and especially the poorest families.3 The 2008 Countdown meeting is in Cape Town, linked to the Inter-Parliamentary Union meeting.6 Countdown is independent, but has a wide ownership, incorporating UN agencies and civil society, individual researchers, and development workers from the country, regional, and international levels. Its mechanisms are mainly international publications and conferences, linked to national action. Countdown is designed to be action-oriented, in that it amalgamates the information needed to assess progress and to work collectively to

Lancet 2008; 371: 1247–58

*Members listed at end of paper

Correspondence to:Jennifer Bryce, 2081 Danby Road, Ithaca, NY 14850, USAjbrycedanby@aol.com

Articles

1248 www.thelancet.com Vol 371 April 12, 2008

accelerate gains and address shortcomings. An important outcome of Countdown is identifi cation of gaps in data and evidence, to thereby catalyse development of methods and instruments to better assess coverage over time.

The Countdown process builds on and complements other important eff orts to track progress toward the MDGs.3 It consists of individual country profi les, with selected information about the demographic and epidemiological contexts of coverage and the key determinants of coverage. We aimed to present the fi ndings for progress in coverage in selected countries in the context of trends in mortality. Other papers in this special issue also relate to progress in health policy and the strengthening of health systems;7 overseas development assistance for maternal, newborn, and child health and nutrition;8 and inequities in coverage across multiple interventions.9

MethodsSelection of priority countries and interventionsWe focused on 68 priority countries that together had 97% of maternal and child deaths in 2005 (fi gure 1). Details on the inclusion criteria for countries are in our technical report.10 Inclusion criteria for interventions were (1) internationally accepted peer-reviewed evidence of mortality reduction in mothers, newborn babies, or children younger than 5 years of age, or evidence of use

in most priority countries as a package or platform for providing at least one intervention (eg, antenatal care, package can deliver tetanus toxoid immunisation and identifi cation and management of sexually transmitted diseases); (2) feasibility for implementation to universal coverage levels in low-income countries; and (3) availability (or expected availability within 3 years) of nationally representative data based on a standard indicator for coverage in most priority countries. We convened a working group on indicators and coverage data, which reviewed new evidence about the interventions that were included in the 2005 Countdown to decide whether any additional interventions or delivery platforms should be included in 2008.10 We focused on interventions that have the potential to avert the greatest number of deaths in mothers, newborn babies, and children. Interventions and defi nitions of the coverage indicators for the 2008 Countdown are in the technical report.10 We defi ned coverage as the proportion of people in a population who need a service and receive it.3 We defi ned women as those aged 15–49 years; mothers as pregnant women or those who had given birth; children as those from birth until 59 months (ie, younger than 5 years of age); and newborn babies as infants aged 0–28 days.10

Data sources and analysisNone of the 68 priority countries had health-information systems that were able to produce coverage estimates for

HaitiMauritania

SenegalGambia

Guinea BissauGuinea

Sierra LeoneLiberia

TogoBenin

Peru

Bolivia

Brazil

Mexico

Guatemala

Mali

Burkina Faso

GhanaCote d’Ivoire

Niger

Nigeria

Morocco

Chad

Cameroon

Eritrea

Egypt

Ethiopia

Djibouti

Central AfricanRepublic

Equatorial GuineaGabon

Congo

Democratic Republic ofthe Congo

RwandaBurundi

Uganda

SomaliaKenya

TanzaniaMalawi

Zambia

ZimbabweBotswana

South Africa

Mozambique

Madagascar

LesothoSwaziland

Afghanistan

TurkmenistanTajikistan

Pakistan

Vietnam

IranNepal

China

North Korea

India

Iraq

Azerbaijan

Bangladesh

Papua NewGuinea

Indonesia

Philippines

YemenCambodia

LaosBurma

Sudan

Angola

Countries added in the Countdown 2008Priority countries in 2005 and 2008

Figure 1: The 2008 countdown priority countries

Articles

www.thelancet.com Vol 371 April 12, 2008 1249

all the standard indicators.10 We used country-specifi c estimates of mortality in children younger than 5 years from tables prepared by UNICEF.11 The methods and limitations associated with these estimates are described elsewhere.12 We assessed progress toward MDG 4 by determining whether the average annual rate of reduction in mortality in children younger than 5 years between 1990 and 2006 was greater than or equal to the rate that would be needed between 2007 and 2015 to achieve the MDG.11 We obtained country-specifi c cause-of-death profi les for children from WHO databases,13,14 with updated neonatal cause-of-death estimates for 2005 based on methods used in 2000.15 We assessed indicators of nutritional status, and adjusted country-specifi c estimates to incorporate WHO’s new growth standards.16 Estimates of the prevalence of low birthweight were adjusted for high levels of under-reporting, especially in sub-Saharan Africa.17

We used 2005 data for country-specifi c estimates of maternal mortality per 100 000 livebirths.18 Because the large margins of uncertainty around these estimates precluded trend analysis,19 maternal mortality ratios were classifi ed as either very high (550 or more); high (300–549); moderate (100–299); or low (below 100).2 We used regional estimates from WHO’s systematic review for maternal cause-of-death profi les.20

With few exceptions, we obtained country-specifi c indicators for coverage of the Countdown interventions from population-based surveys in the 68 priority countries—usually either the UNICEF-supported multiple-indicator cluster surveys (MICS)21 or the USAID-supported demographic and health surveys (DHS).22 These data were available in global databases maintained by UNICEF.23 The exceptions were interventions for which data collection and analysis of coverage indicators were not yet routine or harmonised—eg, special analyses for postnatal visits within 2 days of birth were undertaken by Save the Children, and unmet need for family planning by the United Nations Population Fund (UNFPA). In addition, coverage estimates for vaccinations, vitamin A supplementation, and prevention of mother-to-child transmission of HIV were obtained from routine programme data and data collected through household surveys.

