household survey madagascar 2012 survey report - … · p a g e | ii household survey report...

Evidence for Malaria Medicines Policy

Household Survey Madagascar

2012 Survey Report

MINSTERE DE LA SANTE PUBLIQUE

P a g e | i

www.ACTwatch.info Household Survey Report Madagascar, 2012

Acknowledgements

ACTwatch is funded by the Bill and Melinda Gates Foundation.

This study was implemented by Population Services International (PSI).

ACTwatch’s 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

P a g e | ii


The following individuals contributed as follows to the research study in Madagascar:

Jacky Raharinjatovo ACTwatch Country Program Coordinator and Quantitative Research Coordinator, PSI/Madagascar, was responsible for all aspects of implementation and management of the survey.

Justin Rahariniaina ACTwatch Research Supervisor, PSI/Madagascar, assisted the Country Program Coordinator and was responsible for the coordination of trainings, data collection, and data cleaning.

Sitraka Ramamonjisoa Quantitative Research Supervisor, PSI/Madagascar, assisted the Country Program Coordinator and helped with training and data collection.

Iarimalanto Rabary Director of Research Monitoring and Evaluation, PSI/Madagascar, assisted with advocacy and survey implementation.

Andry Rabemantsoa Senior Research Coordinator, Monitoring and Evaluation, PSI/Madagascar, helped to manage communication.

Dr. Rova Ratsimandisa

Malaria Treatment Coordinator, PSI/Madagascar, provided information on the national malaria context and assisted with the training.

Arsène Ratsimbasoa Deputy Director of NMCP, Ministry of Health Madagascar, contributed to report writing and dissemination of findings.

Stephen Poyer Research Associate, ACTwatch Central, provided overall guidance on the analysis and construction of indicators.

Julius Njogu Research Associate, ACTwatch Central, assisted with the field preparations, and trained field workers.

Hellen Gatakaa Senior Research Associate, ACTwatch Central, provided technical support for the analysis of the data.

Dr. Megan Littrell Senior Malaria Research Advisor, ACTwatch Central, assisted with the survey protocol and questionnaire.

Dr. Kathryn O’Connell Principal Investigator, ACTwatch Central, provided overall technical guidance.

Tanya Shewchuk Project Director, ACTwatch Central, provided overall project oversight and dissemination.

P a g e | iii


The ACTwatch Group is comprised of the following individuals:

PSI ACTwatch Central Tanya Shewchuk, Project Director; Dr. Kathryn O’Connell, Principal

Investigator; Hellen Gatakaa, Senior Research Associate; Stephen Poyer, Illah Evance, Julius Njogu, Research Associates; Meghan Bruce, Policy Advocate and Communications Specialist; Linda K. Ongwenyi, ACTwatch Project Assistant.

PSI ACTwatch Country Program Coordinators

Cyprien Zinsou, ABMS/Benin; Sochea Phok, PSI/Cambodia; Dr. Louis Akulayi, ASF/DRC; Jacky Raharinjatovo, PSI/Madagascar; Ekundayo Arogundade, SFH/Nigeria; Peter Buyungo, PACE/Uganda; Felton Mpasela, SFH/Zambia.

LSHTM Dr. Kara Hanson, Principal Investigator; Edith Patouillard, Dr. Catherine Goodman, Benjamin Palafox, Sarah Tougher, Immo Kleinschmidt, co-investigators. LSHTM is responsible for the supply chain research component of ACTwatch.

Suggested citation:

ACTwatch Group & PSI/Madagascar, 2013. Household Survey, Madagascar, 2012 Household Survey

Report. Washington DC: Population Services International. Available at: www.actwatch.info

ACTwatch Contacts

Madagascar

Mr. Jacky Raharinjatovo

ACTwatch Country Program Coordinator

Population Services International/Madagascar

Immeuble Fiaro Ampefiloha

BP 7748 Antananarivo

Madagascar

Phone: + 261 202262984

Email: [email protected]

ACTwatch Central

Tanya Shewchuk

ACTwatch Director

Malaria & Child Survival Department

Population Services International

Regional Technical Office

Whitefield Place, School Lane, Westlands

P.O. Box 14355-00800 Nairobi, Kenya

Phone: +254 20 4440125/6/7/8

Email: [email protected]

http://www.actwatch.info/

mailto:[email protected]

P a g e | iv


Table of Contents

LIST OF TABLES ........................................................................................................................... VI

LIST OF FIGURES ......................................................................................................................... VII

DEFINITIONS ............................................................................................................................. VIII

ABBREVIATIONS ........................................................................................................................... X

EXECUTIVE SUMMARY ................................................................................................................. 1

Overview of ACTwatch ............................................................................................................. 1

Endline Household Survey Methods .......................................................................................... 1

Key findings from the household survey .................................................................................... 2

1. BACKGROUND ............................................................................................................... 10

1.1 Overview of the ACTwatch Research Project...................................................................... 10

1.2 Affordable Medicines Facility for Malaria .......................................................................... 10

1.3 Country background ......................................................................................................... 12

1.3.1 Overview of the country ...................................................................................................... 12

1.3.2 Description of health care system ....................................................................................... 13

1.3.3 Epidemiology of malaria ...................................................................................................... 14

1.3.4 Antimalarial Policies and Regulatory Environment ............................................................. 15

1.3.5 Malaria control strategy ...................................................................................................... 15

1.3.6 Malaria financing ................................................................................................................. 17

1.3.7 AMFm Phase 1 pilot ............................................................................................................. 18

1.3.8 Other research findings ....................................................................................................... 19

2. METHODS ...................................................................................................................... 20

2.1 Household Survey ............................................................................................................. 20

2.1.1 Sampling Approach .............................................................................................................. 20

2.1.1.1 Target Population ........................................................................................................... 20 2.1.1.2 Sample Size ..................................................................................................................... 20 2.1.1.3 Selection Procedure for Clusters and EAs ...................................................................... 21

2.1.2 Questionnaire ...................................................................................................................... 22

2.1.3 Data Collection ..................................................................................................................... 22

2.1.3.1 Preparatory Phase .......................................................................................................... 22 2.1.3.2 Fieldwork ........................................................................................................................ 23

2.1.5 Data analysis ........................................................................................................................ 24

2.1.5.1 Data analysis process ...................................................................................................... 24 2.1.5.2 Indicators ........................................................................................................................ 24

3. RESULTS ........................................................................................................................ 26

3.1 Characteristics of the sample ............................................................................................ 26

3.2 Treatment for fever .......................................................................................................... 29

3.3 Diagnosis .......................................................................................................................... 30

3.4 Type of antimalarials taken and source of antimalarials ..................................................... 32

3.5 Sources of advice and treatment for fever ......................................................................... 38

3.6 Breakdown of antimalarials acquired ................................................................................ 42

3.7 Caregiver Knowledge and Beliefs ....................................................................................... 43

3.8 Caregiver Awareness of and Exposure to the AMFm .......................................................... 45

4. REFERENCES .................................................................................................................. 52

P a g e | v


5. ACKNOWLEDGEMENTS .................................................................................................. 54

6. APPENDICES .................................................................................................................. 55

6.1 AMFm use indicator for poorest households...................................................................... 55

6.2 Survey team ..................................................................................................................... 56

6.3 Questionnaire ................................................................................................................... 58

P a g e | vi


List of Tables

Table 3.1.1: Results of the household and individual interviews ......................................................... 27

Table 3.1.2: Demographic characteristics ............................................................................................. 28

Table 3.2.1: Treatment of children with fever ...................................................................................... 29

Table 3.3.1: Diagnosis of fever among children under five .................................................................. 30

Table 3.3.2: Source of diagnosis ........................................................................................................... 31

Table 3.4.1: Type of antimalarial medicines taken by children under five ........................................... 32

Table 3.4.2: Type of antimalarial medicines taken promptly by children under five ........................... 33

Table 3.4.3: Type of antimalarial medicines taken among children who received an antimalarial ..... 34

Table 3.4.4: Source of antimalarials ...................................................................................................... 35

Table 3.4.5: Source of ACTs................................................................................................................... 36

Table 3.4.6: Source of ACTs with the AMFm logo, among children under five with fever who received

an AMFm logo ACT ................................................................................................................................ 37

Table 3.5.1: Care seeking behaviour: first place to seek care............................................................... 38

Table 3.5.2: Care seeking behaviour: any source to seek care ............................................................. 39

Table 3.5.3: Treatment at home (Supplementary Table) ..................................................................... 40

Table 3.5.4: Initial treatment source (Supplementary Table) .............................................................. 41

Table 3.6.1: Types of antimalarials acquired ........................................................................................ 42

Table 3.7.1: Caregiver knowledge of malaria and antimalarials ........................................................... 43

Table 3.7.2: Caregiver beliefs about the most effective antimalarial treatment ................................. 44

Table 3.8.1: Caregiver awareness of and exposure to the AMFm logo and initiative .......................... 45

Table 3.8.2: Sources of exposure to the AMFm logo ............................................................................ 46

Table 3.8.3: Sources of exposure to the AMFm initiative ..................................................................... 47

Table 3.8.4: Meaning of the AMFm logo .............................................................................................. 48

Table 3.8.5: Knowledge of the recommended price for AMFm medicine............................................ 49

Table 3.8.6: Knowledge of the use of AMFm medicine ........................................................................ 49

Table 3.8.7: Caregiver reported ever use of ACTs with the AMFm logo .............................................. 50

Table 3.8.8: Source of ACTs with the AMFm logo ................................................................................. 50

Table 3.8.9: Perceptions of the efficacy and affordability of AMFm ACTs ........................................... 51

Table 6.1.1: Treatment of children with fever in the poorest households ........................................... 55

Table 6.2.1: List of staff members involved in the survey .................................................................... 56

P a g e | vii


List of Figures

Figure 1: Treatment of children with fever ............................................................................................. 2

Figure 2: Use of antimalarials among children who received an antimalarial ....................................... 3

Figure 3: Use of antimalarials among children who received an antimalarial by urban/rural strata ..... 4

Figure 4: Care seeking behavior: first place caregivers seek advice or treatment for fever .................. 5

Figure 5: Source of antimalarials, ACTs and ACTs with the AMFm logo ................................................. 6

Figure 6: Caregiver awareness of and exposure to the AMFm logo and initiative ................................. 7

Figure 7: Common sources of exposure to the AMFm logo and initiative ............................................. 8

Figure 8: Meaning of the AMFm logo ..................................................................................................... 9

Figure 1.3.1: Location of Madagascar ................................................................................................... 12

Figure 1.3.2: Geographical distribution of confirmed malaria cases .................................................... 14

Figure 3.1.1: Survey flow diagram ........................................................................................................ 26

P a g e | viii


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. In Madagascar, they were defined as communes.

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 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. In Madagascar they were defined as fokantany.

First-line treatment The government recommended treatment for uncomplicated malaria. Madagascar’s first-line treatment for malaria is artesunate-amodiaquine, 50/153mg.

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.

P a g e | ix


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.

Private for-profit sector

For reporting purposes ACTwatch classifies sources of advice and treatment into two sectors: private for-profit and public/private not for profit. In Madagascar the following outlet types are classified as private for-profit: private for-profit health facility, pharmacy, drug shop (dépôt de médicament), and grocery store (épicerie).

Public/not for profit sector

For reporting purposes ACTwatch classifies sources of advice and treatment into two sectors: private for-profit and public/private not for profit. In Madagascar the following outlet types are classified as public/private not for profit: public health facility, community health worker, and NGO/Mission-based health facility.

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 1.12) 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, 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.

P a g e | x


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

CSB Centre de Santé de Base

DHS Demographic and Health Survey

EA Enumeration Area

Global Fund Global Fund to Fight AIDS, Tuberculosis, and Malaria

ICCM Integrated Community Case Management

INSTAT Institut National de la Statistique

IPTp Intermittent Preventive Treatment in pregnancy

IRS Indoor Residual Spraying

LLIN Long-lasting Insecticidal Net

LSHTM London School of Hygiene and Tropical Medicine

MIS Malaria Indicator Survey

MOH Ministry of Health

NGO Non-governmental organization

NMCP National Malaria Control Program

NMS National Malaria Service

NSA National Strategy Application

OTC Over-the-counter

PDA Personal Digital Assistant

PHF Public Health Facility

PMI President’s Malaria Initiative

PPS Probability proportional to size

PSI Population Services International

RDT Rapid Diagnostic Test

SP Sulfadoxine-pyrimethamine

UGP Unité de Gestion de Projets

UN United Nations

UNDP United Nations Development Program

USD United States Dollar

WHO World Health Organization

P a g e | 1


Executive Summary

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 (O’Connell 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

behaviour for fever in children under 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).

Baseline and endline household surveys have now been conducted in each of the seven ACTwatch

countries. This report covers the endline household survey in Madagascar, which was conducted

between April and June 2012.

Endline Household Survey Methods

This study uses data from a cross-sectional household survey of children’s caregivers. A

nationally-representative sample of households in Madagascar was drawn using three-stage cluster

sampling, with separate samples drawn for urban and rural areas. This allows the estimation of

indicators at the national level, and for robust comparisons to be made between urban and rural

areas.

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 all caregivers of children under five, irrespective of a child’s fever status, were

eligible for a subset of questions regarding exposure to the Affordable Medicines Facility – malaria

(AMFm).

The household survey was designed to monitor 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. Data were entered

using Personal Digital Assistants (PDAs). Stata 11 (Stata Corp, College Station, TX) was used for all

analyses. To obtain the national estimates provided in this report, data were weighted.

P a g e | 2


Key findings from the household survey

Data collection ran from the 23rd of April 2012 until the 10th of June 2012. This period fell within the

rainy season in Madagascar. A total of 15,195 households were selected for inclusion in the study.

1,169 households were not screened for various reasons: 31 refused to participate in screening; 185

dwellings were abandoned; in 74 households an eligible respondent was not available; no one was

present in 865 households at the time of the survey visits (up to three visits were attempted); and 14

households were not screened for other reasons. Overall, 14,026 households agreed to participate in

the survey and were screened. Of the 2,176 households that met the screening criteria and were

eligible for full interview 6 refused to continue and in 8 households an eligible respondent was not

available or the time was not convenient for the full interview. In the 2,161 households that

completed interviews: 2,169 caregivers were interviewed regarding 2,388 children under five with

fever in the previous two weeks. In total 6,781 caregivers of children under five were administered

the subset of questions regarding the AMFm.

Figure 1: Treatment of children with fever

Among children under five with fever in the two weeks preceding the survey the percentage who took any antimalarial medicines/ACTs and any antimalarial medicines/ACTs the same or next day following the onset of fever, and percentage who had blood taken from a finger or heel for testing (n=2,388).

Fewer than 20% of children with fever in the two weeks preceding the survey received any antimalarial medicine and only 8% received an ACT. 16% of children received an antimalarial the same or next day, and 7% of children received an ACT promptly. 12% of children received a diagnostic test, and this was most commonly an RDT (95%) as compared with a microscopic test (3%) (see Table 3.3.2). The majority of children who were tested received the test in the public/not for profit sector (90%) rather than the private sector (8%) (see Table 3.3.2).

19 16

8 7 3

12

0

10

20

30

40

50

60

70

80

90

100

Antimalarial Prompt Antimalarial

ACT Prompt ACT ACT with the AMFm logo

Diagnostic test

P a g e | 3


Figure 2: Use of antimalarials among children who received an antimalarial

Among children under five with fever in the two weeks preceding the survey who received an antimalarial, the percentage who took specific antimalarial medicines (n=427).

60% of children who received an antimalarial received some type of non-artemisinin therapy. More than 1 in 3 children received ineffective chloroquine (36%). Slightly fewer than half of children who received an antimalarial got an ACT (45%), while 16% received an ACT with the AMFm logo. Fewer than 1% of children received an artemisinin monotherapy (Table 3.4.1).

59.4

5.6

36

19.6

44.5

22.9

15.6

0

10

20

30

40

50

60

70

80

90

100

Any non-

artemisinin therapy

SP Chloroquine Quinine Any ACT First-line ACT

(ASAQ)

ACT with

AMFm logo

P a g e | 4


Figure 3: Use of antimalarials among children who received an antimalarial by urban/rural strata

Among children under five with fever in the two weeks preceding the survey who received an antimalarial, the percentage who took specific antimalarial medicines by strata (n=427).

Among children who received an antimalarial, those in rural areas were more likely to receive an ACT than children in urban areas. However, children in urban areas were more likely to receive ACTs with the AMFm logo. None of the differences presented in Figure 3 were significant at the 5% level.

37

15

24

45

24

15

0

10

20

30

40

50

60

70

80

90

100

ACT First-line ACT (ASAQ) ACT with the AMFm logo

Urban areas Rural areas

P a g e | 5


Figure 4: Care seeking behavior: first place caregivers seek advice or treatment for fever

Among children under five with fever in the two weeks preceding the survey for whom advice or treatment was sought, percentage for whom advice or treatment was first sought at a given source (n=2215).

