Scale-up of Malaria Control Interventions and
Reduction in All-Cause Mortality in Children 
Less than 5 Years of Age in Ghana
2003 – 2016
Dr. Frank E. Baiden, Principal Investigator
USAID/Ghana Evaluate for Health 
June 20, 2019
Malaria Impact Evaluation Final Report 6/25/2019 1
Acknowledgements
USAID/Ghana provided funding under its Evaluate for Health (Evaluate) project for this study on 
scale-up of malaria control interventions and reduction in all-cause mortality in children under five 
years of age in Ghana over the period 2003-2016. Dr Frank Baiden, research scientist, MBChB, 
MSc, PhD, served as principal investigator for the study, with support from a Steering Committee 
(SC) composed of leading Ghanaian malaria experts and with advice from the President’s Malaria 
Initiative (PMI) representatives both in Ghana and Washington, D.C. The SC and PMI reviewed the 
design, implementation, findings and conclusions of this study and provided invaluable insights
to the study team. Dr Baiden and the Evaluate team would like to acknowledge the important 
contributions of the following individuals:
Malaria Impact Evaluation Steering Committee members
Dr Abraham Hodgson, Chairperson (Director of Ghana Health Service Research and Development 
Division
Dr Kezia Malm, Program Manager, National Malaria Control Programme, Ghana Health Service
Nana Yaw Peprah, Sr Program Officer, National Malaria Control Programme, Ghana Health 
Service
Dr Samuel Dadzie, Research Scientist, Noguchi Memorial Institute for Medical Research
Dr Godwin Afenyadu, Project Director, Evaluate for Health
Abena Osei-Akoto, Chief Statistician, Ghana Statistical Service
Elizabeth Awini, Research Scientist
PMI Ghana 
Jon Eric Tongren, PMI Ghana
Sixte Zigirumugabe, PMI Ghana 
PMI/CDC Washington
Sergio Salgado, PMI, USAID, M&E Advisor 
Lia Florey, PMI, USAID Malaria Technical Advisor
Anna Bowen, CDC, Medical Epidemiologist
ICF
Yazoumé Ye, Vice President, Malaria Research and Surveillance
Malaria Impact Evaluation Final Report 6/25/2019 2
Contents
Introduction ..............................................................................................................................3
Methods ....................................................................................................................................3
Study site ...............................................................................................................................3
Evaluation Design..................................................................................................................5
Data Sources and Variables ..................................................................................................6
Results .......................................................................................................................................9
Change in all-cause under-five child mortality.....................................................................9
Changes in intervention coverage ......................................................................................11
Trends in morbidity ............................................................................................................. 15
Relationship between household ownership and use of ITN and under-five mortality.19
Discussion................................................................................................................................ 22
Conclusion ...............................................................................................................................24
References ..............................................................................................................................25
Annex: Supplementary Materials .......................................................................................... 27
Malaria Impact Evaluation Final Report 6/25/2019 3
Introduction 
There were an estimated 219 million cases and 435,000 deaths from malaria worldwide according 
to the 2018 World Malaria Report. Nearly all (93%) of the deaths were in sub-Saharan Africa.1 Since 
2003, global investments in malaria control and elimination have increased nearly 20-fold, reaching 
US $2.9 billion in 2015.2 These investments have been made with the expectation that they will 
cause substantial reduction in deaths due to malaria, particularly among children aged less than 
five years. 3-5
Children aged 0-59 months are particularly vulnerable to malaria.6 The major interventions 
deployed to reduce under-five deaths from malaria include the use of insecticide-treated nets 
(ITNs), intermittent preventive treatment of malaria in pregnant women (IPTp), prompt diagnosis 
and treatment of uncomplicated malaria using artemisinin-based combination treatments (ACTs), 
indoor residual spraying (IRS) and, where favourable climatic conditions exist, seasonal malaria 
chemoprevention (SMC). In many areas in malaria-endemic sub-Saharan Africa, the period of scale￾up of malaria control interventions has coincided with substantial decline in under-five mortality 
rates.7-9
Methods
Study site
Ghana occupies a land area of 238,537 sq. km in West Africa. It has an estimated population of 29 
million people, approximately 16% of whom are below five years of age.10 Malaria is endemic in all 
ten regions of Ghana and the entire population is at risk of the disease. There are three distinct 
zones of malaria transmission. These are the intense and seasonal transmission form in the three 
northern regions, a moderate and perennial form in the middle forest belt, and a lower but 
perennial form in the coastal south.10,11 Entomological inoculation rates range from 418 infective 
bites per person per year in the northern regions to 231 in the middle-forest belt to five in the 
southern coastal lowlands regions.12-14 The three northern regions of the country are generally 
regarded as the most under-served and tend to be targeted in the roll-out of health and other 
interventions.15-17
Malaria Impact Evaluation Final Report 6/25/2019 4
Figure 1. Map of Ghana
In 2016, Ghana recorded approximately 10.4 million 
suspected malaria cases through the routine district 
health information management system (DHIMS2), which 
represented about 39% of outpatient department (OPD) 
visits, 25% of total admissions and 4% of total deaths. 
Children under five years old and pregnant women were 
the most affected.18
The highest incidence of malaria in Ghana occurs during 
the rainy season (months). Plasmodium falciparum is 
responsible for more than 95% of malaria cases. The 
dominant vector species are Anopheles gambiae, Anopheles funestus and Anopheles arabiensis.
19,20
The major interventions to control malaria in Ghana are ITNs, IPTp with sulfadoxine￾pyrimethamine (SP), prompt diagnosis and treatment of uncomplicated malaria using ACTs, IRS 
and SMC.18,21,22
While the implementation of ITNs, ACTs and IPTp has been nationwide, the deployment of IRS 
and SMC have largely focused on the three northern regions. SMC, which involves administration 
of full treatment doses (4 rounds) of SP and amodiaquine (AQ) to children under 5 years old at 
monthly intervals from mid-June to mid-September,17 was introduced in 2015 in the Upper West 
Region and extended in 2016 to the Upper East in northern Ghana. In-door residual spraying (IRS) 
is implemented before the rainy season to provide protection once the malaria transmission 
season begins. IRS was initially implemented in one district in the Ashanti Region in 2006. 
Although it has been extended to other regions, mostly in the north, < 5% of the country’s 
population was protected by IRS in 2016.
In 2005, the Government of Ghana, in collaboration with the Ministry of Health and the Ghana 
Health Service, adopted the Community-Based Health Planning and Services (CHPS) as a national 
policy for the provision of primary health care services. The National/District Health Insurance 
Scheme (N/DHIS) is an intervention in national health care financing that was introduced in 2003 
to promote universal access to health care. From a reported coverage for women of 38.8% in 
2008, coverage stood at 58.0% by 2016. 
Malaria Impact Evaluation Final Report 6/25/2019 5
In early 2010, WHO issued revised treatment guidelines that call for a shift from the presumptive 
to the test-based approach. Based on research in Ghana and other countries, and evidence from 
program work, the Ghana National Malaria Control Programme issued revised national treatment 
guidelines that call for implementation of test-based management of malaria in all cases, and 
across all age groups.22
Major non-malarial interventions
Between 2003 and 2014, new vaccines were introduced into the routine immunization programs. 
These included the pentavalent vaccine which contains antigens for diphtheria, pertussis, 
tetanus, pneumococcus and rotavirus. Campaigns were launched to promote use of oral 
rehydration salts (ORS) and zinc the management of diarrhoea. Vitamin A supplementation was 
also initiated as well as intensified promotion of modern family planning methods and HIV testing 
and treatment. Health access also improved through the launch of community health posts and 
the national health insurance scheme. 
Under-five mortality
Under-five mortality rates in Ghana declined 50% between 1993 and 2014, from 119 to 60 deaths 
per 1,000 live births.23 The period of greatest decline was between 2003 and 2014. To date there 
has been no systematic exploration of the relationship between the scale-up of malaria control 
interventions and the reduction in under-five mortality. In 2017, the National Malaria Control 
Programme (NMCP), in partnership with USAID/Ghana’s Evaluate for Health project, 
commissioned an evaluation of the impact of malaria control interventions on all-cause under-five 
mortality in Ghana during 2003 – 2016, when increases in intervention coverage occurred.24,25 This 
paper is an output of the evaluation. 
Evaluation Design
We used a plausibility design26,27 to explore the possible contribution of interventions in malaria 
control to reduction in all-cause under-five mortality in Ghana between 2003 and 2016. All-cause 
under-five mortalities were extracted from national household surveys. Confidence intervals
were constructed to gauge extent of changes in rates over time. The changes in outcome 
measure (ACCM) and malaria control intervention coverage were compared in terms of direction, 
temporality and significance to explore the case for plausible relationship.
The data used in the analysis were the Ghana Demographic and Health Survey (GDHS, 2003, 2008 
and 2014), the Multiple Indicator Cluster Survey (MICS, 2006 and 2011) and the Malaria Indicator 
Malaria Impact Evaluation Final Report 6/25/2019 6
Survey (MIS, 2016). All surveys were conducted by the Ghana Statistical Service and details of the 
various methodologies are captured in respective reports and publications.23 The temporal 
relationship between the introduction of various malaria control interventions and the times when
data was collected are shown in Figure 2.
Figure 2. Timeline of evaluation data sources and malaria control interventions
Data Sources and Variables
Descriptive Analyses
Data sets from the full reports of the surveys were accessed and manipulated to calculate coverage 
of malaria control interventions. Data on under-five mortality were obtained from the surveys 
conducted in 2006 (MICS), 2008 (GDHS), 2011 (MICS), and 2014 (GDHS). Beyond the 2006-2014 
studies, data from the 2016 MIS were used to illustrate additional data trends after the evaluation 
period.