We summarised indicators of coverage for countries in which they were relevant (eg, only 45 of the 68 countries had endemic malaria, defi ned here as documented risk of Plasmodium falciparum transmission nationwide and throughout the year).24 The country profi les in the Countdown report10 present coverage for countries with malaria risk in defi ned geographic areas. We assumed that antiretroviral treatment for pregnant women with HIV, to prevent mother-to-child transmission of the virus, was needed in all Countdown priority countries. The latest available coverage data and methods for estimating coverage for the use of antiretroviral treatment for the prevention of mother-to-child transmission of

HIV were based on harmonised estimates developed by UNAIDS, UNICEF, and WHO. These data were based on denominators derived from unpublished HIV estimates for 2007 by UNAIDS and WHO, and are more recent than those in the State of the World’s Children Report 2008.11 Data for the prevalence of caesarean sections were from analysis of demographic and health surveys.22 To construct summary estimates for each of the coverage indicators, we included point estimates from data collected between 2000 and 2006. Exceptions were estimates for coverage for vitamin A supplementation (which were based on estimates for 2005) and coverage of vaccines for measles; diptheria, pertussis, and tetanus (DPT3); Haemophilus Infl uenzae (Hib3); and neonatal tetanus protection (which refer to estimates for 2006).10

Role of the funding source Some of the funding for this work came from the regular budgets of the authors’ institutions, primarily in the form of in-kind contributions of staff time. These sponsors were therefore involved through their staff in the analysis and interpretation of the data contained in the country profi les, in the writing of the report, and in the decision to submit the paper for publication. However, the views expressed in this report are those of the authors and do not necessarily represent the policies or views of their institutions.

ResultsThe most important Countdown fi ndings are presented as country profi les. Figure 2 shows the country profi le from Benin to illustrate how the coverage data are complemented by information that is useful for their interpretation, such as policy and fi nancial fl ows. We selected Benin as an example because it was the fi rst country profi le (in alphabetical order) in the report for which data were available for all major indicator categories such as malaria and HIV. The full 2008 Countdown report10 provides information for the 68 countries and a detailed analysis of the main trends and issues. These can be summarised as fi ve themes (panel), all of which have implications for continuing eff orts to accelerate progress towards the health MDGs.

The fi rst theme was the inadequate progress towards reduction of maternal and child mortality. Table 1 shows UNICEF’s analysis of trends in mortality in the 68 Countdown priority countries and their progress towards the MDG 4 target, which is to reduce mortality in children younger than 5 years by two-thirds between 1990 and 2015. The number of child deaths worldwide has been reduced to fewer than 10 million per year, although the relative proportion in Africa (now over 50%) continued to rise.11

Data for the 2008 Countdown cycle showed that 16 of 68 countries (24%) were on track to meet MDG 4. These included seven countries (Bangladesh, Brazil, Egypt, Indonesia, Mexico, Nepal, and the Philippines) that had

Articles

1250 www.thelancet.com Vol 371 April 12, 2008

44% (2006)9% (2006)

50% (2006)16% (2001)

Intervention coverage for mothers, newborn babies, and children

Nutrition

Exclusive breastfeeding Vitamin A supplementation

Diarrhoeal disease treatment Pneumonia treatment

Stunting prevalence (moderate and severe) Wasting prevalence (moderate and severe)

Complementary feeding rate (6–9 months) Low birthweight incidence

1999 2000 2001 2002 2003 2004 2005

23% 42%

1996 2001

At least one doseTwo doses

Prevalence of underweight children

20%

2001

22%

2006 2001

38%

70%

2006

56

100% 96% 85%95%

98%95% 89% 92%

94% 94%

1990 1995 2000 2005

93%93%

89%

1996 2001

32% 35%

Causes of deaths in children

20

0

40

60

80

100

Unde

rwei

ght c

hild

ren

youn

ger t

han

5 ye

ars

Child

ren

imm

unise

d (%

)

Child

ren

youn

ger t

han

10 y

ears

who

had

inse

ctici

de-t

reat

ed n

ets (

%)

HIV

-pos

itive

pre

gnan

tw

omen

who

rece

ived

antir

etro

vira

l dru

gs (%

)

Exclu

sivel

y br

east

fed

infa

nts

youn

ger t

han

6 m

onth

s (%

)De

aths

per

100

0 liv

ebirt

hs

Child

ren

aged

6–5

9 m

onth

sw

ho re

ceiv

ed v

itam

in A

(%)

1996

10% 20

0

40

60

80

100

1990 1995 2000 2005 2010 2015

40

0

80

120

160

200

62

148

185

MDG target

Child mortality rate

Malaria prevention

2001 2006

20%7%

Malaria treatment

20062001

54%60%

2005

27%

2006

53%

0 000

Prevention of mother-to-childtransmission of HIV

20

0

40

60

80

100

20

0

40

60

80

100

Child

ren

youn

ger t

han

5 ye

ars w

ithdi

arrh

oea

who

had

ora

l reh

ydra

tion

trea

tmen

t or i

ncre

ased

flui

ds (%

)

20

0

40

60

80

100

Febr

ile ch

ildre

n yo

unge

r tha

n 10

yea

rsw

ho re

ceiv

ed a

ntim

alar

ial d

rugs

(%)

20

0

40

60

80

100

Child

ren

youn

ger t

han

5 ye

ars

with

susp

ecte

d pn

eum

onia

(%)

20

0

40

60

80

100

20

0

40

60

80

100

20

0

40

60

80

100

Child health

8 760 000 (2006)1 488 000 (2006)

358 000 (2006)70% (2006)148 (2006)

88 (2006)38 (2000)

53 000 (2006)840 (2005)

1 in 20 (2005)2900 (2005)

Total populationTotal population younger than 5 yearsBirthsBirth registrationChild mortality rate (per 1000 livebirths)Infant mortality rate (per 1000 livebirths)Neonatal mortality rate (per 1000 live births)Deaths in children younger than 5 yearsMaternal mortality ratio (per 100 000 livebirths)Lifetime risk of maternal deathTotal maternal deaths

Children taken to appropriate health providerChildren given antibiotics (data not available)

Measles vaccineThree doses DPT vaccineThree doses Hib vaccine

Immunisation

Diarrhoea 2%Tetanus 4%Other 5%Congenital 8%Asphyxia 19%

Infection 34%

Pretermbirth 28%

Pneumonia21%

Diarrhoea17%

Measles5%

HIV/AIDS2%

Injuries2%

Other0% Malaria

27%

Neonatal25%

Causes of neonataldeaths

Demographics

Globally more than a third of child deaths are attributable to undernutrition

Figure 2: Country profi le for Benin The profi les for all 68 priority countries are in the Countdown report, with details of sources and methods for each data type.10 DPT=diphtheria, pertussis, and tetanus toxoid. HiB=Haemophilus infl uenzae type B.