The most common first source of advice or treatment was at home (44% of children). Advice or treatment was first sought in the private for-profit sector by caregivers of 35% of children and the most common source of care within the private sector was from general retailers (commonly called épiceries or gargotes). For one in five children the first source of care was the public/not for profit sector (22% of children), and most commonly from public health facilities (16% of children).

At home (44%)

Public health facility (16%)

Public/not for profit sector

(22%)

General retailer (20%) Private sector

(35%)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

First place to seek care: Sector First place to seek care: Outlet type

Public health facility Community health worker

Public/not-for-profit health facility Private health facility

Pharmacy or drug store General retailer

Other private source

P a g e | 6


Figure 5: Source of antimalarials, ACTs and ACTs with the AMFm logo

Among children under five with fever in the two weeks preceding the survey who received an antimalarial, an ACT or an ACT with the AMFm logo, the percentage who sought treatment at a given source.

Most children received antimalarials, ACTs and ACTs with the AMFm logo from the public/not for profit sector, and namely from public health facilities. Around one in three children received ACTs from the private sector (39%), and one in five received ACT with the AMFm logo from the private sector (21%). For 18% of children who received any antimalarial, the product was already present in the home when the fever started, either as a partial treatment leftover from a previous illness episode or as a complete treatment bought in anticipation of illness.

46

62

68

39

29

21 18

10 15

0

10

20

30

40

50

60

70

80

90

100

Any Antimalarial n=427

Any ACT n=167

ACT with the logo n=79

Public not for profit sector Private sector At home

P a g e | 7


Figure 6: Caregiver awareness of and exposure to the AMFm logo and initiative

Among caregivers of all children under five (regardless of fever status), the percentage who have seen or heard of the AMFm logo, or heard of the initiative to reduce the price of ACTs.

Overall, around one in seven caregivers (13%) reported having seen or heard of the AMFm leaf logo. A similar proportion reported having heard of the initiative in Madagascar to reduce the price of ACTs (16%). Together, around a quarter of caregivers (26%) were aware of the AMFm logo or initiative. Awareness was significantly higher in urban areas than in rural areas (p<0.05).

29

12 13

22

15 16

44

24 26

0

10

20

30

40

50

60

70

80

90

100

Urban areas N=3466

Rural areas N=3315

All areas N= 6781

Seen or heard of the AMFm logo

Heard of initiative to reduce the price of ACTs

Either seen/heard of the AMFm logo or the initiative

P a g e | 8


Figure 7: Common sources of exposure to the AMFm logo and initiative

Among caregivers of children under five who have seen or heard of the AMFm logo, or heard of the initiative to reduce the price of ACTs, the percentage citing a given source of exposure.

The most common sources of exposure to the AMFm logo were medicine packaging (27%) and radio broadcasts (28%). By far most common source of exposure to information about the initiative to reduce the price of ACTs was the radio (63%). Caregivers could cite multiple sources and other responses (such as leaflets, general retailers, and community events) were each mentioned by fewer than 6% of respondents.

27

15

28

13

23

12

7

15

63

1

16 14

0

10

20

30

40

50

60

70

80

90

100

Medicine packaging

TV Radio Poster Public Health Facility

Community Health Worker

Sources of the AMFm logo (n=1289)

Sources of the price reduction Initiative (n=1216)

P a g e | 9


Figure 8: Meaning of the AMFm logo

Among caregivers of children under five who have seen or heard of the AMFm logo, the percentage citing a given meaning of the logo (n=1,289).

34% of caregivers did not know the meaning of the logo. Caregivers could cite multiple responses, and for those who did know a meaning the most common responses were antimalarial (36%) and medicine (32%). Only 10% of caregivers reported that the logo signified a cheap or effective antimalarial, one of the key messages of the supporting communication campaign for AMFm.

34

8

6

10

8

36

7

32

0 10 20 30 40 50 60 70 80 90 100

Don't know

Environment

Health

Effective antimalarial

Cheap antimalarial

Antimalarial

Cheap medicine

Medicine

P a g e | 10


1. Background 1.1 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 side of the market. The supply side is evaluated by

collecting outlet level and trend data on antimalarials and rapid diagnostic tests (RDTs) in public and

private sector outlets and wholesalers of antimalarial drugs (O’Connell et al., 2011). To evaluate

demand, data are collected at the household level on consumer treatment-seeking behaviour and

knowledge. In combination, the research components thread together the antimalarial market and

consumer behaviours (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. The research in Madagascar was conducted as follows:

three outlet surveys (2008, 2010 and 2011), supply chain research (2009) and two household surveys

(2009 and 2012).

Information on other ACTwatch studies can be found at www.ACTwatch.info.

1.2 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-

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

http://www.actwatch.info/

P a g e | 11


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 Fund’s 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 in the multi-country AMFm report (ICF Macro and

London School of Hygiene and Tropical Medicine, 2012) and the ACTwatch Madagascar outlet survey

report (ACTwatch Group, PSI/Madagascar and the IE team, 2011).

P a g e | 12


1.3 Country background

1.3.1 Overview of the country

Madagascar is the world’s fourth largest island and is located in Southern Africa, in the Indian Ocean,

east of Mozambique. In 2010, the population of Madagascar was 21.3 million (Population Division,

2010), with about 67% of the population living in rural areas (Population Division, 2011).

The climate of Madagascar is generally subtropical, with a hot and rainy season between November

and April, and a cooler dry season from May to October. The climate is tropical along the coast,

temperate inland and arid in the south. The east coast receives the most rain and is also prone to

cyclones between February and March. Temperatures are much cooler in the highlands and can be

as low as 4 degrees Celsius. Average temperature ranges in Antananarivo are from 9 to 20 degrees

Celsius in July to 16 to 27 degrees Celsius in December.

There are three official languages in Madagascar: Malagasy, French and English. There are a variety

of dialects within the nation, but Malagasy is the common spoken language. Malaria is known locally

as “tazo” or “tazomoka” (Encylopaedia Britannica, 2013).

Figure 1.3.1: Location of Madagascar

Source: http://www.hoveraid.co.uk/news2.html

Since February 2009, Madagascar has been in the throes of a political crisis that has led to a decline

in economic output and job losses. While economic growth had averaged 5% per year from 2002 to

2008, growth collapsed from 7% in 2008 to –5% in 2009; since 2009 GDP has grown at around 1%

per annum. Per capita income has returned to 2003 levels and estimates suggest that the proportion

of the population living below the poverty line may have increased by 10 percentage points since the

political crisis began. The number of out-of-school children has increased by half a million and acute

child malnutrition has increased in some areas by more than 50% (World Bank, 2012). A large

http://www.iexplore.com/topics/Madagascar;$sessionid$Z22AU1YAAGIAGP2MN5XCGWQ

P a g e | 13


fraction of official aid, representing 40% of Madagascar’s budget and 75% of inward investments,

remains on hold. The crisis has led to a decline in the delivery of social services, including health care

and significant cuts in the public investment program: from 2008 to 2010 the government cut its

national health budget by more than 30% (President’s Malaria Initiative [PMI], 2011). Even before

the political crisis Madagascar was one of the poorest countries in the world. Madagascar is

currently ranked 151 out of 186 countries in the 2013 Human Development Index (UNDP, 2013). The

population living below the poverty line has increased from 69% in 2005 to 77% in 2010 (INSTAT and

ICF Macro, 2010). The poverty rate in rural areas is significantly higher than in urban areas.

Administratively, Madagascar is divided into 6 provinces. Provinces are further divided into 22

regions, 111 health districts (119 administrative districts), 1,557 communes and 17,900 fokontany,

which is the smallest administrative unit. A commune is a combination of 6 to 20 fokontany.

1.3.2 Description of health care system

The health system has four distinct functional levels: central, regional, district and community. The

public sector is considered a major source of health care in Madagascar, especially in rural areas

where it accounts for more than 70% of primary contacts. In urban areas, it is estimated that fewer

than 40% of primary contacts occur through public facilities (PMI, 2011). Health facilities in the

public sector are composed of 138 hospitals; 1,335 level 2 health centres (Centre de Santé de Base

2), which are staffed by doctors; and 1,059 level 1 health centres (Centre de Santé de Base 1), which

do not have any doctors on staff (ICF Macro and London School of Hygiene and Tropical Medicine

[LSHTM], 2012). The government is in the process of decentralizing to give more decision-making

power to communes and fokontany.

Through the Global Fund’s National Strategy Application (NSA) Madagascar is in the process of

training 34,000 CHWs on integrated management of childhood illnesses, including malaria. Training

began in late 2011 and was set to be completed by the end of 2012. It is unclear how many CHWs

had been trained at the time of data collection for this survey. Community-based delivery through

CHWs has historically been an important distribution channel for antimalarials in Madagascar. For

example, from 2005 and 2008, CHWs distributed an estimated 600,000 treatment doses annually.

ACTs have been made widely available via public health facilities since 2007 (AMFm Independent

Evaluation Team, 2012).

Madagascar also has a significant private sector, which forms an integral part of the health system.

There are 44 hospitals, 724 private or religiously affiliated health centres and more than 1,500

doctors. There is also a private pharmaceutical sector with a network of 33 pharmaceutical

wholesalers, 200 pharmacies and 2,000 rural drug stores (dépôts de médicaments) (PMI, 2012). The

private sector also includes providers that have little or no qualifications (and are not authorized to

dispense medicines) such as grocery stores, among which data suggest that 20% stock antimalarials

(ACTwatch, 2012).

In the public sector, the central medical store, Centrale d’Achats de Médicaments Essentiels et

Générique, called SALAMA, is the only wholesaler importing and distributing drugs and medical

equipment. Public health facilities are either provided with drugs directly through SALAMA, or

through district public sector pharmacies (which receive medicines directly from SALAMA). Public

P a g e | 14


health facilities are supplied twice in a year, with the exception of facilities that are not accessible

during the rainy season.

Antimalarials are a notable exception to this system of procurement, as only quinine is available

through SALAMA. ACTs do not go through SALAMA and are instead procured and managed by the

Principal Recipient of the Global Fund and then make their way to health facilities via the National

Malaria Service (NMS). Private, not-for-profit health facilities working with community health

workers utilize the same system as the public sector, under a program developed by the MOH which

aims to make ACTs continuously available.

In the private, for-profit sector, distribution channels for ACTs are different but complementary.

Private health facilities are supplied by pharmaceutical wholesalers, most of which are concentrated

in the capital. The source of supply to grocery stores is unknown.

1.3.3 Epidemiology of malaria

Malaria is endemic in 90% of Madagascar, though the entire population is considered to be at risk

(PMI, 2011). 70% of the population lives in low-transmission areas that are prone to epidemics and

30% live in areas of high risk. 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 highlands—regions 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. Over 224,000 clinical malaria cases were reported through the

public health system in 2010 (WHO, 2012).

Figure 1.3.2: Geographical distribution of confirmed malaria cases Confirmed malaria cases per 1000 population, Madagascar, 2010

Source: World Malaria Report, 2010

P a g e | 15


The reported number of outpatient malaria cases decreased from an average of 1.6 million in the

period 2000–2004 to 299,094 in 2009, a reduction of 81%. Over the same period inpatient malaria

cases decreased by 69% and deaths by 75%. The drop in cases may also have been partly due to the

introduction of a new policy to implement parasitological diagnosis in public health facilities as

opposed to presumptive treatment of fever (WHO, 2010) and figures have continued to drop each

year overall (WHO, 2012).

However, while hospital deaths from malaria are estimated to have fallen from 17% in 2003 to 7% in

2011, severe malaria remains among the top five causes of overall mortality. As well, there was a

spike in malaria cases in late 2011 and early 2012 on the southeast coast with 2.5-10 times more

cases reported than in the previous two years. This may indicate that this area is in transition from

an endemic zone with year-round transmission to an epidemic-prone area and may also be linked to

decreasing immunity combined with reduced net use (PMI, 2012).

1.3.4 Antimalarial Policies and Regulatory Environment

In 2006 the National Malaria Control Program (NMCP) adopted artesunate/amodiaquine

combination therapy (ASAQ) as the first-line treatment for uncomplicated malaria. The second-

line treatment is artemether/lumefantrine (AL) with oral quinine as the alternative. Quinine is

recommended for the treatment of severe malaria.

Antimalarials are distributed by the public sector and, since 2008, community health workers

(CHWs) and can also be stocked by pharmacies and drug shops (dépôts de médicaments) that are

legally registered with government authorities as per the 1980 Malagasy public health code. These

outlets are obliged to prominently display their registration certificates and numbers. Except when

delivered by CHWs at the community level, antimalarials require medical prescription and do not

have over-the-counter status. The country’s new 2013-2017 National Strategic Plan requires

confirmatory testing by microscopy or RDT before treatment, with health facility targets of 95%

timely diagnosis and treatment in control districts and 100% in pre-elimination districts. Targets for

case management at the community level include RDT diagnosis of at least 80% of fever cases in

children under five (PMI, 2012).

Despite being six years since the official policy change to treatment with ASAQ, chloroquine remains

readily available in Madagascar, particularly in informal private for-profit outlets such as grocery

stores. The 2011 ACTwatch outlet survey estimates that one in five grocery stores stocks

chloroquine. In addition, almost 80% of drug shops had chloroquine in stock on the day of interview.

After some delay the government signed into force a regulatory note banning the sale of

chloroquine in July 2011. This ban was recalled in December 2011 and was then reintroduced in July

2012 following data collection for this survey. It is not yet clear what, if any, effect this ban has had

on the supply of chloroquine in Madagascar.

1.3.5 Malaria control strategy

Madagascar has moved from its National Strategic Plan for Malaria Control for 2008-2012 to a new

2013-2017 plan, which is based on recommendations made in a National Malaria Programme

Review in 2011. Some areas of Madagascar remain in a malaria control phase, while others have

moved to a pre-elimination phase.

P a g e | 16


The National Malaria Control Strategy tailors prevention and treatment interventions to regions of

the country according to epidemiological risk. Interventions include indoor residual spraying;

universal coverage of long-lasting insecticidal nets (LLIN) with distribution through antenatal clinics,

immunization visits, large-scale campaigns, and social marketing; intermittent preventive treatment

for pregnant women; and case management using parasitological diagnosis in health facilities and

clinical diagnosis and RDTs at the community level.

Between 2009 and the end of 2011, 9.4 million LLINs were distributed in 91 priority health districts

across Madagascar through mass campaigns and routine distribution (PMI, 2012). The 2011 MIS

indicated that 94% of households own at least one LLIN, which is a substantial increase on the

estimated 73% ownership among households reported in the 2008/2009 DHS. Similarly, net use

among children under five has increased from 58% to 89% sleeping under a net the previous night

(INSTAT and ICF Macro, 2010; INSTAT and ICF International, 2012). Under the 2013-2017 National

Strategic Plan, Madagascar has adopted the strategy of one LLIN for every two persons to achieve

universal coverage in all epidemic-prone districts except 20 (out of 112), which have IRS and

comprehensive epidemic surveillance systems.

Figure 1.3.5: Key net indicators, Madagascar 2008-2011

Indicator 2008/2009 DHS (PMI baseline)

MIS 2011

Proportion of all households with at least one ITN

73% 94%

Average number of ITNs per household 1.1 1.8

Proportion of households with at least one ITN for every two people

23% 37%

Proportion of population with access to an ITN within their household

44% 67%

Proportion of children under five years old who slept under an ITN the previous night

58% 89%

Proportion of pregnant women who slept under an ITN the previous night1

58% 85%

Source: INSTAT & ICF Macro, 2010 and INSTAT & ICF International, 2012

From 2008 to 2012, generalized IRS campaigns covered the Central Highlands and surrounding

districts (Fringe) and extended to districts in the South and West. At the end of 2011, 79% of

households in target districts received at least one round of IRS in the past year and this completed

four consecutive years of universal IRS in the Central Highlands and Fringe districts. By the end of the

2012, the South and West districts were slated to have received three consecutive years of IRS. From

2013, all 54 districts in these regions will be transitioned to targeted IRS and vigilant surveillance

(PMI, 2012). In target districts, pregnant women are also treated with IPTp. While the 2008/2009

DHS indicated that 86% of women attended at least one antenatal clinic visit during pregnancy, only

15% of pregnant women received at least one dose of SP for malaria prevention, rising to 31% by

2011 (INSTAT & ICF International, 2012).

According to national policy, consultation and treatment of uncomplicated malaria in children is to

be provided in public health facilities free of charge. In September 2008, a socially-marketed co-

blister pack of artesunate+amodiaquine (AS+AQ) for children under five was launched under the

P a g e | 17


name ACTipal, with formulations for infants and young children. ACTipal was highly subsidized

nationwide at 100 Ariary (about USD $0.05) and distribution channels included CHWs, pharmacies

and drug stores. ACTipal was repackaged in November 2010 and re-priced at 200 Ariary in

pharmacies and medical stores, while remaining at 100 Ariary at the community level. The brand

was discontinued by authorities in late 2011 on the grounds of packaging issues.

Since mid-2009, the first-line treatment (ASAQ) has been available in a co-blistered form to public

health facilities and CHWs. Efforts to increase availability and use of ACTs have been made through

public sector distribution, funded by the Global Fund, as well as initiatives through private and

community channels, funded by the Global Fund and the U.S. President’s Malaria Initiative.