Regression Analyses
We used the data from the 2003, 2008 and 2014 Ghana Demographic Health Surveys (DHS) in 
regression analysis that explored the relationship between possession and use of ITNs and under￾five child survival. The dataset included full birth history of all women within the reproductive age 
(15-49 years). This birth history included all children born alive to women of reproductive age and 
Malaria Impact Evaluation Final Report 6/25/2019 7
their survival status. The children’s data files were merged with household data to obtain a 
complete dataset required for the analysis. The children’s data contain information related to the 
child's antenatal and postnatal care and immunization and health, including the data for the 
mother of each of these children. In addition, the file contains child health indicators such as 
bednet use, immunization coverage, vitamin A supplementation, and recent occurrences of 
diarrhea, fever and cough for young children and treatment of childhood diseases. The household 
data contain information about the sex of the household head, age, place of residence, household 
wealth, sanitation, access to improved water source, etc. Information about the year of the child's 
birth, whether each child was alive at the time of the survey, and how old a child was if s/he died 
was used to define a binary outcome of death among each child less than 5 years old during the 
five years preceding the survey. The exposure variables of interest were household ownership of 
ITN, number of children in the household that slept under an ITN the previous night and whether
the child’s mother slept under an ITN the previous night. Number of months each household had 
owned the bed-net was also included. 
Co-variates
We extracted additional variables that were associated with child survival in other reports and used 
them as covariates in the regression models. These variables were categorized into household 
characteristics (sex of the household head, age of the head, household size, place of residence, 
region, household wealth, household access to improved water and toilet facilities), maternal 
characteristics (mother’s age first birth, current age of mother, marital status, highest educational 
level and body mass index), child characteristics (age of the child, sex of the child, multiple birth, 
birth order and preceding birth interval).
Individual level analysis
Poisson model: We performed an individual-level modified Poisson model with robust standard 
error to assess the association between children that slept under an ITN the previous night and 
mortality in children aged 0–59 months, using data from the 2003, 2008 and 2014 Ghana DHS. We 
assumed that the relationship between ITN use and ITN ownership is fairly linear and that non-use 
of ITN is largely determined by lack of access 28,29. Indeed, households with ITN are likely to use it 
which explains why ITN use among households with ITN is generally high. Although it would have 
been preferable to use data on ITN use by individual children as the exposure variable during the 
survival period, DHS data on ITN use only referred to use in the night preceding the survey visit.
Malaria Impact Evaluation Final Report 6/25/2019 8
Cox-proportional hazard model: We fitted a Cox proportional hazard modelto compare the hazard 
rate of mortality between the children/women who slept under an ITN the previous night and 
children/women that did not sleep under an ITN the previous night. We assumed that 
mothers/children that slept under-ITN the night before the survey may have developed the habit 
of sleeping under ITN long before the survey, and it was;therefore, a good proxy for ITN use before 
the survey. 30 We used life table procedure and Kaplan-Meier survival analysis to compare 
estimates of under-five mortality and cumulative incidence rates between children that used an 
ITN and those who did not.
To obtain an unbiased estimate from both regression models, we accounted for complex survey 
design structure (clustering, stratification and weighting). To estimate population-level mortality 
and assess the relationship between ITN use and mortality using data pooled from different 
surveys, we de-normalized women sampling weights. This was done by dividing the women 
standard weight by the women survey sampling fraction; that is, the ratio of total number of 
women aged 15-49 years interviewed in the survey year over the total number of women aged 15-
49 years in the country at the time of the survey. The total number of women aged 15-49 
interviewed in the survey year was obtained from the DHS datasets, while the total number of 
women aged 15-49 years in the country at the time of the survey was obtained from
OurWorldinData, World Population Growth 2017,
31 which provides annual population estimates by 
country, disaggregated by sex. 
Regional level (Ecological) analysis
The regional level analysis was conducted to assess the impact of coverage of ITN on all-cause 
under-five mortality in Ghana. Regional-based aggregate data on household ownership of bed net, 
underweight, antenatal care attendance (ANC), postnatal care attendance (PNC), and educational 
level of women, unemployment rate, diarrhoea and vitamin A supplementation were extracted 
from the published reports of 2003, 2008 and 2014 DHSs. To determine if the time-fixed effect was 
needed when running the Fixed-Effect (FE) Poisson model, the combined effect of time dummies 
was tested. These were not found to be statistically significant, hence time-fixed effect was
ignored in the final model. The region-fixed effect was included in the model. We performed all 
statistical analyses with Stata MP Version 15 (StataCorp, Texas) and p<0.05 were considered 
statistically significant.
Malaria Impact Evaluation Final Report 6/25/2019 9 
Biological and dose-effect analysis
Given the higher coverage of malaria control interventions in the three northern regions, we 
explored evidence of biological explanation and dose-effect by comparing the trend in 
parasitaemia in children under five years old in the three northern regions with that in the rest of 
the country. 
Results 
Change in all-cause under-five child mortality 
Ghana’s under-five mortality decreased from 111 (95% CI 100-123) per 1,000 live births in 1993 
through 80 (95% CI 69-92) in 2008 to 60 (95% CI 53-68) in 2014. (Figure 3) Significant reductions
were observed in both urban and rural areas, but greater decreases were recorded in rural than 
urban (149 to 75 deaths per 1,000 live births). Mortality decreased among all wealth quintiles, but 
the highest reductions were in the fourth (108 to 55 deaths per 1,000 live births) and middle (111 to 
61 deaths per 1,000 live births) quintiles. Mortality reductions were observed in all regions, but the 
greatest reductions were in the regions that had high malaria parasitaemia (according to MICS 
2011) in under-five children: Upper West (188 to 92 deaths per 1,000 live births), Western (132 to 56 
deaths per 1,000 live births), Brong-Ahafo (95 to 57 deaths per 1,000 live births), Northern (237 to 
111 deaths per 1,000 live births), and Central (128 to 69 deaths per 1,000 live births). (Table 1)
Figure 3. Trend in under-five mortality in Ghana 1993 - 2014
Malaria Impact Evaluation Final Report 6/25/2019 10
Table 1. Trends in All-Cause Mortality in Children Under Five Years of Age in Ghana, 1993-2014
Under-five mortality rates per 1,000 live births
Background 
Characteristic
1989-1993
(1993 DHS)
1994-1998
(1998 DHS)
1999-2003
(2003 DHS)
2004-2008
(2008 DHS)
2010-2014
(2014DHS) Relative Change
# *LCI *UCI # LCI UCI # LCI UCI # LCI UCI # LCI UCI 2003 and 2014
Total (National) 119 109 131 107 96 120 111 100 123 80 69 92 60 53 68 -85
Residence
Urban 90 77 105 77 64 92 93 79 108 75 63 88 64 55 74 -44
Rural 149 139 160 122 111 133 118 108 130 90 80 102 75 67 83 -58
Region
Western 132 108 159 110 87 137 109 83 143 65 44 94 56 42 74 -97
Central 128 102 159 142 110 182 90 65 122 108 79 147 69 54 87 -31
Greater Accra 100 75 132 62 43 88 75 52 105 50 33 74 47 35 62 -59
Volta 116 97 139 98 76 126 113 91 140 50 30 83 61 46 81 -85
Eastern 93 72 120 89 71 111 95 73 122 81 54 120 68 54 86 -39
Ashanti 98 80 119 78 61 100 116 97 139 80 63 102 80 63 101 -46
Brong-Ahafo 95 77 116 129 94 174 91 71 115 76 54 106 57 45 71 -60
Northern 237 207 270 171 139 209 154 126 186 137 118 159 111 90 135 -39
Upper West 188 135 255 156 117 203 208 183 234 142 115 174 92 74 113 -127
Upper East 180 153 210 155 127 189 79 53 115 78 59 102 72 54 95 -9
Wealth Quintile
Lowest n/a n/a n/a n/a n/a n/a 127 110 147 103 89 119 92 80 106 -39
Second n/a n/a n/a n/a n/a n/a 105 87 126 79 65 96 73 61 86 -44
Middle n/a n/a n/a n/a n/a n/a 111 92 132 102 81 127 61 49 75 -83
Fourth n/a n/a n/a n/a n/a n/a 108 90 129 68 52 88 55 44 69 -95
Highest n/a n/a n/a n/a n/a n/a 88 69 113 60 43 83 64 50 81 -39
*LCI and UCL are lower and upper confidence intervals
Malaria Impact Evaluation Final Report 6/25/2019 11
Changes in intervention coverage
Ownership and Use of Insecticide-treated Nets
Household ownership of at least one ITN increased from 3.9% (95% CI 3.3 – 4.8) in 2003 to 73% (95% 
CI 70.6 – 75.3) in 2016. (Figure 4) 
Figure 4. Percentage of households owning at least one ITN, by region 2003 - 2016
Access to ITNs, measured as at least one ITN per two persons in the household, increased from 1% 
(95% CI 0.8-1.3) in 2003 to 50.9% (95% CI 48-53.7) in 2016. Increases in ownership and access 
tended to be higher in the northern regions of the country than in the south (Tables 2 and 3)
Uptake of Intermittent Presumptive Treatment for Pregnant Women
The uptake of more than two doses of SP as part of IPTp rose from 27.5% in 2006 to 67.5% in 2014, 
and further increased to 78.0% in 2016. Uptake was similar across regions. (Table 2)
0
10
20
30
40
50
60
70
80
90
100
Percentage of households
DHS 2003 DHS 2008 DHS 2014 MIS 2016
Malaria Impact Evaluation Final Report 6/25/2019 12
Table 2. Use of Intermittent Preventive Treatment Among Pregnant Women in Ghana by background characteristics, 2003-2016 i.e. 