Articles

www.thelancet.com Vol 371 April 12, 2008 1251

Water and sanitation

1990 2004

57%

73%

57%

78%67%63%

1990 2004

2%

32%

11%

59%

33%

12%

66

80% 81%88%

60% 66%78%

94%

Causes of maternal deaths

Regional estimates for Africa, 1997–2002

Other causes30%

Anaemia 4%

Haemorrhage34%

Abortion 4%Obstructed labour 4%

HypertensiveDisorders 9%

Sepsis or infections,Including AIDS 16%

20

0

40

60

80

100

20

0

40

60

80

100

20

0

40

60

80

100

20

0

40

60

80

100

20

0

40

60

80

100

Rural Urban Total Rural Urban Total

Water Sanitation

Prop

ortio

n of

pop

ulat

ion

usin

g im

prov

ed d

rinki

ngw

ater

sour

ces (

%)

Prop

ortio

n of

pop

ulat

ion

usin

gim

prov

ed sa

nita

tion

facil

ities

(%

)

Equity

Poorest Secondpoorest

Thirdpoorest

Fourthpoorest

Wealthiest

20

0

40

60

80

100

Coverage gapRatio (poorest/wealthiest)

Difference (poorest–wealthiest)

19962001

199648·0% 1·929·0%

200141·0% 1·722·0%

Cove

rage

gap

by

wea

lth q

uint

ile

1986 1991 20011996 20061996 2001 2006 1996 2001 2006

Skilled attendant at deliveryAntenatal care Neonatal tetanus protection

Live

birt

hs a

tten

ded

bysk

illed

hea

lth p

erso

nnel

(%)

New

born

bab

ies p

rote

cted

agai

nst t

etan

us (%

)

Wom

en a

ged

15–4

9 ye

ars a

tten

ded

at le

ast o

nce

by a

skill

ed h

ealth

prov

ider

dur

ing

preg

nanc

y (%

)

Financial flows and human resources

Expenditure on health per person Proportion of total government expenditurespent on health Out-of-pocket expenditure as proportion of total expenditure on healthDensity of health workers (per 1000 population) Official Development Assistance to child health(per child) Official Development Assistance to maternal andneonatal health (per livebirth) National availability of Emergency Obstetric Careservices (proportion of recommended minimum)

US$40 (2007)10% (2007)

49% (2007)

0·9 (2004)US$7 (2005)

US$4 (2005)

66% (2002)

International Code of Marketing of Breastmilk Substitutes New formula for oral rehydration salts and zinc for management of diarrhoeaCommunity treatment of pneumonia with antibioticsIntegrated Management of Childhood Illnesses adaptedto cover newborn babies (0–1 week of age)Costed implementation plan or plans for maternal,newborn, and child health availableMidwives be authorised to administer a core set of life-saving interventions Maternity protection in accordance with InternationalLabour Organisation Convention 183Specific notification of maternal deaths

YesYes

PartialYes

Yes

Partial

No

Yes

27% (2001)62% (2001)

3% (2006)3% (2001)6% (2001)2% (2001)

49% (2001)..

Unmet need for family planningAntenatal visits for woman (four or more visits)Intermittent preventive treatment for malariaRate of caesarean section (total*)Rate of caesarean section (urban)Rate of caesarean section (rural)Early initiation of breastfeeding (within 1 hour of birth)Postnatal visit for baby (within 2 days for home births)

*Target is a minimium of 5% and maximum of 15%

Coverage along the continuum of care

Pre-pregnancy

Pregnancy

Birth

Neonatal period

Infancy

0 20 40 60 80 100

17%

88%

78%

70%

89%Measles

Exclusivebreastfeeding

Skilled attendantat birth

Antenatal visit(one or more)

Contraceptiveprevalence rate

Postnatal care

Maternal and newborn health

Policies Systems

Articles

1252 www.thelancet.com Vol 371 April 12, 2008

been included in the 2005 Countdown report and were judged to be on track for reduction of child mortality at that time.4 Six of the remaining nine countries were included in the Countdown process for the fi rst time in 2008 (Bolivia, Eritrea, Guatemala, Laos, Morocco, and Peru). The other three countries (China, Haiti, and Turkmenistan) had been assessed at the time of the 2005 Countdown and were not on track; therefore they had made demonstrable progress toward MDG 4 between 2005 and 2008. The achievement of China, as the world’s largest country, is important, as are encouraging signs that several countries, many of which are in east Africa, have reduced mortality in children younger than 5 years since 2005. Although these countries have not yet made enough progress to be on track to meet MDG 4, strategic rapid action could bring these countries on track.

Table 2 presents the best available estimates of maternal mortality for the 68 countries, and therefore shows progress towards the MDG 5 target, which is to reduce the maternal mortality ratio by three-quarters between 1990 and 2015. However, we did not have suffi cient data to assess trends in maternal mortality over time. Of the 68 priority countries, 56 (82%) had either high or very high rates of maternal mortality. Only three (Azerbaijan, China, and Mexico) had low maternal mortality. Lifetime risks for maternal death ranged from one in seven (in Niger) to one in 1300 (in China). Sub-Saharan Africa as a whole accounted for 50% of the world’s maternal deaths, and south Asia for another 45%.2 In all the 41 countries in sub-Saharan Africa that were priority countries in the 2008 Countdown, maternal mortality was very high (32 countries) or high (nine countries).

Tables 3 and 4 show the ten countries that made the greatest and the least progress toward MDG 4 and had the lowest and highest maternal mortality as a proxy for progress towards MDG 5. Of 26 countries with no reduction of mortality in children younger than 5 years, all had high or very high maternal mortality. None of the ten countries with either most progress in mortality in children younger than 5 years or lowest maternal mortality were in sub-Saharan Africa. All ten with the least reduction in mortality in children younger than 5 years and nine of the ten countries with the highest maternal mortality were in this region. However, some countries had made progress towards reduction of child mortality but still had high maternal mortality; other countries had fairly low maternal mortality but had not made progress in reducing child mortality.

