-
www.ACTwatch.info Copyright 2012 Population Services
International (PSI). All rights reserved.
Evidence for Malaria Medicines Policy
ACTwatch Baseline and Endline
Household Survey Results 2009-2012: Benin, Democratic Republic
of the Congo,
Madagascar, Nigeria, Uganda, Zambia
-
www.ACTwatch.info i|Page www.ACTwatch.info P a g e | i
Acknowledgements
ACTwatch is funded by the Bill and Melinda Gates Foundation.
This study was implemented by Population Services International
(PSI).
ACTwatchs Advisory Committee:
Mr. Suprotik Basu Advisor to the UN Secretary General's Special
Envoy for Malaria
Mr. Rik Bosman Supply Chain Expert, Former Senior Vice
President, Unilever
Ms. Renia Coghlan Global Access Associate Director, Medicines
for Malaria Venture (MMV)
Dr. Thom Eisele Assistant Professor, Tulane University
Mr. Louis Da Gama Malaria Advocacy & Communications
Director, Global Health Advocates
Dr. Paul Lalvani Executive Director, RaPID Pharmacovigilance
Program
Dr. Ramanan Laxminarayan
Senior Fellow, Resources for the Future
Dr. Matthew Lynch Malaria Program Director, VOICES, Johns
Hopkins University Centre for Communication Programs
Dr. Bernard Nahlen Deputy Coordinator, President's Malaria
Initiative (PMI)
Dr. Jayesh M. Pandit Head, Pharmacovigilance Department,
Pharmacy and Poisons Board-Kenya
Dr. Melanie Renshaw Chief Technical Advisor, ALMA
Mr. Oliver Sabot Vice President, Vaccines Clinton Foundation
Ms. Rima Shretta Senior Program Associate, Strengthening
Pharmaceutical Systems Program, Management Sciences for Health
Dr. Rick Steketee Science Director, Malaria Control and
Evaluation Partnership in Africa (MACEPA)
Dr. Warren Stevens Health Economist
Dr. Gladys Tetteh Deputy Director Country Programs, Systems for
Improved Access to Pharmaceuticals and Services, Management
Sciences for Health
Prof. Nick White, OBE Professor of Tropical Medicine, Mahidol
and Oxford Universities
Prof. Prashant Yadav Director-Healthcare Delivery Research and
Senior Research Fellow, William Davidson Institute, University of
Michigan
Dr. Shunmay Yeung Paediatrician & Senior Lecturer, LSHTM
-
www.ACTwatch.info ii|Page www.ACTwatch.info P a g e | ii
The ACTwatch Group is comprised of the following
individuals:
PSI ACTwatch Central Dr. Kathryn OConnell, Principal
Investigator;
Hellen Gatakaa, Senior Research Associate; Stephen Poyer, Illah
Evance, Julius Njogu, Research Associates; Meghan Bruce, Policy
Advocate and Communications Specialist, Linda Ongwenyi, ACTwatch
Project Assistant, Tanya Shewchuk, Project Director.
PSI ACTwatch Country Program Coordinators
Cyprien Zinsou, ABMS/Benin; Sochea Phok, PSI/Cambodia; Dr. Louis
Akulayi, SFH/DRC; Jacky Raharinjatovo, PSI/Madagascar; Ekundayo
Arogundade, SFH/Nigeria; Peter Buyungo, PACE/Uganda; Felton
Mpasela, SFH/Zambia.
LSHTM Dr. Kara Hanson, Principal Investigator;
Dr. Catherine Goodman, Benjamin Palafox, Edith Patouillard,
Sarah Tougher, Immo Kleinschmidt, co-investigators. LSHTM is
responsible for the supply chain research component of
ACTwatch.
In Benin, the DRC, Madagascar, Nigeria, Uganda, and Zambia:
Association Bninoise pour le Marketing Social (ABMS) - Benin
Cyprien Zinsou Sethson Kassegne Ghyslain Guedegbe Stephen Poyer
Association de Sant Familiale (ASF) - DRC Dr. Louis Akulayi
Tshisungu Willy Mpwate Nsuele Stephen Poyer Lanto Razafindralambo
Population Services International (PSI) - Madagascar Iarimalanto
Rabary Jacky Raharinjatovo Dr. Justin Rahariniaina Sitraka
Ramamonjisoa Angela Camilleri Julius Njogu Society for Family
Health (SFH) - Nigeria Dr. Jennifer Anyanti Dr. Samson Adebayo
Ekundayo D. Arogundade Tsione Solomon Julie Archer
-
www.ACTwatch.info iii|Page www.ACTwatch.info P a g e | iii
Program for Accessible Health, Communication & Education
(PACE) - Uganda Peter Buyungo Dr. Simon Sensalire Dr. Susan
Kambabazi Emily Carter
Mitsuru Toda Society for Family Health (SFH) - Zambia Mr. Felton
Mpasela Nicholas Shiliya Edward Ngoma Illah Evance Additional
support for household survey implementation and/or analysis was
provided by: Dr. Megan Littrell, PSI Research Technical Advisor Dr.
Adam Bennett, Tulane University Cristin Fergus, Tulane University
Rebecca Geary, London School of Hygiene and Tropical Medicine
-
www.ACTwatch.info iv|Page www.ACTwatch.info P a g e | iv
Suggested citation: ACTwatch Group, ABMS/Benin, ASF/DRC,
PSI/Madagascar, SFH/Nigeria, PACE/Uganda, and SFH/Zambia. 2013.
ACTwatch Baseline and Endline Household Survey Results 2009-2012:
Benin, Democratic Republic of Congo, Madagascar, Nigeria, Uganda,
Zambia. Washington, DC: Population Services International.
Available at: www.actwatch.info For further information
contact:
Association Bninoise pour le Marketing Social (ABMS) Cyprien
Zinsou Lot 919 Immeuble Montcho Sikcodji - Cotonou Republic of
Benin Phone: + 229 21327713/14 Email: [email protected]
Association de Sant Familiale (ASF) - DRC Dr. Louis Akulayi
Tshisungu 232, Avenue Tombalbaye Croisement Kasa Democratic
Republic of Congo Phone: +243 998 23 96 80 Email:
[email protected]
PSI - Madagascar Mr. Jacky Raharinjatovo ACTwatch Country
Program Coordinator Immeuble Fiaro Ampefiloha BP 7748 Antananarivo
Madagascar Phone: + 261 202262984
Email: [email protected]
Society for Family Health (SFH) - Nigeria Dr. Jennifer Anyanti
Research Director Society for Family Health/Nigeria 8, Portharcourt
Crescent, Area 11, Garki Abuja, Nigeria Phone: + 234 8033116065
Email: [email protected]
PACE - Uganda Mr. Peter Buyungo Director, Research Department
Uganda Plot 2 Ibis Vale, Kololo P.O. Box 27659, Kampala Phone: +256
312-351100 / +256 414-230080 Email: [email protected]
Society for Family Health (SFH) - Zambia Mr. Felton Mpasela
ACTwatch Country Program Coordinator Plot 549, Ituna Road P.O. Box
50770 Lusaka, Zambia Phone: + 260 211257407 Email:
[email protected]
ACTwatch Central Tanya Shewchuk ACTwatch Project Director
Malaria Control & Child Survival Department Population Services
International P.O. Box 14355-00800 Nairobi, Kenya Phone: +254 20
4440125/6/7/8 Email: [email protected]
http://www.actwatch.info/mailto:[email protected]:[email protected]
-
www.ACTwatch.info v|Page www.ACTwatch.info P a g e | v
Table of Contents
TABLE OF CONTENTS
.......................................................................................................................................
V
LIST OF TABLES
..............................................................................................................................................
VI
LIST OF FIGURES
...........................................................................................................................................
VIII
DEFINITIONS
..................................................................................................................................................
IX
ABBREVIATIONS
............................................................................................................................................
XI
1. EXECUTIVE SUMMARY
.........................................................................................................................
XII
1.1. OVERVIEW OF ACTWATCH
........................................................................................................................
XII 1.2. HOUSEHOLD SURVEY METHODS
.................................................................................................................
XII 1.3. HOUSEHOLD SURVEY DESCRIPTION
............................................................................................................
XIII 1.4. HOUSEHOLD SURVEY KEY FINDINGS
............................................................................................................
XIV
2. BACKGROUND
........................................................................................................................................
1
2.1. OVERVIEW OF THE ACTWATCH RESEARCH PROJECT
........................................................................................
1 2.2. AFFORDABLE MEDICINES FACILITY FOR MALARIA
............................................................................................
1 2.3. COUNTRY BACKGROUNDS
...........................................................................................................................
3
3. HOUSEHOLD SURVEY METHODS
...........................................................................................................
15
3.1. INTRODUCTION
......................................................................................................................................
15 3.2. SAMPLING
.............................................................................................................................................
16 3.3. DATA COLLECTION
..................................................................................................................................
19 3.4. DATA QUALITY ASSURANCE
.......................................................................................................................
19 3.5. DATA PROCESSING AND ANALYSIS
..............................................................................................................
20 3.6. ETHICAL APPROVAL
.................................................................................................................................
20
4. HOUSEHOLD SURVEY CONTEXTUAL
INFORMATION..............................................................................
21
4.1. INTRODUCTION
......................................................................................................................................
21 4.2. METHODS
.............................................................................................................................................