Madagascar was one of seven countries selected to be a recipient of subsidized ACTs through the

Affordable Medicines Facility – Malaria (AMFm), which was intended to scale up the widespread

availability of ACT in the private and public sectors (see Section 1.3.7 below for more).

Parasitological diagnosis of fever is of increasing importance in Madagascar in the context of scaled

prevention efforts and pre-elimination plans for some areas of the country. The 2013-2017 National

Strategic Plan requires confirmatory testing before treatment, either by microscopy or RDT. Training

by the NMCP on use of rapid diagnostic tests (RDTs) began in 2005, and microscopy capabilities have

been strengthened through increased human resources and training in the public and private

sectors. By mid-2011, over 3,700 CHWs were trained to use RDTs and USAID had procured

1 million RDTs - 700,000 for community use and 300,000 for the private sector. Public facility

diagnostics continue to be supported by Global Fund, which ordered 1.7 million tests in 2011. At the

CHW level, the national target for community-based RDT testing before treatment is 80% by 2015,

rising to 100% for health facilities in pre-elimination areas (PMI, 2012).

1.3.6 Malaria financing

Malaria control financing in Madagascar is almost entirely from external sources. Funding increased

massively from 2006 (less than US$5 million) to 2007 (more than US$20 million) and has increased

every year since then, to more than US$70 million in 2011. Funds come primarily from the Global

Fund and the U.S. President’s Malaria Initiative, with some funding also coming from United Nations

agencies such as UNICEF and WHO and other bilateral agencies (WHO, 2012).

The National Malaria Control Program has received several Global Fund grants for malaria: a Round

7 grant worth US$69 million was signed in August 2008; a Rolling Continuation Channel Round 4

grant for US$64 million was signed in October 2009; and a US$73 million National Strategy

Application was signed for 2010-2013. As well, Madagascar’s AMFm grant provided an additional

US$1.4 million in 2010 and US$760,000 in 2011 for subsidized ACTs (PMI, 2012).

As one of 15 PMI countries, Madagascar received US$16.7 million in 2009; US$33.9 million in 2010,

US$28.8 million in 2011 and US$27 million in 2012. US$ 25.92 million has been allocated for 2013.

Although direct support to the government of Madagascar was suspended following the coup in

2009 and will remain so until after the elections slated for October 2013, PMI has continued to

support malaria control through its international and local partners. PMI funds prevention

interventions including LLIN distribution and promotion of LLIN use, indoor residual spraying and

promoting malaria in pregnancy interventions through awareness-raising and education at the

P a g e | 18


community level. As part of integrated community case management (iCCM) of children under 5,

PMI funds RDTs and ACTs for community health workers (PMI, 2012).

Between 2007 and 2009, UNITAID donated more than $5 million to fund ACT procurement.

Madagascar is a World Bank Booster Programme country with funding of $100,000 in 2010 and

2011. World Bank and UNITAID funds were put on hold during the political crisis.

1.3.7 AMFm Phase 1 pilot

Madagascar’s contract for the AMFm began in May 2010. Eight private-sector first-line buyers were

involved in procurement, as well as the public sector Unité de Gestion de Projet (UGP). The first

shipment of AMFm drugs arrived to private sector first-line buyers in October 2010 and to the public

sector in February 2011. Artesunate/amodiaquine (ASAQ) comprised approximately three-quarters

of the doses procured through the AMFm with the remaining quarter being

artemether/lumefantrine (AL). Customs clearance issues created tensions between first-line buyers

and local authorities in the beginning, but ministry officials then streamlined the process of clearing

customs to remove this particular bottleneck. Nonetheless, lead times from order approval to

delivery ranged from several weeks to six months and were particularly long during the second half

of 2011.

While availability of ACTs generally increased, the project experienced widespread stock-outs in the

public and private sectors, exacerbated by malaria epidemics in 2011 and 2012 in the south and

southeast regions. In addition, promotional support for subsidized ACTs was substantially limited by

enforcement of national regulations on prescription drug advertising. Minimal communications

activities included a national launch event in January 2010 and the distribution of AMFm logo

leaflets, pens, posters and prescription pads to 30 to 40 wholesalers, 200 pharmacies, and 2,000

drug stores. TV and radio spots ran for only one month from April to May 2011 before authorities

pulled the campaign from circulation and annulled plans for billboards and other activities. By the

end of 2011, only US$1.3 million had been disbursed by Global Fund to partners implementing

supporting interventions. Despite glitches in the program, ACT orders through the AMFm continued

and treatment volumes increased. By June 2012, a total of 2.6 million ACT treatments were ordered

through the AMFm in Madagascar (AMFm Independent Evaluation Team, 2012).

P a g e | 19


1.3.8 Other research findings

Data from a Demographic and Health Survey, Malaria Indicators Survey and the ACTwatch baseline

household survey (2008/2009) are presented below.

Results from DHS, MIS and ACTwatch surveys in Madagascar Background characteristics

Any antimalarial treatment

ACT treatment

Prompt ACT treatment

Source

Percentage N Percentage N Percentage N Source & date

Madagascar MDHS

Nov 08-Aug 09 Residence

Urban 14.9 164 1.8 164 1.8 164 Rural 20.5 952 0.9 952 0.1 952

Wealth quintiles Lowest 19.1 279 1.1 279 0.0 279 Second 24.2 249 0.3 249 0.0 249 Middle 22.9 198 1.6 198 0.0 198 Fourth 14.7 198 0.3 198 0.3 198 Highest 16.4 192 1.9 192 1.9 192

Total 19.7 1,116 1.0 1,116 0.4 1,116

Madagascar ACTwatch

Dec 08–Jan 09 Residence

Urban 44.0 1,061 4.2 1,061 3.4 1,061 Rural 47.7 1,059 3.3 1,059 3.1 1,059


Total 47.3 2,120 3.4 2,120 3.2 2,120

Madagascar MIS

Mar-May 2011 Residence

Urban 13.8 65 3.7 65 1.9 65 Rural 20.3 873 3.8 873 2.1 873


Total 19.8 938 3.8 938 2.1 938

Madagascar

ACTwatch April-May

2012 Residence

Urban 16.9 1,326 6.2 1,326 4.5 1,326 Rural 19.2 1,062 8.7 1,062 7.3 1,062


Total 19.0 2,388 8.4 2,388 7.0 2,388

P a g e | 20


2. Methods 2.1 Household Survey

The primary objectives of this household survey are to monitor key aspects of treatment seeking

behaviour for fever in children under five between 2008/2009 and 2012 nationally as well as in rural

and urban areas. Key behaviours measured include:

use of diagnostic blood testing;

presumptive antimalarial treatment;

presumptive ACT treatment; and

presumptive treatment with the national first-line antimalarial.

These comparisons provide information on the success of national malaria control efforts

maintained or scaled up between 2009 and 2012, including the AMFm.

A secondary objective of this study was to estimate exposure to AMFm supporting interventions that

were designed to create informed demand and lead to appropriate fever treatment seeking

behavior.

2.1.1 Sampling Approach

2.1.1.1 Target Population

The target population for the primary objectives of this study is caregivers of children under five

living in malaria-endemic areas (urban or rural) who have had fever in the past two weeks. The

target population for the secondary objective is caregivers of children under five that had no

children with fever in the previous two weeks.

2.1.1.2 Sample Size

The household survey is designed to measure differences in indicators over time and between urban

and rural strata. The following paragraphs summarise the methodology for determining the overall

sample size needed to detect statistically significant changes in proportions over time.

The key question for powering the study was: How has presumptive treatment of fever in children

under five changed between 2008/9 and 2012 nationally and in urban and rural areas?

In 2008/9, 47% of children under five with malaria in the past 2 weeks received treatment

with any antimalarial (44% in urban areas, 48% in rural areas). ACT treatment was 3%.

The desired sample size would detect a 10% change over time in treatment nationally, and

within urban and rural areas.

The required number of children for a single domain was calculated using the following formula:

2

21

2

221111

)(

)1()1()1(2

PP

PPPPZPPZDeffn

P a g e | 21


where: Urban Rural P1 0.44 0.48 P2 0.54 0.58 P 0.49 0.53 and:

Deff = 2.0, the estimated design effect of the sampling strategy;

Zα = 1.96, corresponding to an α (type I) error of 5% with a two-sided test

Z1-β = 0.84, corresponding to a power of test at 80% (or a type II error of 20%)

This gave required samples sizes of 782 children in urban areas and 779 children in rural areas. When

estimating how many households would need to be screened in order to achieve this sample size we

began by inflating these figures by 10% to allow for a certain level of non-response during survey

implementation; this gave a required 861 children in urban areas and 857 in rural areas. The next

step was to convert the number of children required to households, and account for estimated fever

prevalence.

According to the 2011 Madagascar Malaria Indicator Survey (MIS) (INSTAT and ICF Macro, 2012), the

average number of children under five per households was 0.6 in urban areas and 0.9 in rural areas.

Fever prevalence estimates of 17.6% in urban and 14.5% in rural areas were used given findings

from the 2011 MIS (where data were collected in November-December). Based on these

assumptions, the number of households to be screened was 8,442 in urban areas and 6,802 in rural

areas in order to achieve a sample of at least 861 fevers in urban and 857 fevers in rural areas.

2.1.1.3 Selection Procedure for Clusters and EAs

The last census in Madagascar was conducted in 1993 and was used as the sample frame for the

2012 household survey. The list of communes from the 1993 census included population size, and

classification as either urban or rural (Institut National de la Statistique, 1999). Geographical areas

that are non-endemic for malaria, such as the capital city of Antananarivo, were excluded from the

sampling frame.

Nationally-representative samples were selected using stratified, three-stage cluster sampling. A

sampling summary is provided in Table 2.1.1. At the first stage, a total of 46 communes (18 urban

and 28 rural) were selected with probability proportional to size (PPS) from the 1993 census. These

primary sampling units were administrative areas that had already been selected as part of the

ACTwatch outlet survey. A list of second stage enumeration areas (EAs), or fokontany, was then

drawn with PPS for each cluster: 4 EAs per commune in urban areas and 3 EAs per commune in rural

areas for a total of 156 EAs. The third stage involved the systematic selection of households from a

list of households in each cluster, where households had been mapped prior to the survey.

Systematic sampling with a specific skip interval was used to select households for screening. 118

households were selected from urban fokontany and 81 from rural fokontany.

P a g e | 22


Table 2.1.1: Sampling summary

Urban areas

Rural areas

Total

Number of communes selected for the outlet survey

18 28 46

Number of enumeration areas (EAs) or fokontany selected per commune

4 3 --

Total number of EAs 72 84 156

Number of households to be screened per EA 118 81 --

Total number of households screened 8,496 6,804 15,300

All caregivers of a child under five with fever in the past two weeks were eligible to be included in

the survey. All caregivers of a child under five who were listed in the household were eligible to

complete a questionnaire module on exposure and awareness of the AMFm.

2.1.2 Questionnaire

Caregivers responded to a series of questions about management of fever occurring among children

in their care in the two weeks preceding the survey. Five modules were used in the household

survey: 1) a screening module, to identify households that were eligible for the full questionnaire or

the supplementary AMFm awareness section; 2) a household listing of all the usual members in the

selected households together with basic information on the characteristics of each person listed,

including age, sex and for children under five, their primary caregiver; 3) a household questionnaire

module, modelled after the Demographic and Health Survey (DHS), to collect information on housing

characteristics and household assets used in assessment of relative wealth status; 4) a treatment

seeking module, which included questions documenting the type, timing, source and cost of

treatments acquired for the child’s fever; and 5) an AMFm awareness section, which included

questions on caregiver awareness of and exposure to the AMFm interventions and activities.

Caregiver recall and recognition of the type of treatment acquired was aided by the use of a

comprehensive antimalarial field guide with photographs and brand names of common

antimalarials, antibiotics and fever reducers available in public and private sector outlets. Modules 1-

4 were administered to caregivers with a child under five with fever in the last two weeks. Module 5

was administered to any caregiver with a child under five.

The questionnaire was translated into Malagasy through a process of forwards and backwards

translation. For survey implementation, questionnaires were programmed into Personal Digital

Assistants (PDAs). PDA programming used Visual Basic running on the Windows Mobile 5.0

operating system. The PDA programming was pretested prior to the main data collection.

2.1.3 Data Collection

2.1.3.1 Preparatory Phase

The study received ethical clearance from Madagascar’s ethical approval committee at the Ministry

of Health on the 30th March 2012.

P a g e | 23


62 candidates participated in an eight-day household survey training in Antananarivo between the

4th and 14th of April. Standardised training materials developed by ACTwatch were adapted to the

national setting, and sessions were led by PSI/Madagascar research staff. Training session topics

covered sampling and identifying households, gaining informed consent, and administration and

completion of the questionnaire modules. Interviewers were trained on how to use the prompt

cards and when to administer different sections of the questionnaire. PDAs were introduced to the

field staff after the main body of the training had been completed. A field practice session was

undertaken to mimic actual data collection. Of the 62 candidates, 37 were selected as interviewers,

9 as supervisors and 9 as quality controllers.

Supervisors and quality controllers received additional training to clarify roles and responsibilities in

the field. This training also included a review of logistical procedures to be followed during data

collection, trouble-shooting PDAs and backing up data.

Training was also held for the household mapping teams responsible for conducting the household

census in each of the selected EAs. This training was conducted on the 4th of April, enabling

household mapping to begin prior to data collection.

2.1.3.2 Fieldwork

Nine teams carried out data collection. Teams consisted of one team supervisor, one quality

controller and four interviewers. Two coordinators from the PSI/Madagascar research team were

responsible for managing the supervisors and ensuring that standardized methods were

implemented. Fieldwork commenced on the 23rd of April and was completed on the 10th of June

2012.

For each household selected, geographic location and the household’s longitude and latitude

coordinates were recorded. The fieldworker then identified the household head or primary caregiver

and administered the screening questions. All caregivers of children under five with fever in the two

weeks preceding the survey were invited to participate in the study. Primary caregivers were

identified based on their responsibilities as the main caregiver for the child with fever (i.e.

responsible for daily care of the child including supervision, bathing and feeding). Primary caregivers

were typically the child’s mother with the exception of orphaned and foster children. In all selected

households, all caregivers of children under five were administered a module to assess caregiver

awareness of and exposure to the AMFm interventions and activities (i.e. irrespective of a child’s

recent fever status).

For households that were eligible for either the full questionnaire or the AMFm module, the

fieldworker read an information sheet to the household head or representative and obtained

witnessed oral consent to proceed with the interview. A full interview lasting approximately one

hour was conducted with each caregiver of an eligible child. Informed consent was sought from each

household member interviewed.

At the end of each day data were synchronized from the PDAs to computers and submitted to a

database acting as a central server.

P a g e | 24


Teams were visited by supervisors and quality controllers in the field during the survey period. Spot

checks were conducted on at least 10% of all households by quality controllers. Supervisors

observed 10% of interviews, verifying adherence to study procedures. Supervisory and quality

control measures also included a review of teams’ PDA records, using a tool in Excel to facilitate data

checks. Data were checked for inconsistencies, irregular skip patterns or large amounts of missing

values. PSI coordinators also uploaded data from interviewers’ PDAs for review, and documented

reasons why certain households could not be accessed.

Throughout the planning, training and implementation of the survey, Standard Operating

Procedures (SOPs) developed by ACTwatch were used to ensure the collection of high quality data

and set out a framework for documenting challenges encountered during fieldwork.

2.1.5 Data analysis

2.1.5.1 Data analysis process

Step by step instructions on how to clean the data using range and consistency checks were utilized

during the analysis process across all ACTwatch countries. Data cleaning and analysis was centralised

to maintain consistency in methods and cleaning decisions. Commands executed during data

cleaning were documented using syntax files, and the results archived in Stata log files.

A tabulation plan was created for the household survey report, and tables were produced using

standardized data management and analysis syntax files in Stata.

2.1.5.2 Indicators

Indicators of treatment seeking behaviour and treatment of fever were constructed from caregiver

reports on treatment sources; type of treatments acquired (brand names); timing of treatments; and

whether or not the child received a diagnostic blood test for malaria. Brand names were used to

categorize medicines according to generic antimalarial type (e.g. chloroquine, quinine, artesunate-

amodiaquine). These were further classified as artemisinin combination therapy (ACT), artemisinin

monotherapy, or non-artemisinin monotherapy. Indicators were calculated using the three classes of

antimalarials above, as well as an overall category for any antimalarial.

Consistent with indicators calculated from the DHS and MIS, antimalarial treatment received the

same or next day after onset of fever was used as a proxy measure for treatment within 24 hours of

onset of fever and is considered prompt treatment.

Treatment sources were categorized as belonging to either the public/not for profit sector or private

for-profit sector. Public health facilities (PHFs), community health workers (CHWs) and non-profit

health facilities were classified as public/not for profit, with PHFs constituting the majority of this

category. The private for-profit sector encompassed all other outlets, with or without qualified

health workers, such as private for-profit health facilities, licensed pharmacies and informal market

stalls.

