Web Appendix: Additional description of methods for estimating malaria deaths
prevented 2001-2010 using the Lives Saved Tool (LiST) model
1. African countries included in this analysis
Thirty-six countries in malaria-endemic Africa were included in the analysis of the impact of vector
control on child mortality (figure 1). Gabon was excluded due to lack of available data. Cape Verde,
Comoros, Botswana, Djibouti, Namibia, Sao Tome and Principe, South Africa, Lesotho and Swaziland
were excluded due to the low number of malaria deaths in these countries [1]. The 36 countries
included in the analysis represent 98.5% of the population at risk of malaria in sub-Saharan Africa (SSA)
or 99.7% of the malaria-caused mortalities in 2000 in SSA [1, 2]. Thirty-one malaria endemic countries
were included in the analysis of the impact of malaria prevention in pregnancy on malaria-caused child
mortality. Cape Verde, Comoros, Eritrea, Burundi, Ethiopia, Mauritania, Botswana, Djibouti, Namibia,
Rwanda, Sao Tome and Principe, South Africa, Lesotho, and Swaziland were excluded because malaria
prevention in pregnancy has little to no effect in countries with low transmission, thus there is no official
policy for malaria prevention in pregnancy with intermittent preventive treatment (IPTp) in these
countries [3, 4]. The 31 countries included in the analysis account for 88.3% of the population in SSA at
risk of malaria and 90.0% of the malaria-caused child deaths in 2000 in SSA [1, 2].
2. LiST model overview
The LiST model used in this analysis (version 4.22) and accompanying documentation can be
downloaded from www.jhsph.edu/dept/ih/IIP/list/. LiST is a computer projection model used to
estimate the number of deaths that can be prevented as a result of scaling up effective child health
interventions. A complete description of the uses of LiST and background on its creation, including
expert technical inputs, are described in detail elsewhere [5]. LiST is programmed as a module in the
demographic projection model SPECTRUM, as described elsewhere [6]. LiST uses a simple cohort model
that follows children through five age bands from birth to five years to estimate the number of neonatal
and child deaths that could be prevented by different intervention scale-up scenarios.
The model can be used to make future projections of deaths prevented from intervention scale-up,
compared to a baseline of the current year, or can be used retrospectively to estimate the number of
deaths that were prevented in the past from intervention scale-up, compared to a historical baseline
year. The model estimates child deaths prevented (within specific cause of death categories) due to
1
intervention scale-up within a specified country as a function of three primary parameters: 1) the
number of child deaths by cause projected to occur in each year (including population growth
parameters over time); 2) the protective effect (PE) on cause-specific mortality (PE = 1-relative risk*100)
for each intervention being scaled-up; and 3) increases in population coverage of each intervention.
After accounting for population growth, the model computes the number of deaths prevented by cause
each year as the difference between the estimated deaths that occur with intervention scale-up and the
estimated deaths that would have occurred had no scale-up occurred beyond the level at a baseline
year. The following basic equation is used within the model to estimate the number of child malaria
deaths 1-59 months that were prevented from increases in vector control coverage (insecticide-treated
mosquito nets [ITNs] and indoor-residual spraying [IRS]), where cause i is malaria and intervention j is
vector control:
Deaths Avertedijs = Deaths Averted Totals * (%RedMorttij / %RedMortsTotal)
%RedMortijs
=
[Iij * (Pjt – Pj0) ]/ (1 - Iij* Pj0),
%RedMortijs
=
% reduction in mortality from cause i by scale-up of intervention j
Iij
=
effectiveness of intervention j in reducing mortality from cause i
P j0
=
baseline coverage of the intervention j
P js
=
scale-up coverage for the intervention j
where
While low birth weight (LBW) is due to either intrauterine growth retardation (IUGR) or preterm
delivery, the effect of malaria prevention interventions during pregnancy (either through ITNs or IPTp)
on LBW in the LiST model acts solely through IUGR. The effect of reducing IUGR has two effects in the
LiST model for estimating <5 child deaths prevented, as noted elsewhere [7, 8]. First, children with IUGR
have a greater relative risk (RR) of dying during the neonatal period, with increased RR of dying due to
diarrhea [RR = 2.0], sepsis/pneumonia (RR =2.0), and asphyxia (RR = 2.3). Second, IUGR increases the
chance that the child will be stunted, which in turn increases the RR for measles, malaria, diarrhea and
pneumonia deaths in the post-neonatal period. In this analysis, the effect of malaria prevention in
pregnancy acted only on deaths from the first 2 pregnancies of women in each country. The following
basic equation is used within the model to estimate the number of child malaria deaths 1-59 months
that can be prevented from malaria prevention in pregnancy, where cause i is deaths from IUGR and
intervention j is composite indicator for ITNs and IPTp in the first 2 pregnancies:
2
%RedIUGRj
=
Ij (Pjt – Pj 0) /(1 – IjPj 0), where
%RedIUGRj
=
percent of reduction in IUGR due to intervention scale-up of j
Ij
=
proportion by which intervention j reduces IUGR
P j0
=
baseline coverage of the intervention j
P js
=
scaled-up coverage for the intervention j
3. Within country estimates of cause-specific child deaths
Within the LiST model, the number of disease/condition-specific deaths among children under 5 years
old in each country is based on country-specific estimates of cause-specific mortality for all low- and
middle-income countries, whereby a disease/condition-specific mortality profile is applied to the
estimated total number of deaths among children <5 estimated for each country each year. These
estimates were developed for the period 2000-2003 for WHO and UNICEF by CHERG [9] and then
reviewed by national programs before being adopted as the official UN estimates of cause-specific
mortality for children <5. The number of malaria deaths in children 1-59 months in our baseline year of
2000 was estimated as the proportion of the 1-59 month all-cause mortality envelope attributable to
malaria by the CHERG for each sub-Saharan African country [10].
4. Estimates of intervention effectiveness
As described in detail elsewhere, the PE of vector control for preventing malaria deaths in children 1-59
months has been estimated to be 55% (range 49% - 60%) based on a systematic review of related trials
and studies [11]. The PE of malaria prevention during pregnancy for preventing low-birth weight has
been estimated to be 35% (95% confidence interval [CI] 23-45%) during the first 2 pregnancies in malaria
endemic areas in the presence of SP resistance based on a systematic review of related trials [11].
5. Methods for obtaining survey point estimates for malaria prevention intervention coverage 20002010
The majority of malaria deaths occur in rural areas [12]. It has also been shown that intervention
coverage in rural areas lags behind urban areas in many countries in Africa, especially prior to 2005[13,
14]. For these reasons and to be most conservative, the level of malaria prevention intervention
coverage in rural survey strata were used for the national-level estimates for each country for
estimating the malaria deaths prevented from intervention scale-up at the national level.