Percentage of women age 15-49 with a live birth in the two years preceding the survey who received at least two doses of sulfadoxine￾pyrimethamine for Intermittent Preventive Treatment (IPTp) during ANC visits during their last pregnancy
Background 
Characteristic
2003 DHS 2008 DHS 2014 DHS MIS 2016 Percentage Point 
% LCI UCI n % LCI UCI n % LCI UCI n % LCI UCI n Change 2003-2016
Total (National) 0.8 0.4 1.4 1421 43.7 40 47.5 1178 67.5 64.5 70.3 2264 78 73.1 82.3 1285 77.2
Residence
Urban 0.7 0.2 2.1 477 46.3 40.2 52.4 455 68.2 63.3 72.7 1009 82.6 77.5 86.7 577 81.9
Rural 0.8 0.4 1.7 944 42.1 37.4 46.8 723 66.9 63.2 70.4 1255 74.3 67 80.5 708 73.5
Region
Western 0 128 45.5 34.2 57.3 111 67.3 57.9 75.6 217 77.3 68 84.5 101 77.3
Central 0.8 0.1 5.8 120 45.7 36.6 55.2 123 68.9 63.2 74 258 84.5 77.4 89.6 131 83.7
Greater Accra 1.6 0.4 5.9 150 29.4 20.3 40.4 133 59.3 50.1 68 332 78.7 68.6 86.3 207 77.1
Volta 1.6 0.4 6.4 134 59.8 50 68.8 107 65.1 56.6 72.7 177 75.1 62 84.8 110 73.5
Eastern 0 142 40.8 30.8 51.5 105 64.2 57.3 70.5 206 89.2 80.9 94.2 100 89.2
Ashanti 1 0.2 4.2 245 50.8 40.5 61.1 215 73.2 65.4 79.9 397 79.6 69.4 87.1 238 78.6
Brong-Ahafo 0.3 0 1.8 158 63.7 48.5 76.5 107 80.7 72.1 87.2 214 85 77.3 90.5 111 84.7
Northern 0 208 27.9 20.7 36.3 177 60.7 49 71.3 304 61 48 72.6 211 61
Upper West 1.5 0.4 5.6 49 52.5 42.3 62.6 36 73.8 67.4 79.4 64 82.2 76.1 87 30 80.7
Upper East 2.6 0.7 8.7 86 26 15.7 39.9 63 67.7 58.8 75.5 95 90.8 81.4 95.7 45 88.2
Wealth Quintile
Lowest 0.7 0.2 2.5 373 31.2 25.2 37.8 283 64.7 57.4 71.4 519 69.1 58.9 77.7 282 68.4
Second 0.7 0.2 2 319 42.6 35.9 49.5 261 70.8 65.2 75.8 474 74.9 66.8 81.6 269 74.2
Middle 1.1 0.3 3.6 284 50.3 42.6 58 222 64.1 58.8 69.1 433 78.3 71.4 83.9 265 77.2
Fourth 0.4 0.1 2.8 235 49.2 41.5 57 243 63.2 56.5 69.4 444 83.5 76.4 88.8 237 83.1
Highest 1.1 0.3 4.3 210 49.8 40.6 59 169 75.6 68.9 81.2 393 86.6 77.2 92.5 231 85.5
Age (in years)
15-19 0 96 44.2 32.7 56.4 80 66.8 55.5 76.4 143 79.3 64.8 88.8 91 79.3
20-24 1.4 0.5 3.8 308 43.9 36.9 51.1 278 61.3 55.4 66.8 441 73.3 64.8 80.4 289 71.9
25-29 1.5 0.7 3.4 384 44.5 38 51.1 342 69.8 64.2 74.8 614 77.1 67.9 84.3 314 75.6
30-34 0 296 46.6 39.5 53.8 223 70.4 65.2 75.2 516 82 74.3 87.8 313 82
35-39 0 225 42.1 33.9 50.7 169 67.4 60.2 73.8 379 82 73.8 88 198 82
40-44 1.3 0.2 8.9 74 38.4 26.7 51.7 68 65.7 56.2 74.1 137 72.5 55.6 84.7 72 71.2
45-49 0 38 21.3 7.3 48.3 17 73.3 58.7 84.1 34 66.2 36.9 86.8 8 66.2
Note: n=Weighted number of women (denominator); IPTp: Intermittent Preventive Treatment during pregnancy is preventive treatment with two or more doses of SP/Fansidar.
Malaria Impact Evaluation Final Report 6/25/2019 13
Use of ITNs in children under five also increased from 3.9% (95% CI 3.1-4.9) in 2003 to 52.2% (95% CI 
48.8-55.5) in 2016. (Figure 5) ITN use by pregnant women also increased from 2.7% (95% CI 1.6-4.6) 
in 2003 to 50% (95% CI 42.3-57.6) in 2016. (Figure 6) Increases in ITN use for both under-five 
children and pregnant women tended to be higher in the northern regions of the country than 
the rest of the country. (Figures 5 and 6, Northern, Upper West and Upper East Regions)
Figure 5. Use of insecticide-treated net the night before the survey by under-five 
children by regions 2003-2016
0
10
20
30
40
50
60
70
80
90
100
%
DHS 2003 DHS 2008 DHS 2014 MIS 2016
Malaria Impact Evaluation Final Report 6/25/2019 14
Figure 6. Use of insecticide-treated nets by pregnant women by regions 2003-2016
Case management
The percentage of children under 5 years old with recent fever for whom advice or care was sought 
increased from 51% (95% CI 45.5-55.9) in 2008 to 72% (95% CI 64.5-78.1) in 2016. Among children with 
fever for whom care was sought and who received antimalarials, the percentage who reported 
receiving the recommended first-line treatment for uncomplicated malaria, increased from 48% 
(95% CI 39.2-56.54) in 2008 to 78% (95% CI 72.1-83.3) in 2014 but decreased to 59% (95% CI 50-66) in 
2016. The greatest decreases between 2014-2016 were among the poorest quintiles (75-48% lowest; 
84-58% second; 77-49% middle) and in rural areas (81-59%). 
Among all children with reported fever in the previous two weeks,the proportion reported to have 
received a finger or heel stick (proxy for receiving a malaria diagnostic test) was 34.2% (95% CI 29.6-
39.4) in 2014 and 30.3% (95% CI 25.6-35.4) in 2016. 
Major non-malaria factors
Specific child health indicators that registered improvements over the period of evaluation are: 
prevalence of underweight, defined as low weight for age, among children less than 5 years old 
0
10
20
30
40
50
60
70
80
90
100
Percentage of Pregnent women
DHS 2003 DHS 2008 DHS 2014 MIS 2016
Malaria Impact Evaluation Final Report 6/25/2019 15
decreased from 17.6% in 2006 to 11.0% in 2014. In the same period, the uptake of vitamin A 
supplementation rose from 60.2% to 65.2% and the proportion of children who completed the 
WHO-recommended schedule of basic immunization by age 12 months rose from 64.4% to 71.1%. 
Rates of exclusive breastfeeding remained stable from 2006 (54.4%) to 2014 (52.3%). Annual GDP 
per capita increased 121% from $US 625 in 2003 to $US 1,384 in 2016. 
Indoor Residual Spraying (IRS)
Following IRS implementation, malaria parasitaemia among under-five children decreased as 
follows: Upper West Region, 51.2% in 2011 to 37.8% in 2014 to 22% in 2016; Upper East Region, 44% in 
2011, 11.7% in 2014, and 15.5% in 2016; Northern Region (where a subset of districts was sprayed), 
48.3% in 2011, 40% in 2014, and 24.6% in 2016. (Figure 7). 
Trends in morbidity
Malaria Prevalence and Anaemia in Children
Malaria prevalence measured by microscopy among under-five children decreased from 27.5% in 
2011 to 26.7% in 2014, and to 20.6% in 2016. (Table 6) Decreases occurred in all regions, wealth 
quintiles, age categories, and residence (urban or rural). No parasitaemia data are available for a 
2006 baseline. 