Household wealth status was assessed relative to other households using measures of housing

P a g e | 25


characteristics, water, sanitation and household assets modelled after the DHS household

questionnaire (www.measuredhs.com). A wealth index was constructed from the individual

indicators, which were assigned a weight through principal component analysis and standardized in

relation to a standard normal distribution. Each child was categorized according to the value of their

household’s wealth index, and placed in one of five wealth quintiles, ranging from poorest to least

poor.

P a g e | 26


3. Results 3.1 Characteristics of the sample

Figure 3.1.1: Survey flow diagram

Households selected [15,195]

Eligible respondent not available [74]

Households not screened [1,169]

Entire household absent [865]

Dwelling abandoned [185]

Households screened [14,026]

Refused [31]

Other [14]

Households which

did not meet either screening

criteria [7,361]

Households which met

screening criteria1

1 = [2,176]

Eligible respondent not available

[8]

Households not interviewed

[15]

Entire household absent [0]

Refused [6]

Households interviewed

[2,161]

Other [1]

Number of eligible

children2

[2,440]

Caregivers of

eligible children [2,217]

Eligible children

interviewed3

[2,388]

Caregivers of eligible children

interviewed3

[2,169]

1 Household includes at least one child under five with fever in the past two weeks.

2 Eligible children means a child was under five years of age and had fever in the past two weeks.

3 Interviewed means the final interview status was completed or partial. Partial interviews are counted if the relevant Section of the question was at least begun: Section 4 for eligible children; Section 5 for caregivers of eligible children; the AMFm section for all caregivers.

Source: ACTwatch Household Survey, Madagascar, 2012.

P a g e | 27


Table 3.1.1: Results of the household and individual interviews

Number of households, number of interviews and response rates, according to location of residence (unweighted).

Residence

Urban Rural Total

Household interviews

Households selected 8,438 6,757 15,195

Households occupied 8,351 6,659 15,010

Households screened 7,874 6,152 14,026

Eligible households (criteria 1)1

1,217 959 2,176

Eligible households (criteria 2)2

2,170 2,319 4,489

Household response rate3 94.3 92.4 93.4

Interviews conducted about children under five with fever (criteria 1)

Eligible children 1,362 1,078 2,440

Eligible children for whom fever treatment information was recorded

1,326 1,062 2,388

“Child” response rate4

97.4 98.5 97.9

1 Household includes at least one child under five with fever in the past two weeks.

2 No children under five with fever in the past two weeks, but at least one child under five in the household.

3 Households screened / households occupied.

4 Children for whom fever treatment information was complete or interrupted / eligible children.


P a g e | 28


Table 3.1.2: Demographic characteristics

Characteristics of children under five with fever in the two weeks preceding the survey, of children’s caregivers, and of households for interviewed cases (unweighted).

Number (Percentage)

Children under five

with fever Caregivers of children under five with fever

Households

N=2,388 N=2,164 N=2,161

Strata

Urban 1,326 (55.5) 1,209 (55.9) 1,206 (55.8)

Rural 1,062 (44.5) 955 (44.1) 955 (44.2)

Household wealth index

Lowest 483 (20.2) 427 (19.7) 425 (19.7)

Second 481 (20.1) 430 (19.9) 424 (19.6)

Middle 475 (19.9) 430 (19.9) 425 (19.7)

Fourth 458 (19.2) 429 (19.8) 424 (19.6)

Highest 473 (19.8) 432 (20.0) 424 (19.6)

Missing 18 (0.8) 16 (0.7) 39 (1.8)

Age (years)

Infants (<1 year) 502 (21.0) - -

1 510 (21.4) - -

2 452 (18.9) - -

3 450 (18.8) - -

4 474 (19.9) - -

Sex

Male 1,165 (48.8) - -

Female 1,223 (51.2) - -

Education

No education - 456 (21.1) -

Some primary - 880 (40.7) -

Primary or higher - 824 (38.1) -

Missing - 4 (0.2)


P a g e | 29


3.2 Treatment for fever

Table 3.2.1: Treatment of children with fever

Among children under five with fever in the two weeks preceding the survey the percentage who took antibiotic medicines, the percentage who took antipyretic medicines, the percentage who took antimalarial medicines, and the percentage who took the antimalarial medicines the same or next day following the onset of fever, by background characteristics.

Percentage who took antibiotic medicines

Percentage who took antipyretic medicines

Percentage who took antimalarial medicines

Percentage who took antimalarial medicines

same or next day

Number of children

with fever

Age (in years)

<1 41.4 (33, 50.4) 57.4 (48.7, 65.7) 13.1 (8.0, 20.7) 11.4 (7.0, 18.0) 502

1 39.8 (31.2, 49.0) 59.8 (52, 67.1) 18.2 (11.0, 28.6) 15.6 (8.9, 26.1) 510

2 44.1 (34.9, 53.8) 57.0 (48.4, 65.2) 19.7 (14.7, 26.0) 15.0 (10.7, 20.8) 452

3 37.5 (28.1, 48.0) 59.0 (48.9, 68.5) 20.5 (12.9, 31.0) 16.4 (9.6, 26.6) 450

4 33.4 (23.7, 44.9) 58.0 (48.1, 67.2) 24.0 (16.9, 32.9) 19.4 (13.7, 26.6) 474

Strata

Urban 43.5 (39.7, 47.5) 76.7 (72.7, 80.2) 16.9 (13.5, 20.8) 13.2 (10.2, 17.0) 1,326

Rural 38.8 (32, 46.2) 56.4 (50.5, 62.2) 19.2 (13.8, 26.0) 15.8 (11.2, 21.7) 1,062

Caregiver’s education

No education 31.2 (23.7, 39.9) 46.3 (36.7, 56.2) 17.3 (10.5, 27.3) 13.0 (8.0, 20.4) 509

Some primary 43.4 (34.8, 52.5) 60.2 (54.4, 65.8) 20.9 (14.3, 29.4) 17.1 (11.1, 25.4) 974

Primary completed + 41.2 (33, 49.9) 70.8 (62.6, 77.9) 17.1 (11.4, 24.7) 15.4 (9.9, 23.3) 894

Wealth index

Poorest 32.3 (24.5, 41.2) 46.5 (38.9, 54.4) 20.5 (12.8, 31.2) 16.3 (9.6, 26.5) 483

Second 39.8 (31.2, 49.0) 55.0 (45.5, 64.2) 20.4 (13.7, 29.3) 17.0 (11.6, 24.1) 481

Middle 46.7 (35.8, 57.9) 67.9 (61.8, 73.3) 17.0 (11.3, 24.8) 14.2 (9.7, 20.4) 475

Fourth 42.8 (36.6, 49.3) 75.2 (65.5, 82.8) 14.7 (9.5, 21.9) 12.3 (7.7, 19) 458

Richest 42.8 (34.0, 52.2) 77.3 (66.3, 85.6) 17.1 (11.6, 24.6) 12.5 (8.2, 18.6) 473

All children 39.3 (33.0, 45.9) 58.3 (52.8, 63.5) 19.0 (14.1, 25.1) 15.5 (11.3, 20.9) 2,388


P a g e | 30


3.3 Diagnosis

Table 3.3.1: Diagnosis of fever among children under five

Among children under five with fever in the two weeks preceding the survey the percentage that had blood taken from finger or heel for testing; among children who had blood taken, the type of test received and the test result, by background characteristics.

Among children under five with fever who had blood taken for testing:

Percentage who reported having

blood taken from finger or heel for

testing

Number of

children

Percentage who received an RDT

Percentage who received

microscopy

Percentage who don’t know /

missing test type

Percentage who self report positive for

malaria

Percentage who self report

negative for malaria

Percentage who don’t know /

missing test result

Number of children

tested

Age (in years)

<1 11.4 (7.4, 17.2) 499 97.7 (87.4, 99.6) 0.2 (<0.1, 1.9) 2.1 (0.3, 13.4) 12.7 (4.7, 30.2) 40.2 (21, 62.9) 47.1 (27.1, 68.0) 44

1 13.9 (8.8, 21.2) 506 97.0 (86.6, 99.4) 3.0 (0.6, 13.4) 0 35.6 (18.9, 56.7) 35.8 (22.5, 51.6) 28.7 (16.0, 45.9) 66

2 11.1 (6.5, 18.2) 449 84.5 (61.0, 95.0) 5.1 (0.9, 23.3) 10.4 (2.5, 34.5) 40.1 (17.8, 67.4) 23.3 (10.6, 43.8) 36.6 (15.8, 64.0) 47

3 10.7 (4.8, 22.0) 447 97.7 (86.5, 99.7) 1.8 (0.2, 14.9) 0.4 (0.1, 2.0) 81.2 (56.5, 93.5) 11.4 (3.8, 29.6) 7.3 (1.5, 28.9) 43

4 12.3 (7.3, 20.2) 472 97.6 (87.6, 99.6) 2.4 (0.4, 12.4) 0 60.8 (42.3, 76.6) 25.9 (12.7, 45.6) 13.3 (5.8, 27.7) 58

Strata

Urban 9.9 (7.9, 12.5) 1,317 93.3 (85.1, 97.1) 5.1 (1.9, 12.8) 1.6 (0.5, 5.3) 26.6 (17.6, 38.2) 50.1 (38.6, 61.6) 23.2 (17.3, 30.4) 132

Rural 12.2 (8.2, 17.7) 1,056 95.4 (90.3, 97.8) 2.3 (0.9, 5.9) 2.3 (0.7, 8.0) 45.9 (32.5, 59.8) 26.7 (19.3, 35.8) 27.4 (17.1, 40.8) 126


No education 6.6 (3.5, 12.0) 506 93.8 (73.8, 98.8) 6.0 (1.1, 26.4) 0.2 (<0.1, 1.9) 33.9 (15.9, 58.1) 34.4 (19.1, 53.7) 31.7 (12.7, 59.7) 38

Some primary 14.3 (9.7, 20.7) 966 96.0 (90.8, 98.4) 2.0 (0.6, 6.3) 2.0 (0.4, 8.6) 51.0 (33.6, 68.2) 18.6 (11.6, 28.5) 30.4 (17.1, 48.1) 117

Primary completed + 13.9 (8.8, 21.2) 890 94.0 (78.4, 98.6) 1.6 (0.6, 4.2) 4.4 (0.7, 24.0) 34.4 (22.4, 48.8) 49.1 (35.8, 62.5) 16.5 (6.5, 36.1) 100

Wealth index

Poorest 11.7 (6.8, 19.6) 482 98.7 (89.3, 99.8) 0 1.3 (0.2, 10.7) 46.9 (26.5, 68.4) 25.2 (14.1, 41.1) 27.9 (15.5, 45.0) 50

Second 15.1 (9.9, 22.3) 480 94.7 (85.7, 98.2) 3.4 (0.9, 11.6) 1.9 (0.2, 13.3) 47.4 (30.1, 65.4) 25.3 (13.6, 42.3) 27.2 (15.3, 43.8) 61

Middle 11.7 (6.9, 19.2) 472 91.8 (74.3, 97.7) 3.6 (0.8, 14.8) 4.6 (0.6, 27.0) 38.9 (22.0, 59.0) 31.8 (17.6, 50.5) 29.3 (12.2, 55.4) 51

Fourth 5.1 (2.4, 10.4) 452 90.6 (73.7, 97.1) 7.6 (2.0, 24.9) 1.8 (0.4, 8.1) 59.8 (32.4, 82.2) 21.4 (9.6, 41.2) 18.8 (7.9, 38.7) 40

Richest 9.7 (6.3, 14.6) 469 96.9 (81.9, 99.5) 1.8 (0.2, 12.5) 1.4 (0.2, 9.8) 13.6 (6.3, 27.1) 71.0 (52.8, 84.3) 15.3 (5.7, 35.3) 55

All children 12.0 (8.3, 16.9) 2,373 95.2 (90.6, 97.6) 2.5 (1.1, 5.6) 2.3 (0.7, 7.3) 44.4 (31.8, 57.7) 28.5 (21.3, 37.1) 27.1 (17.5, 39.4) 258


P a g e | 31


Table 3.3.2: Source of diagnosis

Among children under five with fever in the two weeks preceding the survey who received a diagnosis, source of diagnostic test, by background characteristics.

Public / not for profit sector Private sector At home

Public health facility

Community health worker

Private not-for-profit

health facility Total

Private health facility

Pharmacy / drug store

General retailer Other Total At

home

Missing or

don’t know

Number of

children tested

Age (in years)

<1 90.8 (78.8, 96.3) 2.1 (0.3, 13.4) 0.9 (0.3, 3.1) 93.7 (80.9, 98.1) 6.3 (1.9, 19.1) 0 0 0 6.3 (1.9, 19.1) 0 0 44

1 84.5 (66.0, 93.8) 11.9 (3.6, 33.0) 0.7 (0.3, 2.1) 95.7 (88.0, 98.5) 4.3 (1.5, 12.0) 0 0 0 4.3 (1.5, 12.0) 0 0 66

2 73.0 (58.0, 84.1) 10.6 (4.3, 23.8) 4.3 (1.0, 17.5) 87.7 (65.3, 96.5) 12.3 (3.5, 34.7) 0 0 0 12.3 (3.5, 34.7) 0 0 47

3 70.6 (40.1, 89.6) 19.4 (6.2, 47.0) 6.0 (1.4, 22.6) 96.0 (89.3, 98.6) 4.0 (1.4, 10.7) 0 0 0 4.0 (1.4, 10.7) 0 0 43

4 40.4 (21.0, 63.3) 33.5 (13.9, 61.3) 3.6 (0.7, 17.1) 77.6 (61.8, 88.1) 14.4 (7.2, 26.9) 0 0 0 14.4 (7.2, 26.9) 8.3 (2.0, 28.3) 0 58

Strata

Urban 54.4 (43.8, 64.7) 1.7 (0.4, 6.8) 9.3 (4.7, 17.6) 65.0 (54.9, 73.9) 34.1 (25.2, 44.4) 0 0 0 34.1 (25.2, 44.4) 1.6 (0.4, 7.0) 0 132

Rural 74.0 (60.2, 84.3) 16.5 (9.3, 27.5) 2.3 (0.7, 7.3) 92.4 (85.0, 96.3) 5.9 (2.6, 12.9) 0 0 0 5.9 (2.6, 12.9) 1.7 (0.4, 7.2) 0 126


No education 91.5 (78.5, 96.9) 5.3 (1.1, 22.2) 2.2 (0.3, 13.0) 99.0 (96.3, 99.7) 0.9 (0.3, 3.2) 0 0 0 0.9 (0.3, 3.2) 0.5 (0.1, 3.6) 0 38

Some primary 70.7 (54.1, 83.2) 17.0 (8.0, 32.4) 2.8 (0.6, 11.3) 89.7 (80.5, 94.9) 9.1 (4.3, 18.2) 0 0 0 9.1 (4.3, 18.2) 1.2 (0.2, 8.0) 0 117

Primary Completed +

63.8 (42.5, 80.9) 18.5 (6.6, 42.1) 3.5 (1.0, 11.6) 85.8 (70.2, 93.9) 10.6 (4.5, 23.2) 0 0 0 10.6 (4.5, 23.2) 3.6 (0.4, 23.6) 0 100

Wealth index

Poorest 85.3 (75.9, 91.5) 11.3 (6.2, 19.7) 0 95.3 (82.8, 98.9) 2.3 (0.4, 13.0) 0 0 0 2.3 (0.4, 13.0) 2.4 (0.3, 15.5) 0 50

Second 68.0 (51.7, 80.8) 23.3 (11.1, 42.5) 3.6 (0.7, 17) 94.9 (86.2, 98.2) 5.1 (1.8, 13.8) 0 0 0 5.1 (1.8, 13.8) 0 0 61

Middle 70.2 (45.2, 87.1) 9.6 (2.3, 33.0) 1.6 (0.3, 9.5) 81.5 (61.2, 92.5) 14.3 (5.3, 32.9) 0 0 0 14.3 (5.3, 32.9) 4.3 (0.6, 24.7) 0 51

Fourth 56.3 (26.7, 82.0) 0 19.7 (4.9, 53.9) 75.0 (51.5, 89.5) 26.5 (11.1, 51.1) 0 0 0 26.5 (11.1, 51.1) 0 0 40

Richest 50.8 (31.4, 70.0) 13.0 (1.9, 53.8) 6.9 (2.6, 16.9) 70.6 (54.6, 82.8) 29.4 (17.2, 45.4) 0 0 0 29.4 (17.2, 45.4) 0 0 55

All children 72.5 (60.0, 82.3) 15.4 (8.7, 25.7) 2.8 (1.2, 6.7) 90.3 (83.4, 94.5) 8.1 (4.4, 14.4) 0 0 0 8.1 (4.4, 14.4) 1.7 (0.4, 6.5) 0 258


P a g e | 32


3.4 Type of antimalarials taken and source of antimalarials

Table 3.4.1: Type of antimalarial medicines taken by children under five

Among children under five with fever in the two weeks preceding the survey, percentage who took specific antimalarial medicines after developing fever, by background characteristics.