3
5.1. Estimating yearly vector control intervention coverage (ITNs/IRS) 2000-2010
As outlined elsewhere [11], estimates of the proportion of households protected by vector control in the
Lives Saved Tool (LiST) is defined as a household owning either ≥1 ITN or a long lasting ITN (LLIN;
henceforth referred to as ITN), or having received IRS in the past 12 months. Vector control coverage
estimates from 2000 through 2010 were used in the LiST analysis to estimate the malaria deaths
prevented 2001-2010, compared to a baseline of 2000. These estimates were obtained from reports of
national household surveys that included the Demographic and Health Survey (DHS), the Multiple
Indicator Cluster Survey (MICS), the Malaria Indicator Survey (MIS), and AIDS Indicator Survey (AIS). If
the survey was conducted across 2 calendar years (i.e. 2005-2006), the estimate was applied to the
earlier year.
Empirically-estimated standard errors adjusted for correlated data at the cluster level were ascertained
from all available survey datasets to obtain 95% CIs about survey estimates. For those surveys without
available datasets and where standard errors were not reported (33 of 74), standard errors were
imputed using the point estimate, sample size and assuming a design effect of 2 (Equation 1).
standard error = 2 * square root [1 / (point estimate * 1 – point estimate)]
Formula 1
Most DHS and MICS from 2000-2001 did not ascertain information on the number of ITNs per
household, only on the proportion of children <5 sleeping under an ITN the previous night. To estimate
household possession of an ITN from such surveys without household possession information, we used
the empirical relationship (ratio) between ITN household possession and child ITN use reported in
surveys within the same country at a later time point. Empirically-estimated standard errors for these
estimates were then imputed using Model 1. This method of household ITN estimation was used for the
following countries 2000-2001: Angola, Benin, Burundi, Cameroon, Central African Republic, Cote
d’Ivoire, Democratic Republic of the Congo, Gambia, Guinea-Bissau, Kenya, Malawi, Niger, Rwanda,
Senegal, Sierra Leone, Tanzania, Togo, Uganda and Zambia (Table 1).
The bulk of the scale-up of ITNs/IRS occurred after 2004 across Africa [15]. To avoid using a linear
interpolation between survey data points, for the period of 2004-2008ITN procurement data from
manufactures for each country were used to inform ITN coverage estimate changes between survey
4
estimates as follows. Predicted estimates of the proportion of households with ≥1 ITN within each
country 2004-2008 were ascertained from a basic linear regression model. The model quantified the
observed relationship between net delivery data from manufacturers, aggregated yearly, and ITN
household coverage estimates from household surveys. Data on the delivery of ITNs and LLINs to
countries from manufacturers 2004-2008 were obtained from the Net Mapping Project (John Millner
USAID). Under this project, yearly data on the number of ITNs and LLINs purchased and shipped to
countries were obtained from Sumitomo/A-Z, Vestergaard-Frandsen, Clarke, BASF, Intection/BestNet
and Tana Netting. Estimates of national ITN household possession from nationally-representative
household surveys 2004-2008 were obtained as outlined above. ITNs reported delivered by
manufacturers older than 3 years (i.e. those nets delivered to countries more than 3 years ago) were
subtracted from the cumulative number of nets each year. To account for the lag between ITN
manufacturer delivery to the country and availability to the household, yearly cumulative ITN delivery
estimates were lagged by one year. The 3-year lagged cumulative ITN delivery estimates were then
standardized across countries by the number of households within the country, with the final variable
henceforth referred to as yearly corrected ITN delivery estimates. There were 47 data points across
countries between 2004 and 2008 with both a survey-derived estimate of ITN household possession and
yearly corrected ITN delivery estimates. In addition to the yearly corrected ITN delivery estimates, the
following covariates were included in the regression model and tested for model fit: year, proportion of
households within each country at risk for malaria, as defined by MARA http://www.mara.org.za/, the
gross national income for 2006 (GNI), and a dichotomized variable (0,1) indicating if the country was a
Presidents Malaria Initiative (PMI) country in a particular year. The fit of models with the squaring,
cubing, log, exponent and quadratic of yearly corrected ITN delivery estimates was assessed using an Ftest. The following model had the best fit (R2 = 0.7099, F-test = 20.07, df = 46, P-value < 0.0001):
Proportion of households with ≥1 ITN = -8612.68154 + 18.74738*(corrected net delivery) +
-0.00095456*(proportion of households at risk for malaria) + 4.30381*(year) + -0.00466*(GNI) +
9.40987*(PMI)
Model 1
The predicted estimates from Model 2 were not actually used in the LiST analysis reported here in place
of survey estimates (Table 2). Rather, the percent changes year to year (slope of the line) from the
predicted results of Model 2 were used to inform changes in household ITN possession between
household survey estimates 2004-2008. In the case that this percent change showed a decline in
5
coverage, ITN coverage was held constant until it began to rise again. An actual decline in ITN coverage
is unlikely due to the large increase in funding for malaria interventions throughout this time period.
This occurred in 4 instances: Cote d’Ivoire, Nigeria, Togo and Zambia.
Coverage from years 2000-2003 was estimated using linear interpolation from the earliest survey point
estimate to the 2004 estimate obtained using the above method. For those countries without a survey
estimate in 2000, the 2000 estimate was set to the survey estimate in 2001 with its respective 95%
confidence intervals. If the country did not have a 2001 survey, the estimate in 2000 was set to 1.9, the
mean estimate across all 2000-2001 survey estimates with the exception of Gambia 2000 (Gambia had
very high net coverage in 2001 [20.2%] due to the ITN trials that occurred, and was therefore excluded
from the estimation of the mean ITN coverage in 2000). The standard error from the earliest survey was
then applied to the midpoint estimate 2000-2003 to obtain uncertainty intervals. A survey in 2003 in
Mauritania estimated household ITN possession to be 0.6, and so this value with its respective 95%
confidence interval was used for 2000-2002. This process yields the most conservative increase in ITN
coverage. Although perhaps overestimating the proportion of rural households owning an ITN in 2000,
it will yield the most conservative impact throughout the decade.
Coverage for years 2009-2010 were projected by applying the linear slope between the most recent
survey and the earliest survey. This approach yielded the most conservative slope for this period while
maintaining the likely increase in coverage due to increasing funding.