21
5. HOUSEHOLD SURVEY RESULTS
.............................................................................................................
22
5.1. DESCRIPTION OF SAMPLES AND CHARACTERISTICS OF HOUSEHOLDS
...................................................................
22 5.2. TREATMENT OF FEVER
.............................................................................................................................
30 5.3. DIAGNOSTICS
.........................................................................................................................................
36 5.4. TYPE OF ANTIMALARIALS TAKEN AND SOURCE OF ANTIMALARIALS
.....................................................................
39 5.5 INITIAL SOURCES OF CARE
.........................................................................................................................
58 5.6 AWARENESS AND EXPOSURE TO THE AMFM
.....................................................................................................
61
6. DISCUSSION
..........................................................................................................................................
69
6.1 LIMITATIONS
..............................................................................................................................................
70 6.2 FURTHER RESEARCH
.....................................................................................................................................
70
7. REFERENCES
.........................................................................................................................................
71
8. ACKNOWLEDGEMENTS
.........................................................................................................................
77
9. APPENDICES
.........................................................................................................................................
78
9.1. TREATMENT OF CHILDREN WITH FEVER IN THE POOREST HOUSEHOLDS
AT ENDLINE ............................................... 78 9.2.
TIMING OF DATA COLLECTION FOR HOUSEHOLD SURVEYS AND ARRIVAL OF
COPAID (AMFM) ACTS BY COUNTRY ....... 79 9.3. TREATMENT OF FEVER
AMONG CHILDREN UNDER FIVE YEARS BY SELECTED BACKGROUND
CHARACTERISTICS .............. 80
-
www.ACTwatch.info vi|Page www.ACTwatch.info P a g e | vi
List of Tables Table 1.1. 1: Description of the baseline and
endline household surveys
...........................................................
xiii
Table 2.3. 1: Timing of data collection for household surveys
and arrival of copaid ACTs in Madagascar ............ 7 Table
2.3.2: Timing of data collection for household surveys and arrival
of copaid ACTs in Nigeria ..................... 9 Table 2.3.3:
Timing of data collection for household surveys and arrival of
copaid ACTs in Uganda .................. 12
Table 3.1. 1: Timing of data collection for household surveys
.............................................................................
15
Table 3.2. 1: Definition of cluster and enumerations areas, by
country
.............................................................. 16
Table 3.2.2: Stratification, by country
..................................................................................................................
17 Table 3.2.3: Number of EAs selected by strata for each country
.........................................................................
18
Table 5.1.1: Results of the household and individual interviews
.........................................................................
22 Table 5.1. 2: Demographic characteristics,
Benin.................................................................................................
24 Table 5.1. 3: Demographic characteristics, DRC
...................................................................................................
25 Table 5.1. 4: Demographic characteristics, Madagascar
......................................................................................
26 Table 5.1. 5: Demographic characteristics, Nigeria
..............................................................................................
27 Table 5.1. 6: Demographic characteristics, Uganda
.............................................................................................
28 Table 5.1. 7: Demographic characteristics, Zambia
..............................................................................................
29
Table 5.2.1: Treatment of children with fever
......................................................................................................
30 Table 5.2. 2: Prompt treatment of children with fever
........................................................................................
33
Table 5.3.1: Diagnosis of fever among children under five 36
Table 5.4.1: Treatment with ACT among children under
five...............................................................................
39 Table 5.4. 2: Treatment with ACT with the AMFm logo among
children under five ............................................ 41
Table 5.4. 3: Prompt treatment with ACT among children under five
.................................................................
43 Table 5.4. 4: Prompt treatment with ACT with the AMFm logo among
children under five ............................... 45 Table 5.4.
5: Treatment with ACT among children under five who received an
antimalarial .............................. 47 Table 5.4. 6: Source
of antimalarials
....................................................................................................................
49 Table 5.4. 7: Source of ACTs
.................................................................................................................................
52 Table 5.4. 8: Source of ACTs with the AMFm logo, among children
under five with fever who received an AMFm
logo ACT
......................................................................................................................................................
55
Table 5.5. 1: Care seeking behaviour: first place to seek
care..............................................................................
58
Table 5.6. 1: Caregiver awareness of and exposure to the AMFm
logo and initiative (endline only) 61 Table 5.6. 2: Sources of
exposure to the AMFm logo 62 Table 5.6. 3: Sources of exposure to
the AMFm initiative 63 Table 5.6. 4: Meaning of the AMFm logo 64
Table 5.6. 5: Knowledge of the recommended price for AMFm medicine
65 Table 5.6. 6: Knowledge of the use of AMFm medicine 66 Table
5.6. 7: Caregiver use of ACTs with the AMFm logo 67 Table 5.6. 8:
Source of ACTs with the AMFm logo 68
Table 5.6. 1: Caregiver awareness of and exposure to the AMFm
logo and initiative (endline only) .................. 61 Table 5.6.
2: Sources of exposure to the AMFm logo
...........................................................................................
62 Table 5.6. 3: Sources of exposure to the AMFm initiative
....................................................................................
63 Table 5.6. 4: Meaning of the AMFm logo
.............................................................................................................
64 Table 5.6. 5: Knowledge of the recommended price for AMFm
medicine
........................................................... 65
Table 5.6. 6: Knowledge of the use of AMFm medicine
.......................................................................................
66
-
www.ACTwatch.info vii|Page www.ACTwatch.info P a g e | vii
Table 5.6. 7: Caregiver use of ACTs with the AMFm logo
.....................................................................................
67 Table 5.6. 8: Source of ACTs with the AMFm
logo................................................................................................
68
-
www.ACTwatch.info viii|Page www.ACTwatch.info P a g e | viii
List of Figures Figure 1: Diagnosis of fever among children
under five
.......................................................................................
xiv Figure 2: Treatment with any antimalarial among children under
five
................................................................
xiv Figure 3: Treatment with Artemisinin-based Combination Therapy
(ACT) among children under five with fever
....................................................................................................................................................................
xv Figure 4: Treatment with ACT among children under five who
received an antimalarial .................................... xv
Figure 5: Source of antimalarials among children under five who
received an antimalarial ............................... xvi Figure
6: Source of ACTs among children under five who received an ACT
......................................................... xvi
Figure 7: Caregiver awareness of and exposure to the AMFm logo and
initiative .............................................. xvii
Figure 2.3. 1: Location of the AMFm Phase 1 countries
.........................................................................................
3
-
www.ACTwatch.info ix|Page www.ACTwatch.info P a g e | ix
Definitions Antimalarial Any medicine recognized by the WHO for
the treatment of malaria.
Affordable Medicines Facility malaria (AMFm)
The AMFm is a financing mechanism designed to increase the
provision of affordable ACTs through the public, private
not-for-profit (e.g., NGO) and private for-profit sectors. The AMFm
is being evaluated in a first phase that includes 9 pilots in 8
countries: Cambodia, Ghana, Kenya, Madagascar, Niger, Nigeria,
Republic of Tanzania (mainland and Zanzibar) and Uganda.
Artemisinin-based Combination Therapy (ACT)
An antimalarial that combines artemisinin or one of its
derivatives with an antimalarial or antimalarials of a different
class. Refer to Combination Therapy (below).
AMFm logo All AMFm co-paid ACT packaging bears a logo (the "ACTm
leaf logo") to facilitate communication campaigns and product
identification. The logo is applied to all quality-assured ACTs
purchased through the AMFm.
Artemisinin monotherapy An antimalarial medicine that has a
single active compound, where this active compound is artemisinin
or one of its derivatives.
Cluster The primary sampling unit in the multi-stage sample
drawn for the household survey. For each ACTwatch country the
household survey clusters are the same units as those selected for
sampling in the Outlet Survey. The clusters were defined as follows
in each of the countries:
Benin: arrondissements DRC: aires de sant Madagascar: communes
Nigeria: localities Uganda: subcounties (parishes in Kampala)
Zambia: wards
Combination therapy
The use of two or more classes of antimalarial drugs/molecules
in the treatment of malaria that have independent modes of
action.
Dosing/treatment regimen The posology or timing and number of
doses of an antimalarial used to treat malaria. This schedule often
varies by patient weight.
Enumeration Area (EA) The secondary sampling unit for the
household survey. It is an administrative unit that generally has a
population size of 250-500. These units frequently are defined by
geographical, health or political boundaries. The enumeration areas
were defined as follows in each of the countries:
Benin: census enumeration areas (most often villages/quartiers)
DRC: villages in rural areas and quartiers in urban areas
Madagascar: fokontany Nigeria: supervisory areas (SAs) Uganda:
villages
Zambia: standard enumeration areas (SEAs)
-
www.ACTwatch.info x|Page www.ACTwatch.info P a g e | x
First-line treatment The government recommended treatment for
uncomplicated malaria. The first-line treatment is as follows in
each of the countries:
Benin: artemether-lumefantrine, 20mg/120mg DRC:
artesunate-amodiaquine, 50mg/153mg Madagascar:
artesunate-amodiaquine, 50mg/153mg Uganda: artemether-lumefantrine,
20mg/120mg Nigeria: artemether-lumefantrine, 20mg/120mg and
artesunate-amodiaquine, 50mg/153mg Zambia: artemether-lumefantrine,
20mg/120mg
Monotherapy An antimalarial medicine that has a single mode of
action. This may be a medicine with a single active compound or a
synergistic combination of two compounds with related mechanisms of
action. Antimalarial monotherapies include amodiaquine, quinine,
chloroquine, and sulphadoxine-pyrimethamine.