Any non-

artemisinin therapy

SP Chloroquine Quinine Other non-artemisinin

therapy

Artemisinin monotherapy

Any ACT First-line ACT

(ASAQ) ACT with

AMFm logo

Number of children

with fever

Age (in years)

<1 8.9 (4.5, 16.5) 0.6 (0.1, 2.5) 5.7 (2.5, 12.7) 2.6 (0.9, 7.3) 0 0 4.6 (2.5, 8.2) 2.3 (1.0, 5.4) 0.8 (0.3, 2.4) 502

1 9.6 (6.1, 14.8) 0.3 (0.1, 1.5) 6.2 (3.8, 10.0) 3.2 (1.3, 7.4) 0 0 9.2 (3.9, 20.4) 5.0 (1.5, 14.8) 2.2 (0.8, 6.2) 510

2 12.9 (9.0, 18.2) 0.6 (0.1, 3.0) 8.6 (5.1, 14.0) 3.4 (1.7, 6.7) 0.3 (<0.1, 2.0) 0 7.6 (3.6, 15.1) 1.7 (0.6, 5.1) 3.5 (1.5, 8.1) 452

3 12.2 (8.3, 17.4) 1.7 (0.5, 5.1) 7.1 (4.1, 1.02) 3.1 (1.4, 6.9) 0.3 (<0.1, 2.0) 0.1 (<0.1, 0.5) 9.4 (3.6, 22.1) 6.3 (2.5, 15.0) 2.7 (1.0, 7.2) 450

4 13.4 (7.2, 23.6) 2.3 (0.4, 11.5) 6.8 (3.1, 14.5) 6.4 (2.7, 14.6) 0.1 (<0.1, 0.3) 0.1 (<0.1, 0.3) 11.6 (6.6, 19.4) 6.5 (3.5, 11.6) 5.8 (3.0, 10.8) 474

Strata

Urban 10.7 (8.0, 14.1) 0.4 (0.1, 1.0) 7.5 (5.3, 10.4) 2.9 (1.9, 4.4) 0.1 (<0.1, 0.7) 0.4 (0.2, 1.2) 6.2 (4.5, 8.6) 2.5 (1.7, 3.8) 4.0 (2.7, 5.8) 1,326

Rural 11.3 (7.8, 16.1) 1.1 (0.5, 2.7) 6.8 (4.4, 10.3) 3.8 (2.0, 7.1) 0.1 (<0.1, 0.8) 0 8.7 (4.8, 15.2) 4.5 (2.4, 8.4) 2.9 (1.5, 5.5) 1,062


No education 14.2 (7.9, 24.1) 2.4 (0.8, 7.0) 6.7 (2.9, 14.7) 6.2 (2.7, 13.4) 0.2 (<0.1, 1.1) 0 3.4 (1.4, 8.1) 1.2 (0.5, 2.7) 0.6 (0.2, 1.5) 509

Some primary 10.8 (7.7, 15.0) 0.7 (0.2, 2.0) 7.1 (4.9, 10.0) 3.2 (1.7, 6.1) <0.1 (<0.1, 0.1) <0.1 (<0.1, 0.1) 11.4 (6.1, 20.4) 7.2 (3.5, 14.2) 3.1 (1.4, 6.9) 974

Primary completed + 8.4 (4.5, 14.9) 0.1 (<0.1, 0.3) 6.5 (3.0, 13.8) 1.5 (0.8, 3.0) 0.3 (<0.1, 1.6) 0.2 (0.1, 0.4) 8.6 (5.2, 14.0) 2.1 (0.9, 4.8) 6.0 (3.4, 10.3) 894

Wealth index

Poorest 13.8 (8.6, 21.5) 0.7 (0.2, 3.2) 9.5 (5.6, 15.7) 3.6 (1.6, 7.6) 0.2 (<0.1, 1.1) 0 7.7 (2.9, 18.6) 4.7 (1.5, 13.8) 1.6 (0.6, 4.1) 483

Second 11.3 (6.6, 18.7) 2.0 (0.6, 6.9) 4.8 (2.6, 8.6) 5.8 (2.6, 12.3) 0 0 10.1 (5.1, 19.3) 5.6 (2.8, 10.9) 3.2 (1.2, 8.2) 481

Middle 9.9 (5.9, 16.1) 0.3 (<0.1, 1.7) 6.6 (3.2, 13.2) 2.8 (1.1, 6.9) 0.2 (<0.1, 1.7) 0 7.8 (4.8, 12.4) 3.8 (2.0, 7.3) 2.5 (1.1, 5.6) 475

Fourth 8.3 (4.4, 15.1) 1.4 (0.2, 8.9) 6.0 (2.8, 12.3) 0.9 (0.4, 2.0) 0 0.2 (0.1, 0.7) 6.2 (3.1, 11.9) 1.9 (0.6, 6.2) 4.5 (1.9, 10.3) 458

Richest 5.7 (3.6, 8.9) 0.4 (0.1, 1.6) 3.7 (2.0, 6.7) 1.7 (0.8, 3.4) 0.2 (<0.1, 1.4) 0.4 (0.1, 1.7) 11.3 (6.3, 19.6) 2.0 (0.9, 4.3) 9.4 (4.7, 18.0) 473

All children

11.3 (8.1, 15.6) 1.1 (0.5, 2.5) 6.8 (4.6, 10.0) 3.7 (2.1, 6.6) 0.1 (<0.1, 0.7) <0.1 (<0.1, 0.1) 8.4 (4.9, 14.3) 4.4 (2.4, 7.8) 3.0 (1.6, 5.3) 2,388


P a g e | 33


Table 3.4.2: Type of antimalarial medicines taken promptly by children under five

Among children under five with fever in the two weeks preceding the survey, percentage who took specific antimalarial medicines the same or next day after developing fever, by background characteristics.

Any non-

artemisinin therapy


therapy



(ASAQ) ACT with

AMFm logo

Number of children

with fever

Age (in years)

<1 7.3 (3.7, 14.0) 0.3 (<0.1, 2.5) 5.6 (2.4, 12.7) 1.3 (0.4, 3.8) 0 0 4.4 (2.4, 8.1) 2.3 (0.9, 5.4) 0.7 (0.2, 2.4) 502

1 8.2 (4.9, 13.3) <0.1 (<0.1, 0.2) 5.8 (3.4, 9.7) 2.3 (0.8, 6.4) 0 0 7.5 (3.0, 17.4) 4.6 (1.3, 15.0) 1.4 (0.5, 3.8) 510

2 9.4 (6.2, 14.0) 0.2 (<0.1, 1.4) 6.1 (3.3, 10.9) 2.9 (1.3, 6.2) 0.3 (<0.1, 2.0) 0 5.7 (2.6, 11.7) 1.4 (0.4, 5.1) 2.4 (0.9, 6.1) 452

3 9.6 (6.3, 14.4) 1.6 (0.5, 5.1) 5.7 (3.0, 10.5) 2.0 (0.8, 5.3) 0.3 (<0.1, 2.0) 0.1 (<0.1, 0.5) 7.8 (2.6, 21.0) 4.9 (1.6, 13.7) 1.5 (0.5, 4.5) 450

4 10.5 (5.9, 18.1) 1.7 (0.3, 7.7) 5.8 (2.7, 12.2) 4.6 (1.9, 10.8) <0.1 (<0.1, 0.4) 0.1 (<0.1, 0.3) 9.9 (5.8, 16.4) 5.2 (2.6, 9.9) 5.4 (2.9, 9.8) 474

Strata

Urban 8.6 (6.2, 11.8) 0.2 (0.1, 0.9) 6.7 (4.7, 9.4) 1.7 (0.9, 3.1) 0.1 (<0.1, 0.8) 0.4 (0.1, 1.2) 4.5 (3.0, 6.8) 1.8 (1.1, 3.0) 3.0 (1.9, 4.8) 1,326

Rural 9.0 (6.4, 12.5) 0.8 (0.3, 1.9) 5.7 (3.8, 8.5) 2.7 (1.3, 5.5) 0.1 (<0.1, 0.8) 0 7.3 (3.9, 13.2) 3.9 (1.8, 8.0) 2.2 (1.1, 4.2) 1,062


No education 10.4 (6.1, 17.4) 1.8 (0.6, 5.0) 5.3 (2.6, 10.3) 4.1 (1.4, 11.1) 0.2 (<0.1, 1.1) 0 2.8 (1.1, 7.0) 0.8 (0.3, 2.4) 0.4 (0.1, 1.3) 509

Some primary 8.6 (6.0, 12.0) 0.4 (0.1, 1.5) 5.9 (4.1, 8.6) 2.4 (1.2, 4.8) 0 <0.1 (<0.1, 0.1) 9.4 (4.8, 17.6) 6.2 (2.7, 13.7) 2.1 (0.9, 4.7) 974

Primary completed + 7.8 (4.0, 14.5) <0.1 (<0.1, 0.2) 6.3 (2.8, 13.7) 1.2 (0.5, 2.7) 0.3 (<0.1, 1.6) 0.1 (<0.1, 0.4) 7.6 (4.3, 13.1) 1.9 (0.7, 4.7) 5.1 (2.7, 9.4) 894

Wealth index

Poorest 10.0 (6.3, 15.4) 0.6 (0.1, 2.2) 7.3 (4.5, 11.8) 2.1 (0.7, 5.9) 0.2 (<0.1, 1.2) 0 7.0 (2.5, 18.4) 4.5 (1.4, 13.9) 1.2 (0.4, 3.3) 483

Second 9.5 (5.6, 15.7) 1.3 (0.3, 4.9) 4.1 (2.1, 7.9) 4.9 (2.2, 10.6) 0 0 8.3 (4.2, 15.9) 4.7 (2.1, 10.0) 2.0 (0.7, 5.7) 481

Middle 8.2 (4.6, 14.3) <0.1 (<0.1, 0.1) 6.5 (3.1, 13.1) 1.5 (0.4, 4.8) 0.2 (<0.1, 1.7) 0 6.0 (3.5, 10.1) 2.7 (1.3, 5.6) 2.4 (1.1, 5.5) 475

Fourth 6.6 (3.3, 12.6) 1.4 (0.2, 9.3) 4.7 (2.2, 9.9) 0.6 (0.2, 1.4) 0 0.2 (<0.1, 0.6) 5.5 (2.6, 11.4) 1.4 (0.3, 6.6) 4.0 (1.5, 10.0) 458

Richest 4.7 (2.8, 7.7) 0.4 (0.1, 1.6) 3.0 (1.5, 6.1) 1.1 (0.5, 2.4) 0.2 (<0.1, 1.4) 0.4 (0.1, 1.7) 7.7 (3.9, 14.6) 1.8 (0.8, 4.1) 6.4 (2.9, 13.4) 473

All children

8.9 (6.5, 12.1) 0.7 (0.3, 1.7) 5.8 (4.0, 8.3) 2.6 (1.3, 5.1) 0.1 (<0.1, 0.7) <0.1 (<0.1, 0.1) 7.0 (3.9, 12.4) 3.7 (1.8, 7.4) 2.2 (1.2, 4.0) 2,388


P a g e | 34


Table 3.4.3: Type of antimalarial medicines taken among children who received an antimalarial

Among children under five with fever in the two weeks preceding the survey who received an antimalarial, percentage who took specific antimalarial medicines, by background characteristics.

Any non-

artemisinin therapy


therapy



(ASAQ) ACT with

AMFm logo

Number of children who received an antimalarial

Age (in years)

<1 67.6 (45.0, 84.2) 4.5 (1.0, 17.6) 43.6 (23.0, 66.7) 19.5 (6.9, 44.0) 0 0 34.9 (17.4, 57.7) 17.8 (7.0, 38.5) 6.1 (1.8, 18.6) 63

1 52.6 (32.4, 72.0) 1.5 (0.3, 6.7) 33.9 (21.2, 49.5) 17.5 (6.8, 37.9) 0 0 50.7 (27.8, 73.4) 27.3 (10.0, 56.1) 12.2 (4.3, 30.0) 85

2 65.4 (43.5, 82.3) 3.3 (0.7, 14.1) 43.5 (25.5, 63.5) 17.3 (8.3, 32.5) 1.4 (0.2, 10.3) 0 38.3 (19.5, 61.4) 8.6 (2.8, 23.6) 17.8 (8.0, 35.1) 88

3 59.4 (38.0, 77.8) 8.2 (2.5, 23.5) 34.8 (17.7, 56.9) 15.2 (7.6, 28.1) 1.4 (0.2, 10.1) 0.5 (0.1, 2.4) 45.7 (22.6, 70.9) 30.8 (16.5, 50.1) 13.2 (5.9, 26.8) 87

4 55.9 (33.7, 75.9) 9.6 (1.7, 39.8) 28.5 (13.7, 50.0) 26.9 (12.3, 49.2) 0.3 (0.1, 1.4) 0.4 (0.1, 1.4) 48.3 (25.6, 71.6) 27.0 (13.4, 46.8) 24.1 (12.2, 42.1) 104

Strata

Urban 63.3 (53.0, 72.5) 2.1 (0.8, 5.3) 44.2 (34.3, 54.7) 17.4 (12.2, 24.3) 0.8 (0.2, 3.8) 2.6 (0.9, 6.9) 36.9 (27.5, 47.3) 15.0 (10.1, 21.7) 23.5 (15.9, 33.2) 239

Rural 59.0 (42.8, 73.5) 5.9 (2.4, 13.7) 35.3 (23.8, 48.7) 19.8 (10.7, 33.7) 0.6 (0.1, 4.4) 0 45.2 (28.2, 63.4) 23.7 (13.7, 37.7) 14.9 (8.0, 25.8) 188


No education 81.6 (59.2, 93.1) 14 (4.4, 36.5) 38.5 (19.8, 61.3) 35.7 (17.6, 59.1) 1.0 (0.1, 7.5) 0 19.5 (7.5, 41.9) 7.0 (2.7, 16.8) 3.2 (1.1, 9.3) 84

Some primary 51.9 (37.5, 66.1) 3.2 (1.2, 8.6) 33.9 (22.7, 47.3) 15.3 (8.5, 26.0) <0.1 (<0.1, 0.3) 0.1 (<0.1, 0.5) 54.7 (36.6, 71.7) 34.7 (20.7, 51.9) 14.8 (6.6, 30.1) 198

Primary completed +

49.1 (30.8, 67.7) 0.5 (0.1, 1.6) 38.4 (20.3, 60.4) 9.0 (4.0, 18.9) 1.7 (0.3, 9.0) 0.9 (0.3, 2.7) 50.6 (32.1, 69.0) 12.3 (5.0, 26.9) 35.2 (21.4, 52.1) 144

Wealth index

Poorest 67.2 (41.8, 85.4) 3.6 (1.0, 12.8) 46.1 (26.6, 66.9) 17.4 (8.9, 31.4) 0.9 (0.1, 5.9) 0 37.3 (16.7, 63.8) 22.9 (8.3, 49.2) 7.9 (3.2, 18.3) 98

Second 55.2 (33.5, 75.2) 9.8 (2.7, 30.0) 23.2 (12.9, 38.3) 28.2 (13.5, 49.9) 0 0 49.6 (26.7, 72.7) 27.6 (13.9, 47.5) 15.6 (6.4, 33.4) 103

Middle 58.2 (43.9, 71.2) 1.6 (0.3, 9.3) 38.9 (23.1, 57.5) 16.5 (6.2, 37.4) 1.3 (0.2, 9.6) 0 45.8 (31.6, 60.8) 22.4 (12.4, 37.0) 14.5 (6.0, 31.2) 70

Fourth 56.8 (33.3, 77.5) 9.8 (1.4, 44.7) 40.9 (21.2, 64.0) 6.4 (2.8, 13.6) 0 1.5 (0.4, 4.7) 42.1 (21.5, 65.9) 13.2 (3.9, 36.3) 30.8 (13.1, 56.8) 66

Richest 33.4 (18.6, 52.4) 2.1 (0.5, 9.4) 21.5 (10.3, 39.4) 9.9 (4.8, 19.6) 1.1 (0.1, 8.1) 2.4 (0.6, 9.5) 66.2 (46.4, 81.5) 11.8 (5.2, 24.4) 54.9 (34.1, 74.1) 88

All children

59.4 (44.4, 72.8) 5.6 (2.4, 12.7) 36.0 (25.3, 48.3) 19.6 (11.1, 32.3) 0.6 (0.1, 3.7) 0.2 (0.1, 0.6) 44.5 (28.8, 61.4) 22.9 (13.7, 35.9) 15.6 (9.1, 25.4) 427


P a g e | 35


Table 3.4.4: Source of antimalarials

Among children under five with fever in the two weeks preceding the survey who received an antimalarial treatment, the source of antimalarial treatment, by background characteristics.