Many of the countries in which mortality in children younger than 5 years increased between 1990 and 2006 had high HIV prevalence (eg, Botswana, Swaziland, Lesotho, South Africa, and Zimbabwe). HIV was also a factor in the countries with the highest absolute mortality in children younger than 5 years and in the 26 countries with no progress on mortality.10 Ten of the 26 countries with no progress towards reduction of child mortality

Mortality in children younger than 5 years

MDG 4 target for 2015

Average annual rate of reduction (1990–2006)

Rate of reduction needed to meet MDG 4 (2007–15)

Progress towards the MDG 4 target

1990 2006

Afghanistan 260 257 87 0·1% 12·1% No progress

Angola 260 260 87 0·0% 12·2% No progress

Azerbaijan 105 88 35 1·1% 10·2% Insuffi cient

Bangladesh 149 69 50 4·8% 3·6% On track

Benin 185 148 62 1·4% 9·7% Insuffi cient

Bolivia 125 61 42 4·5% 4·2% On track

Botswana 58 124 19 −4·7% 20·7% No progress

Brazil 57 20 19 6·5% 0·6% On track

Burkina Faso 206 204 69 0·1% 12·1% No progress

Burma 130 104 43 1·4% 9·7% Insuffi cient

Burundi 190 181 63 0·3% 11·7% No progress

Cambodia 116 82 39 2·2% 8·3% Insuffi cient

Cameroon 139 149 46 −0·4% 13·0% No progress

Central African Republic

173 175 58 −0·1% 12·3% No progress

Chad 201 209 67 −0·2% 12·6% No progress

China 45 24 15 3·9% 5·2% On track

Congo 103 126 34 −1·3% 14·5% No progress

Congo, Democratic Republic

205 205 68 0·0% 12·2% No progress

Côte d’Ivoire 153 127 51 1·2% 10·1% Insuffi cient

Djibouti 175 130 58 1·9% 8·9% Insuffi cient

Egypt 91 35 30 6·0% 1·6% On track

Equatorial Guinea 170 206 57 −1·2% 14·3% No progress

Eritrea 147 74 49 4·3% 4·6% On track

Ethiopia 204 123 68 3·2% 6·6% Insuffi cient

Gabon 92 91 31 0·1% 12·1% No progress

Gambia 153 113 51 1·9% 8·8% Insuffi cient

Ghana 120 120 40 0·0% 12·2% No progress

(Continues on next page)

Panel: Countdown to 2015—themes from the data

1 Progress towards the Millennium Development Goals for maternal and child survival has been inadequate in the 68 priority countries, despite some hopeful signs

2 Coverage was uneven between countries and between interventions: interventions delivered through a scheduled approach had greater coverage than clinical-care interventions, which progressed more slowly and had higher levels of inequity

3 Coverage varied along the continuum of care, with especially low coverage for family planning, care in childbirth, postnatal care, and case management of illnesses in newborn babies and children

4 Opportunities have been missed to increase coverage by combined delivery of interventions throughout the continuum of care

5 The eff ectiveness of Countdown depends on the availability of reliable and timely data to guide decision making in countries

Articles

www.thelancet.com Vol 371 April 12, 2008 1253

were countries in confl ict or post-confl ict situations (eg, Chad and Congo).

The second theme was that coverage of interventions was uneven between countries and between interventions (panel). Table 5 shows the average change in coverage of selected interventions in 68 countries that had at least two comparable data points since 2000. The wide ranges in table 5 show diff erences in coverage for each intervention in the 68 countries that can also be seen from a comparison of coverage between individual country profi les.10 Skilled attendance, oral rehydration therapy, and care-seeking for pneumonia increased by 2% or less compared with increases of 4–7% for tetanus immunisation in pregnancy, antenatal care (at least one visit), and insecticide-treated bednets.

The third theme was that coverage varied across the continuum of care (panel). Figure 3 shows the median coverages for a selection of Countdown interventions along the continuum of care from before pregnancy to early childhood.5 Services that can be scheduled—notably antenatal care and immunisation—had high coverage. Figure 3 emphasises three priority gaps in the continuum of care: contraceptive prevalence, skilled attendance at birth, and clinical case management of newborn and child illnesses. Contraceptive prevalence diff ers from other interventions in that the target coverage is not 100%, but varies according to desired family size. However, coverage was low, for both demand and supply reasons. Cultural and social factors inhibit use, but investment in family-planning programmes has declined, related to global policy swings. Reinvigoration of family planning is a highly cost-eff ective approach for reduction of maternal and child deaths.

Skilled attendance at birth only reached half of women and babies in these 68 countries, and is aff ected by logistical factors (such as long distances to health facilities and high transport costs), and by supply factors, including non-existent or poor-quality services. The importance of postnatal care for both mothers and newborn babies has only recently been recognised fully.5 Technical issues related to the programmatic aspects of postnatal care (timing, and provision at home and in facility settings) need further research and translation into guidelines and protocols. Clinical case management of child illnesses was another gap in coverage; these interventions reached only about a third of ill children. Coverage of care for illnesses in newborn babies was not routinely tracked. According to updated fi nancial tracking analysis, Integrated Management of Childhood Illness (IMCI) continues to receive less than 1% of investment in maternal, newborn, and child health, so it is not surprising that progress is slow.8

The fourth theme was that opportunities to increase coverage through combined delivery of interventions have been missed (panel). The Countdown process tracks progress in coverage of single interventions (eg, immunisation) and in coverage of platforms or packages

(eg, antenatal care, skilled attendance at birth, and postnatal care) that can be used to deliver many interventions. Separate tracking of single interventions has shown that coverage rates sometimes vary within

Mortality in children younger than 5 years

MDG 4 target for 2015

Average annual rate of reduction (1990–2006)

Rate of reduction needed to meet MDG 4 (2007–15)

Progress towards the MDG 4 target

1990 2006

(Continued from previous page)