Figure 7. National and regional prevalence of parasitaemia among children 6 – 59 months old in 
Ghana: 2011 to 2016
Malaria Impact Evaluation Final Report 6/25/2019 16
The prevalence of severe anaemia (Hb<8g/dL) in children under five years old also decreased from 
14.3% (95% CI 17-21.4) in 2003 to 6.9% (95% CI 5.4-8.7) in 2016. Significant decreases were noted 
across all wealth quintiles (greatest decrease in poorest), age categories, urban/rural areas, and 
regions. (Table 3)
Malaria Impact Evaluation Final Report 6/25/2019 17
Table 3. Prevalence of Severe Anaemia in Children Aged 6-59 Months in Ghana by background characteristics
Background 
Characteristic
2003 DHS 2008 DHS 2014 DHS MIS 2016 Percentage Point 
% LCI UCI n % LCI UCI n % LCI UCI n % LCI UCI n Change 2003-2016
Total (National) 14.3 12.8 15.9 2992 19.1 17 21.4 2342 8.3 7 9.9 2568 6.9 5.4 8.7 2874 -7.4
Residence
Urban 9 7.1 11.3 984 13.1 10.4 16.4 894 4.4 3 6.6 1180 4.1 2.2 7.4 1276 -4.9
Rural 16.9 15 18.9 2008 22.8 20 25.9 1448 11.6 9.7 13.9 1388 9.1 7.1 11.7 1598 -7.8
Region
Western 15.4 11.1 21.2 293 26.5 19.3 35.3 221 8 5.3 11.8 273 3.9 1.6 9.3 213 -11.5
Central 14.9 10.8 20.2 267 19.2 14 25.7 225 10.7 6.6 17 304 14 8 23.4 281 -0.9
Greater Accra 8 4.7 13.1 324 6 3 11.9 267 4.2 1.4 12.1 389 1.3 0.3 4.6 406 -6.7
Volta 10.2 6.6 15.6 255 16.7 11.5 23.7 203 8.4 5.8 12 189 8.7 5.5 13.5 217 -1.5
Eastern 11.8 8.4 16.4 292 11.2 6.9 17.8 211 5.8 3.3 9.8 238 8.6 5.1 14.1 224 -3.2
Ashanti 14.6 10.8 19.3 553 20.8 14.9 28.2 460 5 2.6 9.5 432 3.7 1 12.9 656 -10.9
Brong-Ahafo 16.8 12.8 21.8 333 19.1 13.6 26.3 250 6.4 3.7 10.8 260 4.4 2 9.4 233 -12.4
Northern 20.1 15.3 26 403 27.8 22.7 33.5 332 18.2 14 23.4 313 12.4 9.1 16.7 464 -7.7
Upper West 11.7 8.1 16.6 86 31.2 24.5 38.8 63 16.5 11.1 23.8 66 9.1 5.9 13.8 75 -2.6
Upper East 15 10.1 21.6 186 15 10.3 21.3 109 6.7 4.1 10.7 105 7.4 4.9 11 105 -7.6
Wealth Quintile
Lowest 19.9 17 23 774 26.2 22.5 30.2 585 15.8 12.7 19.6 588 12.1 9.8 14.8 645 -7.8
Second 16.5 13.5 19.9 660 23 19 27.5 543 12.6 9.7 16.2 530 12.1 7.8 18.3 593 -4.4
Middle 13.7 11 17 597 20.4 16.4 25.2 425 7.1 4.5 10.9 523 5 3.2 7.8 581 -8.7
Fourth 11.3 8.3 15.1 521 13.8 10.4 18.2 463 3.2 1.8 5.6 483 2.6 1.5 4.8 588 -8.7
Highest 5.4 3.3 8.6 441 5.7 3.4 9.4 326 0.3 0 1.9 445 0.7 0.2 2.8 466 -4.7
Age (in months)
6-11 months 20.8 16.3 26.2 348 28.4 22.7 34.9 245 9.6 6.4 14 260 10 6.2 15.7 310 -10.8
12-23 months 23.1 19.7 26.8 661 24.3 20.5 28.5 521 14 10.6 18.2 587 11.8 9 15.5 648 -11.3
24-35 months 13.6 10.8 16.9 635 23.7 19.4 28.5 492 7.3 5.3 9.9 573 6.3 4.5 8.7 670 -7.3
36-47 months 10.1 8 12.7 716 12.7 9.5 16.8 517 8.5 6 11.9 570 4.5 2.8 7.1 591 -5.6
48-59 months 6.9 5.1 9.2 632 12.2 9.4 15.6 566 2.9 1.8 4.6 578 3.2 1.8 5.9 655 -3.7
6-23 months 22.3 19.5 25.4 1009 25.6 22.2 29.3 767 12.6 9.8 16.1 847 11.3 8.5 14.8 958 -11
24-59 months 10.2 8.8 11.8 1983 15.9 13.7 18.4 1575 6.2 5 7.7 1721 4.7 3.6 6.1 1916 -5.5
Note: n=Weighted number of children (denominator)
Malaria Impact Evaluation Final Report 6/25/2019 18
Dose-response relationship
The decline in malaria parasitaemia among children less than 5 years old in the three northern 
regions was much faster than it was for the rest of the country. It was an average decline of 13.7% 
per year compared to 1.1% in the rest of the country) from 2011 to 2016. Although parasitaemia level 
was approximately twice as high in the northern regions as in the rest of the country in 2011, by 
2016, the malaria parasitaemia levels in these regions were comparable (20% versus 22%). (Figure 
8) 
Figure 8. Mean percentage of children less than 5 years old with malaria parasites at the time 
of the survey in the three northern regions versus the remaining seven regions, 2011-2016
*The three northern regions are the Upper West Region, Upper East Region and the Northern 
Regions
0
10
20
30
40
50
60
MICS2011 GDHS2014 MICS2016
Mean parasitaemia (%)
Seven other regions Three Northern Regions
Malaria Impact Evaluation Final Report 6/25/2019 19
Relationship between household ownership and use of ITN and under-five mortality
Individual level analyses
All-cause mortality among under-five children was consistently lower among households with at 
least one ITN at the time of the survey, compared to households with no ITN at the time of the 
survey for each of the surveys studied. The results from 2014 DHS showed that there were 
approximately 8 (95% CI: 6-10) deaths per 1000 person-years among children who live in households 
with at least one ITN compared to 10 (95% CI: 6-20) deaths per 1000 person-years among children 
who live in households with no ITN (Table 4).
Table 4: Deaths among children <5 years old who live in households with at least one ITN or no 
ITN at the time of the survey, by person-time observed or number of live births: 2003, 2008 and 
2014 DHS.
2003 DHS
Ref. period: 
1999-2003
2008 DHS
Ref. period: 
2004-2008
2014 DHS
Ref. period: 
2010-2014
2003-2014 DHS
Ref. period: 1999-
2014
Cumulative deaths per 1000 person￾years among children <5 years old
Household with at least one ITN 14.1 (9.9- 20.7) 11.6 (8.9-15.4) 7.5 (5.9-9.6) 9.8 (8.3-11.6)
Household with no ITN 18.7 (15.3-23.0) 18.7 (13.1-27.6) 10.2 (5.8-20.0) 17.1 (14.420.4)
Overall 17.6 (14.7 -21.1) 13.7 (11.1-17.2) 8.0 (6.4-10.2) 12.7 (11.3-14.3)
Deaths among children <5 years old
per 1000 live births
Household with at least one ITN 104.0 (80.0-134.7) 73.2 (61.6-87.0) 58.5 (50.7-
67.4)
68.2 (61.6-75.6)
Household with no ITN 109.5 (94.8- 126.4) 109.2 (82.3-144.0) 75.3 (55.4-
101.9) 
101.8 (90.4-114.6)
Overall 108.1 (95.2- 122.6) 83.3 (71.5-96.8) 61.6 (54.1-70.1) 80.7 (74.7-87.2)
Total number of children sampled 3,844 2,992 5,884 12,720
Person-time at risk for children who 
live in households with at least one 
ITN (child-years)
1,948.0 4,614.5 10,569.5 12,380,396.0
Person-time at risk for children who 
live in households with at no ITN
(child-years)
6,223.0 1,958.2 2,487.5 8,005,462.5
Data source: Children's Data - Children's Recode (KR) for 2003, 2008 and 2014 DHS: (www.dhsprogrammme.com). Ref: Reference 
period.
Malaria Impact Evaluation Final Report 6/25/2019 20
Estimates obtained using a modified Poisson model showed that among children < 5 years old, a 
child’s reported or inferred use of an ITN was significantly associated lower mortality during each 
time period assessed. Household ownership of an ITN was non-significantly associated with lower 
all-cause mortality during each period assessed. The sensitivity analysis based on the Cox￾proportional hazard model showed similar results (Table 5)
Table 5: Association between ITN ownership and use at the time of the survey and all￾cause under five mortality during the 5 years preceding the survey, Ghana, 2003-2104 
2003 DHS
Ref. period:
1999-2003
2008 DHS
Ref. period:
2004-2008
2014 DHS
Ref. period: 
201042014
2003-2014 
DHS
Ref. period:
1999-2014
aRR (95%CI) aRR (95%CI) aRR (95%CI) aRR (95%CI)
Modified Poisson model 
Child use of ITN* (ref 
children not using ITN )
0.56*(0.36-
0.88)
0.39***(0.25-
0.60)
0.56**(0.40-
0.78)
0.52***(0.41-
0.65)
Child in household 
owning at least one ITN
(Ref. child in household 
not owning net)
0.86 (0.61-
1.23)
0.70 (0.48-1.04) 0.84 (0.58-1.23) 0.83 (0.66-
1.03)
Cox-proportional hazard 
model
Child use of ITN* (ref 
child not using ITN)
0.53**(0.34-
0.80)
0.38**(0.26-
0.57)
0.57***(0.42-
0.78)
0.51***(0.41-
0.64)
Child in household 
owning at least one ITN
(Ref child in household 
not owning net)
0.84 (0.61-
1.17)
0.69* (0.48-
0.98)
0.88 (0.54-1.45) 0.81 (0.66-
1.00)
P-value notation: * indicates p<.05; ** indicates p<.01;*** indicates p<.001. GDHS: Ghana Demographic Health Survey, aRR: Adjusted 
Relative Risk estimates based on all births in the 1-59 months preceding the date of interview; N=3760 for the 2003 GDHS, N=2933 for 
the 2008 GHDS, N= 5847 for the 2014 GDHS, N=12720 for the combined dataset (2003-2014). Adjusted relative risk estimates control for 
characteristics of the household (sex and age of household head, urban residence, wealth quintile, region, household size, access to an 
improved water source, and access to an improved non-shared toilet), characteristics of the mother (current age of the mother, age of 
mother at first child's birth, marital status, and level of education) and characteristics of the child (sex, birth order, and multiple births). 
For the GDHS 2003, 2008, 2014 and the combined dataset (2003-2014), Body Mass Index, Preceding birth interval, Antenatal care 
attendance and Tetanus vaccination status were not included in the adjusted model because of empty cells. We further adjusted for 
time period in fitting the model based on the full dataset (2003-2014 combined).
*for children who died we used mother ITN use as proxy for children ITN use.
Ecological (regional) level analyses:
An increase in percentage of households with at least one ITN was associated reducing under-five 
mortality rate. The results from the fixed-effect Poisson model with robust standard error showed 
that a unit increase in the percentage of households with at least one ITN is associated with a 0.49% 
reduction in under-five mortality rate, controlling for other time-varying covariates in the model 
(adjusted relative risk [aRR] 0.995, 95% CI: 0.992-0.998). (Table 6).