Non-artemisinin therapy An antimalarial treatment that does not
contain artemisinin or any of its derivatives.
Oral artemisinin monotherapy
Artemisinin or one of its derivatives in a dosage form with an
oral route of administration. These include tablets, suspensions,
and syrups and exclude suppositories and injections, which are used
in the treatment of severe malaria.
Rapid Diagnostic Test (RDT) for malaria
Malaria rapid diagnostic tests, sometimes called "dipsticks" or
malaria rapid diagnostic devices, assist in the diagnosis of
malaria by providing evidence of the presence of malaria parasites
in human blood. RDTs do not require laboratory equipment, and can
be performed and interpreted by non-clinical staff.
Screened A household that was administered the screening
questions (1.11 and 1.11a at baseline, and 1.11 and 1.12 at
endline) of the household survey questionnaire.
Screening criteria The set of requirements that must be
satisfied before the full questionnaire is administered. In this
survey a household met the screening criteria if it included a
child under five who had experienced fever in the two weeks prior
to the interview. In addition, at endline a series of questions to
capture awareness of the AMFm was administered to 1) caregivers of
children under five in households that did not meet the main
screening criteria, and 2) any caregivers of children under five
with no reported fever in households that met the screening
criteria.
Treatment/dosing regimen The posology or timing and number of
doses of an antimalarial used to treat malaria. This schedule often
varies by patient weight.
-
www.ACTwatch.info xi|Page www.ACTwatch.info P a g e | xi
Abbreviations ACT Artemisinin-based Combination Therapy
AL Artemether-lumefantrine
AMFm Affordable Medicines Facility malaria
ASAQ Artesunate-amodiaquine
CHW Community Health Worker
CI Confidence Interval
CQ Chloroquine
DfID Department for International Development
DHA-PPQ Dihydroartemisinin-piperaquine
DHS Demographic and Health Survey
DRC Democratic Republic of Congo
EA Enumeration Areas
Global Fund Global Fund to Fight AIDS, Tuberculosis, and
Malaria
GPS Global Positioning System
IMF International Monetary Fund
IPTp Intermittent Preventive Treatment in pregnancy
IQR Interquartile Range
IRS Indoor Residual Spraying
ITN Insecticide Treated Net
LGA Local Government Authority
LLIN Long-lasting Insecticidal Net
LSHTM London School of Hygiene and Tropical Medicine
MIS Malaria Indicator Survey
MOH Ministry of Health
nAT Non-artemisinin Therapy
NGO Non-governmental organization
NMCP National Malaria Control Program
NMS National Medical Stores
PMI Presidents Malaria Initiative
PPS Probability proportional to size
PSI Population Services International
RDT Rapid Diagnostic Test
SP Sulfadoxine-pyrimethamine
UN United Nations
UNICEF United Nations Children Emergency Fund
USAID United States Agency for International Development
USD United States Dollar
WHO World Health Organization
-
www.ACTwatch.info xii|Page www.ACTwatch.info P a g e | xii
1. Executive Summary
1.1. Overview of ACTwatch
The ACTwatch Household Survey is a population-based survey
conducted in each of the seven
ACTwatch countries (Benin, Cambodia, the Democratic Republic of
Congo [DRC], Madagascar,
Nigeria, Uganda and Zambia). The survey is one of three ACTwatch
research components, and is led
by Population Services International (PSI). The other elements
of ACTwatch research are Outlet
Surveys led by PSI (OConnell et al., 2011) and Supply Chain
Research led by the London School of
Hygiene & Tropical Medicine (LSHTM) (Shewchuk et al.,
2011).
The objective of the household survey component is to monitor
consumer treatment-seeking
behavior for fever in children under the age of five, including
types of medicines (specifically,
antimalarials) and diagnoses obtained, sources visited for
advice, treatment and diagnosis, the price
paid for antimalarials, and caregiver knowledge and awareness of
fever symptoms and antimalarials
(Littrell et al., 2011).
1.2. Household Survey Methods
This study uses data from cross-sectional household surveys of
childrens caregivers. Nationally-
representative samples of households in Benin, the DRC,
Madagascar, Nigeria, Uganda and Zambia
were drawn using multi-staged cluster sampling. Stratification
and sampling varied by country. In
Madagascar and Zambia, stratification was conducted with urban
and rural domains. In the DRC the
sample was stratified across four geographic regions and in
Nigeria across six geo-political regions. In
Uganda, two strata were included for low/moderate and high
malaria transmission areas. In Benin, a
single domain was used. At endline in Uganda and Nigeria
stratification was by urban and rural
areas. Estimated fever prevalence was used to identify the
number of households necessary to
screen in order to yield the desired number of children with
fever.
All caregivers with a child under five who had experienced fever
in the two weeks prior to interview
were eligible for inclusion and were asked questions about their
treatment of the recent fever
episode. In addition, in Madagascar, Nigeria and Uganda at
endline all caregivers of children under
five, irrespective of a childs fever status, were eligible for a
subset of questions about the Affordable
Medicines Facility malaria (AMFm).
The household survey was designed to monitor all key malaria
treatment indicators, as well as
additional indicators addressing sources of treatment for fever,
antimalarials and diagnostics.
Validation and data checking steps occurred during and after
data collection. Double data entry was
conducted using Microsoft Access (Microsoft Cooperation,
Seattle, WA, USA). Stata 11 (Stata Corp,
College Station, TX) was used for all analyses. To obtain the
national estimates provided in this
report, data were weighted.
-
www.ACTwatch.info P a g e | xiii
1.3. Household Survey Description
The survey dates, number of households enumerated, and number of
children under five included in the analysis is presented in table
1.1.1.
Table 1.1.1: Description of the baseline and endline household
surveys
Benin DRC Madagascar Nigeria Uganda Zambia
Baseline Endline Baseline Endline1 Baseline Endline Baseline
Endline Baseline Endline Baseline Endline
Survey dates April 15
th
May 08th
2009
June 24th
July 10
th
2011
April 27th
June 21
st
2010
Dec1st
2008 - January 12
th
2009
April 23rd
- June 10
th
2012
August 11th
Sept 23
rd
2009
June 18th
July 28
th
2012
March 16th
-April 20th
2009
April 19th
May 11
th
2012
April 14th
- July 4
th
2009
May 25th
- Sept 18
th
2011
Households selected 2,352 3,322 8,426 n/a 5,085 15,195 4,649
8,258 2,064 5,306 2,234 8,998
Households screened 2,312 3,275 8,369 n/a 4,394 14,026 4,616
7,863 2,603 5,067 2,208 8,207
Children under five with fever for whom treatment-seeking
behaviour was recorded.
927 1,015 2,665 n/a 2,120 2,388 3,273 1,551 1,752 2,273 1,886
1,327
Caregivers completing the AMFm section
2.
n/a n/a n/a n/a n/a 6,781 n/a 3,681 n/a 3,193 n/a n/a
1 In the DRC, a household follow-up survey was not
conducted.
2 The AMFm section was only administered at endline for the
pilot AMFm countries (Madagascar, Nigeria and Uganda).
Source: ACTwatch Group et al., 2013. ACTwatch Baseline and
Endline Household Survey Results 2009-2012: Benin, Democratic
Republic of the Congo, Madagascar, Nigeria, Uganda, and Zambia.
-
www.ACTwatch.info xiv|Page www.ACTwatch.info P a g e | xiv
1.4. Household Survey Key findings
Data collection took place during the annual rainy season in
2009 for the baseline surveys (except for
DRC, which took place in 2010), and 2012 for endline surveys in
all countries except Benin and
Zambia (2011) and DRC where no follow-up survey was conducted.
This section provides an
overview of the survey results over time.
Figure 1: Diagnosis of fever among children under five
Diagnostic testing of children with fever remained low across all
countries, ranging between 5% in Benin to
37% in Zambia at endline. However, slight increases were
observed in all countries with Madagascar and
Uganda almost doubling the proportion of children with fever who
receive diagnosis: 7% to 12% and 10% to
18% respectively.
Figure 2: Treatment with any antimalarial among children under
five The percentage of children with fever who received any
antimalarial ranged from 19% in Madagascar to 54% in
Uganda at endline with slight increases observed between
baseline and endline in Benin, Nigeria, Uganda and
Zambia. In Madagascar, there was a decrease from 47% at baseline
to 19% at endline with the decrease
observed in both urban and rural and across all wealth quintiles
(data not shown).
4
14
7 6 10
34
5 n/a
12 7
18
37
0
10
20
30
40
50
60
70
80
90
100
Benin DRC Madagascar Nigeria Uganda Zambia
Baseline Endline
42 40
47
32
50
39
51
n/a
19
37
54
41
0
10
20
30
40
50
60
70
80
90
100
Benin DRC Madagascar Nigeria Uganda Zambia
Baseline Endline
-
www.ACTwatch.info xv|Page www.ACTwatch.info P a g e | xv
Figure 3: Treatment with Artemisinin-based Combination Therapy
(ACT) among children under five with fever The percentage of
children with fever who received an ACT increased over time in all
countries, with increases
evident in both urban and rural areas. In Benin, there was a
substantial increase in use from 10% to 28%
between 2009 and 2011. In Uganda, 44% of children with fever
received an ACT at endline compared to 20% at
baseline. Overall levels of treatment were lower in Madagascar
and Nigeria, but the percentage of children
with fever receiving an ACT more than doubled in both countries
(Madagascar 3% to 8%; Nigeria 4% to 11%).