The 2004 DHS conducted in Chad did not estimate household ITN possession, but did however show the
proportion of households owning a mosquito net to be 61% among rural households. The ratio of the
proportion of households owning a mosquito net to the proportion of households owning ≥ 1 ITN was
ascertained for Burkina Faso and then applied to Chad. The standard error was then imputed using
model 1 to achieve 95% CIs. Burkina Faso was used because of the similarity in climate, mosquito net
culture, and proximity of survey date. Sudan’s household survey in 2006 did not specify between urban
and rural populations, thus the overall proportion of households owning ≥1 ITN was used. Eritrea’s
estimate of ITN coverage in 2003 was taken from the published results of a sub-national survey [16] that
included 3 of the 4 zobas surveyed in the 2008 MIS. Information was unavailable for mainland
Equatorial Guinea. However, considering that Bioko Island is 28.6% of the population of Equatorial
Guinea we applied the coverage reported from malaria prevention interventions on the island [17] and
6
multiplied the results by 28.6%. The national survey in Niger 2010 did not report a rural estimate. To
estimate the rural estimate from the total estimate, the ratio of total ITN coverage to rural ITN coverage
in the 2009 Niger survey was applied to the 2010 estimate.
Resultant ITN/IRS coverage changes for each country 2000-2010, with uncertainty, are presented in
Figure 2.
5.2. Estimating yearly intervention coverage for malaria prevention in pregnancy (IPTp/ITN) 2000-2010
As outlined elsewhere,[11] protection by malaria prevention in pregnancy, defined as the higher of the
estimates of either proportion of pregnant women using an ITN the previous night or the proportion of
women who had a live birth in the past 2 years who received 2+ doses of sulfadoxamine-pyremethamine
(SP) during an ante-natal care (ANC) visit (IPTp), were used in the LiST analysis. These estimates were
obtained from the final reports of DHS, MICS, MIS and Office of National Statistics (ONS) surveys (Table
3). If the survey did not specify the number of doses of SP or where it was received, the estimate of IPTp
taken during pregnancy without specification of number of doses or if taken at ANC was used. This
occurred in the following surveys: Congo 2005, Liberia 2005, and Malawi 2000.
Standard errors were ascertained from survey datasets to obtain 95% CIs about survey point estimates.
In the case that standard errors about the estimate were unavailable, the standard errors were imputed
using formula 1.
The year prior to the country declaring IPTp with SP as national policy was set to 0%, for those countries
using IPTp as the coverage indicator[4, 18, 19]. It was unclear what year Liberia adopted the policy, and
so the year 2000 was set to 0 in this case. Where the higher indicator for malaria prevention in
pregnancy was ITN use by pregnant women, the coverage of ITNs among pregnant women was set to
0% in 2000. While this may lead to a slight overestimate of the impact of malaria prevention in
pregnancy the overall contribution of malaria prevention in pregnancy to the total number of lives saved
is quite small.
Linear interpolation was used between the first year of measured coverage and the next available
survey point estimate. Linear interpolation was also used between multiple surveys where available.
The slope between the most recent household survey and the earliest household survey was used to
7
inform the increase for years beyond the most recent household survey through 2010. In the case that
there was only 1 household survey in the country, the year set to 0 was used in place of the earliest
household survey. Coverage was assumed to never decrease unless surveys estimated otherwise.
Resultant IPTp/ITN coverage changes for each country 2000-2010, with uncertainty, are presented in
Figure 3.
6. Uncertainty estimation of LiST estimates
Uncertainty bounds about total estimated malaria deaths prevented from vector control 2001-2010
were estimated by projecting lives saved when varying the three primary model parameters: 1)
estimated malaria deaths within each country; 2) the estimated PE of vector control on malaria
mortality; and 3) intervention coverage changes 2000-2010. Using this approach, lower and upper
uncertainty bounds were estimated using a high impact / low impact scenario. The uncertainty about
the number of malaria deaths among children 1-59 months was derived from the 95% confidence
intervals about the proportion of all deaths due to malaria in this age group in 2000 estimated by the
CHERG [1]. The reported range of 49-60% about the 55% PE of vector control for preventing malaria
mortality was used as the uncertainty about this parameter in this analysis [11]. The uncertainty about
ITN scale-up is dependent on the percent change, or the slope of the vector control coverage curve,
from 2000-2010. Under the largest-increase scenario, the percent change in coverage was set to that
from the lower bound of the 95% CI in 2000 to the upper bound in 2010, resulting in the greatest slope
during this period. Under the smallest-increase scenario, the percent change in coverage was set to that
from the upper bound of the 95% CI in 2000 to the lower bound in 2010, resulting in the least slope
during this period. The slope of each uncertainty bound between 2000-2010 was then calculated in the
same manner as the slope of the midpoint over this period (see Figure 1 for resultant uncertainty of
vector control scale-up 2000-2010).
Uncertainty bounds about total estimated child deaths prevented from malaria prevention during
pregnancy 2001-2010 were estimated by projecting lives saved while two primary model parameters: 1)
the estimated PE of malaria prevention in pregnancy on preventing LBW; and 2) intervention coverage.
The 95% CI of 23-45% about the 35% PE of malaria prevention in pregnancy for preventing LBW was
used as the uncertainty about this parameter in this analysis [11]. The uncertainty about IPTp/ITN scaleup under the largest-increase / smallest-increase scenario followed the methodology for vector control
8
outlined above. Linear interpolation was used between bounds (see Figure 2 for resultant uncertainty
of malaria prevention in pregnancy scale-up 2000-2010).
7. Estimating total continental coverage
Yearly coverage estimates for each country were weighted according to each country’s 2005 population
estimate (mid-point of 2000-2010) to estimate the continental coverage rates of vector control in rural
areas and malaria in pregnancy prevention in rural areas[20].
8. Coverage estimation of other interventions in this analysis
Using the LiST model, this analysis accounted for the scale-up of other interventions that affect child
mortality such as vaccinations and antenatal care. Coverage estimates for these interventions were
derived from nationally representative household surveys. Coverage for the years after the survey was
held constant. Other LiST inputs not included in the survey or not included in this analysis, one being
case management of malaria, were held constant during this time period. As LiST is driven by a change
in coverage, these indicators were then assumed to have no effect on child mortality throughout the
time period. This method may affect the number of deaths occurring each year, but would have only a
minimal effect on the number of deaths prevented due to malaria prevention interventions.
9
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Rowe AK, Rowe SY, Snow R, Korenromp EL, Armstrong-Schellenberg JR, Stein C, Nahlen BL, Bryce J, Black RE, Steketee
R: Estimates of the burden of mortality directly attributable to malaria for children under 5 years of age in Africa
for the year 2000 Final report. For the Child Health Epidemiology Reference Group (CHERG). In. Washington DC:
Child Health Epidemiology Reference Group (CHERG); 2006.
Hay SI, Guerra CA, Gething PW, Patil AP, Tatem AJ, Noor AM, Kabaria CW, Manh BH, Elyazar IR, Brooker S et al: A
world malaria map: Plasmodium falciparum endemicity in 2007. PLoS Med 2009, 6(3):e1000048.
Steketee RW, Nahlen BL, Parise ME, Menendez C: The burden of malaria in pregnancy in malaria-endemic areas. The
American journal of tropical medicine and hygiene 2001, 64(1-2 Suppl):28-35.