Malaria Impact Evaluation Final Report 6/25/2019 21
Table 6: Ecological analysis of the effect of ITN ownership on all-cause under-five 
mortality using fixed-effect modified Poisson with robust standard error.
aRR (95% CI)
Predictor
Household ownership of at least one 
ITN 0.995** (0.992-0.998)
Co-variates
Underweight 1.005 (0.995-1.016)
No ANC 1.010 (0.974-1.046)
No PNC 1.002 (0.997-1.01)
No Education 0.982* (0.967-0 .998)
Unemployment 1.005 (0.996-1.015)
No vaccination 1.014 (0.961-1.070)
Diarrhea 1.019** (1.005-1.034)
Vitamin A supplement 1.004 (0.998-1.009)
Data source: 2003, 2008, 2014 Ghana Demographic Health Survey Report; Abbreviation: aRR, 
Adjusted Relative Risk -Definitions: Underweight measured using weight-for-age which is a 
composite index of height-for-age and weight-for-height. It considers both acute and chronic 
malnutrition. Children whose weight-for-age is below -2 SD from the median of the reference 
population are classified as underweight; No ANC: % of women 15-49 with a live birth in the 5yrs 
preceding the survey with no ANC; No PNC: % of women 15-49 with a birth in the 5yrs preceding 
the survey with no PNC check; No Education: Percent of women age 15-49 with no level of 
schooling; Unemployment: % of women 15-49 who were not employed in the 12 months 
preceding the survey; No vaccination: % of children 12-23 months who didn't received any 
vaccination at any time before the survey; Diarrhea: % of children under five years with diarrhea
in 2 weeks preceding the survey; Time fixed effect was excluded from the final model since the 
combined effect of time was not statistically significant P-value notation: ***p<0.001, **p<0.01, 
*p<0.05. 
Malaria Impact Evaluation Final Report 6/25/2019 22
Discussion
Decline in under-five mortality in Ghana has aligned very closely with the expanded coverage of 
other malaria control interventions. A 46% reduction in under-five mortality was observed along
with 44.3, 11.0 and 50.5 percentage point increase ITN use by children aged less than 5 years, access 
to the first-line antimalarial for children with fever and uptake of at least 2 doses of ITP. Malaria 
parasite prevalence in under-five children also recorded by 7.1 percentage points. 
In this study, we applied a range of plausibility analyses that included using regression analysis that 
related household ITN possession and use and death of an under-five child within the household. 
We found a significant association that implied that possession and use of ITN was associated with 
reduced mortality. Based on the robustness and concurrence of the findings using different 
methodological approaches, we believe that the significant increase in ITN use in Ghana from 2006 
to 2014 contributed substantially to decline in under-five mortality over the same period. Our 
finding that a unit increase in the percentage of households with at least one ITN is associated with 
a 0.49% reduction in the risk of under-five mortality rate is consistent with published evidence on 
the effect of ITN reducing malaria morbidity and mortality in sub-Saharan Africa.
3,32-34 It strengthens 
evidence for the continued deployment of ITN in Ghana and malaria-endemic sub-Saharan Africa. 
In our analysis, we found that malaria parasitaemia in under-five children in the three northern 
regions declined at a rate more than 10 times that observed in the rest of the country. This finding 
lends credence to the case of impact of malaria control interventions in two important ways. It 
represents a dose-response effect given the concentration of activities in the three northern 
regions and the evident positive effect this has had on the coverages of the various interventions. 
The findings also represent consistency with theoretical and biological models that explain how
increased coverage in interventions should lead to less malaria parasitaemia and this ultimately 
leading to reduced overall mortality. Unfortunately, data quality precluded analysis using routine 
health service data on reported malaria cases in the country. 
Our finding that coverage in IPTp was less well correlated with reduction in under-five mortality is 
consistent with the theoretical consideration that prevention of malaria in pregnant women was 
not quite as proximal to under-five mortality as a child sleeping under an ITN or child being given 
ACT to treat fever. 
Malaria Impact Evaluation Final Report 6/25/2019 23
Malaria control activities in Ghana have led to remarkable increases in the coverage of the major 
interventions. The findings in this study are consistent with finding from an analysis of routine 
Ghanaian health service data. In an analysis of records of malaria cases and deaths and availability 
of ACT in 88 health facilities across the country, Aregawi et al found that compared to the period 
before 2005, the number of outpatient malaria cases declined by 57% between 2005-2015. While 
the number of malaria fell significantly by 65%, the number of deaths among children aged less 
than 5 years decreased by 50%.21 In analysis using the 2008 DHS, Afoakwah et al demonstrated that 
under-five mortality among children who sleep under treated bed nets was about 18.8% lower than 
among children who do not sleep under treated bed nets.35
In this study we found that uptake of interventions was particularly high in the three northern 
regions. Although the period of substantial increase in the coverage of malaria control 
interventions has coincided with the remarkable decline in under-five mortality, it needs to be 
acknowledged that evidence of decline in under-five mortality began before the period when 
malaria interventions were scaled-up. This suggests the existence of other important contributory 
factors. 
A major limitation in ecological analysis is the fact that it does not establish causality. The accepted 
principle is therefore to build the case of possible causal relationship on the basis of evidence of 
correlation, consistency (internal and external), biological plausibility and dose-effect. 36 The case 
is further strengthened when the relationship between interventions and effects holds up across 
different populations and epidemiological zones and is reasonably consistent with findings made 
in similar other settings. 4
Our efforts at relating the scale-up of malaria control interventions to decline in under-five 
mortality has been complex for a number of well-described limitations. First, all the available data 
is derived from observational studies (surveys) and not from purposefully-designed experiments. 
There was thus mismatch (in some cases) between the period covered by the available data on 
coverages and under-five mortality. For example, while under-five mortality trend could be traced 
to 1993 to establish a clear decline trend, nationally-representative data on malaria parasitaemia 
existed only from 2011. Similarly, IRS was deployed in specific selected districts. This situation was 
encountered in previous similar work and led to an evaluation of complex interventions assuming 
a body of science on its own. 37-39
Malaria Impact Evaluation Final Report 6/25/2019 24
Conclusion
The scale-up of malaria control interventions in Ghana between 2003 and 2016 coincided with 
substantial decline in under-five mortality during the same period. Our analyses have 
demonstrated close association between the two events, with scale-up in the use of ITNs and ACTs 
being the most firmly-correlated. Our findings support the case of impact and the need to sustain 
support for malaria control interventions in Ghana. 
Malaria Impact Evaluation Final Report 6/25/2019 25
References
1. World Health Organization. World malaria report. Geneva, Switzerland: World Health 
Organization; 2017. p. volumes.
2. World Health Organization. World malaria report. Geneva, Switzerland: World Health 
Organization; 2018. p. volumes.
3. Akachi Y, Atun R. Effect of investment in malaria control on child mortality in sub-Saharan Africa in 
2002-2008. PLoS One 2011; 6(6): e21309.
4. Hershey CL, Bhattarai A, Florey LS, et al. Implementing Impact Evaluations of Malaria Control 
Interventions: Process, Lessons Learned, and Recommendations. Am J Trop Med Hyg 2017; 97(3_Suppl): 20-
31.
5. Patouillard E, Griffin J, Bhatt S, Ghani A, Cibulskis R. Global investment targets for malaria control 
and elimination between 2016 and 2030. BMJ Glob Health 2017; 2(2): e000176.
6. Black RE, Cousens S, Johnson HL, et al. Global, regional, and national causes of child mortality in 
2008: a systematic analysis. Lancet 2010; 375(9730): 1969-87.
7. Eckert E, Florey LS, Tongren JE, et al. Impact Evaluation of Malaria Control Interventions on 
Morbidity and All-Cause Child Mortality in Rwanda, 2000-2010. Am J Trop Med Hyg 2017; 97(3_Suppl): 99-110.
8. O'Meara WP, Mangeni JN, Steketee R, Greenwood B. Changes in the burden of malaria in sub￾Saharan Africa. Lancet Infect Dis 2010; 10(8): 545-55.
9. Snow RW, Marsh K. Malaria in Africa: progress and prospects in the decade since the Abuja 
Declaration. Lancet 2010; 376(9735): 137-9.
10. Awine T, Malm K, Peprah NY, Silal SP. Spatio-temporal heterogeneity of malaria morbidity in 
Ghana: Analysis of routine health facility data. PLoS One 2018; 13(1): e0191707.
11. Kusi KA, Manu EA, Manful Gwira T, et al. Variations in the quality of malaria-specific antibodies with 
transmission intensity in a seasonal malaria transmission area of Northern Ghana. PLoS One 2017; 12(9): 
e0185303.
12. Appawu M, Owusu-Agyei S, Dadzie S, et al. Malaria transmission dynamics at a site in northern 
Ghana proposed for testing malaria vaccines. Trop Med Int Health 2004; 9(1): 164-70.
13. Dery DB, Brown C, Asante KP, et al. Patterns and seasonality of malaria transmission in the forest￾savannah transitional zones of Ghana. Malar J 2010; 9: 314.
14. Ononye N. Characterization of Malaria Transmission and Insecticide Susceptibility Status of 
Anopheles Gambiae Sensu Lato Gilles (Diptera: Culicidae) In Shai-Osudoku District of Southern Ghana. 
http://ugspace.ug.edu.gh/handle/123456789/8455: University of Ghana; 2015.
15. Adua E, Frimpong K, Li X, Wang W. Emerging issues in public health: a perspective on Ghana's 
healthcare expenditure, policies and outcomes. EPMA J 2017; 8(3): 197-206.
16. Akazili J, Welaga P, Bawah A, et al. Is Ghana's pro-poor health insurance scheme really for the 
poor? Evidence from Northern Ghana. BMC Health Serv Res 2014; 14: 637.