Figure 4: Treatment with ACT among children under five who
received an antimalarial Treatment with ACT among children under
five who received any antimalairal increased substantially in
all
countries, in both urban and rural areas. Absolute increases
were lowest in Nigeria (19% to 33% urban; 10% to
28% rural) and greatest in Uganda (24% to 75% urban; 41% to 84%
rural). In Zambia at endline, 90% of children
in rural areas and 75% of children in urgan areas who received
any antimalarial got an ACT. Overall levels at
endline were lowest in Nigeria and Madagascar.
10
4 3 4
20 21
28
n/a
8 11
44
35
0
10
20
30
40
50
60
70
80
90
100
Benin DRC Madagascar Nigeria Uganda Zambia
Baseline Endline
29
20
9 11 10 7
19
10
24
41
50
56
47
58
n/a n/a
37
45
33 28
75
84
75
90
0
10
20
30
40
50
60
70
80
90
100
Urban Rural Urban Rural Urban Rural Urban Rural Urban Rural
Urban Rural
Benin DRC Madagascar Nigeria Uganda Zambia
Baseline Endline
-
www.ACTwatch.info xvi|Page www.ACTwatch.info P a g e | xvi
Figure 5: Source of antimalarials among children under five who
received an antimalarial The public/not-for-profit sector remained
the main source of antimalarials in Benin (for 49% of children
who
received any antimalarial) and Zambia (80%) while in Uganda and
Nigeria, the private sector was the main
source of antimalarials at both baseline and endline. In
Madagascar, there was a decrease in percentage of
children receiving an antimalarial from the private sector (61%
to 39%) and an increase in the public/not-for-
profit sector (27% to 46%).
Figure 6: Source of ACTs among children under five who received
an ACT The public/not-for-profit sector played an important role in
ACT distribution among children under five with
fever in Benin, Madagascar and Zambia at both baseline and
endline while the private sector remained the
main source of ACTs in Nigeria. The percentage of children
receiving ACTs from the private sector increased
from 29% to 42% in Uganda although close to 40% still received
ACTs from the public not-for-profit sector.
36
49
33
n/a
27
46
25 22
26
36
83 80
27 27
62 61
39
51
67
51 47
11 13
0
10
20
30
40
50
60
70
80
90
100
Baseline Endline Baseline Endline Baseline Endline Baseline
Endline Baseline Endline Baseline Endline
Benin DRC Madagascar Nigeria Uganda Zambia
Public/not for profit sector Private sector
59 62
44
n/a
49
62
30
41 44
39
89
84
19 22
52
48
29
45
50
29
42
8 10
0
10
20
30
40
50
60
70
80
90
100
Baseline Endline Baseline Endline Baseline Endline Baseline
Endline Baseline Endline Baseline Endline
Benin DRC Madagascar Nigeria Uganda Zambia
Public/not for profit sector Private sector
-
www.ACTwatch.info xvii|Page www.ACTwatch.info P a g e | xvii
Figure 7: Caregiver awareness of and exposure to the AMFm logo
and initiative Less than 50% of caregivers of children under five,
regardless of the childs fever status, had seen or heard of
the AMFm logo or initiative in the three AMFm pilot countries.
Among those who had seen the logo, less than
10% mention it is a cheaper/effective antimalarial and only 8%
in Nigeria and 12% in Uganda said there was a
recommended price for the logo drug (data not shown).
39
20
13 18
9
16
48
26 26
0
10
20
30
40
50
60
70
80
90
100
Uganda Nigeria Madagascar
Seen or heard of the AMFm logo
Heard of initiative to reduce the price of ACTs
Either seen/heard of the AMFm logo or initiative
-
www.ACTwatch.info 1|Page www.ACTwatch.info P a g e | 1
2. Background
2.1. Overview of the ACTwatch Research Project
In 2008, Population Services International (PSI) in partnership
with the London School of Hygiene
and Tropical Medicine (LSHTM) launched a five-year multi-country
research project called ACTwatch
(Shewchuk et al., 2011). The project is designed to provide a
comprehensive picture of the
antimalarial market to inform national and international
antimalarial drug policy evolution. The
research is designed to detect changes in the availability,
price and use of antimalarials over time
and between sectors, and to monitor the effects of policy or
intervention developments at country
level.
ACTwatch addresses both the supply and demand sides of the
market. The supply side is evaluated
by collecting level and trend data on antimalarials and rapid
diagnostic tests (RDTs) in public and
private sector outlets and wholesalers of antimalarial drugs
(OConnell et al., 2011). To evaluate
demand, data are collected at the household level on consumer
treatment-seeking behavior and
knowledge. In combination, the research components thread
together the antimalarial market and
consumer behavior (Littrell et al., 2011). Findings can help
determine where and to what extent
interventions may positively impact access to and use of
quality-assured ACTs and RDTs as well as
resistance containment efforts.
The first phase of the project was conducted in seven
malaria-endemic countries between 2008 and
2012: Benin, Cambodia, the Democratic Republic of Congo,
Madagascar, Nigeria, Uganda and
Zambia. Countries were selected with the aim of studying a
diverse range of markets from which
comparisons and contrasts could be made.
Information on other ACTwatch studies can be found at
www.ACTwatch.info.
2.2. Affordable Medicines Facility for Malaria
The success of malaria control efforts depends on a high level
of coverage in the use of effective
antimalarials such as artemisinin-based combination therapies
(ACTs). Although these antimalarials
have been procured in large amounts by countries, evidence
suggests that ACT use still remains far
below target levels. Reasons suggested for the low uptake of
ACTs include interruptions in public
sector supply; limited availability outside major urban centres;
the high prices of the drugs,
particularly in the private sector; lack of provider adherence
to new recommendations; and patient
self-treatment with other more common and cheaper antimalarials
(Sabot et al., 2009). Lowering
the cost of ACTs to the end user through a subsidy mechanism
could be an effective way to increase
their uptake (Arrow et al., 2004).
In response to this issue, the Affordable Medicines Facility
malaria (AMFm) was established,
hosted by The Global Fund to Fight AIDS, Tuberculosis and
Malaria (Global Fund). As described by
Adeyi and Atun (2010), AMFm is a financing mechanism designed to
incorporate three elements: (1)
price reductions through negotiations with manufacturers of
ACTs; (2) a buyer subsidy, via a co-
http://www.actwatch.info/
-
www.ACTwatch.info 2|Page www.ACTwatch.info P a g e | 2
payment at the top of the global supply chain by AMFm on behalf
of eligible buyers from the public,
private for-profit and private not-for-profit sectors; and (3)
support of interventions to promote
appropriate use of ACTs. Examples of these supporting
interventions include training providers
and outreach to communities to promote ACT utilization. AMFm was
tested in a first phase that
includes nine pilots in eight countries: Cambodia, Ghana, Kenya,
Madagascar, Niger, Nigeria,
Republic of Tanzania (mainland and Zanzibar) and Uganda.
The AMFm pilot was being evaluated on the outcome of four
components related to the availability,
affordability, market share and use of ACTs. These questions are
formulated as follows:
1. Has the AMFm mechanism helped increase the availability of
quality-assured ACTs to
patients across public, private for-profit and not-for-profit
sectors, in rural/urban areas?
2. Has the AMFm mechanism helped to reduce the cost of
quality-assured ACTs to patients at
public, private for-profit and not-for-profit outlets in
rural/urban areas to a price comparable
to the price of most popular antimalarials?
3. Has the AMFm mechanism helped increase use of quality-assured
ACTs, including among
vulnerable groups, such as poor people, rural residents and
children?
4. Has the AMFm mechanism helped increase the market share of
quality-assured ACTs
relative to all antimalarial treatments in the public, private
for-profit and not-for-profit
sectors in rural/urban areas?
The final independent evaluation report on the AMFm pilot was
released in October 2012,
concluding that the AMFm had had a significant impact in the
private-for-profit sector in six of eight
countries though less impact in the public sector of most. In
November 2012, the Global Fund Board
decided that the AMFm would not continue as a stand-alone
programme but would instead
integrate some of its key elements into the Global Funds core
funding processes. In addition, it was
decided that the pilot phase of the AMFm would be extended for
another year (to the end of 2013)
to allow for a smooth transition and continuity of access to
affordable ACTs. The independent
evaluation and the Global Fund Board based their conclusions in
part on evidence gathered through
ACTwatch outlet and household surveys.
Additional information on the AMFm can be found on the AMFm
website (Global Fund, 2012). More
information on the independent evaluation can be found in the
final report (AMFm Independent
Evaluation Team, 2012a).
-
www.ACTwatch.info 3|Page www.ACTwatch.info P a g e | 3
2.3. Country backgrounds
2.3.1 Overview of ACTwatch Countries ACTwatch undertook
household surveys in seven countries: Benin, Cambodia, DRC,
Madagascar,
Nigeria, Uganda and Zambia (Figure 2.3.1). This trends report
covers the surveys in the six sub-
Saharan countries. An overview of each is given below, and
additional detail is provided in the
individual country-specific household survey reports (available
at www.actwatch.info).