Public / not for profit sector Private sector At home N. of

children who

received an antimalarial








home1

Missing or

don’t know

Age (in years)

<1 27.7 (13.1, 49.2) 13.0 (4.1, 34.2) 0.3 (<0.1, 2.2) 41.0 (21.1, 64.3) 4.0 (1.3, 11.5) 0.8 (<0.1.2, 2.6) 19.6 (7.2, 43.5) 0 24.4 (10.7, 46.6) 35.9 (14.9, 64.3) 0 63

1 37.3 (24.7, 51.9) 11.4 (4.1, 28.1) 0.2 (<0.1, 1.8) 49.0 (31.8, 66.4) 8.6 (3.5, 19.6) 0.7 (0.2, 2.2) 33.6 (12.8, 63.5) 0 42.8 (21.2, 67.5) 11.7 (4.7, 26.1) 0 85

2 36.7 (24.0, 51.6) 11.1 (4.1, 26.8) 5.1 (1.7, 14.4) 52.8 (40.1, 65.2) 10.3 (3.9, 24.5) 13.6 (6.2, 27.5) 10.9 (3.8, 27.4) 0 34.8 (21.7, 50.7) 18.4 (8.8, 34.4) 0 88

3 37.8 (15.7, 66.5) 7.3 (2.6, 18.9) 0 45.1 (22.5, 70.0) 16.5 (5.6, 39.8) 1.8 (0.8, 4.1) 24.0 (11, 44.6) 0.2 (<0.1, 1.2) 42.5 (22.3, 65.5) 12.5 (4.1, 32.2) 0 87

4 19.9 (10.7, 34) 20.4 (10.2, 36.6) 2.0 (0.4, 10.1) 42.2 (25.6, 60.9) 14.4 (7.2, 26.8) 22.2 (9.3, 44.4) 6.7 (2.6, 16.2) 1.1 (0.2, 7.9) 43.4 (29.2, 58.8) 16.1 (9.3, 26.4) 0 104

Strata

Urban 14.7 (10.1, 21.0) 1.6 (0.4, 5.9) 3.9 (1.8, 8.5) 19.9 (14.4, 26.9) 28.4 (19.8, 38.8) 16.1 (10.7, 23.5) 10.1 (4.7, 20.5) 0.4 (<0.1, 2.8) 54.7 (45.3, 63.9) 26.4 (17.1, 38.5) 0 239

Rural 33.1 (24.4, 43.2) 14.0 (8, 23.4) 1.3 (0.3, 5.7) 48.5 (37.7, 59.5) 9.7 (4.9, 18.5) 8.1 (3.7, 16.9) 19.5 (11.2, 31.6) 0.3 (<0.1, 2.0) 37.3 (28.0, 47.6) 16.9 (9.7, 27.9) 0 188


No education 27.5 (12.0, 51.2) 11.9 (4.8, 26.3) 0.1 (<0.1, 0.9) 39.3 (21.7, 60.2) 8.5 (3.7, 18.1) 12.9 (5.5, 27.4) 21.0 (10.6, 37.5) 0 42.4 (28.4, 57.9) 18.3 (9.1, 33.5) 0 84

Some primary 35.3 (26.1, 45.7) 13.8 (5.6, 30.1) 2.7 (0.8, 8.9) 51.8 (41.4, 62.0) 12.8 (5.4, 27.3) 6.8 (2.2, 18.7) 21.9 (10.8, 39.4) 0.5 (0.1, 3.6) 41.4 (28.6, 55.6) 11.1 (5.4, 21.2) 0 198

Primary completed +

27.1 (14.9, 44.1) 12.5 (5.2, 27.1) 0.4 (0.1, 1.6) 40 (23.1, 59.7) 10.8 (4.8, 22.7) 8.3 (3.7, 17.4) 6.2 (2.2, 16.5) 0.2 (<0.1, 1.3) 25.4 (16.4, 37.3) 35.7 (19.1, 56.8) 0 144

Wealth index

Poorest 33.5 (20.1, 50.3) 12.1 (6.4, 21.7) 1.3 (0.2, 8.1) 46.9 (33.3, 61.0) 5.9 (1.8, 18.1) 8.6 (4.3, 16.5) 25.6 (14.8, 40.5) 0.8 (0.1, 5.1) 40.1 (28.6, 52.8) 14.3 (6.5, 28.5) 0 98

Second 38.4 (27.1, 51.1) 17.4 (8.5, 32.3) 2.5 (0.4, 14.4) 58.3 (42.2, 72.8) 11.5 (5.8, 21.6) 4.8 (0.9, 22.0) 13.3 (5.6, 28.5) 0.1 (<0.1, 0.7) 29.7 (19.2, 42.8) 13.3 (6.9, 24.3) 0 103

Middle 26.6 (12.6, 47.7) 11.3 (3.2, 32.8) 0.5 (0.1, 2.2) 38.3 (20.0, 60.6) 10.9 (3.8, 27.5) 10.4 (3.6, 26.4) 17.9 (6.3, 41.4) 0 39.1 (26.1, 53.9) 26.2 (10.7, 51.4) 0 70

Fourth 13.3 (3.2, 41.9) 12.0 (2.9, 38.3) 0.6 (0.1, 4.4) 25.9 (10.2, 51.7) 25.3 (9.4, 52.5) 16.9 (4.8, 44.8) 14.4 (3.6, 43.2) 0 56.6 (34.0, 76.7) 28.0 (11.8, 53.1) 0 66

Richest 24.4 (8.1, 54.0) 0 3.1 (1.0, 9.4) 27.5 (10.6, 54.9) 30.7 (16.9, 49.0) 21.3 (5.6, 55.3) 1.0 (0.1, 7.6) 0 52.9 (31.1, 73.7) 24.1 (12.3, 41.9) 0 88

All children

31.6 (23.5, 41.0) 13 (7.5, 21.6) 1.6 (0.5, 4.9) 46.1 (36.2, 56.4) 11.3 (6.5, 18.9) 8.8 (4.5, 16.5) 18.7 (11.0, 29.9) 0.3 (0.1, 1.7) 38.7 (30.1, 48.1) 17.7 (10.8, 27.6) 0 427

1 The most common original sources for antimalarials obtained from home are: pharmacy or drug shop (48%, n=44), general retailer (39%, n=36), and public health facility (8%, n=7) (unweighted).


P a g e | 36


Table 3.4.5: Source of ACTs

Among children under five with fever in the two weeks preceding the survey who received an ACT, the source of treatment, by background characteristics.


children who









home1

Missing or

don’t know

Age (in years)

<1 * * * * * * * * * * * 20

1 27.3 (15.8, 42.8) 15.5 (3.8, 45.6) 0.5 (0.1, 3.9) 43.2 (19.5, 70.5) 8.7 (1.7, 34.4) 0.4 (0.1, 2.1) 41.5 (14.2, 75.2) 0 50.5 (22.7, 78.0) 6.3 (1.4, 23.8) 0 36

2 48.7 (28.4, 69.4) 13.4 (2.8, 45.2) 7.1 (1.8, 23.9) 69.1 (49.3, 83.7) 3.5 (1.2, 10.0) 19.9 (6.4, 47.5) 0 0 23.4 (8.5, 50.0) 16.9 (5.0, 44.2) 0 34

3 61.6 (23.1, 89.6) 12.0 (2.9, 38.4) 0 73.6 (38.1, 92.6) 6.9 (2.1, 20.4) 1.6 (0.4, 6.5) 13.6 (2.9, 45.2) 0 22.2 (5.8, 57.1) 4.5 (0.6, 26.1) 0 33

4 27.0 (13.3, 47.2) 29.2 (12.7, 53.8) 3.5 (0.6, 18.7) 59.8 (30.9, 83.1) 5.1 (1.8, 13.4) 23.8 (5.5, 62.3) 0 0 28.9 (9.2, 61.9) 11.3 (3.5, 31.3) 0 44

Strata

Urban 11.4 (5.3, 22.9) 4.4 (1.0, 17.3) 6.3 (2.5, 15.1) 22.2 (12.9, 35.3) 42.4 (28.6, 57.4) 13.4 (6.5, 25.7) 2.8 (0.6, 11.6) 0 58.6 (46.2, 69.9) 20.1 (11.1, 33.7) 0 86

Rural 42.5 (31.8, 54.0) 20.8 (10.1, 38.0) 2.0 (0.3, 11.8) 65.4 (50.9, 77.5) 3.0 (1.0, 8.8) 10.0 (2.6, 31.3) 14.0 (5.2, 32.7) 0 27 (14.1, 45.6) 9.3 (4.1, 19.5) 0 81


No education * * * * * * * * * * * 19

Some primary 39.6 (27.2, 53.4) 19.7 (7.2, 44.0) 3.5 (0.8, 14.8) 62.8 (47.4, 76.0) 4.8 (1.6, 13.3) 10.4 (2.8, 31.8) 16.4 (5.3, 40.6) 0 31.5 (16.5, 51.6) 8.1 (3.0, 20.1) 0 85

Primary completed +

41.4 (25.9, 58.8) 15.7 (5.2, 38.8) 0 57.1 (36.0, 75.8) 8.3 (3.9, 16.7) 10.1 (2.2, 36.4) 6.2 (1.2, 26.7) 0 24.5 (11.2, 45.5) 18.4 (6.6, 42.1) 0 62

Wealth index

Poorest * * * * * * * * * * * 24

Second 40.5 (26.3, 56.5) 29.8 (13.4, 53.9) 5.1 (1.0, 22.0) 75.4 (54.7, 88.6) 3.9 (1.6, 9.1) 0.1 (<0.1, 0.9) 10.1 (1.3, 48.5) 0 14.1 (3.7, 41.5) 10.5 (3.9, 25.5) 0 42

Middle 37.2 (13.6, 69.0) 17.8 (5.0, 47.2) 0.4 (0.1, 3.4) 55.5 (25, 82.3) 0.2 (<0.1, 1.9) 19.3 (4.6, 54.2) 12.8 (3.1, 39.9) 0 32.3 (11.4, 63.8) 12.2 (3.0, 38.3) 0 27

Fourth * * * * * * * * * * * 24

Richest 24.2 (4.6, 67.7) 0 3.6 (0.9, 13.6) 27.7 (6.7, 67.2) 26.8 (11.0, 52.0) 30 (7.1, 70.7) 1.5 (0.2, 11.7) 0 58.3 (26.9, 84.1) 14 (5.1, 33.0) 0 50

All children

40.4 (30.1, 51.6) 19.7 (9.9, 35.5) 2.3 (0.5, 9.9) 62.4 (49.3, 74.0) 5.7 (2.9, 11.0) 10.2 (3.0, 29.3) 13.3 (4.9, 31.2) 0 29.2 (16.8, 45.7) 10 (4.8, 19.6) 0 167

1 The most common original sources for ACTs obtained from home are: pharmacy or drug shop (58%, n=15), public health facility (19%, n=5) and general retailer (12%, n=3), and (unweighted).

*Where n<25 data are not shown.


P a g e | 37


Table 3.4.6: Source of ACTs with the AMFm logo, among children under five with fever who received an AMFm logo ACT

Among children under five with fever in the two weeks preceding the survey who received an ACT with the AMFm logo, the source of treatment, by background characteristics.


children who









home1

Missing or

don’t know

Age (in years)

<1 * * * * * * * * * * * 7

1 * * * * * * * * * * * 15

2 * * * * * * * * * * * 16

3 * * * * * * * * * * * 17

4 * * * * * * * * * * * 24

Strata

Urban 9.9 (3.7, 23.8) 0.9 (0.1, 7.2) 10.0 (4.0, 22.8) 20.8 (11.5, 34.5) 41.2 (25.5, 58.9) 8.8 (2.8, 24.3) 4.4 (1.0, 17.3) 0 54.3 (39.1, 68.8) 24.9 (12.3, 44.2) 0 50

Rural 54.6 (35.0, 72.8) 20.4 (8.4, 41.7) 0 75.0 (51.9, 89.3) 2.5 (0.5, 10.8) 10.3 (1.2, 51.0) 3.7 (0.4, 25.1) 0 16.5 (4.3, 46.2) 13.7 (5.9, 28.6) 0 29


No education * * * * * * * * * * * 4

Some primary 47.9 (23.9, 72.8) 19.0 (4.1, 56.1) 2.4 (0.7, 8.2) 69.2 (46.7, 85.2) 6.1 (2.2, 15.4) 9.2 (1.1, 47.4) 0 0 15.2 (4.4, 41.1) 24.3 (9.3, 50.0) 0 30

Primary completed +

48.6 (27.4, 70.4) 14.9 (4.2, 41.2) 0 63.5 (34.8, 85.0) 9.7 (4.2, 20.9) 12.5 (2.2, 48.0) 8.9 (1.4, 39.4) 0 31.0 (11.8, 60.2) 5.4 (1.8, 15.6) 0 45

Wealth index

Poorest * * * * * * * * * * * 7

Second * * * * * * * * * * * 12

Middle * * * * * * * * * * * 12

Fourth * * * * * * * * * * * 15

Richest 27.8 (4.9, 74.1) 0 4.3 (1.0, 17) 32.1 (7.2, 74.1) 20.9 (7.2, 47.4) 31.9 (6.3, 76.6) 1.8 (0.2, 14.2) 0 54.6 (20.2, 85.1) 13.3 (3.8, 37.2) 0 33

All children

49.1 (32.0, 66.3) 18.0 (7.4, 37.5) 1.2 (0.4, 3.7) 68.3 (48.6, 83.1) 7.3 (4.1, 12.5) 10.1 (1.5, 44.6) 3.8 (0.6, 20.4) 0 21.1 (8.2, 44.7) 15.1 (7.5, 27.8) 0 79

1 The most common original sources for ACTs with the AMFm logo obtained from home are: pharmacy (59%, n=10), general retailer (18%, n=3) and public health facility (12%, n=2) (unweighted).

*Where n<25 data are not shown.


P a g e | 38


3.5 Sources of advice and treatment for fever

Table 3.5.1: Care seeking behaviour: first place to seek care

Among children under five with fever in the two weeks preceding the survey for whom advice or treatment was sought1, percentage for whom advice or treatment was first sought at a given outlet type, by

background characteristics.

Public / not for profit sector Private sector At home

Number of children








home1

Missing / don’t know

Age (in years)

<1 25.6 (18.6, 34.3) 4.0 (1.8, 8.5) 0.4 (0.2, 0.8) 30.0 (22.8, 38.4) 3.8 (2.3, 6.2) 6.7 (3.8, 11.4) 14.9 (9.6, 22.4) 6.4 (3.0, 13.0) 31.7 (24.6, 39.9) 38.3 (31.4, 45.7) 0 467

1 15.9 (10.6, 23.1) 5.4 (3.0, 9.5) 0.2 (0.1, 0.5) 21.5 (15.5, 28.9) 2.9 (1.5, 5.5) 5.5 (3.3, 9.1) 20.2 (11.6, 32.8) 2.2 (1.1, 4.4) 30.9 (21.9, 41.6) 47.7 (37.1, 58.4) 0 474

2 17.3 (12.2, 23.8) 2.8 (1.3, 6.1) 0.3 (0.2, 0.6) 20.4 (14.6, 27.9) 3.3 (1.6, 6.9) 7.8 (4.5, 13.1) 22.1 (14.4, 32.4) 1.0 (0.3, 3.8) 34.2 (26.4, 43.0) 45.4 (35.8, 55.3) 0 418

3 13.9 (7.6, 24.1) 3.6 (1.9, 6.7) 0.1 (<0.1, 0.5) 17.6 (10.5, 28.0) 5.7 (2.6, 11.9) 6.5 (3.1, 13.2) 26.7 (17.9, 37.9) 3.3 (1.3, 8.2) 42.3 (32.9, 52.3) 40.1 (30.0, 51.1) 0 423

4 8.2 (4.9, 13.4) 7.6 (2.9, 18.5) 1.0 (0.2, 4.9) 16.8 (10.5, 25.8) 4.3 (2.1, 8.5) 15.2 (9.2, 24) 15.9 (10.3, 23.6) 0.8 (0.3, 2.4) 36.2 (28.6, 44.5) 47.0 (37.7, 56.5) 0 433

Strata

Urban 9.2 (7.4, 11.2) 0.5 (0.2, 1.3) 2.7 (1.6, 4.3) 12.3 (10.3, 14.6) 8.6 (6.8, 10.7) 9.8 (6.4, 14.9) 9.1 (5.6, 14.5) 0.4 (0.2, 0.9) 27.9 (21.5, 35.3) 59.8 (52.6, 66.7) 0 1,272

Rural 17.2 (13.3, 21.9) 5.2 (3.2, 8.3) 0.2 (<0.1, 1.3) 22.5 (18.1, 27.6) 3.5 (1.9, 6.1) 8.0 (5.4, 11.6) 21.0 (15, 28.6) 3.1 (1.8, 5.1) 35.5 (29.7, 41.8) 42.0 (35.1, 49.2) 0 943

Caregiver’s educ.