WHO: World Malaria Report 2009. In. Edited by WHO. Geneva: WHO; 2009.
Boschi-Pinto C, Young M, Black RE: The Child Health Epidemiology Reference Group reviews of the effectiveness of
interventions to reduce maternal, neonatal and child mortality. Int J Epidemiol 2010, 39 Suppl 1:i3-6.
Stover J, McKinnon R, Winfrey B: Spectrum: a model platform for linking maternal and child survival interventions
with AIDS, family planning and demographic projections. Int J Epidemiol 2010, 39 Suppl 1:i7-10.
Bhutta ZA, Ahmed T, Black RE, Cousens S, Dewey K, Giugliani E, Haider BA, Kirkwood B, Morris SS, Sachdev HP et al:
What works? Interventions for maternal and child undernutrition and survival. Lancet 2008, 371(9610):417-440.
Friberg IK, Bhutta ZA, Darmstadt GL, Bang A, Cousens S, Baqui AH, Kumar V, Walker N, Lawn JE: Comparing modelled
predictions of neonatal mortality impacts using LiST with observed results of community-based intervention trials
in South Asia. Int J Epidemiol 2010, 39 Suppl 1:i11-20.
Bryce J, Boschi-Pinto C, Shibuya K, Black RE: WHO estimates of the causes of death in children. Lancet 2005,
365(9465):1147-1152.
Rowe AK, Rowe S, Snow RW, Korenromp E, Armstrong Schellenberg J, Stein C, Nahlen BL, Bryce J, Black RE, Steketee
RW: The burden of malaria mortality among African children in the year 2000. International Journal of Epidemiology
2006, 35(3):691-704.
Eisele TP, Larsen D, Steketee RW: Protective efficacy of interventions for preventing malaria mortality in children in
Plasmodium falciparum endemic areas. Int J Epidemiol 2010, 39(suppl_1):i88-101.
Hay SI, Guerra CA, Tatem AJ, Atkinson PM, Snow RW: Urbanization, malaria transmission and disease burden in
Africa. Nature reviews 2005, 3(1):81-90.
Steketee RW, Eisele TP: Is the scale up of malaria intervention coverage also achieving equity? PLoS ONE 2009,
4(12):e8409.
Korenromp EL, Miller J, Cibulskis RE, Kabir Cham M, Alnwick D, Dye C: Monitoring mosquito net coverage for malaria
control in Africa: possession vs. use by children under 5 years. Trop Med Int Health 2003, 8(8):693-703.
Unicef: Malaria & children: Progress in intervention coverage. In. New York: Unicef and RBM; 2007.
Macintyre K, Keating J, Okbaldt YB, Zerom M, Sosler S, Ghebremeskel T, Eisele TP: Rolling out insecticide treated nets
in Eritrea: examining the determinants of possession and use in malarious zones during the rainy season. Trop Med
Int Health 2006, 11(6):824-833.
Kleinschmidt I, Schwabe C, Benavente L, Torrez M, Ridl F, Segura J, Ehmer P, Nchama G: Marked increase in child
survival after four years of intensive malaria control. Am J Trop Med Hyg 2009, 80(6):882-888.
WHO: World Malaria Report 2005. In.: World Health Organization; 2005.
WHO: World Malaria Report 2008. In. Edited by WHO. Geneva: WHO; 2008.
U.N.: The World Population Prospect: The 2008 Revision. In. New York: UN Population Division; 2008.
10
Figure legend
Figure 1: Countries included in the analysis of the impact of malaria prevention intervention scale-up.
Figure 2: Country specific estimates of vector control (ITNs/IRS) coverage in rural areas from 20002010. The blue line is the predicted vector control scale-up. The red line is the scenario of the lowest
coverage increase used for lower bound estimate for uncertainty; the green line is the scenario of the
highest coverage increase used for the upper bound estimate for uncertainty. Circles represent MICS
surveys, diamonds represent DHS surveys, triangles represent MIS surveys, and squares represent other
types of household surveys.
Figure 3: Country specific estimates of malaria prevention in pregnancy (IPTp/ITNs) coverage in rural
areas from 2000-2010. The blue line is the predicted malaria in pregnancy prevention intervention
scale-up. The red line is the scenario of the lowest coverage increase used for lower bound estimate for
uncertainty; the green line is the scenario of the highest coverage increase used for the upper bound
estimate for uncertainty. Circles represent MICS surveys, diamonds represent DHS surveys, triangles
represent MIS surveys, and squares represent other types of household surveys.
11
12