Figure 2.3.1: Location of the ACTwatch study countries
2.3.2 Benin
Benin is located in the West Africa sub-region and is bordered
by Niger and Burkina Faso to the
north, Nigeria to the east and Togo to the west. In 2010 the
population was estimated at 8.8 million,
with 1.5 million children under 5 years of age (Population
Division, 2011). Approximately 60% of
people are live in rural areas (Population Division, 2010). In
the past decade per capita gross
domestic product (GDP) has increased from $1,300 to just over
$1,400; 47% of the population still
live below the standard poverty line of $1.25 a day. In 2011
Benin was ranked 167th out of 187
countries in the Human Development Index (UNDP, 2011). One in
eight children die before reaching
their fifth birthday (INSAE, 2007), and life expectancy at birth
in 2009 was 54 years for men and 60
years for women (WHO, 2011a). Malaria is considered to be the
leading cause of morbidity and
mortality among children under five, accounting for 41% of
outpatient visits and 29% of
-
www.ACTwatch.info 4|Page www.ACTwatch.info P a g e | 4
hospitalizations for this age group (INSAE, 2007). Government
figures for 2010 report 900,000
confirmed malaria cases across all age groups, and a further
350,000 probable cases (WHO, 2010b).
The public health system operates through three tiers, linked to
the three levels of health care and
their associated structures. Healthcare consumers at government
facilities are expected to pay for
consultations, diagnostic tests, procedures and medicines. In
line with the general policy at the time
of data collection, treatment of uncomplicated malaria is
provided at public facilities for a fee (policy
changes were introduced after the household survey was
conducted). In addition to government-run
facilities, the health sector includes private-for-profit
facilities, NGO- and faith-based clinics and
hospitals, licensed pharmacies, and unlicensed drug vendors
selling from permanent sites or
hawking their products from site to site.
Malaria epidemiology in Benin can be characterized as stable
endemic and, as such, the risk of an
epidemic is considered to be low. The entire population is at
risk of infection. Transmission is more
intense in the (more populated) southern third of the country,
while a single seasonal peak is
observed in the north. The predominant parasite species is P.
falciparum.
Benin adopted artemether-lumefantrine (AL) as the first-line
treatment for uncomplicated malaria in
March 2004 (PNLP, 2005). Artesunate-amodiaquine (ASAQ) is
recommended for children under six
months of age, when AL is not available or when patients cannot
tolerate AL. Parenteral quinine is
recommended for the treatment of severe malaria and as
pre-referral treatment. Artesunate
injection and suppositories are included in the national policy
as a pre-referral treatment for severe
malaria. Oral artemisinin monotherapies have been banned in
Benin since 2008 (WHO, 2010b).
Although the policy change to ACT occurred in 2004, it was a
further four years before this policy
was truly active at the national level, and ACTs only became
widely available in the public sector in
2009 (PMI, 2009). Prior to 2009, ACT was most commonly available
through the private sector,
which has historically been an important distribution channel of
antimalarials in Benin (Tougher et
al, 2009). Procurement of ACT is currently supported by a number
of donors, including the Global
Fund, Presidents Malaria Initiative (PMI) and the United Nations
Childrens Fund (UNICEF).
At the time of data collection the treatment policy in place for
children under five was based
presumptive diagnosis, while treatment for older patients was
recommended only for those with a
positive diagnostic test. In line with World Health Organization
(WHO) guidelines, the malaria case
management policy was modified in February 2011 to recommend
universal diagnostic testing for
malaria. Although signaled in February 2011, the change was not
widely implemented until late
2011. With donor support, RDTs have been extensively scaled-up
alongside ACTs, although
implementation and commodity management have been similarly
challenging. In mid-2010 the
government had 460,000 RDTs for use in the public health system,
divided between health facilities
and the central medical stores. However, a health facility
survey conducted in early 2011 estimated
that 30% of facilities had expired RDTs in their stocks (PMI,
2011). RDTs are to be used at all levels of
the health system except the community level, and to be provided
free of charge. In addition to
RDTs, hospitals and larger health centres are supposed to
provide microscopy, although such
facilities often lack functional equipment and laboratory
workers skills are often sub-optimal.
-
www.ACTwatch.info 5|Page www.ACTwatch.info P a g e | 5
2.3.2 DRC
DRC has a population of 68.7 million (CIA World Fact Book, 2009)
and the per capita GDP is $300
(2008 estimate), ranking DRC second lowest in the world. An
estimated 80% of the population lives
below the poverty line. The country has experienced two decades
of civil war and there is continuing
violence in the east of the country that can be characterized as
a humanitarian crisis (International
Crisis Group, 2010). It is estimated that one in five children
die before reaching their fifth birthday
(UNICEF, 2009). Malaria is one of the leading causes of
mortality in DRC and the country has one of
the highest malaria burdens in Africa. There were an estimated
23.6 million malaria cases in 2006,
and annually 96,000 deaths are attributed to malaria (WHO,
2008).
97% of the population lives in areas with stable, perennial
malaria transmission (with seasonal peaks
in the countrys tropical zone) while the remaining 3% lives in
areas with unstable malaria
(Programme National de Lutte contre le Paludisme, Ministere de
la Sante RDC, 2007).
In 2005, the national malaria control program adopted
artesunate-amodiaquine (ASAQ) as the first-
line treatment for uncomplicated malaria. Quinine is the
recommended treatment should the first-
line treatment fail, and for the treatment of severe malaria.
Sulfadoxine-pyremethamine (SP) is used
for Intermittent Preventive Treatment of malaria in pregnancy
(IPTp). The Ministry of Health (MOH)
introduced community case management guidelines in 2007 that
included community-level
deployment of ACTs. The policy on diagnosis is quite narrow,
restricting its focus to treatment
failure and complicated malaria. At the peripheral level (health
centre), there is generally no
diagnostic capacity and suspected malaria cases are treated
based on clinical signs. At the referral
level, the guideline states that laboratory confirmation is
obligatory where there is no response to
first-line treatment of clinically diagnosed malaria and for
complicated malaria cases.
The health sector in DRC is largely unregulated and scant
information exists regarding the private
sector. The public sector is built around the health zone (zone
de sant), an administrative area at
the periphery that typically covers a population of 100,000
people. Each health zone is organized
around a central office, which includes a general referral
hospital and is divided into health areas
where health centres operate. There are currently more than
6,000 health centres scattered across
515 health zones covering the country (Programme National de
Lutte contre le Paludisme, Ministere
de la Sante RDC, 2007). The public health system relies on a
cost-recovery model and antimalarial
treatment is available to patients for a fee that varies across
public health facilities. ACTs are
procured through external funding from the Global Fund, United
States Agency for International
Development (USAID), and the World Bank amongst other
donors.
-
www.ACTwatch.info 6|Page www.ACTwatch.info P a g e | 6
2.3.3 Madagascar
Madagascar is the fourth largest island in the world with a
population of approximately 21 million
(Institut National de la Statistique, 2010). In February 2009, a
political crisis destabilized
Madagascars government and negatively impacted economic growth.
The GDP decreased by 4.6%
in 2009, while the population living below the poverty line
increased from 68.7% in 2005 to 76.5% in
2010. A substantial portion of financial and technical aid was
suspended in 2009 and the
government cut its national health budget by more than 30% from
2008 to 2010 (PMI, 2011). The
most common causes of death for children under five are malaria,
diarrheal diseases and respiratory
infections, often related to high rates of undernutrition
(Institut National de la Statistique, 2009).
Malaria is endemic in 90% of Madagascar, though the entire
population is at risk (PMI, 2011).
Transmission patterns vary across the island with year-round
transmission in the north, stable
perennial transmission on the east and west coasts and unstable
seasonal transmission in the south
and central highlandsregions prone to epidemics. Rainy season
lasts from late October to April
with cyclone season from December to April, accompanied by
flooding and increased risk of malaria.
Approximately 75% of the population lives in low and unstable
transmission regions, while 25% live
in areas of high and intense transmission (AMFm Independent
Evaluation Team, 2012b). Over
220,000 probable and confirmed cases of malaria were reported in
2011 (WHO, 2012).
The public health system operates across 111 health districts
and is divided into central, regional,
district and community levels (PMI, 2011). The public health
sector provides 70% of primary health
contacts in rural areas and 40% in urban areas, distributed
across of 138 hospitals, 1,335 secondary
health centers, 1,059 primary health centers, and 14,989
community health workers (CHW) (AMFm
Independent Evaluation Team, 2012b). Public sector drugs are
imported and distributed through the
central medical store, with the exception of ACTs, which are
purchased by the Unit de Gestion de
Projets and distributed by the National Malaria Program. The
private sector consists of 44 hospitals
and 724 private health centers, as well as 22 pharmaceutical
wholesalers, 200 pharmacies and more
than 1,000 rural pharmacies.
Malaria prevention strategies include LLIN distribution and
targeted IRS and IPTp in endemic districts
(PMI, 2011). Campaign and routine distribution of LLINs has
increased household ownership of at
least one net from 72% in the 2008/2009 DHS to 94% in the 2011
MIS (INSTAT and ICF Macro, 2010
and INSTAT and ICF International, 2012). Net use has similarly
increased from 58% to 89% of children
under five sleeping under a net in the previous night. In 2011,
79% of households in target districts
had been treated with IRS in the past year. While only 35% of
pregnant women receive at least one
dose of SP for malaria prevention despite antenatal care
attendance over 85% (INSTAT and ICF
International, 2012).