Guatemala 82 41 27 4·3% 4·5% On track

Guinea 235 161 78 2·4% 8·0% Insuffi cient

Guinea-Bissau 240 200 80 1·1% 10·2% Insuffi cient

Haiti 152 80 51 4·0% 5·1% On track

India 115 76 38 2·6% 7·6% Insuffi cient

Indonesia 91 34 30 6·2% 1·3% On track

Iraq 53 46 18 0·9% 10·6% No progress

Kenya 97 121 32 −1·4% 14·7% No progress

Laos 163 75 54 4·9% 3·6% On track

Lesotho 101 132 34 −1·7% 15·2% No progress

Liberia 235 235 78 0·0% 12·2% No progress

Madagascar 168 115 56 2·4% 8·0% Insuffi cient

Malawi 221 120 74 3·8% 5·4% Insuffi cient

Mali 250 217 83 0·9% 10·6% No progress

Mauritania 133 125 44 0·4% 11·5% No progress

Mexico 53 35 18 2·6% 7·6% On track

Morocco 89 37 30 5·5% 2·4% On track

Mozambique 235 138 78 3·3% 6·3% Insuffi cient

Nepal 142 59 47 5·5% 2·5% On track

Niger 320 253 107 1·5% 9·6% Insuffi cient

Nigeria 230 191 77 1·2% 10·1% Insuffi cient

North Korea 55 55 18 0·0% 12·2% No progress

Pakistan 130 97 43 1·8% 9·0% Insuffi cient

Papua New Guinea 94 73 31 1·6% 9·4% Insuffi cient

Peru 78 25 26 7·1% −0·4% On track

Philippines 62 32 21 4·1% 4·8% On track

Rwanda 176 160 59 0·6% 11·1% No progress

Senegal 149 116 50 1·6% 9·4% Insuffi cient

Sierra Leone 290 270 97 0·4% 11·4% No progress

Somalia 203 145 68 2·1% 8·5% Insuffi cient

South Africa 60 69 20 −0·9% 13·8% No progress

Sudan 120 89 40 1·9% 8·9% Insuffi cient

Swaziland 110 164 37 −2·5% 16·6% No progress

Tajikistan 115 68 38 3·3% 6·4% Insuffi cient

Tanzania 161 118 54 1·9% 8·7% Insuffi cient

Togo 149 108 50 2·0% 8·6% Insuffi cient

Turkmenistan 99 51 33 4·1% 4·8% On track

Uganda 160 134 53 1·1% 10·2% Insuffi cient

Yemen 139 100 46 2·1% 8·6% Insuffi cient

Zambia 180 182 60 −0·1% 12·3% No progress

Zimbabwe 76 105 25 −2·0% 15·8% No progress

Data are deaths per 100 000 livebirths, unless otherwise specifi ed. MDG=millennium development goal. Data from UNICEF.1

Table 1: Progress toward the MDG 4 target for reduction of child mortality in 68 priority countries

Articles

1254 www.thelancet.com Vol 371 April 12, 2008

these packages. For example, median coverage of interventions for prevention of mother-to-child transmission of HIV and intermittent preventive treatment of malaria in pregnancy, both of which should be delivered during antenatal care visits, was much lower than that for antenatal care, indicating that opportunities to integrate these interventions with antenatal care are being missed.25 Countdown does not monitor indicators of the quality or content of service packages such as antenatal care, skilled birth attendance, or postnatal care. The eff ectiveness of such packages depends on whether they include proven interventions, such as use of a partograph in labour or active management of the third

stage of labour. Similarly, programmatic links between diff erent elements of the continuum of care for maternal, newborn, and child health are often not being promoted or provided. For example, immunisation visits off er an opportunity to counsel women on feeding of babies and children or on family planning, and to provide vitamin A supplementation and insecticide-treated nets. Another missed opportunity relates to maternal and child undernutrition. Undernutrition has been shown to be underlying cause of 3·5 million child deaths every year, and at least 35% of the disease burden in children younger than 5 years.26 Maternal undernutrition increases the risk of death of the mother at birth, and might be

Maternal mortality ratio (2005, adjusted)

Lifetime risk of maternal death (2005)

Maternal mortality

Afghanistan 1800 1 in 8 Very high

Angola 1400 1 in 12 Very high

Azerbaijan 82 1 in 670 Low

Bangladesh 570 1 in 51 Very high

Benin 840 1 in 20 Very high

Bolivia 290 1 in 89 Moderate

Botswana 380 1 in 130 High

Brazil 110 1 in 370 Moderate

Burkina Faso 700 1 in 22 Very high

Burundi 1100 1 in 16 Very high

Burma 380 1 in 110 High

Cambodia 540 1 in 48 High

Cameroon 1000 1 in 24 Very high

Central African Republic

980 1 in 25 Very high

Chad 1500 1 in 11 Very high

China 45 1 in 1300 Low

Congo 740 1 in 22 Very high

Congo, Democratic Republic

1100 1 in 13 Very high

Côte d’Ivoire 810 1 in 27 Very high

Djibouti 650 1 in 35 Very high

Egypt 130 1 in 230 Moderate

Equatorial Guinea 680 1 in 28 Very high

Eritrea 450 1 in 44 High

Ethiopia 720 1 in 27 Very high

Gabon 520 1 in 53 High

Gambia 690 1 in 32 Very high

Ghana 560 1 in 45 Very high

Guatemala 290 1 in 71 Moderate

Guinea 910 1 in 19 Very high

Guinea-Bissau 1100 1 in 13 Very high

Haiti 670 1 in 44 Very high

India 450 1 in 70 High

Indonesia 420 1 in 97 High

Iraq 300 1 in 72 High

Kenya 560 1 in 39 Very high

(Continues in next column)

Maternal mortality ratio (2005, adjusted)

Lifetime risk of maternal death (2005)

Maternal mortality

(Continued from previous column)

Laos 660 1 in 33 Very high

Lesotho 960 1 in 45 Very high

Liberia 1200 1 in 12 Very high

Madagascar 510 1 in 38 High

Malawi 1100 1 in 18 Very high

Mali 970 1 in 15 Very high

Mauritania 820 1 in 22 Very high

Mexico 60 1 in 670 Low

Morocco 240 1 in 150 Moderate

Mozambique 520 1 in 45 High

Nepal 830 1 in 31 Very high

Niger 1800 1 in 7 Very high

Nigeria 1100 1 in 18 Very high

North Korea 370 1 in 140 High

Pakistan 320 1 in 74 High

Papua New Guinea 470 1 in 55 High

Peru 240 1 in 140 Moderate

Philippines 230 1 in 140 Moderate

Rwanda 1300 1 in 16 Very high

Senegal 980 1 in 21 Very high

Sierra Leone 2100 1 in 8 Very high

Somalia 1400 1 in 12 Very high

South Africa 400 1 in 110 High

Sudan 450 1 in 53 High

Swaziland 390 1 in 120 High

Tajikistan 170 1 in 160 Moderate

Tanzania 950 1 in 24 Very high

Togo 510 1 in 38 High

Turkmenistan 130 1 in 290 Moderate

Uganda 550 1 in 25 Very high

Yemen 430 1 in 39 High

Zambia 830 1 in 27 Very high

Zimbabwe 880 1 in 43 Very high

Data are deaths per 100 000 livebirths, unless otherwise specifi ed. MDG=millennium development goal. Data from UNICEF.1

Table 2: Progress toward the MDG 5 target for maternal mortality in 68 priority Countdown countries

Articles

www.thelancet.com Vol 371 April 12, 2008 1255

associated with at least 20% of maternal mortality.26 In 33 of the 68 priority countries, at least 20% of children were moderately or severely underweight, and in 62 countries, stunting prevalence was greater than 20%. Although the prevalence of exclusive breastfeeding has shown some improvement, the median coverage in the priority countries was low (28%, range 1–88%). Low prevalences of timely initiation of breastfeeding10 also show a need for continued eff orts to promote breastfeeding and improve the delivery of essential nutrition interventions.