Figure 2: Total estimated vector control coverage by year for 36 countries in sub-Saharan Africa, weighted by population size
100
90
80
70
60
50
40
30
20
10
0
Lower bound
Estimated ITN coverage
20002001200220032004200520062007200820092010
Upper bound
13
Angola
Benin
100
100
80
80
60
60
40
40
27.5
20
1.4
20
0
Burundi
100
100
80
80
60
60
40
50.4
40
15.2
20
3.2
0
20
1.2
8.5
0
Cameroon
Central African Republic
100
100
80
80
60
60
40
40
0
4.7
0
Burkina Faso
20
21.4
0.9
1.0
3.5
20
11.2
1.4
0
14
Chad
Congo (Brazzaville)
100
100
80
80
60
60
36.5
40
20
40
20
4.7
0
0
Cote d'Ivoire
Democratic Republic of the Congo
100
100
80
80
60
60
40
40
20
0
8.0
2.1
2.2
6.3
48.3
20
0
0.2
7.1
15
Equatorial Guinea
100
Eritrea
100
87.0
94.0
80
77.0
60
40
20
20
0
0
Gambia
100
100
80
80
56.2
60
40
20
20
3.1
0.0
21.9
Guinea
100
100
80
80
60
60
37.5
40
0
61.6
0
Ghana
20
94.1
60
40
0
66.8
60
40
Ethiopia
70.7
80
25.8
4.8
40
20
9.3
2.7
0
16
Guinea-Bissau
Kenya
100
100
80
80
54.1
60
53.0
0
20
2.9
3.3
Madagascar
100
100
80
80
60
60
51.8
40
20
6.7
0
0
Malawi
Mali
100
100
80
80
60.2
60
37.5
40
24.8
60
76.5
48.4
40
20
3.1
0
0
Mauritania
Mozambique
100
100
80
80
60
60
40
40
0
56.4
40
20
20
5.8
0
Liberia
20
47.7
40
40
20
60.7
60
0.6
8.5
28.8
20
0
17
Niger
Nigeria
100
80.5
80
60
100
80
60
44.2
40
20
78.4
45.0
40
20
3.6
0
7.6
3.1
0
Rwanda
Senegal
100
100
80
80
60
69.9
60
53.8
40
40
20
20
38.4
22.2
1.8
11.8
0
0
Sierra Leone
Somalia
100
100
80
80
60
60
36.7
40
20
0
4.8
0.9
6.5
40
20
11.0
0
18
Sudan
Tanzania
100
100
80
80
60
60
50.6
40
18.4
0
Togo
0.1
Uganda
100
100
80
80
60
60
42.5
40
40
20
20
1.5
47.7
14.0
0
Zambia
Zimbabwe
100
100
80
66.3
53.7
60
37.9
40
0
13.8
20
0
20
32.6
40
20
0
63.4
10.9
3.3
72.9
80
60
33.5
40
20
7.2
0
19
Figure 3: Total estimated malaria prevention in pregnancy coverage by year for 31 countries in sub-Saharan Africa, weighted
by population size
100
90
80
70
60
50
40
30
20
10
0
Smallest IPTp coverage change
Estimated IPTp coverage
Largest IPTp coverage change
20002001200220032004200520062007200820092010
20
Angola
Benin
100
100
80
80
60
60
40
26.4
40
20
20
0
0
Burkina Faso
Cameroon
100
100
80
80
60
60
40
40
20
0.3
20
0
8.9
0
Central African Republic
Chad
100
100
80
80
60
60
40
40
20
20
0
2.3
4.7
17.7
0
21
Congo (Brazzaville)
Cote d’Ivoire
100
100
80
80
60
60
40
40
20
20
3.2
0
0
Democratic Republic of the Congo
Equatorial Guinea
100
100
80
80
60
42.5
60
40
40
20
20
4.4
0
19.0
0
Gambia
Ghana
100
100
80
67.2
80
60
60
40
7.2
33.8
40
20
20
0
0
26.1
0.9
22
Guinea
Guinea-Bissau
100
100
80
80
60
60
40
40
20
1.2
0
20
3.3
Liberia
100
100
80
80
48.3
60
60
44.8
40
40
12.8
4.4
0
20
4.3
0
Madagascar
Malawi
100
100
80
80
60
7.1
0
Kenya
20
29.3
45.6
60
40
40 28.3
20
20
0
0
50.7
42.5 45.7
23
Mali
Mozambique
100
100
80
80
60
41.9
40
40
20
60
16.1
20
5.6
0
0
Niger
Nigeria
100
71.3
80
100
80
60
60
40
40
20
20
4.9
0
0
40.8
0.2
3.7
Senegal
100
80
60
46.5
54.2
40
20
]
10.0
0
24
Sierra Leone
Somalia
100
100
80
80
60
60
40
20
29.3
40
20
1.1
0.7
0
0
Sudan
Tanzania
100
100
80
80
60
60
31.6
40
59.2
27.9
40
21.7
20
20
0
0
Togo
Uganda
100
100
80
80
60
60
40
40
20
0
18.5
20
43.5
16.2
0
25
Zambia
Zimbabwe
100
100
80
60
59.1 58.1
52.5
65.2
80
60
40
40
20
20
0
0
7.4
26
Table 1: Vector control coverage estimates from household surveys
27
Country
Year
Angola
Angola
Benin
Benin
Burkina Faso
Burkina Faso
Burundi
Burundi
Burundi
Cameroon
Cameroon
Cameroon
CAR
CAR
Chad
Chad
Congo (Brazzaville)
Cote d’Ivoire
Cote d’Ivoire
Cote d’Ivoire
DRC
DRC
DRC
Equatorial Guinea4
Equatorial Guinea4
Equatorial Guinea4
Eritrea
Eritrea
Ethiopia
Ethiopia
Ethiopia
Gambia
Gambia
Gambia5
Ghana
Ghana
Ghana
Guinea
Guinea
Guinea-Bissau
Guinea-Bissau
Guinea-Bissau
Kenya
Kenya
Kenya5
Kenya
Liberia
Liberia
Madagascar
Malawi
Malawi
Malawi
Malawi
Mali
Mali
Mauritania
Mauritania
Mozambique5
Niger
Niger
2001
2006
2001
2006
2003
2006
2001
2005
2010
2000
2004
2006
2000
2006
2004
2010
2005
2000
2005
2006
2001
2007
2010
2005
2006
2008
2003
2008
2000
2005
2007
2000
2006
2010
2003
2006
2008
2005
2007
2000
2006
2010
2000
2003
2007
2009
2005
2009
2008
2000
2004
2006
2010
2006
2010
2003
2007
2007
2000
2006
Rural HH
estimate
1.41
25.9
4.71
21.4
3.2
15.2
1.21
8.5
50.4
0.9 1
1.0
3.5
1.41
11.2
20.73
36.5
2.1
2.11
2.2
6.3
0.21
7.1
48.3
87.0
77.0
94.0
66.8