In 2006, Madagascar adopted ASAQ as the first-line treatment for
uncomplicated malaria and AL as
the second-line treatment (PNLP, 2005). Public health facilities
provide ACTs free of charge and
began to introduce RDTs in 2005. Community distribution of ACTs
by CHWs was initiated in late
2008, though treatment has been largely presumptive as RDTs were
only permitted at the
community level in 2010 (PMI, 2011). Drugstores and pharmacies
can legally dispense ACTs in
Madagascar with a prescription. Despite changes in national
treatment protocols, chloroquine has
-
www.ACTwatch.info 7|Page www.ACTwatch.info P a g e | 7
been widely available on the market, leading to a recent ban on
distribution outside of hospitals
which went into effect July 2012 (PNLP, direct correspondence,
2012).
Generally, health facility utilization is low in Madagascar. For
malaria, treatment seeking for children
with a fever was only 44% in 2011 and prompt treatment with an
ACT was only 0.4% in the
2008/2009 DHS and 2% in the 2011 MIS (INSTAT and ICF Macro, 2010
and INSTAT and ICF
International, 2012). Efforts to increase availability and use
of ACTs have been made through public
sector distribution, as well as initiatives through private and
community channels. A social marketing
project funded by the Global Fund distributed subsidized ASAQ,
branded as ACTipal, through
CHWs, pharmacies and drugstores from 2008 until discontinuation
of the brand in 2012 (PMI, 2011).
USAID and UNICEF have also supported scale-up of cIMCI through
CHWs in rural villages.
Madagascar participated in the AMFm program providing
manufacturer-level subsidy for ACTs
beginning in May 2010 (AMFm Independent Evaluation Team, 2012b).
The first shipment of AMFm
drugs arrived to private sector first-line buyers in October
2010 and to the public sector in February
2011. While availability of ACTs generally increased, the
project experienced wide-spread stock-outs
in the public and private sectors, exacerbated by malaria
epidemics in 2011 and 2012 in the South
and South East regions. In addition, promotional support for
subsidized ACTs was substantially
limited by enforcement of national regulations on drug
advertising. By June 2012, a total of 2.6
million ACT treatments were ordered through the AMFm in
Madagascar (CHAI, direct
correspondence, 2012).
Table 2.3.1 presents the timing of the baseline and endline
household surveys, and the time
between the midpoint of fieldwork and the first arrival of
copaid drugs in each country.
Table 2.3.1: Timing of data collection for household surveys and
arrival of copaid ACTs in Madagascar
Start date End date
Date of first arrival of copaid drugs
in country
Time prior to the arrival of AMFm drugs
in country
Baseline
Madagascar Dec 1, 2008 Jan 12, 2009 Oct 14, 2010 22 months
Endline Time after arrival of
AMFm drugs in country
Madagascar Apr 23, 2012 June 10, 2012 Oct 14, 2010 18 months
Additional information on the country background, AMFm context
and events is described elsewhere (see: AMFm Independent Evaluation
Team, 2012b; ACTwatch Group and PSI/Madagascar 2010, 2011,
2013).
-
www.ACTwatch.info 8|Page www.ACTwatch.info P a g e | 8
2.3.4 Nigeria Nigeria is the most populous country in Africa
with a population of 162.5 million ((UN Population
Division, 2010). The country is divided into six geopolitical
zones, 36 states (plus the capital Abuja)
and 774 Local Government Authorities (LGAs) (PMI, 2011).
Authority is largely decentralized with
substantial autonomy for state governments and limited influence
of the federal government over
local affairs. Despite rich oil resources and economic growth
over the past decade, 55% of the
population continues to live below the poverty line (AMFm
Independent Evaluation Team, 2012b).
The south of the country is better on most socioeconomic
indicators than the north, including lower
under five and maternal mortality rates.
Malaria is a major public health issue in Nigeria, with 97% of
the population at risk of malaria (PMI,
2011). Nigeria has five ecological strata ranging from desert
savannah to rainforest to mangrove
swampeach with varied patterns of malaria transmission. The
south experiences year-round
transmission while the north faces seasonal transmission during
its three-month rainy season.
Malaria is the leading cause of death in children under five and
leads to an estimated 225,000 deaths
annually (NMCP, 2009). It accounts for 50% of the total disease
burden and health expenditures in
Nigeria and 30% of hospitalizations.
The public sector health system operates at the federal, state
and local level with the Federal
Ministry of Health responsible for policy, research and tertiary
care and state and local authorities
overseeing secondary and primary care services (AMFm Independent
Evaluation Team, 2012b).
Supply chain management in the public sector is complex and
fragmented with multiple donors and
government authorities managing separate procurement and
distribution processes leading to weak
coordination and frequent stock-outs of essential medicines
(PMI, 2011).
Nigeria has a robust private sector that provides 65% of health
care across the country, particularly
at proprietary patent medicine vendors (PPMVs), which are often
the first choice of health care for
the poor (AMFm Independent Evaluation Team, 2012b). Among
patients with suspected malaria,
more than half seek care first in the private sector. Medicines
in the private sector are distributed
through a vast network of wholesalers, distributors,
pharmacists, medical representatives and
PPMVs (PMI, 2011).
Prevention efforts in Nigiera include LLIN distribution,
targeted IRS and IPTp. A mass campaign
distributed 34 million LLINs to 22 of Nigerias 37 states by May
2011 (PMI, 2011). A total of 63 million
nets are needed to reach universal coverage targets of two nets
per household nationwide. The MIS
indicated that household ownership of at least one ITN has
increased to 42% in 2010 from only 8% in
the 2008 DHS (NPC, NMCP and ICF International, 2012; NPC and ICF
Macro, 2009). In the same
survey, 29% of children under five slept under an ITN the
previous nighta substantial increase from
6% in 2008. Coverage for the 250,000 households targeted for IRS
in seven states is only 15%, while
IPTp efforts have similarly struggled with only 13.2% of
pregnant women receiving two or more
doses of SP in 2010 (NPC, NMCP and ICF International, 2012).
The Federal Ministry of Health adopted AL as the first-line
treatment for uncomplicated malaria in
2005 with ASAQ as the alternate first-line after detecting high
levels of failure for chloroquine and SP
in drug efficacy trials (PMI, 2011). ACTs were classified as
over-the-counter medicines in 2006 and
-
www.ACTwatch.info 9|Page www.ACTwatch.info P a g e | 9
oral artemisinin monotherapies were banned for import or
manufacturing in the same year. Policy
states that ACTs are provided in public facilities free of
charge. While local pharmaceutical
manufacturing is abundant in Nigeria, none of the producers are
WHO pre-qualified to manufacture
ACTs (AMFm Independent Evaluation Team, 2012b). In 2010, Nigeria
aligned its national treatment
guidelines with the WHO to recommend universal diagnostic
testing for suspected malaria (PMI,
2011). Microscopy capacity and RDT availability is low, however,
and it is estimated that a total of 50
to 60 million RDTs would be required each year to achieve
universal access to malaria diagnosis in
the public sector.
Several initiatives have provided ACTs to Nigeria, including
programs supported by the World Bank,
PMI and the Global Fund. Nigeria signed an agreement for the
AMFm pilot in September 2010 and
received the first shipment of drugs in January 2011 (AMFm
Independent Evaluation Team, 2012b).
The National Malaria Control program procured ACTs through the
Voluntary Pooled Procurement
(VPP) system, while the Society for Family Health (SFH) and the
28 of the 54 registered first-line
buyers placed orders directly with the manufacturers. National
advocacy and promotion efforts
were successful in the first year and Nigerias ACT demand
outstripped international supply. By the
end of 2011, a total of 78 million treatments had been ordered
in Nigeria, of which about 39 million
had been delivered. This represents 40% of the total ACTs
delivered through the AMFm.
Table 2.3.2 presents the timing of the baseline and endline
household surveys, and the time
between the midpoint of fieldwork and the first arrival of
copaid drugs in each country.
Table 2.3.2: Timing of data collection for household surveys and
arrival of copaid ACTs in Nigeria
Start date End date
Date of first arrival of copaid drugs
In-country
Time prior to the arrival of AMFm drugs
in-country
Baseline
Nigeria Aug 11, 2009 Sept 23, 2009 Jan 25, 2011 17 months
Endline Time after arrival of
AMFm drugs in-country
Nigeria June 18, 2012 July 28, 2012 Jan 25, 2011 15 months
Additional information on the country background, AMFm context
and events is described elsewhere (see: AMFm Independent Evaluation
Team, 2012b;; ACTwatch Group and SFH/Nigeria 2010, 2011, 2013).
-
www.ACTwatch.info 10|Page www.ACTwatch.info P a g e | 10
2.3.5 Uganda Uganda is located in East Africa, bordered by South
Sudan to the north, Kenya to the east, Tanzania
and Rwanda to the south and Democratic Republic of Congo (DRC)
to the West. In 2010 the
population was estimated at 34.5 million, 24% of whom were five
years of age or younger
(Population Division, 2011). Approximately 84% of people live in
rural areas (UN Population Division,
2010). Though the GDP has increased by 5.8% from 2009 to 2010,
over one-third of the population
continues to live below the poverty line (AMFm Independent
Evaluation Team, 2012b). Inflation
rates in Uganda reached 30% in 2011, leading to increases in the
price of fuel, food and other goods,
including medicine (USAID, direct correspondence, 2012).