The fi fth theme is that the eff ectiveness of the Countdown depends on the availability of reliable and timely data to guide decision making in countries (panel). Since the fi rst Countdown report in 2005, an unprecedented number of household surveys have been undertaken, with new MICS data from 54 countries, new DHS data for 35 countries, and recent surveys for AIDS indicators and prevention of mother-to-child transmission of HIV. Although a large amount of coverage data was compiled for the 68 countries, other than data for immunisation, nearly a third of all data were either lacking, did not use comparable defi nitions, were not nationally representative, or were from before 2000. Of this missing third, more than half was from 19 countries and pertained to data for

postnatal care, childhood treatment for pneumonia, intermittent presumptive treatment of malaria in pregnancy, antenatal care visits (four or more), and caesarean sections. Given the preponderance of maternal and newborn deaths at birth and within the fi rst few days after birth, fi lling the data gaps for maternal care and for postnatal care should be a top priority, and would be achievable with fairly minor changes to fi eld survey questionnaires.

Trends in coverage in the 68 priority countries could not be calculated for 40% of country-specifi c estimates for the nine interventions being tracked, which limited our ability to assess the eff ectiveness of programmes. Although many trend data were missing because of recent initiation of country-wide programmes (such as insecticide-treated nets and case management for pneumonia), about 60% were missing because national data for established programmes (such as exclusive breastfeeding, contraception, antenatal care, and skilled attendance at birth) had been assessed only infrequently. Of the 45 countries with endemic malaria that were included in the Countdown study, a third of data for coverage of preventive and curative treatment for malaria and for use of insecticide-treated nets were unavailable.

Rank Mortality in children younger than 5 years (1990)

Mortality in children younger than 5 years (2006)

Change (1990–2006)

Ten with most progress

Peru 1 78 25 68%

Brazil 2 57 20 65%

Indonesia 3 91 34 63%

Egypt 4 91 35 62%

Nepal 5 142 59 58%

Morocco 6 89 37 58%

Laos 7 163 75 54%

Bangladesh 8 149 69 54%

Bolivia 9 125 61 51%

Guatemala 10 82 41 50%

Ten with least progress

Chad 59 201 209 –4%

Cameroon 60 139 149 –7%

South Africa 61 60 69 –15%

Equatorial Guinea 62 170 206 –21%

Congo 63 103 126 –22%

Kenya 64 97 121 –25%

Lesotho 65 101 132 –31%

Zimbabwe 66 76 105 –38%

Swaziland 67 110 164 –49%

Botswana 68 58 124 –114%

Data are rank or deaths per 100 000 livebirths, unless otherwise specifi ed. MDG=millennium development goal.

Table 3: Ten best and ten worst performing countries on progress towards MDG 4 since 1990

Rank Maternal mortality

Ten with lowest mortality

China 1 45

Mexico 2 60

Azerbaijan 3 82

Brazil 4 110

Egypt 5 130

Turkmenistan 6 130

Tajikistan 7 170

Philippines 8 230

Peru 9 240

Morocco 10 240

13 with highest mortality

Congo, Democratic Republic 56= 1100

Malawi 56= 1100

Nigeria 56= 1100

Guinea-Bissau 56= 1100

Burundi 56= 1100

Liberia 61 1200

Rwanda 62 1300

Somalia 63= 1400

Angola 63= 1400

Chad 65 1500

Niger 66= 1800

Afghanistan 66= 1800

Sierra Leone 68 2100

Data are rank or deaths per 100 000 livebirths. MDG=millennium development goal.

Table 4: Ranking of selected countries for maternal mortality as a proxy for progress towards MDG 5

Articles

1256 www.thelancet.com Vol 371 April 12, 2008

This gap is already being addressed with increased investment in tracking coverage of treated bednets.

Other gaps in data that have been emphasised by the Countdown process include the need for improved tracking of deaths, especially regular comparable data for maternal and newborn deaths. Despite over 3·2 million stillbirths every year, these deaths remain invisible on the global agenda, even though several of the Countdown interventions could reduce such deaths.27 Insuffi cient data for the prevalence of preterm births also reduce their visibility, despite at least 1 million deaths every year and associated risks from neonatal infections.28 Data about quality of care are also needed, because high coverage of care that is low quality might not be eff ective in saving lives. The 2008 report has included policy tracking, but

many gaps remain in these data. Work on fi nancial tracking is still largely limited to donor fl ows, yet in many cases country governments are investing more, requiring more analysis and attention.

DiscussionProgress towards MDGs 4 and 5 varies between countries. Important positive milestones include China’s progress; it is now on track to meet MDG 4. High mortality in many countries provides a continuing and urgent rationale for the Countdown eff ort. The interventions and approaches included in Countdown are those with proven eff ectiveness for reduction of mortality. Factors such as human resources for health, armed confl ict, and high HIV burden have prevented implementation of these interventions, and contributed to stagnating or deteriorating progress towards reduction of mortality in women, newborn babies, and children. The Countdown results indicated that inequities (both between and within countries) and insuffi cient or unreliable fi nancial resources were important underlying constraints to progress.8,9

Understanding the reasons for gaps in coverage of diff erent interventions, and the factors associated with high and low coverage in specifi c country settings, will be imperative for future progress. Coverage of preventive interventions (such as insecticide-treated nets, vitamin A supplementation, and immunisation for measles, maternal and newborn tetanus, and H infl uenza) increased more than that for most curative interventions (treatment of pneumonia, diarrhoea, malaria, and neonatal illness) and those that depend on availability of 24-hour clinical care (skilled attendance at birth and caesarean section). Moreover, progress on interventions that require behavioural and social change, such as early initiation of breastfeeding and complementary feeding, was mixed and often insuffi cient to achieve the MDGs. Contraceptive prevalence was also lagging in coverage.