No education 15.5 (10.4, 22.3) 3.7 (1.5, 8.9) 0.7 (0.1, 3.7) 19.8 (13.6, 27.9) 3.1 (1.7, 5.7) 6.2 (3.4, 10.8) 22.8 (15.5, 32.3) 4.9 (2.5, 9.5) 37.0 (29.3, 45.4) 43.2 (34.0, 52.9) 0 444

Some primary 18.1 (12.8, 24.8) 5.2 (2.6, 10.1) 0.1 (<0.1, 0.2) 23.3 (17.8, 30.0) 4.5 (2.4, 8.2) 8.1 (5.5, 11.8) 22.7 (16.1, 31.0) 2.6 (1.5, 4.5) 37.9 (30.8, 45.5) 38.8 (31.7, 46.5) 0 899

Primary completed +

13.7 (10.2, 18.2) 5.0 (2.4, 9.9) 0.8 (0.4, 1.5) 19.5 (14.0, 26.6) 3.9 (2.3, 6.4) 10.8 (7.0, 16.5) 9.7 (5.5, 16.6) 0.7 (0.2, 2.8) 25.1 (19.9, 31.1) 55.4 (47.2, 63.3) 0 862

Wealth index

Poorest 19.7 (13.8, 27.3) 4.5 (2.7, 7.4) 0 24.2 (17.7, 32.2) 3.3 (1.8, 6.0) 3.9 (2.2, 6.7) 28.6 (20.5, 38.2) 3.7 (2.1, 6.2) 39.4 (31.5, 47.9) 36.4 (26.8, 47.2) 0 417

Second 15.9 (11.6, 21.3) 6.8 (3.4, 13.0) <0.1(<0.1, 0.2) 22.7 (16.9, 29.7) 4.1 (2.0, 8.2) 7.9 (4.8, 12.9) 22.5 (14.1, 33.8) 2.9 (1.1, 7.4) 37.4 (29.4, 46.2) 39.9 (31.5, 49.0) 0 434

Middle 16.7 (10.3, 26.0) 4.6 (2.2, 9.4) 0.9 (0.2, 3.7) 22.2 (15.9, 30.2) 2.4 (0.8, 7.5) 12.2 (8.5, 17.2) 14.5 (8.9, 22.7) 2.6 (1.0, 6.7) 31.7 (24.7, 39.7) 46.1 (39.0, 53.3) 0 449

Fourth 10.5 (5.4, 19.4) 1.2 (0.2, 7.5) 0.8 (0.3, 2.0) 12.6 (7.0, 21.6) 6.9 (3.4, 13.7) 12.8 (7.3, 21.5) 9.5 (4.7, 18.2) 0.1 (<0.1, 0.9) 29.3 (20.6, 39.9) 58.1 (44.6, 70.5) 0 438

Richest 9.2 (5.0, 16.4) 1.4 (0.2, 9.1) 2.3 (1.2, 4.4) 13.0 (7.8, 20.8) 6.8 (4.6, 9.8) 9.3 (4.5, 18.1) 1.1 (0.4, 2.8) 0 17.2 (11.5, 24.9) 69.9 (61.2, 77.3) 0 462

All children

16.4 (12.8, 20.7) 4.7 (2.9, 7.5) 0.4 (0.2, 0.9) 21.5 (17.5, 26.1) 4.0 (2.5, 6.2) 8.2 (5.8, 11.4) 19.8 (14.4, 26.7) 2.8 (1.7, 4.7) 34.8 (29.5, 40.4) 43.7 (37.4, 50.2) 0 2,215

1 Excludes caregivers of children under five with fever who reported they did not do anything to treat the fever.

Subtotals by background characteristics may not sum to the value given here due to missing values for some background characteristics.


P a g e | 39


Table 3.5.2: Care seeking behaviour: any source to seek care

Among children under five with fever in the two weeks preceding the survey for whom advice or treatment was sought1, percentage for whom advice or treatment was sought at a given outlet type, by background

characteristics.

Public / not for profit sector Private sector At home N. of children

who received an antimalarial








home1

Missing / don’t know

Age (in years)

<1 35.5 (27.4, 44.5) 5.0 (2.5, 9.5) 1.3 (0.6, 2.5) 41.7 (33.8, 50.0) 6.7 (4.2, 10.4) 13.4 (7.0, 24.0) 18.1 (11.7, 26.8) 8.0 (4.1, 14.9) 42.6 (33.8, 51.9) 40.1 (33.1, 47.5) 0 467

1 27.5 (21.1, 34.8) 6.0 (3.2, 10.7) 0.6 (0.3, 1.0) 33.5 (26.8, 40.9) 5.9 (3.8, 9.0) 10.3 (6.9, 15.1) 22.6 (13.4, 35.6) 4.8 (2.6, 8.6) 41.4 (31.5, 52.0) 49.5 (38.6, 60.5) 0.6 (0.1, 4.2) 474

2 22.1 (16.9, 28.4) 4.3 (2.3, 7.9) 1.5 (0.6, 3.6) 27.8 (21.6, 35.1) 8.4 (5.1, 13.6) 10.7 (6.9, 16.3) 26.9 (19.1, 36.5) 1.0 (0.3, 3.7) 44.8 (35.5, 54.5) 48.0 (38.0, 58.1) 0 418

3 19.2 (11.2, 30.9) 4.5 (2.3, 8.6) 1.0 (0.4, 2.3) 24.8 (15.7, 36.8) 9.9 (4.9, 19.1) 7.9 (4.1, 14.6) 29.6 (20.6, 40.5) 5.3 (2.7, 10.2) 48.9 (39.6, 58.2) 42.1 (31.6, 53.4) 0 423

4 15.2 (10.3, 22.0) 8.9 (3.9, 19.0) 1.7 (0.6, 5.0) 25.8 (19.1, 34.0) 7.1 (4.1, 11.9) 17.6 (10.7, 27.6) 17.2 (11.7, 24.7) 1.6 (0.8, 3.4) 42.2 (35.0, 49.8) 48.1 (38.9, 57.5) <0.1 (<0.1, 0.3) 433

Strata

Urban 16.3 (13.7, 19.3) 0.6 (0.2, 1.7) 6.0 (4.1, 8.6) 22.6 (19.9, 25.6) 20.1 (16.5, 24.2) 15.9 (11.7, 21.2) 11.0 (6.8, 17.3) 1.7 (1.0, 2.9) 44.5 (38.5, 50.6) 62.0 (54.5, 69.0) 0.1 (<0.1, 0.6) 1,272

Rural 25.3 (20.8, 30.3) 6.3 (4.1, 9.5) 0.7 (0.3, 1.5) 32.1 (27.3, 37.3) 6.1 (3.5, 10.4) 11.5 (7.9, 16.5) 24.0 (17.5, 32) 4.6 (2.9, 7.1) 43.7 (37.5, 50.2) 43.8 (36.7, 51.2) 0.1 (<0.1, 1.1) 943

Caregiver’s educ.

No education 19.7 (14.3, 26.5) 5.1 (2.6, 10.0) 0.9 (0.2, 3.5) 25.7 (19.4, 33.1) 5.1 (3.2, 8.0) 8.5 (4.9, 14.5) 24.7 (16.4, 35.6) 7.6 (4.3, 13.1) 44.5 (35.9, 53.5) 45.4 (36.1, 55.0) <0.1 (<0.1, 0.2) 444

Some primary 27.1 (21.2, 33.8) 6.3 (3.4, 11.2) 1.0 (0.4, 2.2) 34.0 (28.3, 40.3) 7.2 (4.0, 12.6) 12.6 (8.3, 18.7) 25.8 (18.8, 34.4) 3.4 (2.0, 5.9) 46.1 (38.2, 54.2) 40.2 (32.9, 47.9) 0 899

Primary completed +

24.1 (18.2, 31.1) 5.3 (2.7, 10.2) 2.0 (1.2, 3.3) 31.3 (23.6, 40.3) 11.1 (7.7, 15.8) 14.7 (9.9, 21.2) 13.3 (8.6, 20.1) 2.0 (0.8, 4.9) 38.0 (30.9, 45.6) 58.1 (50.0, 65.9) 0.6 (0.1, 4.5) 862

Wealth index

Poorest 24.5 (18.3, 32.1) 6.5 (4.2, 9.8) 0.5 (0.1, 2.0) 31.1 (24.4, 38.8) 4.6 (2.5, 8.3) 5.1 (3.0, 8.6) 31.1 (22.1, 41.8) 5.1 (3.0, 8.4) 44.1 (35.4, 53.2) 37.2 (27.1, 48.5) <0.1 (<0.1, 0.2) 417

Second 24.8 (19.5, 30.9) 7.4 (3.9, 13.5) 0.7 (0.1, 3.7) 32.8 (25.5, 41.0) 6.7 (3.9, 11.2) 12.5 (7.7, 19.8) 25.9 (16.9, 37.7) 5.8 (3.0, 10.8) 48.5 (40.3, 56.8) 42.4 (33.9, 51.3) 0 434

Middle 25.7 (16.6, 37.6) 5.5 (2.7, 10.7) 1.2 (0.4, 3.5) 32.3 (23.7, 42.4) 7.4 (2.3, 20.9) 15.9 (11.4, 21.6) 16.2 (10.2, 24.7) 2.7 (1.1, 6.8) 39.8 (30.5, 49.9) 48.3 (40.5, 56.3) 0 449

Fourth 23.9 (15.0, 35.9) 2.0 (0.5, 7.4) 2.9 (1.5, 5.7) 28.7 (19.3, 40.5) 11.5 (7.2, 17.9) 21.2 (14.4, 30.2) 11.8 (6.6, 20.3) 0.6 (0.3, 1.3) 42.0 (35.5, 48.8) 61.5 (48.6, 73.0) 1.4 (0.2, 9.9) 438

Richest 17.8 (12.0, 25.6) 1.4 (0.2, 9.1) 4.7 (2.9, 7.6) 23.7 (17.0, 32.0) 21.0 (15.6, 27.6) 15.4 (9.2, 24.7) 4.9 (1.2, 17.8) 1.4 (0.4, 4.2) 38.3 (29.7, 47.8) 71.0 (62.3, 78.5) 0 462

All children

24.4 (20.3, 28.9) 5.7 (3.8, 8.6) 1.2 (0.7, 1.9) 31.2 (26.8, 35.9) 7.5 (5.0, 11.1) 11.9 (8.6, 16.3) 22.7 (16.8, 30.0) 4.3 (2.8, 6.6) 43.8 (38.1, 49.7) 45.6 (39.1, 52.3) 0.1 (<0.1, 1.0) 2,215

1 Excludes caregivers of children under five with fever who reported they did not do anything to treat the fever.

Subtotals by background characteristics may not sum to the value given here due to missing values for some background characteristics.


P a g e | 40

www.ACTwatch.info Household Survey Report, Madagascar, 2012 Survey

Table 3.5.3: Treatment at home (Supplementary Table)

Among children under five with fever in the two weeks preceding the survey who only received treatment at home, caregiver reasons for not seeking treatment outside the home.

Reasons treatment was not sought outside the home

Percent1

N=694

Illness not serious 36.5 (29.4, 44.2)

Illness went away / child got better 39.2 (32.6, 46.3)

No money for treatment 14.2 (10.1, 19.5)

No transportation 1.6 (0.5, 5.5)

Place for treatment was too far away 9.4 (4.7, 18.2)

No one in the household had time to obtain treatment

1.2 (0.4, 3.7)

Did not know where to get treatment 0.5 (0.1, 3.1)

Medicines / drugs not available at outlets 8.1 (3.8, 16.4)

Still ill, waiting for the fever to worsen 2.7 (1.3, 5.7)

Other 4.6 (1.9, 10.6)

Don’t know 0.2 (<0.1, 1.4)

1 Caregivers could state multiple reasons and total may sum to more than 100%.


P a g e | 41


Table 3.5.4: Initial treatment source (Supplementary Table)

Among children under five with fever in the two weeks preceding the survey for whom the first treatment source was outside the home, caregiver reasons for seeking treatment at this source.

Main reason for seeking treatment from initial source

Percent

N=1,058

Close by or easy to reach 37.6 (32.3, 43.2)

Reputation for quality treatment 13.8 (10.5, 18)

Availability of inexpensive treatment 8.3 (5.4, 12.6)

Availability of modern medicine 21.4 (16.4, 27.3)

Availability of traditional medicine 0.9 (0.3, 2.8)

Source provides credit 0.8 (0.3, 2.8)

Source has a short waiting time 0.2 (0.1, 0.6)

Fever wasn’t serious 2.4 (1.2, 4.8)

Source open at night 2.0 (1.0, 3.6)

Illness was serious/had persisted 0.6 (0.2, 2)

“Habit” 8.5 (5.3, 13.1)

Other 0.6 (0.2, 1.9)

Don’t know 2.9 (1.5, 5.6)


P a g e | 42


3.6 Breakdown of antimalarials acquired

Note that unlike other tables in the report, this section shows information at the antimalarial level, rather than the child or caregiver level.

Table 3.6.1: Types of antimalarials acquired

Percentage distribution of antimalarials acquired for children under five with fever in the past two weeks.

Urban Rural Total

Type of antimalarial % % %

N=256 N=206 N=462

Any non-artemisinin therapy 61.7 57.9 58.4

SP 2.0 5.2 4.9

Chloroquine 42.2 34.0 34.7

Quinine 16.7 18.2 18.1

Other non-artemisinin therapy 1.0 0.5 0.7

Artemisinin monotherapy 2.5 - 0.2

Any ACT 35.9 42.1 41.5

First-line ACT (ASAQ)1

14.1 21.7 21.1

Any ACT with AMFm logo 22.7 14.3 15.0 1 Of the 171 ACT cases, caregivers responded “ACT” for 34 cases and “ACTm” for 28 cases. These appear in the ‘Any ACT’ row but

as the precise generic ingredients in the ACT are not know they are not included in the First-line ACT row.


P a g e | 43


3.7 Caregiver Knowledge and Beliefs

Table 3.7.1: Caregiver knowledge of malaria and antimalarials

Percentage of caregivers of children under five with fever in the two weeks preceding the survey who state that fever is the main symptom of malaria in children under five, and who spontaneously name given antimalarial types or brands when asked what medicines can be used to treat malaria.

Malaria symptoms Knowledge of antimalarials

State that fever is the main symptom

of malaria in children under five

Number of

caregivers

Name ACT or ACT brand

Name CQ or a CQ brand

Name SP or an SP brand

Number of

caregivers

Strata

Urban 76.4 (70.0, 81.7) 1,209 25.5 (20.8, 30.8) 31.6 (27.4, 36.2) 14.9 (12.2, 18.1) 1,203

Rural 58.7 (50.3, 66.6) 955 13.9 (9.0, 20.8) 16.1 (12.3, 20.7) 4.1 (2.5, 6.5) 954


No education 39.7 (30.7, 49.4) 456 3.8 (1.8, 8.0) 8.2 (4.8, 13.6) 2.1 (0.8, 5.2) 454

Some primary 64.8 (56.4, 72.5) 880 15.6 (9.9, 23.7) 16.7 (12.4, 22.1) 4.1 (2.3, 7.3) 877

Primary completed + 77.6 (70.1, 83.7) 824 28.4 (20.1, 38.6) 31.9 (27, 37.2) 11.4 (8.4, 15.2) 822

All caregivers 60.4 (52.7, 67.6) 2,164 15.0 (10.4, 21.1) 17.5 (14, 21.7) 5.1 (3.6, 7.2) 2,157


P a g e | 44


Table 3.7.2: Caregiver beliefs about the most effective antimalarial treatment

Percentage of caregivers of children under five with fever in the two weeks preceding the survey who state a given antimalarial type as most effective for treating malaria in children under five and in adults.

Percentage of caregivers who cite antimalarial type as being most effective for:

Children under five

Number of caregivers = 2,136 Adults

Number of caregivers = 2,131

Type of antimalarial

Any non-artemisinin therapy 10.1 (7.3, 13.9) 10.8 (7.6, 15.1)

SP 1.1 (0.6, 1.8) 3.6 (2.3, 5.6)

Chloroquine 8.3 (5.6, 12.1) 5.2 (3.2, 8.3)

Quinine 0.7 (0.3, 1.8) 2.0 (1.2, 3.3)

Other non-artemisinin therapy 0.1 (<0.1, 0.3) 0.1 (<0.1, 0.4)

Artemisinin monotherapy <0.1 (<0.1, 0.1) 0

Any ACT 11.7 (8.2, 16.3) 5.3 (3.4, 8.0)

First-line ACT (ASAQ) 2.6 (1.5, 4.3) 0.7 (0.3, 1.7)

Non-antimalarial 14.4 (10.5, 19.3) 14.7 (10.4, 20.3)

Don’t know 63.8 (55.8, 71.1) 69.2 (61.6, 75.9)


P a g e | 45


3.8 Caregiver Awareness of and Exposure to the AMFm

Table 3.8.1: Caregiver awareness of and exposure to the AMFm logo and initiative

Among caregivers of children under five, the percentage who have seen or heard of the AMFm logo, or heard of the initiative to reduce the price of ACTs, by strata.

Have seen or heard of the AMFm logo

Number of

caregivers

Have heard of an initiative to reduce

the price of ACTs

Number of

caregivers

Have either seen/heard of the

AMFm logo or heard of the AMFm initiative

Number of

caregivers

Strata

Urban 29.0 (25.0, 33.3) 3,466 22.0 (19.2, 25.1) 3,466 44.0 (39.5, 48.6) 3,466

Rural 11.8 (8.3, 16.5) 3,315 15.0 (11.4, 19.5) 3,315 24.1 (18.3, 31.0) 3,315

All caregivers 13.4 (10.1, 17.5) 6,781 15.7 (12.3, 19.7) 6,781 25.9 (20.6, 32.1) 6,781


P a g e | 46


Table 3.8.2: Sources of exposure to the AMFm logo

Among caregivers of children under five who have seen or heard of the AMFm logo, the percentage citing the following sources of exposure.