Malaria is a major public health problem in Uganda. Perennial
transmission occurs in 95% of Uganda
with the remaining 5% experiencing unstable and epidemic-prone
transmission (PMI, 2011d). The
2009 Malaria Indicator Survey indicates that 99% of malaria
cases are caused by P. falciparum (UBOS
et al, 2010). Malaria is the most frequently reported disease in
both public and private health
facilitiesaccounting for 25-40% of outpatient visits at health
facilities and 9-14% of all hospital
deaths (PMI, 2011d). Among children under five, about half of
all deaths are attributed to malaria,
while across Uganda malaria is estimated to account for
approximately 70,000-110,000 deaths
annually (NMCP, 2010).
Ugandas public sector health system is decentralized and
consists of five levels of service: national,
regional and district-level hospitals; sub-district-level health
centers, subcounty-level health centers,
parish-level health centers and Village Health Teams (AMFm
Independent Evaluation Team, 2012b).
Public health facilities procure medicines through the National
Medical Stores (NMS). In early 2010,
the Ministry of Health introduced a push system of supply to
lower-level health facilities operating
at the subcounty and parish level (PMI, 2011d). Each health
center at this level receives a standard
kit of essential drugs and health supplies on a bimonthly basis.
One study shows that the system has
resulted in a 69% decrease in the number of days facilities are
stocked out of five tracer drugs,
including ACTs (MOH, 2010). At a facility level, there are some
reports of excess stocks as well as
splitting or stacking doses to correct imbalances in
age-specific treatments (USAID, direct
correspondence, 2012). Hospitals and health centers at the
district level continue to order based on
determined needs directly from the NMS.
The private sector makes up 46% of Ugandas health facilities
(PMI, 2011d). An estimated 82% of
households first seek care from drug shops, private clinics and
private not-for-profit providers, while
55% of children with fever receive treatment from private
facilities. Private not-for-profit facilities
are predominantly faith-based organizations which procure drugs
and medical supplies from the
Joint Medical Stores (AMFm Independent Evaluation Team, 2012b).
Private for-profit facilities
include primary and secondary-care health facilities, retail
pharmacies, licensed drug shops and
unregistered drug shops that operate illegally. Pharmacies are
permitted to dispense prescription-
only medicines under the supervision of a registered pharmacist,
while drug shops are only
permitted to sell over-the-counter medicines. For-profit
providers purchase medicines and supplies
through a private sector supply chain that includes 250
wholesale pharmacies, 70 drug importers
and distributors and 15 local manufacturers. In December 2010,
the Kampala manufacturing site for
Quality Chemicals Industries Limited became Africas first local
supplier to receive WHO pre-
-
www.ACTwatch.info 11|Page www.ACTwatch.info P a g e | 11
qualification to produce ACTs. The company is licensed to
produce artemether/lumefantrine (AL)
through the company Cipla Limited, based in India.
Malaria prevention strategies include LLIN distribution,
targeted IRS and IPTp. A national campaign
in 2010/2011 distributed 7.2 million LLINs to pregnant women and
children under five with plans for
a follow-on campaign of 10.4 million in mid-2012 to reach
universal coverage (PMI, 2011d).
Campaign and routine distribution of LLINs has increased
household ownership of at least one LLIN
from 9% in the 2009 MIS to 39% in the 2011 PMI bed net coverage
survey. Net use has similarly
increased from 11% to 43% of children under five sleeping under
a net in the previous night.
IRS efforts target four high-transmission districts in northern
Uganda (Kitgum, Pader, Apac, and
Oyam) with large populations of internally displaced persons
(PMI, 2011d). The program has
achieved over 95% coverage in targeted households through
several spraying rounds annually. IPTp
is implemented nationally through focused antenatal care,
however only 45% of pregnant women
received at least one dose of SP in 2009 despite ANC attendance
over 90%.
The first-line treatment for malaria has been AL since 2004 and
the current second-line treatment is
dihydroartemisinin piperaquine and quinine (AMFm Independent
Evaluation Team, 2012b). Oral
artemisinin monotherapies were banned in 2007 and ACTs became
classified as an over-the-counter
drug in 2008. Although public health facilities in Uganda
provide diagnosis and ACTs free of charge,
only 36% of children under five were treated with an
antimalarial the same or next day after onset of
fever, according to the 2009 MIS (UBOS et al, 2010). In the same
survey, only 13.7% were treated in
the first 24 hours with an ACT. This may reflect the chronic
stock-outs across Uganda in this period
as a result of grant interruptions, procurement delays and weak
supply chains in-country (PMI,
2011d).
Efforts to increase availability and use of ACTs have been made
through public sector distribution, as
well as initiatives through private and community channels. In
2008, a pilot led by the Ministry of
Health and Medicines for Malaria Venture introduced subsidized
ACTs to the private sector in four
districts (Budaka, Pallisa, Kaliro and Kamuli) (AMFm Independent
Evaluation Team, 2012b). The ACT
was Coartem (AL) repackaged with a green leaf logo which served
as the prototype for the global
AMFm logo. The pilot ended in 2010 and stocks likely remained on
the market at the time of
baseline surveys.
Uganda was one of the first countries in Africa to introduce
community management of fever,
distributing ACTs through Community Medicine Distributors in
rural areas. Despite early adoption,
the program has been plagued by shortages of AL, with only 9% of
households in 2009 reporting
Community Medicine Distributors in their community with
available malaria treatment (UBOS,
2010). The Global Fund, Canadian International Development
Agency, Gates Foundation and UNICEF
currently support pilot efforts to scale-up integrated community
case management in rural districts
(PMI, 2011d).
In February 2011, Uganda signed its AMFm grant, making it the
last of participating countries to
receive the manufacturer-level subsidy (AMFm Independent
Evaluation Team, 2012b). Negotiations
were held up over concerns that the inexpensive AMFm drugs would
cause financial harm to local
-
www.ACTwatch.info 12|Page www.ACTwatch.info P a g e | 12
manufacturer, Quality Chemicals Industries Limited. Once WHO
granted pre-qualification for local AL
production in December of 2010, negotiations advanced. The
initial order of AMFm drugs was
further delayed by quantification discussions and confusion over
public sector suppliers. The first
order of copaid ACTs was finally delivered to the NMS in July
2011 and a total of 20.7 million
treatment courses were approved and delivered to the public
sector in 2011. DfID and PMI made
emergency procurements to re-supply dwindling stocks to the
public and not-for-profit sector during
AMFm negotiations.
The private not-for-profit sector procured its AMFm drugs
through three registered first line buyers,
receiving its first supply of copaid ACTs in April 2011 (AMFm
Independent Evaluation Team, 2012b).
A total of 1.1 million treatments were ordered and 0.6 million
treatments were delivered to the Joint
Medical Stores by the end of 2011. The private for-profit sector
made orders through four registered
first line buyers. A total of 15 orders were approved in 2011
for 7.9 million treatment courses and
6.9 million treatments were delivered by the end of the year.
Supporting interventions for
promotion of AMFm drugs had not begun as of August 2012 (USAID,
direct correspondence, 2012).
Table 2.3.3 presents the timing of the baseline and endline
household surveys, and the time
between the midpoint of fieldwork and the first arrival of
copaid drugs in each country.
Table 2.3.3: Timing of data collection for household surveys and
arrival of copaid ACTs in Uganda
Start date End date
Date of first arrival of copaid drugs
In-country
Time prior to the arrival of AMFm drugs
in-country
Baseline
Uganda March 16, 2009 April 20, 2009 Apr 23, 2011 24 months
Endline Time after arrival of
AMFm drugs in-country
Uganda Apr 19, 2012 May 11, 2012 Apr 23, 2011 12 months
Additional information on the country background, AMFm context
and events is described elsewhere (see: AMFm Independent Evaluation
Team, 2012b;) ACTwatch Group and PSI/Uganda and PACE/Uganda 2010,
2011, 2013).
-
www.ACTwatch.info 13|Page www.ACTwatch.info P a g e | 13
2.3.6 Zambia Zambias population was approximately 13.9 million
in 2010 (Population Division at Department of
Economic and Social Affairs at the United Nations, 2011), of
which an estimated 65% live in rural
areas (Population Division,, 2011). Zambia has experienced
positive economic growth with a real
growth rate of 6.7% (US Department of State, 2012). The per
capita GDP was estimated $1,600 in
2011 and 64% of the population were estimated to live under the
poverty line in 2006 (CIA, 2012).
Malaria transmission occurs across the country throughout the
year, with the peak occurring during
the rainy season from November to April. Transmission intensity
varies across the country and has
been categorized as follows: (i) areas that have less than 1%
parasite prevalence (Lusaka), (ii) areas
with parasite prevalence that ranges between 1% and less than
10% (Central, Copperbelt, North-
Western, Southern, and Western Provinces) and (iii) areas with
parasite prevalence greater than
20% (Eastern, Luapula, and Northern Provinces) (Zambia National
Malaria Control Center, 2011).