For some interventions, such as protection against neonatal tetanus and insecticide-treated nets to prevent malaria, high levels of political commitment and investment seem to have contributed to substantial increases, which are perhaps also facilitated by the simplicity of these interventions and existing vertical systems for delivery. Our data also suggest that interventions that can be scheduled for delivery in advance and that are based on simple technologies, such as immunisation, have achieved and sustained higher rates of coverage than have those that require 24-hour on-demand availability, such as correct treatment of childhood illnesses or skilled attendance at birth. Urgent action and investment will be needed to accelerate progress for coverage of clinical-care interventions, which depend on adequate access, human resources, and essential supplies. Countries that scale up innovative delivery approaches can achieve

0

20

40 Cove

rage

(%)

60

80

100

Pre-pregnancy Pregnancy Birth Postnatal Childhood

Contraceptive prevalence

One or more antenatal visits

Skilled attendant at delivery

Postnatal visit within 2 days

Exclusive breastfeeding

Case management of pneumonia

Measles immunisation

Figure 3: Coverage estimates for interventions across the continuum of care in the 68 priority countries (2000–06)

Number of countries

Average 3-year change in percentage points

Nutrition

Exclusive breastfeeding (0–5 months) 36 3 (–11 to 29)

Maternal and newborn health

Antenatal care coverage (at least one visit to skilled provider) 42 4 (–21 to 19)

Births attended by skilled health personnel 45 2 (–5 to 12)

Neonatal tetanus protection 64 5 (–11 to 31)

Contraceptive prevalence 39 2 (–7 to 10)

Child health

Care seeking for pneumonia 33 1 (–10 to 18)

Oral rehydration therapy (oral rehydration salts or recommended home fl uids) or increased fl uids, with continued feeding

31 2 (–17 to 23)

Children sleeping under insecticide-treated nets 19 7 (2 to 18)

Data are number of countries or median (range). Vitamin A is not included because changes in defi nitions aff ected comparability.

Table 5: Changes for selected interventions since 2000 for Countdown countries with at least two measurements

Articles

www.thelancet.com Vol 371 April 12, 2008 1257

rapid progress; for example, case management of pneumonia in Nepal has increased through large-scale community-based management.

Reductions in mortality in children younger than 5 years do not necessarily go hand in hand with reductions in maternal mortality. Examination of these variations in light of the coverage of interventions for maternal, newborn, and child health shows that some progress towards reduction of mortality in children younger than 5 years can generally be achieved through schedulable services that can be delivered through improved community-based and outreach approaches. Similarly, mortality in newborn babies can be reduced to some extent with schedulable services such as tetanus-toxoid immunisation and home preventive care. However, reductions in maternal mortality, and further reduction in child mortality and newborn deaths, will require 24-hour services. The priority gaps, in which interventions have low coverage and little progress, are contraceptive services, childbirth and postnatal care, and clinical case management of newborn and childhood illnesses.

In some countries, HIV prevalence has peaked and is beginning to drop,29 and rapid increases in coverage for eff ective HIV interventions in recent years might not yet have been fully captured in estimates of coverage and mortality in children younger than 5 years.30 In southern Africa, however, HIV is still having a major eff ect on both maternal mortality and that in children younger than 5 years. Expanded and sustained eff orts are needed to scale up comprehensive programmes for prevention of mother-to-child transmission of HIV, including treatment for pregnant women and those who have recently given birth31 and care and treatment for paediatric HIV. Scale-up in these areas will require a functional continuum of care, especially during the critical childbirth and postnatal period. The full eff ect of confl ict and transition occuring after confl ict might not yet be captured by current mortality estimates. For example, in 2006, Afghanistan had an offi cial estimate of mortality in children younger than 5 years of 257 deaths per 1000 livebirths, but research suggests that this might be decreasing rapidly.32 Innovative programme models, such as performance-based con-tracting to accelerate implementation, are urgently required in such environments.

An adequately functioning health system was crucial to progress towards the MDGs. Variations in coverage across the continuum of care showed where focused eff ort will be needed to meet the MDGs. Priority attention in health-system strengthening should be given to establishment of a functional continuum of care that encompasses women before pregnancy, pregnancy, childbirth, the postnatal period, and the fi rst 24 months of a child’s life.

Countries need better data to assess progress and make results-based decisions. Donors also need such

information, especially given recent unprecedented rises in global health investments. Despite recent increases in the availability of data on coverage of essential inter-ventions, important gaps remain, and many countries still use data that are 5, 10, or even 15 years old. The Countdown process reinforce recommendations that national data collection needs to be done as part of routine activities, which will require investment in strength ening country-owned health-information systems.33,34 More assessment of implementation will be needed to determine how interventions can be delivered in diff erent settings to large populations but with high coverage and high quality.

The 2008 Countdown report suggests that many of the necessary ingredients are in place to accelerate progress towards achievement of the health-related MDGs. These factors include consensus on the priority interventions and countries, commitment from countries and donors, programmes in place, data for decision making, and some increases in funding. A framework for assessment of maternal, newborn, and child survival that is relevant to countries and accepted by all new global initiatives might help to cement agreement on priorities and processes, especially for results-based strengthening of health systems. In the 7 years until 2015, the next 2 years before the next Countdown report will be the most crucial. With strategic decisions and investments, and a focus on partnerships for results, we have the opportunity to see unprecedented progress in these 68 countries. Or, will the 2010 report show more of the same gaps and lives lost?Countdown Coverage Writing Group (alphabetical order) for the Countdown Core GroupJennifer Bryce (Johns Hopkins School of Public Health, USA); Bernadette Daelmans (WHO, Switzerland); Archana Dwivedi (UNICEF, New York, USA); Vincent Fauveau (UNFPA, Geneva, Switzerland); Joy E Lawn (Save the Children-US, Cape Town, South Africa); Elizabeth Mason (WHO, Switzerland); Holly Newby (UNICEF, New York, USA); Jennifer Requejo (PMNCH); Peter Salama (UNICEF, New York, USA); Anuraj Shankar (WHO, Switzerland); Ann Starrs (Family Care International, New York, USA); Tessa Wardlaw (UNICEF, New York, USA).

ContributorsAll authors designed the study in collaboration with the 2008 Countdown Working Group. JB, JL, ASt, ASh, and PS drafted the manuscript, with review and revisions by other named authors. JR, JB, HN, AD, TW, JL, and ASh analysed the data. BD, VF, and EM reviewed and commented on drafts of the manuscript. All authors participated in the interpretation of the results and have seen and approved the fi nal version.

Confl ict of interest statement AD, HN, PS, and TW are employees of UNICEF; BD, EM, and ASh are employees of WHO; and VF is an employee of UNFPA. JB has worked as a consultant to both UNICEF and WHO on this and related work in child survival; JR worked as a consultant for the Secretariat of the Partnership for Maternal, Newborn and Child Health in the preparation of this manuscript. Other authors declare that they have no confl ict of interest.