70.7
0.0
3.1
56.2
21.91
61.6
94.1
4.8
25.8
37.5
2.7
9.3
2.91
54.1
53.0
3.31
5.8
47.7
55.0
6.7
51.8
56.4
3.11
24.8
37.5
60.2
48.4
76.5
0.6
8.5
28.8
3.61
44.2
Sample Size
1968
1500
3185
10279
6898
5010
3325
7020
7817
2071
5616
4808
8963
7628
4295
9987
2028
4919
2473
4348
6559
5188
7014
1274
2142
2478
2341
1530
14072
10055
6154
2300
3230
4335
3733
3822
6603
4562
1669
3571
3071
6059
1439
5662
5929
6147
6337
2278
13364
2528
11939
27711
1091
8859
9400
2218
6393
2740
3227
5298
Rural HH standard
error
0.5
3.5
0.72
0.8
0.4
0.6
0.42
0.3
0.42
0.42
0.2
0.3
0.22
0.72
1.52
1.08
0.3
0.42
0.5
0.5
0.12
1.0
2.0
2.0
1.5
1.0
3.0
2.32
0.52
0.6
1.32
1.72
0.9
0.9
0.6
0.8
1.2
0.3
0.7
0.62
0.9
1.1
0.92
0.6
1.42
1.8
0.62
3.2
1.0
0.72
0.8
0.4
3.02
1.5
0.92
0.32
0.4
2.12
0.72
1.5
95% confidence
interval
0.4 – 2.4
19.0 – 32.7
3.2 – 6.2
20.0 – 22.9
2.4 – 4.1
14.1 – 16.4
0.5 – 2.0
7.8 – 9.1
48.2 – 52.6
0.1 – 1.7
0.6 – 1.5
2.9 – 4.1
0.9 – 1.8
9.8 – 12.6
17.7 – 23.7
34.4 – 38.7
1.5 – 2.6
1.3 – 2.9
1.3 – 3.1
5.3 – 7.4
0.0 – 0.4
5.2 – 9.0
44.3 – 52.3
80.0 – 91.0
74.0 – 80.0
92.0 – 95.0
60.9 – 72.7
66.1 – 75.3
0 – 0.9
1.9 – 4.2
53.7 – 58.7
18.5 – 25.3
59.9 – 63.3
92.3 – 96.0
3.5 – 6.1
24.3 – 27.4
35.2 – 39.9
2.1 – 3.3
7.9 – 10.8
1.8 – 4.0
52.3 – 56.0
50.7 – 55.2
1.5 – 5.2
4.6 – 7.0
45.0 – 50.4
51.5 – 58.5
5.5 – 7.9
45.4 – 58.2
54.5 – 58.4
1.7 – 4.4
23.2 – 26.5
36.8 – 38.3
54.4 – 66.0
45.3 – 51.4
74.8 – 78.2
0.0 – 1.2
7.8 – 9.3
24.6 – 33.0
2.3 – 4.9
41.1 – 47.2
Survey
MICS
MIS
DHS
DHS
DHS
MICS
MICS
MICS
DHS
MICS
DHS
MICS
MICS
MICS
DHS
MICS
DHS
MICS
DHS
MICS
MICS
DHS
MICS
[17]
[17]
[17]
[16]
MIS
DHS
DHS
MIS
MICS
MICS
MICS
DHS
MICS
DHS
DHS
ONS
MICS
MICS
MICS
MICS
DHS
MIS
DHS
MIS
MIS
DHS
DHS
DHS
MICS
MIS
DHS
MICS
DHS
MICS
MIS
MICS
DHS
28
Niger
Niger
Nigeria
Nigeria
Nigeria
Rwanda
Rwanda
Rwanda
Senegal
Senegal
Senegal
Senegal
Sierra Leone
Sierra Leone
Sierra Leone
Somalia
Sudan
Sudan (Southern)
Tanzania
Tanzania
Tanzania
Tanzania
Togo
Togo
Uganda
Uganda
Uganda
Zambia
Zambia
Zambia5
Zambia
Zambia5
Zambia5
Zimbabwe
Zimbabwe
2009
20107
2003
2008
2010
2000
2005
2007
2000
2005
2006
2008
2000
2005
2008
2006
2006
2009
1999
2004
2007
2010
2000
2006
2000
2006
2009
2000
2001
2006
2007
2008
2010
2005
2009
80.5
78.4
3.1
7.6
45.0
1.81
11.8
53.8
4.81
22.2
38.4
69.9
0.91
6.5
36.7
11.0
18.46
50.6
0.1
13.8
32.6
63.4
1.51
42.5
0.31
14.0
47.7
3.31
10.9
37.9
53.7
66.3
72.9
7.2
28.2
10652
6450
4291
23346
4175
2410
8165
6229
6500
4296
1807
6304
1961
2375
4328
3768
24036
2316
2669
7576
6662
7414
2411
4312
6793
7480
3761
2896
5112
1786
4469
2882
2766
6229
7420
0.8
0.82
0.6
0.4
1.5
0.52
0.7
1.32
0.52
1.1
2.3
1.6
0.4
0.5
1.5
0.5
0.52
2.12
0.12
1.0
1.2
1.1
0.52
0.9
0.72
0.9
2.2
0.72
0.9
3.0
1.8
2.7
1.72
0.8
1.0
79.0 – 82.0
76.4 – 80.4
1.9 – 4.2
6.8 – 8.5
42.0 – 48.0
0.7 – 2.9
10.5 – 13.1
51.3 – 56.3
3.8 – 5.8
20.1 – 24.3
33.9 – 43.0
66.7 – 73.0
0.0 – 1.7
5.4 – 7.5
33.7 – 39.7
10.0 – 12.1
17.4 – 19.4
46.5 – 54.7
0.0 – 0.4
11.8 – 15.8
30.2 – 35.0
6.2 – 65.6
0.5 – 2.4
40.8 – 44.2
0 – 1.7
12.2 – 15.9
43.3 – 52.1
2.0 – 4.6
9.2 – 12.6
32.1 – 43.7
50.1 – 57.3
60.9 – 71.7
69.6 – 76.2
5.6 – 8.7
26.2 – 30.2
ONS
ONS
DHS
DHS
MIS
MICS
DHS
MIS
MICS
DHS
MIS
MIS
MICS
MICS
DHS
MICS
ONS
MIS
DHS
DHS
AIS
DHS
MICS
MICS
DHS
DHS
MIS
MICS
DHS
MIS
DHS
MIS
MIS
DHS
DHS
Vector control: Houses protected by either ITNs and/or IRS
CAR: Central African Republic
DRC: Democratic republic of Congo
1. These estimates are derived from the ratio of children sleeping under an ITN the night before
2. These standard errors are estimated using Formula 1
3. Estimate was derived using the ratio of mosquito nets to ITNs of Burkina Faso 2003
4. Data available were only for Bioko Island
5. Estimated with households protected by either ITNs or IRS; all others just use ITN household possession
6. Estimate is total country, not specific to rural
7. Estimated using ratio of total ITN coverage to rural ITN coverage from previous year
29
Table 2: Model-derived estimates of proportion of households with ≥1 ITN, estimated from cumulative
ITN procured model 1 (Table 1). These estimates are not used directly in the LiST analysis; they are only
used to derive the percent increase in ITN coverage between survey rounds 2004-2008.