Malaria incidence was estimated at 358 cases per 1,000 persons
in 2007 (Mouzin, et al., 2011) and
4.3 million cases of malaria were reported with 6,000 deaths
(40% of all infant mortality) (Mouzin, et
al., 2011). In 2010, the number of malaria cases was lower with
an estimated 4.2 million cases of
malaria and 4,834 deaths (Zambia Ministry of Health National
Malaria Control Center, 2011). The
MIS showed a decline in malaria prevalence from 22% in 2006 to
10.2% in 2008 (Zambia Ministry of
Health, 2011).
The Zambian health care system is decentralized and the public
sector plays an important role in the
delivery and management of health services. The public sector is
divided into several tiers: health
posts and community outreach; health centers; and three levels
of hospitals (PMI, 2011e). Only 20-
30% of the sick patients seek treatment from private
practitioners in Zambia (Sipilanyambe et al.,
2008). Health services that are operated by the Ministry of
Health are provided free of charge.
Private drug stores are common and only dispense medicines
registered as over-the-counter, but in
practice, they often dispense prescription-only medicines.
Zambia was the first country in Africa to change its first-line
treatment to an ACT in 2002 (Mouzin et
al., 2011; Sipilanyambe, et al., 2008). AL is the first-line
treatment for uncomplicated malaria. SP is
the alternative first-line treatment in pregnant women and
children less than 5kg1. Oral quinine is
the second-line treatment in cases of failure of first-line
drugs in all age groups. Severe malaria is
treated with quinine injection (Zambia Ministry of Health,
2011). In Zambia, ACTs are classified as
prescription-only medications and are therefore not sold legally
through unlicensed private sector
providers.
Zambia has greatly strengthened the capacity to diagnose and
confirm malaria cases with laboratory
microscopy and RDTs (Zambia National Malaria Control Center,
2011). RDTs have been made
available since 2007, and free distribution of RDTs in health
centers was piloted in 2008 and scaled
up nationally after an intense training period (Mouzin et al.,
2011). Approximately 2 million RDTs are
1During the early implementation of the treatment policy in
2004, AL was not recommended for children less than 10kg. SP was
recommended instead. The policy was shifted in late 2005, where AL
was recommended for children weighing 5 to 9kg (Zurovac, et al.,
2007).
-
www.ACTwatch.info 14|Page www.ACTwatch.info P a g e | 14
distributed yearly (Mouzin et al., 2011). According to the 2010
MIS, the treatment of febrile children
with antimalarial drugs declined from 52.8% in 2006 to 43.1% in
2008 to 34% in 2010, due to
increasing availability of RDTs. The percentage of febrile
children under age 5 with reported finger
stick for presumed diagnostic testing was 10.9% in 2008 and
16.7% in 2010 (Zambia National Malaria
Control Center, 2011).
-
www.ACTwatch.info 15|Page www.ACTwatch.info P a g e | 15
3. Household Survey Methods
3.1. Introduction
Household surveys were carried out in Benin, DRC, Madagascar,
Nigeria, Uganda and Zambia with
the objectives of assessing ACT use for fevers in children under
five, source of ACTs, and diagnostic
practices. Baseline and endline surveys were conducted in all
countries except DRC where only one
survey was conducted, in 2010. ACTwatch also conducted surveys
in Cambodia and these results are
reported elsewhere (ACTwatch Group and PSI/Cambodia 2009,
2011).
The timing of the baseline and endline household surveys
conducted in the six malaria endemic
countries included in this report is presented in Table
3.1.1.
Table 3.1.1: Timing of data collection for household surveys
Start date End date
Baseline
Benin April 15, 2009 May 08, 2009
DRC Apr 27, 2010 June 21, 2010
Madagascar Dec 1, 2008 Jan 12, 2009
Nigeria Aug 11, 2009 Sept 23, 2009
Uganda March 16, 2009 April 20, 2009
Zambia April 14, 2009 July 4, 2009
Endline
Benin June 24, 2011 July 10, 2011
Cambodia Nov 6, 2011 Dec 12, 2011
DRC -- --
Madagascar Apr 23, 2012 June 10, 2012
Nigeria June 18, 2012 July 28, 2012
Uganda Apr 19, 2012 May 11, 2012
Zambia May 25, 2011 Sept, 18 2011
Source: ACTwatch Group et al., 2013. ACTwatch Baseline and
Endline Household Survey Results 2009-2012: Benin, Democratic
Republic of the Congo, Madagascar, Nigeria, Uganda, Zambia.
-
www.ACTwatch.info 16|Page www.ACTwatch.info P a g e | 16
3.2. Sampling
Nationally representative samples were selected using
multi-stage sampling. The primary sampling
units, clusters, were administrative areas that had already been
selected as part of the ACTwatch
outlet survey in each country. A list of second stage units,
generically called enumeration areas (EA),
was drawn for each cluster. EAs are administrative units such as
actual census enumeration areas,
villages, or supervisor areas, with an average of 250-500
inhabitants. Clusters and EAs were selected
with probability proportional to population size (PPS) at each
stage. (Littrell et al., 2011) Table 3.2.1
provides a list of the clusters and EAs used in each
country.
Table 3.2.1: Definition of cluster and enumerations areas, by
country
Baseline Clusters Enumeration Areas
Benin Arrondissement Census enumeration areas
DRC Aires de sant Villages/quartiers
Madagascar Communes Fokontany
Nigeria Localities Supervisory areas (SAs)
Uganda Subcounties (parishes in Kampala)
Villages
Zambia Wards Standard enumeration areas (SEAs)
Source: ACTwatch Group et al., 2013. ACTwatch Baseline and
Endline Household Survey Results 2009-2012: Benin, Democratic
Republic of the Congo, Madagascar, Nigeria, Uganda, Zambia.
At baseline, the samples were stratified depending on the
country context (Table 3.1.3) (Shewchuk
et al., 2011). In Madagascar and Zambia, stratification was
conducted with urban and rural domains.
In the DRC the sample was stratified across four geographic
regions, and in Nigeria, across six geo-
political regions. In Uganda, two strata were included for
low/moderate and high malaria
transmission areas. In Benin, a single domain was used. At
endline, stratification was modified for
Nigeria and Uganda, to urban and rural, as a means to coordinate
efforts with the Independent
Evaluation of the AMFm and Outlet Survey sampling and
stratification guidelines.
Full household listings were not available at baseline and so
enumeration area size, together with
the average household size from DHS figures, was used to derive
the sampling interval applied
during systematic random sampling of households. At endline a
household census was conducted in
each EA and households were then randomly selected.
-
www.ACTwatch.info 17|Page www.ACTwatch.info P a g e | 17
Table 3.2.2: Stratification, by country
Country Baseline Endline
Benin n/a n/a
DRC
1. North-East Provinces: Oriental, Nord Kivu, Sud Kivu, and
Maniema
2. North West Provinces: Bas-Congo, Bandundu, and Equateur
3. Centre-South Provinces: Katanga, Kasai Oriental, and Kasai
Occidental
4. Capital city: Kinshasa
n/a
Source: Localized censuses in health zones conducted between
2001 and 2004, with the support of various INGOS
Madagascar 1. Urban areas 2. Rural areas
1. Urban areas 2. Rural areas
Source: Cartographie censitaire de la population 2009. Institut
National de Statistique.
Nigeria
1. North-Central states: Benue, Abuja-FCT, Kogi, Kwara,
Nasarawa, Niger, Plateau
2. North East states: Adamawa, Bauchi, Borno, Gombe, Taraba,
Yobe
3. North West states: Jigawa, Kaduna, Kano, Katsina, Kebbi,
Sokoto, Zamfara
4. South East states: Abia, Anambra, Ebonyi, Enugu, Imo
3. South South states: Akwa ibom, Bayelsa, Cross rivers, Delta,
Edo, Rivers
4. South West states: Ekiti, Lagos, Ogun, Ondo, Osun, Oyo
1. Urban areas 2. Rural areas
Source: Nigeria National Population Commission
Uganda 1. Low malaria transmission areas 2. Medium to high
malaria transmission
areas
1. Urban areas 2. Rural areas
Source: Malaria Control Programme, Ministry of Health,
Uganda
Zambia 1. Urban areas 2. Rural areas
1. Urban areas 2. Rural areas
Source: Zambia Population Census Frame, 2000. Central
Statistical Office, Lusaka, Zambia
For all surveys the sample size was estimated for each country
to be able to detect a 10 percentage
point change between the baseline and endline values of the
indicator: the proportion of children
under five years with fever in the two weeks preceding the
survey who received any antimalarial
treatment. The sample size was calculated based on the following
parameters, using the most recent
figures from the DHS and MIS at the time of data collection for
the baseline and endline, and
powered to detect differences between strata.
Estimated fever prevalence was used to calculate the number of
households necessary to
screen in order to yield the desired number of children with
fever.
The estimated number of children under five in a household was
used in collaboration with
the estimated fever prevalence above to calculate the number of
households necessary to
screen in order to yield the desired number of children with
fever.
The samples were designed to detect differences with 95%
significance and 80% power.
-
www.ACTwatch.info 18|Page www.ACTwatch.info P a g e | 18
A design effect of two was assumed, to account for the clustered
survey design.
Based on the approach and parameters described above, it was
estimated that 3,181 households in
Benin, 2,168 in DRC, 8,250 in Madagascar, 9,171 in Nigeria,
3,777 in Uganda and 6,441 in Zambia
would need to be visited at baseline in order to find a
sufficient number of febrile cases. At endline,