AcknowledgmentsThe Countdown eff ort was funded by many partners, including the Bill & Melinda Gates Foundation, DFID, NORAD, PMNCH, Save the Children US and UK, UNICEF, UNFPA, and WHO. We thank all members of the Countdown Core Group, who are listed in the

Articles

1258 www.thelancet.com Vol 371 April 12, 2008

Countdown report—specifi cally Irene Deineko for administrative support; Christa Fischer Walker for preparation of the maps; Kate Kerber for analysis to provide the postnatal care data; and Monir Islam for helpful comments on the manuscript. We also thank the DevInfo intitiative for development of the database and production of the country profi les.

References1 Victora CG, Bryce J, Black RE. Learning from new initiatives in

maternal and child health. Lancet 2007; 370: 1113–14.2 UNICEF. Progress for children. A world fi t for children: statistical

review. New York, USA: UNICEF, 2007. http://www.unicef.org/progressforchildren/2007n6/index_41854.htm (accessed Jan 4, 2008).

3 Bryce J, Terreri N, Victora CG, et al. Countdown to 2015: tracking intervention coverage for child survival. Lancet 2006; 368: 1067–76.

4 The Countdown Core Group. Tracking progress in child survival: the 2005 report. New York, USA: UNICEF, 2006. http://www.countdown2015mnch.org/reports (accessed March 28, 2008).

5 Kerber K, de Graft-Johnson J, Bhutta Z, Okong P, Starrs A, Lawn J. Continuum of care for maternal, newborn, and child health: from slogan to service delivery. Lancet; 370: 1358–69.

6 Bustreo F, Johnsson AB. Parliamentarians: leading the change for maternal, newborn, and child survival? Lancet 2008; 371: 1221–22.

7 Countdown Working Group on Health Policy and Health Systems. Assessment of the health system and policy environment as a critical complement to tracking intervention coverage for maternal, newborn, and child health. Lancet 2008; 371: 1284–91.

8 Greco G, Powell–Jackson T, Borghi J, Mills A. Assessment of donor assistance to maternal, newborn, and child health between 2003 and 2006. Lancet 2008; 371: 1268–75.

9 Countdown 2008 Equity Analysis Group. Mind the gap: equity and trends in coverage of maternal, newborn, and child health services in 54 Countdown countries. Lancet 2008; 1259–67.

10 Bryce J, Requejo J, and the 2008 Countdown Working Group. Tracking progress in maternal, newborn, and child survival: the 2008 report. http://www.countdown2015mnch.org (accessed April, 2008).

11 UNICEF. The State of the World’s Children, 2008. New York, USA: UNICEF, 2008.

12 UNICEF, WHO, World Bank, UNPD. Levels and trends of child mortality in 2006: Estimates developed by the Inter-Agency Group for Child Mortality Estimation [Draft]. New York, USA: UNICEF 2007. http://www.childinfo.org/ (accessed February 8, 2008).

13 WHO. World Health Statistics 2007. Geneva, Switzerland: World Health Organization, 2007.

14 Bryce J, Boschi-Pinto C, Shibuya K, Black RE; WHO Child Health Epidemiology Reference Group. WHO estimates of the causes of death in children. Lancet 2005; 365: 1147–52.

15 Lawn JE, Wilczynska-Ketende K, Cousens SN. Estimating the causes of 4 million neonatal deaths in the year 2000. Int J Epidemiol 2006; 35: 706–18.

16 WHO. WHO Child Growth Standards: Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age: methods and development. Geneva, Switzerland: World Health Organization, 2006. http://www.who.int/childgrowth/publications/technical_report_pub/en/index.html (accessed Feb 19, 2008).

17 Blanc A, Wardlaw T. Monitoring Low birthweight: an evaluation of international estimates and updated estimation procedure. Bull World Health Organ 2005; 83: 178–85.

18 Hill, K, Thomas K, AbouZahr C, Walker N, Say L, Suzuki E; Maternal Mortality Working Group. Estimates of maternal mortality worldwide between 1990 and 2005: an assessment of available data. Lancet 2007; 370: 1311–19.

19 WHO, UNICEF, UNFPA, World Bank. Maternal mortality in 2005: estimates developed by WHO, UNICEF, UNFPA and the World Bank. Geneva, Switzerland: World Health Organization, 2007. http://www.who.int/whosis/mme_2005.pdf (accessed March 16, 2008).

20 Khan KS, Wojdyla D, Say L, Gulmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet 2006; 367: 1066–74.

21 UNICEF. Multiple indicator cluster surveys. New York, USA: UNICEF. http://childinfo.org/ (accessed Feb 8, 2008).

22 Measure DHS, MACRO International. Demographic and health surveys. Calverton, MD, USA: Measure DHS, MACRO International. http://www.measuredhs.com/aboutsurveys/dhs/start.cfm (accessed Feb 8, 2008).

23 UNICEF. Childinfo. New York, USA: UNICEF.www.childinfo.com. (accessed March 19, 2008).

24 WHO. International Travel and Health, 2007. Geneva, Switzerland: World Health Organization, 2007. http://www.who.int/malaria (accessed March 15, 2008).

25 Lawn JE, Kerber K, eds. Opportunities for Africa’s newborns: practical data, policy and programmatic support for newborn care in Africa. Cape Town, South Africa: PMNCH, Save the Children, UNFPA, UNICEF, USAID, WHO, 2006.

26 Black RE, Allen LH, Bhutta ZA, et al; Maternal and Child Undernutrition Study Group. Maternal and child undernutrition: global and regional exposures and health consequences. Lancet 2008; 371: 243–60.

27 Stanton C, Lawn JE, Rahman H, Wilczynska-Ketende K, Hill K. Stillbirth rates: delivering estimates in 190 countries. Lancet 2006; 367: 1487–94.

28 Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: when? Where? Why? Lancet 2005; 365: 891–900.

29 UNAIDS. Report on the global AIDS epidemic. Geneva, Switzerland: Joint United Nations Programme on HIV/AIDS, 2006.

30 McIntyre J. Maternal Health and HIV. Reprod Health Matters 2004; 13: 129–35.

31 Eyakuze C, Jones D, Starrs A, Sorkin N. From PMTCT to a more comprehensive AIDS response for women: a much-needed shift. Dev World Bioeth 2008; 8: 33–42.

32 John Hopkins, Indian Institute of Health Management Research and Afghanistan Ministry of Public Health. Afghanistan health survey 2006: estimates of priority health indicators for rural Afghanistan.

33 Setel PW, Macfarlane SB, Szreter S, et al. A scandal of invisibility: making everyone count by counting everyone. Lancet 2007; 370: 1569–77.

34 AbouZahr C, Boerma T. Health information systems: the foundations of public health. Bull World Health Organ 2005; 83: 578–83.