17. Nonvignon J, Aryeetey GC, Issah S, et al. Cost-effectiveness of seasonal malaria chemoprevention 
in upper west region of Ghana. Malar J 2016; 15: 367.
18. Malm KL, Segbaya S, Forson I, et al. Implementation of the Affordable Medicines for Malaria 
Facility (AMFm) in Ghana: processes, challenges and achievements. J Public Health Policy 2013; 34(2): 302-14.
19. Coleman S, Dadzie SK, Seyoum A, et al. A reduction in malaria transmission intensity in Northern 
Ghana after 7 years of indoor residual spraying. Malar J 2017; 16(1): 324.
20. Dadzie SK, Brenyah R, Appawu MA. Role of species composition in malaria transmission by the 
Anopheles funestus group (Diptera: Culicidae) in Ghana. J Vector Ecol 2013; 38(1): 105-10.
21. Aregawi M, Malm KL, Wahjib M, et al. Effect of anti-malarial interventions on trends of malaria 
cases, hospital admissions and deaths, 2005-2015, Ghana. Malar J 2017; 16(1): 177.
22. Baiden F, Malm K, Bart-Plange C, et al. Shifting from presumptive to test-based management of 
malaria - technical basis and implications for malaria control in Ghana. Ghana Med J 2014; 48(2): 112-22.
23. Kayode GA, Grobbee DE, Koduah A, et al. Temporal trends in childhood mortality in Ghana: 
impacts and challenges of health policies and programs. Glob Health Action 2016; 9: 31907.
24. Hemingway J. Malaria: Fifteen years of interventions. Nature 2015; 526(7572): 198-9.
25. Tanner M, Greenwood B, Whitty CJ, et al. Malaria eradication and elimination: views on how to 
translate a vision into reality. BMC Med 2015; 13: 167.
26. Rowe AK. Assessing the Health Impact of Malaria Control Interventions in the MDG/Sustainable 
Development Goal Era: A New Generation of Impact Evaluations. Am J Trop Med Hyg 2017; 97(3_Suppl): 6-8.
Malaria Impact Evaluation Final Report 6/25/2019 26
27. Rowe AK, Steketee RW, Arnold F, et al. Viewpoint: evaluating the impact of malaria control efforts 
on mortality in sub-Saharan Africa. Trop Med Int Health 2007; 12(12): 1524-39.
28. Eisele TP, Keating J, Littrell M, Larsen D, Macintyre K. Assessment of insecticide-treated bednet 
use among children and pregnant women across 15 countries using standardized national surveys. The 
American journal of tropical medicine and hygiene 2009; 80(2): 209-14.
29. Koenker H, Kilian A. Recalculating the net use gap: a multi-country comparison of ITN use versus 
ITN access. PloS one 2014; 9(5): e97496.
30. Takem E, D’Alessandro U. Malaria in pregnancy. Mediterr J Hematol Infect Dis 5: e2013010. 2013.
31. OurWorldinData. World Population Growth. https://ourworldindata.org/world-population￾growth#population-size-vs-population-growth-rate. 2017.
32. Bhatt S, Weiss DJ, Cameron E, et al. The effect of malaria control on Plasmodium falciparum in 
Africa between 2000 and 2015. Nature 2015; 526(7572): 207-11.
33. Steketee RW, Campbell CC. Impact of national malaria control scale-up programmes in Africa: 
magnitude and attribution of effects. Malar J 2010; 9: 299.
34. Walker N, Friberg IK. Introduction: reporting on updates in the scientific basis for the Lives Saved 
Tool (LiST). BMC Public Health 2017; 17(Suppl 4): 774.
35. Afoakwah C, Deng X, Onur I. Malaria infection among children under-five: the use of large-scale 
interventions in Ghana. BMC Public Health 2018; 18(1): 536.
36. Rothman KJ, Greenland S. Causation and causal inference in epidemiology. Am J Public Health
2005; 95 Suppl 1: S144-50.
37. Campbell M, Fitzpatrick R, Haines A, et al. Framework for design and evaluation of complex 
interventions to improve health. BMJ 2000; 321(7262): 694-6.
38. Hasson H. Systematic evaluation of implementation fidelity of complex interventions in health and 
social care. Implement Sci 2010; 5: 67.
39. Bryce J, Victora CG, Advisors M-IT. Ten methodological lessons from the multi-country evaluation 
of integrated Management of Childhood Illness. Health Policy Plan 2005; 20 Suppl 1: i94-i105.
Malaria Impact Evaluation Final Report 6/25/2019 27
Annex: Supplementary Materials
Table 1. Trends in All-Cause Mortality in Children Under Five Years of Age in Ghana, 1993-2014
Background 
Characteristic
2003 DHS 2008 DHS 2014 DHS
Relative 
Change
# LCI UCI # LCI UCI # LCI UCI 2003 and 
2014
Total (National) 111 100 123 80 69 92 60 53 68 -85
Residence
Urban 93 79 108 75 63 88 64 55 74 -44
Rural 118 108 130 90 80 102 75 67 83 -58
Region
Western 109 83 143 65 44 94 56 42 74 -97
Central 90 65 122 108 79 147 69 54 87 -31
Greater Accra 75 52 105 50 33 74 47 35 62 -59
Volta 113 91 140 50 30 83 61 46 81 -85
Eastern 95 73 122 81 54 120 68 54 86 -39
Ashanti 116 97 139 80 63 102 80 63 101 -46
Brong-Ahafo 91 71 115 76 54 106 57 45 71 -60
Northern 154 126 186 137 118 159 111 90 135 -39
Upper West 208 183 234 142 115 174 92 74 113 -127
Upper East 79 53 115 78 59 102 72 54 95 -9
Wealth Quintile
Lowest 127 110 147 103 89 119 92 80 106 -39
Second 105 87 126 79 65 96 73 61 86 -44
Middle 111 92 132 102 81 127 61 49 75 -83
Fourth 108 90 129 68 52 88 55 44 69 -95
Highest 88 69 113 60 43 83 64 50 81 -39
Malaria Impact Evaluation Final Report 6/25/2019 28
Table 2. Percentage of households with at least one insecticide-treated net by background characteristics 2003-2016
Percentage of households with at least one insecticide-treated net (ITN) among all the households interviewed, by background characteristics
Background Characteristic
2003 DHS 2008 DHS 2014 DHS MIS 2016 Percentage 
Point Change 
% LCI UCI n % LCI UCI n % LCI UCI n % LCI UCI n 2003-2016
Total (National) 3.9 3.3 4.6 6,251 41.7 40.1 43.4 11,777 68.3 66.5 70.1 11,835 73 70.6 75.3 5,841 69.1
Residence
Urban 2.8 2.1 3.8 2,870 34.7 32.7 36.8 5627 60.1 57.4 62.7 6503 65.3 62.2 68.3 3195 62.5
Rural 4.8 3.9 5.9 3,381 48.1 45.7 50.6 6150 78.4 76.2 80.4 5332 82.4 79.6 84.8 2646 77.6
Region
Western 2.3 1.2 4.6 612 41.1 37 45.3 1184 67.4 62.2 72.3 1298 66.9 58.9 74 482 64.6
Central 2.5 1.2 5.2 587 41.8 35.8 48.1 1279 69.7 65.4 73.7 1180 83 74.9 88.8 646 80.5
Greater Accra 2.2 1.4 3.5 890 29.8 26.2 33.6 1951 52.8 47.2 58.4 2457 60.9 54.4 66.9 1177 58.7
Volta 2.7 1.5 4.7 538 42.8 36.4 49.5 991 76.3 71.9 80.3 1015 76.1 71 80.5 423 73.4
Eastern 0.7 0.3 1.8 732 36 31.6 40.7 1260 73.1 68.8 77 1255 71.6 67 75.9 574 70.9
Ashanti 2.1 1.3 3.4 1313 39.7 35.8 43.8 2263 70.3 65.9 74.4 2216 69.7 64.6 74.4 1278 67.6
Brong-Ahafo 2.9 1.9 4.5 665 51.3 47.1 55.6 1154 80.8 77.7 83.6 1028 80.6 75.5 84.8 490 77.7
Northern 8.5 5.8 12.3 487 53.5 47.7 59.2 928 71.3 64.7 77.1 742 83.7 77.5 88.4 464 75.2
Upper West 5.5 3.1 9.7 147 70.8 65 76 228 77.4 71 82.8 265 89.7 85.8 92.6 126 84.2
Upper East 28 20.7 36.7 280 52.9 44.6 61.1 540 72.8 67.8 77.4 378 94.5 91.6 96.4 180 66.5
Wealth Quintile
Lowest 7.8 5.9 10.2 971 50.2 46.5 53.8 1813 79.6 76.6 82.2 1600 86.3 82.8 89.2 906 78.5
Second 2.7 1.8 4 1168 46 42.7 49.3 2250 77.9 75.1 80.5 2211 81 77.2 84.3 1143 78.3
Middle 3 2.1 4.2 1315 40.1 37.2 43.1 2548 69.7 66.7 72.6 2647 73.3 70.6 75.8 1203 70.3
Fourth 2.7 1.9 3.7 1452 36.5 34 39 2646 62.9 59.9 65.7 2686 66.3 62.3 70 1310 63.6
Highest 4.4 3.3 5.9 1345 39 36.2 41.9 2520 57.9 54.3 61.4 2690 63.2 58.7 67.4 1278 58.8
Note: n=Weighted number of households (denominator); An insecticide-treated net (ITN) is (1) a factory-treated net that does not require any further treatement (LLIN) or (2) a pretreated net obtained within the past 12 
months of (3) a net that has been soaked with insecticide within the past 12 months.