Country
Angola
Benin
Burkina Faso
Burundi
Cameroon
CAR
Chad
Congo
Cote d’Ivoire
DRC
Ethiopia
Gambia
Ghana
Guinea
Guinea-Bissau
Kenya
Liberia
Madagascar
Malawi
Mali
Mauritania
Mozambique
Namibia
Niger
Nigeria
Rwanda
Senegal
Sierra Leone
Somalia
Sudan
Tanzania
Togo
Uganda
Zambia
Zimbabwe
2004
11
8
9
12
4
10
8
8
5
3
5
11
5
9
12
6
11
10
9
9
9
7
0
9
0
11
10
11
11
4
13
16
16
8
9
2005
16
13
14
17
8
14
13
12
9
8
10
15
10
13
18
11
16
15
23
13
15
20
1
14
0
26
26
16
16
9
18
27
20
13
14
2006
26
26
21
25
12
19
18
17
14
14
27
23
24
20
22
30
34
34
27
29
19
27
10
39
0
34
33
22
25
15
22
32
26
33
18
2007
41
33
27
40
17
26
23
25
21
20
44
30
39
30
39
54
50
43
33
43
25
33
20
44
0
55
44
56
33
22
26
38
37
43
24
2008
55
57
36
48
21
39
29
31
28
28
49
49
46
36
42
60
70
58
43
75
28
43
44
49
1
63
63
62
39
35
31
31
45
68
30
CAR: Central African Republic
DRC: Democratic republic of Congo
30
Table 3: malaria in pregnancy intervention coverage estimates from household surveys 2000-2010
Country
Year
Angola
Benin
Burkina Faso
Cameroon
CAR
Chad
Congo (Brazzaville)
Cote d’Ivoire
DRC
DRC
Equatorial Guinea4
Gambia
Ghana
Gambia
Ghana
Ghana
Guinea
Guinea
Guinea-Bissau
Guinea-Bissau
Kenya
Kenya
Liberia
Liberia
Malawi
Malawi
Malawi
Malawi
Mali
Mali
Mozambique
Niger
Niger
Nigeria
Nigeria
Nigeria
Senegal
Senegal
Senegal
Sierra Leone
Sierra Leone
Somalia
Sudan (Southern)
Tanzania
Tanzania
Tanzania
Togo
Uganda
Uganda
Zambia
Zambia
Zambia
Zambia
2006
2006
2006
2006
2006
2010
2005
2006
2007
2010
2008
2006
2003
2010
2006
2008
2005
2007
2006
2010
2003
2007
2005
2009
2000
2004
2006
2010
2006
2010
2007
2006
2010
2003
2008
2010
2005
2006
2008
2005
2008
2006
2009
2004
2007
2010
2006
2006
2009
2006
2007
2008
2010
Rural HH
estimate
26.41
2.31
0.3
8.9
4.7
17.7
3.23
7.2
4.4
42.51
19.0
33.8
0.9
67.2
26.1
43.5
1.2
3.3
7.1
29.31
4.4
12.8
4.33
44.8
28.3
42.53
45.7
50.7
5.61
41.9
16.1
4.9
71.31
0.2
3.7
38.7
10.0
46.5
54.2
1.1
29.31
0.7
31.6
18.6
27.9
59.21
18.5
16.2
43.61
52.5
59.1
58.1
65.2
Sample Size
269
1962
156
685
2510
1616
933
2258
2130
1024
503
2052
1076
2760
944
811
2122
1669
531
540
1802
1644
510
1009
4180
4246
9279
213
1896
1986
1016
1311
91989
1544
8311
586
3126
1352
4450
103
238
1426
421
1908
2550
788
1213
2956
368
1030
1830
1658
1595
Rural HH
standard error
1.6
0.2
0.2
1.3
0.62
1.2
0.92
0.7
0.8
2.6
2.6
1.1
0.3
1.2
1.6
2.5
0.3
0.92
0.7
2.5
0.6
1.02
1.02
2.9
0.9
1.0
0.7
6.92
0.5
2.2
1.32
0.3
0.72
0.1
0.3
4.0
0.8
2.3
1.4
1.02
1.3
0.2
4.52
1.0
1.5
2.5
1.3
0.9
4.0
1.9
1.8
1.7
2.4
95% confidence
interval
23.3 – 29.5
1.9 – 2.7
0.0 – 0.7
6.4 – 11.5
3.5 – 5.9
15.4 – 20.0
1.5 – 4.9
5.9 – 8.5
2.9 – 5.9
37.2 – 47.7
16.0 – 24.0
31.8 – 35.9
0.3 – 1.5
64.8 – 69.6
22.9 – 29.3
38.6 – 48.3
0.7 – 1.7
1.6 – 5.0
5.8 – 8.4
24.3 – 34.3
3.3 – 5.5
10.4 – 14.4
10.4 – 14.4
39.0 – 50.5
26.6 – 30.0
40.5 – 44.4
44.3 – 47.1
37.3 – 64.1
4.5 – 6.6
37.6 – 46.2
13.6 – 18.6
4.4 – 5.4
70.6 – 71.9
0.0 – 0.4
3.1 – 4.3
30.8 – 46.6
8.4 – 11.7
41.9 – 51.1
51.3 – 57.0
0.0 – 3.0
26.8 – 31.8
0.3 – 1.1
22.7 – 40.5
16.6 – 20.7
25.0 – 30.8
54.3 – 64.0
15.9 – 21.1
14.4 – 18.0
36.0 – 51.6
48.6 – 56.3
55.5 – 62.7
54.7 – 61.5
60.5 – 69.9
Survey
MIS
DHS
MICS
MICS
MICS
MICS
DHS
MICS
DHS
MICS
[17]
MICS
DHS
MICS
MICS
DHS
DHS
ONS
MICS
MICS
DHS
MIS
MIS
MIS
DHS
DHS
MICS
MIS
DHS
MICS
MIS
DHS
ONS
DHS
DHS
MIS
DHS
MIS
MIS
MICS
DHS
MICS
MIS
DHS
AIS
MIS
MICS
DHS
MIS
MIS
DHS
MIS
MIS
31
Zimbabwe
2005
7.4
1642
1.0
5.5 – 9.4
DHS
Malaria in pregnancy interventions: IPTp or ITNs, whichever is higher
CAR: Central African Republic
DRC: Democratic republic of Congo
1. These estimates are pregnant women sleeping under an ITN the night before the survey; all others are IPTp
received among women giving birth in the past 2 years
2. These standard errors are estimated using formula 1
3. These estimates were not specified as 2+ doses of SP received at ANC
4. Data available were only for Bioko Island
32
Table 4: Estimated malaria deaths in children 1-59 months prevented by vector control scale-up 2001-2010
Uncertainty
Country
Malaria
deaths
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Angola
13,991
3
274
565
937
1,443
2,502
4,182
5,967
6,732
7,543
Benin
10,759
6
57
110
166
423
1,220
1,625
3,162
3,502
Burkina Faso
20,089
58
118
178
628
1,163
1,885
2,601
3,767
4,284
1
38
77
118
189
294
510
21,332
4
8
12
11
171
350
537
CAR
5,488
34
70
106
142
225
318
466
712
770
Chad
18,594
75
155
242
332
634
1,012
1,417
1,903
3,272
4,863
21
43
66
90
185
290
443
566
610
Cote d’Ivoire
19,210
0
0
0
0
16
504
878
1,271
DRC
78,901
0
319
655
1,054
1,761
2,593
3,502
267
0
0
0
0
143
129
1,705
231
463
712
752
783
Ethiopia
20,658
70
143
218
294
Gambia
2,101
20
41
62
Ghana
26,873
151
305
464
Guinea
16,687
21
33
55
2,130
84
172
265
Kenya
20,761
113
225
344
Liberia
4,583
23
46
Madagascar
14,235
153
Malawi
10,545
Mali
Burundi
Cameroon
Congo (Brazzaville)
Equatorial Guinea
(Bioko Island)
Eritrea
Guinea-Bissau
Mauritania
Mozambique
Total
Lower
Bound
Upper
Bound
30,148
20,103
42,128
3,841
14,112
9,425
19,801
4,496
5,247
20,141
13,955
27,758
634
1,024
1,428
4,313
3,166
5,574
730
815
883
3,521
1,365
6,640
831
3,674
2,458
5,147
4,692
13,734
8,006
21,165
658
2,972
987
5,321
1,374
1,494
5,537
3,076
8,856
4,595
14,970
25,721
55,170
37,543
76,542
148
167
171
175
934
677
1,185
811
839
861
891
918
7,261
141
8,634
379
4,177
7,125
7,915
9,102
10,326
39,749
30,087
48,133
85
253
589
866
1,128
1,162
1,198
5,404
3,726
7,353
590
1,394
3,922
4,976
6,025
7,254
8,490
33,571
22,972
46,327
67
91
225
772
964
1,333
1,712
5,273
2,992
8,289
359
546
698
708
718
727
734
5,011
3,675
6,454
367
872
3,353
6,387
7,124
7,868
8,906
35,559
25,662
48,200
70
96
143
252
358
490
1,588
1,983
5,049
2,696
7,533
311
464
629
1,078
2,644
3,517
4,790
5,392
5,995
24,973
16,122
33,510
330
676
1,036
1,408
1,851
2,319
2,918
3,922
4,046
4,170
22,676
16,550
29,536
22,029
405
831
1,292
1,764
2,899
6,653
10,425
11,255
11,544
11,821
58,889
42,160
76,497
1,212
0
0
0
20
33
45
60
72
88
105
423
233
606
31,019
187
381
579
775
3,031
4,140
5,285
7,055
7,948
8,893
38,274
20,436
55,421
33
Niger
13,840
138
290
455
642
1,154
4,110
4,938
5,739
8,825
8,876
35,167
25,199
46,139
334,546
823
1,672
2,541
2,663
2,784
2,903
4,075
12,805
53,745
95,587
179,598
116,497
258,931
Rwanda
1,997
10
20
31
43
135
442
744
874
1,008
1,142
4,449
3,615
5,729
Senegal
12,006
64
132
202
276
1,328
2,623
3,655
5,315
6,091
6,868
26,554
18,579
35,766
Sierra Leone
8,289
43
88
136
189
299
658
1,783
2,036
2,347
2,676
10,255
7,080
14,074
Somalia
2,273
11
23
35
48
78
137
197
242
272
303
1,346
928
1,796
Sudan
35,621
161
324
488
647
1,728
3,376
5,087
8,247
10,145
12,439
42,642
30,245
57,163
Tanzania
41,517
826
1,672
2,560
3,492
4,689
6,112
9,000
10,867
14,902
19,184
73,304
54,025
93,550
5,627
161
327
499
675
1,173
1,405
1,697
1,715
1,978
2,252
11,882
8,692
15,427
Uganda
43,518
518
1,065
1,619
2,231
2,963
3,906
5,870
7,404
14,724
16,620
56,920
37,793
75,542
Zambia
14,517
641
661
678
694
1,104
3,174
4,685
5,966
5,995
6,787
30,385
21,884
42,498
Nigeria
Togo
Zimbabwe
Total
151
1
1
2
2
4
6
9
12
22
27
86
39
150
886,218
5,386
10,982
16,815
22,283
37,141
69,772
102,278
137,007
216,722
290,509
908,896
612,789
1,243,375
Vector control: Houses protected by either ITNs and/or IRS
CAR: Central African Republic
DRC: Democratic Republic of Congo
34
Table 5: Child deaths prevented due to scale-up of malaria prevention in pregnancy across sub-Saharan Africa
Uncertainty
Angola
Neonatal
deaths in 2000
35,765
11
22
34
47
61
75
90
106
124
143
713
395
961
Benin
9,942
0
1
1
1
2
2
2
3
3
4
19
9
25
Burkina Faso
19,352
0
0
0
0
0
1
1
2
2
3
9
3
30
Cameroon
17,463
0
0
0
3
8
12
17
22
27
32
121
61
188
CAR
7,450
0
0
0
1
2
4
5
6
8
9
35
19
53
Chad
18,779
0
0
0
3
7
11
15
20
25
30
111
73
148
Congo (Brazzaville)
3,496
0
0
0
0
1
1
2
2
3
4
13
2
27
Cote d’Ivoire
25,092
0
0
0
0
9
18
29
39
51
62
208
111
317
DRC
Equatorial Guinea
(Bioko Island)
Gambia
123,220
0
0
0
6
14
21
29
110
201
297
678
411
952
1,169
0
0
0
0
0
0
1
1
1
1
4
2
8
2,503
0
1
3
4
5
7
9
10
13
15
67
43
85
Ghana
22,418
0
0
1
14
27
41
56
72
89
106
406
248
558
Guinea
14,667
0
0
0
0
1
3
4
6
8
9
31
9
58
Guinea-Bissau
2,524
0
0
0
0
1
3
5
7
9
12
37
22
52
Kenya
32,642
2
3
5
7
10
12
15
38
61
72
225
122
324
Liberia
5,919
1
1
2
3
4
11
20
30
40
51
163
90
231
Madagascar
18,524
14
28
35
47
59
72
84
97
111
125
672
355
911
Malawi
13,359
7
14
24
31
36
42
47
53
56
61
371
192
634
Mali
32,298
4
8
12
17
22
27
70
116
167
221
664
378
906
Mozambique
49,257
0
0
0
0
0
1
1
1
1
1
615
355
898
Niger
19,853
2
4
7
9
12
15
67
126
191
263
696
446
900
Nigeria
188,,910
0
2
4
16
30
46
64
83
435
847
1,527
792
2,169
Senegal
14,455
0
0
0
6
12
57
65
73
95
118
426
0
559
Sierra Leone
11,971
0
0
0
0
1
1
2
3
4
6
17
276
485
Somalia
15,567
0
0
0
0
1
1
2
3
4
5
16
219
34
Sudan
48,404
17
33
51
68
87
105
124
143
163
183
974
7
1,255
Tanzania
39,691
14
29
47
50
60
71
82
116
151
189
809
478
1,067
Country
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
Total
Lower
Bound
Upper
Bound
35
Togo
7,537
0
0
0
0
4
9
14
19
24
29
99
56
147
Uganda
33,991
6
12
19
27
35
45
72
102
134
155
607
332
883
Zambia
14,584
20
40
64
88
113
140
163
168
184
198
1,178
714
1,495
Zimbabwe
9,043
0
0
0
1
2
3
4
5
6
7
28
15
46
Total
875,200
98
198
309
449
626
896
1,240
1,703
2,555
3,465
11,539
6,425
14,635
Malaria prevention in pregnancy: Woman received IPTp during last 2 pregnancies or currently pregnant women used ITNs previous night
CAR: Central African Republic
DRC: Democratic Republic of Congo
Countries without stable malaria transmission or a policy of IPTp were excluded
36