Malaria Impact Evaluation Final Report 6/25/2019 29
Table 3. Percentage of households with at least one insecticide-treated net for every two persons by background characteristics 2003-2016
Percentage of households with at least one ITN for every two people, by background characteristics and survey year, Ghana
Background Characteristic
2003 DHS 2008 DHS 2014 DHS MIS 2016 Percentage 
Point Change 
2003-2016 % LCI UCI n % LCI UCI n % LCI UCI n % LCI UCI n
Total (National) 1 0.8 1.3 6213 17 15.9 18.1 11716 45.2 43.5 46.9 11743 50.9 48 53.7 5770 49.9
Residence
Urban 0.8 0.5 1.3 2852 14.8 13.4 16.4 5596 41.3 38.8 43.8 6444 45.8 42 49.6 3151 45
Rural 1.2 0.8 1.7 3361 18.9 17.3 20.7 6120 50 47.7 52.3 5299 57 53.4 60.5 2619 55.8
Region
Western 0.7 0.2 2.2 607 15.7 13.3 18.5 1174 44.8 40.3 49.4 1293 46.1 40.5 51.9 472 45.4
Central 1.5 0.7 3.4 585 19.1 15 24.1 1265 44.4 41.1 47.9 1167 61.7 52.3 70.4 643 60.2
Greater Accra 0.7 0.3 1.7 885 12.6 10.2 15.6 1935 35.1 30.2 40.4 2431 41.9 34.6 49.4 1151 41.2
Volta 0.7 0.2 2.1 533 18.4 13.8 24.1 987 55.2 49.8 60.4 1008 52.3 45.4 59.2 418 51.6
Eastern 0.2 0 1.5 727 13.5 10.7 16.9 1258 52 47.5 56.4 1249 44.6 40.3 49 571 44.4
Ashanti 0.6 0.3 1.3 1304 15.9 13.8 18.3 2254 46.7 42.9 50.6 2194 47.2 40.5 54 1267 46.6
Brong-Ahafo 1 0.5 2 661 22.2 19.3 25.4 1149 59 55.1 62.7 1021 58 53.1 62.8 482 57
Northern 1.4 0.7 2.9 485 18.2 14.6 22.4 926 37.5 31.7 43.6 740 59.9 54.4 65.3 461 58.5
Upper West 1.5 0.7 3 145 28.9 25.1 33 228 42.6 37.4 47.8 263 65.2 57.5 72.2 125 63.7
Upper East 6 3.6 9.8 280 22 17 27.9 539 36.5 31.6 41.6 376 72.4 65.2 78.6 179 66.4
Wealth Quintile
Lowest 0.7 0.4 1.3 970 15.4 13.2 17.9 1810 42.6 39.6 45.6 1599 59.2 54.1 64.2 902 58.5
Second 0.8 0.4 1.5 1161 18 15.8 20.4 2236 50.9 48 53.7 2197 59.6 55.2 63.8 1129 58.8
Middle 1.1 0.6 1.9 1307 17 15 19.3 2535 48.7 45.8 51.6 2622 50.5 46.7 54.2 1181 49.4
Fourth 1.1 0.7 1.7 1436 15.2 13.5 17.1 2632 42.9 39.8 46.1 2661 45.1 40.8 49.4 1300 44
Highest 1.4 0.8 2.4 1338 19 16.8 21.4 2502 41.1 37.5 44.8 2665 43.4 38.2 48.9 1258 42
Note: n=Weighted number of households (denominator); An insecticide-treated net (ITN) is (1) a factory-treated net that does not require any further treatement (LLIN) or (2) a pretreated net obtained within the past 12 
months or (3) a net that has been soaked with insecticide within the past 12 months.
Malaria Impact Evaluation Final Report 6/25/2019 30
Table 4. Use of Insecticide-treated Nets Among Under Five Children in Ghana by background characteristics, 2003-2016
Background Characteristic
2003 DHS 2008 DHS 2014 DHS MIS 2016 Percentage 
Point Change 
% LCI UCI n % LCI UCI n % LCI UCI n % LCI UCI n 2003-2016
Total (National) 3.9 3.1 4.9 3593 38.7 36.1 41.5 2906 46.6 44.1 49.2 5801 52.2 48.8 55.5 3234 48.3
Residence
Urban 3.8 2.5 5.8 1202 32.6 29.6 35.6 2229 36.1 32.7 39.6 2639 40.8 36.3 45.4 1466 37
Rural 4 2.9 5.3 2391 42.6 39.9 45.3 3561 55.4 51.7 59 3163 61.7 56.5 66.7 1768 57.7
Region
Western 2.2 0.9 5.4 346 37.5 31.2 44.4 534 48 39.8 56.4 583 45.5 33.5 58.1 241 43.3
Central 1.4 0.5 3.8 306 29.4 24.6 34.7 569 51.2 44.7 57.7 621 61.2 46.4 74.2 310 59.8
Greater Accra 3.1 1.2 7.6 390 29.6 24.6 35.1 679 25.9 19.4 33.6 906 32.6 26.7 39.2 490 29.5
Volta 2.5 1.1 5.6 303 43.7 37.3 50.4 474 66.3 60.4 71.8 464 52.5 45.2 59.7 252 50
Eastern 0.3 0 2.2 372 39.2 33.8 44.9 513 49 41.8 56.1 559 48.2 39.7 56.8 264 47.9
Ashanti 1.2 0.5 2.8 661 38.1 33.4 43.1 1060 47.2 39.6 54.9 1043 51.2 44.3 58 705 50
Brong-Ahafo 2.9 1.5 5.6 388 56.9 51.7 61.9 611 60.8 54.7 66.6 524 60.5 49.5 70.4 261 57.6
Northern 7.1 4.2 11.9 488 31.8 26.3 37.9 869 43.2 36.1 50.5 709 61 48.9 71.9 511 53.9
Upper West 2.7 0.8 8.2 108 63.8 56.9 70.1 165 54.5 47.7 61.2 154 60.7 47 72.9 83 58
Upper East 22.1 14.4 32.4 231 41.7 33 50.9 317 37.4 30.8 44.6 238 75.5 63.2 84.7 118 53.4
Wealth Quintile
Lowest 6.4 4.4 9.2 918 42.6 38..2 47.2 1427 55.3 50.8 59.6 1306 66.6 60.8 72 728 60.2
Second 2 1.1 3.3 797 40.1 36.2 44.1 1252 59.5 54.9 63.9 1219 59.4 52 66.3 666 57.4
Middle 2.8 1.7 4.6 717 38.4 34.7 42.3 1128 48.1 44.1 52.2 1145 50.9 44.1 57.7 645 48.1
Fourth 2.8 1.5 5.3 625 36.2 32.5 40.1 1110 35.2 30.3 40.4 1108 43.3 36.7 50.2 657 40.5
Highest 5.4 3.3 8.7 537 34 29.7 38.6 874 30.9 25.8 36.5 1024 36.3 30.3 42.7 539 30.9
Age (in months)
<12 6.1 4.5 8.3 709 49 45.7 52.3 1156 51.9 48.2 55.6 1173 55.3 49.6 60.8 635 49.2
12-23 5 3.6 6.8 711 44.6 41.1 48.1 1072 47.4 43.1 51.7 1156 54.5 49.7 59.1 655 49.5
24-35 3.3 2.1 5 698 36.4 33.3 39.5 1110 46.7 42.4 50.9 1143 51.4 46.1 56.7 678 48.1
36-47 3.2 2.2 4.7 791 35.8 32.6 39.2 1193 46 42.1 49.9 1149 49.9 44.1 55.8 598 46.7
48-59 2 1.2 3.5 685 29.2 26.3 32.3 1260 41.2 37.3 45.2 1181 49.9 44.5 55.2 667 47.9
Note: n=Weighted number of children (denominator); An insecticide-treated net (ITN) is (1) a factory-treated net that does not require any further treatment (LLIN) or (2) a pretreated net obtained within the past 12 months 
or (3) a net that has been soaked with insecticide within the past 12 months.
Malaria Impact Evaluation Final Report 6/25/2019 31
Table 5. Use of Intermittent Preventive Treatment by Pregnant Women by background characteristics, 2003-2016
Background Characteristic
2003 DHS 2008 DHS 2014 DHS MIS 2016 Percentage 
Point Change 
% LCI UCI n % LCI UCI n % LCI UCI n % LCI UCI n 2003-2016
Total (National) 0.8 0.4 1.4 1421 43.7 40 47.5 1178 67.5 64.5 70.3 2264 78 73.1 82.3 1285 77.2
Residence
Urban 0.7 0.2 2.1 477 46.3 40.2 52.4 455 68.2 63.3 72.7 1009 82.6 77.5 86.7 577 81.9
Rural 0.8 0.4 1.7 944 42.1 37.4 46.8 723 66.9 63.2 70.4 1255 74.3 67 80.5 708 73.5
Region
Western 0 128 45.5 34.2 57.3 111 67.3 57.9 75.6 217 77.3 68 84.5 101 77.3
Central 0.8 0.1 5.8 120 45.7 36.6 55.2 123 68.9 63.2 74 258 84.5 77.4 89.6 131 83.7
Greater Accra 1.6 0.4 5.9 150 29.4 20.3 40.4 133 59.3 50.1 68 332 78.7 68.6 86.3 207 77.1
Volta 1.6 0.4 6.4 134 59.8 50 68.8 107 65.1 56.6 72.7 177 75.1 62 84.8 110 73.5
Eastern 0 142 40.8 30.8 51.5 105 64.2 57.3 70.5 206 89.2 80.9 94.2 100 89.2
Ashanti 1 0.2 4.2 245 50.8 40.5 61.1 215 73.2 65.4 79.9 397 79.6 69.4 87.1 238 78.6
Brong-Ahafo 0.3 0 1.8 158 63.7 48.5 76.5 107 80.7 72.1 87.2 214 85 77.3 90.5 111 84.7
Northern 0 208 27.9 20.7 36.3 177 60.7 49 71.3 304 61 48 72.6 211 61
Upper West 1.5 0.4 5.6 49 52.5 42.3 62.6 36 73.8 67.4 79.4 64 82.2 76.1 87 30 80.7
Upper East 2.6 0.7 8.7 86 26 15.7 39.9 63 67.7 58.8 75.5 95 90.8 81.4 95.7 45 88.2
Wealth Quintile
Lowest 0.7 0.2 2.5 373 31.2 25.2 37.8 283 64.7 57.4 71.4 519 69.1 58.9 77.7 282 68.4
Second 0.7 0.2 2 319 42.6 35.9 49.5 261 70.8 65.2 75.8 474 74.9 66.8 81.6 269 74.2
Middle 1.1 0.3 3.6 284 50.3 42.6 58 222 64.1 58.8 69.1 433 78.3 71.4 83.9 265 77.2
Fourth 0.4 0.1 2.8 235 49.2 41.5 57 243 63.2 56.5 69.4 444 83.5 76.4 88.8 237 83.1
Highest 1.1 0.3 4.3 210 49.8 40.6 59 169 75.6 68.9 81.2 393 86.6 77.2 92.5 231 85.5
Age (in years)
15-19 0 96 44.2 32.7 56.4 80 66.8 55.5 76.4 143 79.3 64.8 88.8 91 79.3
20-24 1.4 0.5 3.8 308 43.9 36.9 51.1 278 61.3 55.4 66.8 441 73.3 64.8 80.4 289 71.9
25-29 1.5 0.7 3.4 384 44.5 38 51.1 342 69.8 64.2 74.8 614 77.1 67.9 84.3 314 75.6
30-34 0 296 46.6 39.5 53.8 223 70.4 65.2 75.2 516 82 74.3 87.8 313 82
35-39 0 225 42.1 33.9 50.7 169 67.4 60.2 73.8 379 82 73.8 88 198 82
40-44 1.3 0.2 8.9 74 38.4 26.7 51.7 68 65.7 56.2 74.1 137 72.5 55.6 84.7 72 71.2
45-49 0 38 21.3 7.3 48.3 17 73.3 58.7 84.1 34 66.2 36.9 86.8 8 66.2
Note: n=Weighted number of women (denominator); IPTp: Intermittent Preventive Treatment during pregnancy is preventive treatment with two or more doses of SP/Fansidar.
Malaria Impact Evaluation Final Report 6/25/2019 32
Table 6. Percentage of children (6-59 months) with malaria parasites, by background 
characteristics 2014-2016 
Background 
Characteristic
2014 DHS MIS 2016
Percentage 
Point 
Change 
% LCI UCI n % LCI UCI n 2003-2016
Total (National) 26.7 23.8 29.9 2558 20.6 17.4 24.3 2874 -6.1
Residence
Urban 13.5 10.8 16.9 1175 11.2 8.5 14.6 1276 -2.3
Rural 37.9 33.7 42.4 1384 28.2 22.4 34.9 1598 -9.7
Region
Western 38.9 30.6 47.9 272 23.5 15.6 33.8 213 -15.4
Central 37.9 27.3 49.8 304 30.2 23.2 38.2 281 -7.7
Greater Accra 11.2 7 17.3 383 4.8 2.8 8.1 406 -6.4
Volta 25.2 14.6 39.8 189 27.5 19.7 36.8 217 2.3
Eastern 29.5 21.8 38.6 237 31.3 21.4 43.2 224 1.8
Ashanti 16.6 11.1 24.2 432 16.6 11.2 23.9 656 0
Brong-Ahafo 26.5 18.8 36.1 259 22.4 12 38 233 -4.1
Northern 40 30.1 50.8 313 25.2 12.2 44.9 464 -14.8
Upper West 37.8 27.5 49.5 66 21.5 13.8 31.9 75 -16.3
Upper East 11.7 8 16.8 105 14.7 10 21 105 3
Wealth Quintile
Lowest 42.1 36.1 48.5 586 37.2 29.8 45.2 645 -4.9
Second 39.5 34.3 44.9 529 29 21.6 37.7 593 -10.5
Middle 24.6 19.3 30.8 520 17 12.4 22.9 581 -7.6
Fourth 13.9 9.5 19.9 481 12.5 9.2 16.7 588 -1.4
Highest 7.5 4.6 12 443 1.9 0.9 4 466 -5.6
Age (in months)
6-11 months 18.9 14.2 24.7 260 17 11.8 23.9 310 -1.9
12-23 months 23.3 19.1 28.1 585 17.8 14.2 22.1 648 -5.5
24-35 months 25.4 21.1 30.4 570 19.7 15.1 25.2 670 -5.7
36-47 months 28.9 24.2 34.2 568 21.6 16.8 27.3 591 -7.3
48-59 months 32.9 27.7 38.5 575 25.3 20.6 30.6 655 -7.6
6-23 months 22 18.6 25.7 845 17.6 14.1 21.6 958 -4.4
24-59 months 29.1 25.7 32.8 1713 22.2 18.5 26.4 1916 -6.9
Note: n=Weighted number of children (denominator)
Malaria Impact Evaluation Final Report 6/25/2019 33
Table 7. Prevalence of Severe Anaemia in Children Aged 6-59 Months in Ghana
Background 
Characteristic
2003 DHS 2008 DHS 2014 DHS MIS 2016 Percentage 
Point 
Change 
% LCI UCI n % LCI UCI n % LCI UCI n % LCI UCI n 2003-2016
Total (National) 14.3 12.8 15.9 2992 19.1 17 21.4 2342 8.3 7 9.9 2568 6.9 5.4 8.7 2874 -7.4
Residence
Urban 9 7.1 11.3 984 13.1 10.4 16.4 894 4.4 3 6.6 1180 4.1 2.2 7.4 1276 -4.9
Rural 16.9 15 18.9 2008 22.8 20 25.9 1448 11.6 9.7 13.9 1388 9.1 7.1 11.7 1598 -7.8
Region
Western 15.4 11.1 21.2 293 26.5 19.3 35.3 221 8 5.3 11.8 273 3.9 1.6 9.3 213 -11.5
Central 14.9 10.8 20.2 267 19.2 14 25.7 225 10.7 6.6 17 304 14 8 23.4 281 -0.9
Greater Accra 8 4.7 13.1 324 6 3 11.9 267 4.2 1.4 12.1 389 1.3 0.3 4.6 406 -6.7
Volta 10.2 6.6 15.6 255 16.7 11.5 23.7 203 8.4 5.8 12 189 8.7 5.5 13.5 217 -1.5
Eastern 11.8 8.4 16.4 292 11.2 6.9 17.8 211 5.8 3.3 9.8 238 8.6 5.1 14.1 224 -3.2
Ashanti 14.6 10.8 19.3 553 20.8 14.9 28.2 460 5 2.6 9.5 432 3.7 1 12.9 656 -10.9
Brong-Ahafo 16.8 12.8 21.8 333 19.1 13.6 26.3 250 6.4 3.7 10.8 260 4.4 2 9.4 233 -12.4
Northern 20.1 15.3 26 403 27.8 22.7 33.5 332 18.2 14 23.4 313 12.4 9.1 16.7 464 -7.7
Upper West 11.7 8.1 16.6 86 31.2 24.5 38.8 63 16.5 11.1 23.8 66 9.1 5.9 13.8 75 -2.6
Upper East 15 10.1 21.6 186 15 10.3 21.3 109 6.7 4.1 10.7 105 7.4 4.9 11 105 -7.6
Wealth Quintile
Lowest 19.9 17 23 774 26.2 22.5 30.2 585 15.8 12.7 19.6 588 12.1 9.8 14.8 645 -7.8
Second 16.5 13.5 19.9 660 23 19 27.5 543 12.6 9.7 16.2 530 12.1 7.8 18.3 593 -4.4
Middle 13.7 11 17 597 20.4 16.4 25.2 425 7.1 4.5 10.9 523 5 3.2 7.8 581 -8.7
Fourth 11.3 8.3 15.1 521 13.8 10.4 18.2 463 3.2 1.8 5.6 483 2.6 1.5 4.8 588 -8.7
Highest 5.4 3.3 8.6 441 5.7 3.4 9.4 326 0.3 0 1.9 445 0.7 0.2 2.8 466 -4.7
Age (in months)
6-11 months 20.8 16.3 26.2 348 28.4 22.7 34.9 245 9.6 6.4 14 260 10 6.2 15.7 310 -10.8
12-23 months 23.1 19.7 26.8 661 24.3 20.5 28.5 521 14 10.6 18.2 587 11.8 9 15.5 648 -11.3
24-35 months 13.6 10.8 16.9 635 23.7 19.4 28.5 492 7.3 5.3 9.9 573 6.3 4.5 8.7 670 -7.3
36-47 months 10.1 8 12.7 716 12.7 9.5 16.8 517 8.5 6 11.9 570 4.5 2.8 7.1 591 -5.6
48-59 months 6.9 5.1 9.2 632 12.2 9.4 15.6 566 2.9 1.8 4.6 578 3.2 1.8 5.9 655 -3.7
6-23 months 22.3 19.5 25.4 1009 25.6 22.2 29.3 767 12.6 9.8 16.1 847 11.3 8.5 14.8 958 -11
24-59 months 10.2 8.8 11.8 1983 15.9 13.7 18.4 1575 6.2 5 7.7 1721 4.7 3.6 6.1 1916 -5.5
Note: n=Weighted number of children (denominator)
Malaria Impact Evaluation Final Report 6/25/2019 34
Figure 1 Changes in ACCM vs. Malaria Coverage