Source Percent1

N=1,289

Medicine packaging 27.4 (22.1, 33.5)

TV 15.3 (10.8, 21.1)

Radio 27.7 (20.1, 36.9)

Newspaper 0.4 (0.1, 1.3)

Poster 13.1 (9.2, 18.4)

Billboard 1.3 (0.6, 2.5)

Leaflet 0.5 (0.2, 1.4)

Cap / T-Shirt / Clothing 5.4 (3.1, 9.4)

Community event 2.9 (1.7, 4.7)

Public health facility 23.0 (17.7, 29.4)

Community health worker 12.2 (8.8, 16.7)

NGO/Mission health facility 1.5 (0.9, 2.5)

Private for-profit health facility 2.7 (1.9, 3.8)

Pharmacy 6.2 (4.2, 9.1)

Drug store 1.1 (0.2, 5.2)

General Retailer 1.8 (1.0, 3.4)

Friend or neighbor 5.5 (3.3, 9.1)

Other 2.1 (1.1, 3.9)

Don’t know 0.9 (0.4, 1.9)

1 Caregivers could state multiple sources and total may sum to more than 100%.


P a g e | 47


Table 3.8.3: Sources of exposure to the AMFm initiative

Among caregivers of children under five who have heard of the AMFm initiative, the percentage citing the following sources of exposure.

Source Percent1

N=1,216

Medicine packaging 6.7 (3.3, 13.2)

TV 14.8 (11.8, 18.5)

Radio 63.1 (55.8, 69.9)

Newspaper 0.6 (0.1, 2.1)

Poster 0.9 (0.3, 2.6)

Billboard <0.1 (<0.1, 0.1)

Leaflet <0.1 (<0.1, 0.1)

Cap / T-Shirt / Clothing 0.3 (0.1, 1.6)

Community event 3.6 (2.1, 6.2)





Pharmacy 3.6 (2.0, 6.3)

Drug store 0.2 (<0.1, 1.4)


Friend or neighbor 5.6 (3.8, 8.4)

Other 0.7 (0.3, 1.6)

Don’t know 0.2 (<0.1, 1.0)

1 Caregivers could state multiple sources and total may sum to more than 100%.


P a g e | 48


Table 3.8.4: Meaning of the AMFm logo

Among caregivers of children under five who have seen or heard of the AMFm logo, the reported meaning of the logo.

Meaning Percent1

N=1,289

Medicine 31.9 (24.3, 40.6)

Cheap medicine 6.8 (2.0, 20.8)

Effective medicine 1.2 (0.4, 3.3)

Readily available medicine 0.3 (<0.1, 2.3)

Medicine recommended by the government 0.3 (0.1, 2.1)

Antimalarial 36.2 (28.0, 45.4)

Cheap antimalarial 8.2 (3.2, 19.8)

Effective antimalarial 9.5 (4.3, 19.5)

Readily available antimalarial 1.5 (0.4, 5.7)

Antimalarial recommended by the government 1.1 (0.3, 4.5)

“Health” 5.7 (4.1, 7.9)

“Environment” 7.6 (5.3, 10.7)

Other 2.3 (1.3, 4.1)

Don’t know 33.5 (27.3, 40.4) 1

Caregivers could state multiple responses and total may sum to more than 100%.


P a g e | 49


Table 3.8.5: Knowledge of the recommended price for AMFm medicine

Among caregivers of children under five who have seen or heard of the AMFm logo, the percentage who state there is a recommend price for medicine with the AMFm logo, and of these, percentage that correctly state the recommended price.

State that there is a recommended price

for medicine with the AMFm logo

Number of

caregivers

Correctly state the recommended price

Number of

caregivers

Strata

Urban N/A* N/A

Rural N/A N/A

All caregivers N/A N/A

* Not applicable: In Madagascar, no recommended retail price was negotiated for the AMFm medicine.


Table 3.8.6: Knowledge of the use of AMFm medicine

Among caregivers of children under five who have seen or heard of the AMFm logo, the percentage that cited malaria when asked what illnesses are treated with medicine with the logo.

Cite Malaria only Number

of caregivers

Strata

Urban 67.9 (60.7, 74.3) 932

Rural 69.1 (59.7, 77.1) 357

All caregivers 68.8 (61.2, 75.5) 1,289


P a g e | 50


Table 3.8.7: Caregiver reported ever use of ACTs with the AMFm logo

Among caregivers of children under five who have seen or heard of the AMFm logo, the percentage who report ever purchasing or been given medicine with the AMFm logo.

Ever purchased or been given medicine with

the AMFm logo

Number of

caregivers

Strata

Urban 24.7 (21.2, 28.5) 932

Rural 30.3 (22.0, 40.1) 357

All caregivers 29.2 (22.3, 37.1) 1,289


Table 3.8.8: 1Source of ACTs with the AMFm logo

Among caregivers of children under five who have ever purchased or been given an ACT with the AMFm logo, the percentage who report the following sources for the AMFm-logo ACT.

Source Percent1

N=329





Pharmacy 33.9 (20.2, 50.9)

Drug store 3.9 (1.3, 11.4)


Other 1.3 (0.2, 6.8)

Don’t know 0.1 (<0.1, 0.6) 1

Caregivers could state multiple sources and total may sum to more than 100%.


P a g e | 51

Table 3.8.9: Perceptions of the efficacy and affordability of AMFm ACTs

Among caregivers of children under five who have seen or heard of the AMFm logo and know that AMFm medicines are antimalarials, the perceptions of affordability and efficacy of ACTs with the AMFm logo.

Affordability Effectiveness

Cheaper than most other antimalarial medicines

Same price as most other antimalarial medicines

More expensive than most other

antimalarial medicine

Don’t know / Missing

Less effective than most other

antimalarial medicines

As effective as most other antimalarial medicines

More effective than most other

antimalarial medicines

Don’t know / Missing

Number of

children

Strata

Urban 67.2 (61.3, 72.5) 14.7 (9.9, 21.3) 6.2 (4.4, 8.6) 12.0 (8.4, 16.8) 8.5 (5.6, 12.8) 12.9 (9.6, 17.0) 63.6 (56.8, 69.9) 15.0 (10.7, 20.6) 610

Rural 59.9 (51.6, 67.8) 11.2 (7.4, 16.6) 10.1 (6.8, 14.8) 18.7 (12.2, 27.8) 5.6 (2.7, 11.3) 4.9 (2.6, 9.2) 74.9 (61.4, 84.8) 14.6 (8.0, 24.9) 227

All caregivers

61.4 (54.6, 67.8) 11.9 (8.6, 16.2) 9.3 (6.5, 13.1) 17.4 (12.0, 24.5) 6.2 (3.6, 10.5) 6.5 (4.2, 9.9) 72.6 (61.9, 81.2) 14.6 (9.1, 22.7) 837


P a g e | 52


4. References

ACTwatch Group, PSI/Madagascar and the Independent Evaluation Team. Outlet survey report 2010

(Round 2) for the Baseline Independent Evaluation of the Affordable Medicines Facility – malaria

(AMFm), Madagascar. Nairobi and Antananarivo: ACTwatch/Population Services International and

PSI/Madagascar. Available at: http://www.actwatch.info/countries/evidence.php?id_country=106

ACTwatch Group, PSI/Madagascar and the Independent Evaluation Team, 2012. ACTwatch Outlet Survey

Report 2011 (Round 3). Endline Outlet Survey Report for the Independent Evaluation of Phase 1 of

the Affordable Medicines Facility - malaria (AMFm), Madagascar. Nairobi and Antananarivo:

ACTwatch/Population Services International and PSI/Madagascar. Available at:

http://www.actwatch.info/countries/evidence.php?id_country=106

ACTwatch Group and PSI/Madagascar, 2010. Household Survey Report (Baseline) Madagascar, 2009.

Washington, DC: PSI. Nairobi and Antananarivo: ACTwatch/Population Services International and

PSI/Madagascar. Available at: http://www.actwatch.info/countries/evidence.php?id_country=106

Adeyi O, Atun R, 2010. Universal access to malaria medicines: innovation in financing and delivery. The

Lancet, 376:9755, 1869-1871.

AMFm Independent Evaluation Team, 2012. Independent Evaluation of Phase 1 of the Affordable

Medicines Facility - malaria (AMFm), Multi-Country Independent Evaluation Report. Calverton, MD

and London: ICF International and London School of Hygiene and Tropical Medicine. Available at:

http://www.theglobalfund.org/en/amfm/independentevaluation/ [accessed 20 March 2012]

Arrow, KJ, CB Panosian and H Gellband, eds. 2004. Saving Lives, Buying Time: Economics of Malaria

Drugs in an Age of Resistance. Washington, DC: National Academies Press.

Encylopaedia Britannica, 2013. Madagascar. [online] London: Encyclopaedia Britannica. Available at: <

http://www.britannica.com/EBchecked/topic/355562/Madagascar/> [accessed 20 March 2013]

INSTAT, 1993. Recensement général de la population et de l’habitat. Antananarivo : Institut National de la

Statistique.

INSTAT, 2010. Enquête périodique auprès des ménages. Antananarivo, Madagascar: Institut National de la Statistique.

INSTAT and ICF Macro, 2010. Enquête Démographique et de Santé de Madagascar 2008-2009.

Antananarivo, Madagascar and Calverton, MD: Institut National de la Statistique (INSTAT) et ICF

Macro. Available at: http://www.measuredhs.com/pubs/pdf/FR236/FR236.pdf [Accessed 20 March

2013]

INSTAT and ICF International, 2012. Enquête sur les indicateurs du paludisme à Madagascar (EIPMD)

2011. Antananarivo, Madagascar and Calverton, MD: INSTAT et ICF International. Available at:

http://www.measuredhs.com/pubs/pdf/MIS9/MIS9.pdf [Accessed 20 March 2013]

http://www.theglobalfund.org/en/amfm/independentevaluation/

http://www.measuredhs.com/pubs/pdf/FR236/FR236.pdf

P a g e | 53


Littrell M, Gatakaa H, Phok S, Allen H, Yeung S, Chuor CM, Dysoley L, Socheat D, Spiers A, White C,

Shewchuk T, Chavasse D, O’Connell KA, 2011. Case management of malaria fever in Cambodia:

results from national anti-malarial outlet and household surveys. Malaria Journal 10:328

O'Connell KA, Gatakaa H, Poyer S, Njogu J, Evance I, Munroe E, Solomon T, Goodman C, Hanson K,

Zinsou C, Akulayi L, Raharinjatovo J, Arogundade E, Buyungo P, Mpasela F, Adjibabi CB, Agbango JA,

Ramarosandratana BF, Coker B, Rubahika D, Hamainza B, Chapman S, Shewchuk T, and Chavasse D,

2011. Got ACTs? Availability, price, market share and provider knowledge of anti-malarial medicines

in public and private sector outlets in six malaria- endemic countries. Malaria Journal 10:326.

Population Division of the Department of Economic and Social Affairs, 2010. World Population Prospects:

The 2010 Revision. [online] New York, NY: United Nations Secretariat. Available at:

<http://esa.un.org/unpp> [accessed 10 January 2013].

Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat,

2011. World Urbanization Prospects: The 2011 Revision. [online]

Available at: <http://esa.un.org/wup2009/unup> [accessed 24 October 2011].

President’s Malaria Initiative (PMI), 2010. Malaria operational plan FY2011, Madagascar. Washington DC: U.S. Agency for International Development. Available at: http://www.fightingmalaria.gov/countries/mops/index.html [accessed 20 March 2011]



Shewchuk T, O'Connell KA, Goodman C, Hanson, K, Chapman S, and Chavasse D, 2011. The ACTwatch

project: methods to describe anti-malarial markets in seven countries. Malaria Journal 10:325.

UNDP, 2013. Human Development Report 2013. New York: United Nations Development Programme.

Available at: http://hdr.undp.org/en/statistics/ [accessed 20 March 2013]

World Bank, 2012a. Madagascar Overview. Washington, DC: The World Bank Group. Available at:

http://www.worldbank.org/en/country/madagascar/overview [accessed 20 March 2013]

WHO, 2010. World Malaria Report 2010. Geneva: World Health Organization. Available at:

http://www.who.int/malaria/world_malaria_report_2010/en/index.html [accessed 20 March 2013]

WHO, 2012. World Malaria Report 2012. Geneva: World Health Organization. Available at:

http://www.who.int/malaria/publications/world_malaria_report_2012/en/index.html [accessed 20

March 2013]

http://esa.un.org/wup2009/unup

http://www.fightingmalaria.gov/countries/mops/index.html



http://hdr.undp.org/en/statistics/

http://www.worldbank.org/en/country/madagascar/overview

http://www.who.int/malaria/world_malaria_report_2010/en/index.html

http://www.who.int/malaria/publications/world_malaria_report_2012/en/index.html

P a g e | 54


5. Acknowledgements

This report presents the results of the 2012 Madagascar Household Survey. It is a comprehensive,

nationally representative household survey designed to address key treatment seeking indicators, as

well as exposure to the Affordable Medicines Facility for malaria.

This report would not have been possible without the efforts of a large number of people who assisted

with the data processing, data analysis, and preparation of the report, as well as those who worked

tirelessly to collect the survey data. We particularly appreciate the efforts of the Madagascar National

Malaria Control Program for providing overall support for the survey. Additional thanks are expressed to

PSI/Madagascar for providing project management support and for implementing the field work and

PDA programming. We would also like to thank the Global Fund and the Bill and Melinda Gates

Foundation for their support.

We would like to express our thanks to the field teams and individuals involved in the survey. Their

names are presented in Appendix 6.2.

Finally, we would like to thank the thousands of caregivers who took time to complete the interview.

P a g e | 55


6. Appendices

6.1 AMFm use indicator for poorest households

Table 6.1.1: Treatment of children with fever in the poorest households

Among children under five with fever in the two weeks preceding the survey from the poorest households (lowest two wealth quintiles), the percentage who received treatment with any antimalarials, who received ACT treatment, and who received ACT treatment the same/next, by background characteristics.

Percentage who took

antimalarial medicines Percentage who

took ACTs Percentage who took

ACTs same or next day

Number of children

with fever

Urban 28.0 (17.4, 41.9) 3.2 (1.0, 10.0) 1.9 (0.6, 6.3) 763

Rural 20.3 (14.0, 28.7) 9.0 (4.4, 17.7) 7.8 (3.6, 16.1) 201

All children 20.5 (14.2, 28.6) 8.9 (4.4, 17.4) 7.7 (3.6, 15.9) 964

Source: ACTwatch Household Survey Madagascar, 2012.

P a g e | 56


6.2 Survey team

Table 6.2.1: List of staff members involved in the survey

Household mapping

RAVAHISON Fidèle Rindra Albert RAHERISOAMALALA Angeline ANDRIANIRINA FANANTENANA Patrick AMBININTSIHOARANA Zo RASOLOARIMAHEFA Michel RABETAFIKA Tantelinirina RAHERISOAMALALA Angeline IMBOALAHY Harison Francis Hervé AMBININTSIHOARANA Zo RAKOTOHARISOA Donatien RABETAFIKA Tantelinirina RABEMANANJARA Charmant D'Achise IMBOALAHY Harison Francis Hervé RAZAFIMANDIMBY Roméo RAKOTOHARISOA Donatien RANDRIAMASINORO HARINIAINA Nadia RABEMANANJARA Charmant D'Achise MIANGALISEHENO Temirova Vahatriniaina RAZAFIMANDIMBY Roméo RAKOTOMALALA Andomanitra RANDRIAMASINORO HARINIAINA Nadia BINAHARITOVONTSOA Mazurain Viot MIANGALISEHENO Temirova Vahatriniaina KAMISY Karl Sergio RAKOTOMALALA Andomanitra RAVELONJATO Antsa Nofy Hanitrarilala BINAHARITOVONTSOA Mazurain Viot BOTOMAVO Norlain KAMISY Karl Sergio RAFANOMEZANTSOA Onitiana Daniella RAVELONJATO Antsa Nofy Hanitrarilala ANDRIAMAROLAZA Tovo BOTOMAVO Norlain RAVAHISON Fidèle Rindra Albert RAFANOMEZANTSOA Onitiana Daniella ANDRIANIRINA FANANTENANA Patrick RASOLOARIMAHEFA Michel

Date collection

South East team Supervisor RABOTOVAO SOLO ANDRIAMANJAKONY Elmard Quality Controller ANDRIAMANALINA Claudio Lawrence Interviewer RAKOTOARISOA RINDRA NY AINA Mike RAVELOMANJAKA Iavantsoa RAKOTONJANAHARINIAINA Antsa Mialy RAVONINJATOVO MBELO Hery South West team Supervisor ANDRIANARINTSALAMA Romy Seheno Quality Controller RAZAFITSALAMA SOLONIAINA MAMY Tahina Interviewer RODERA Claudia Fenohasina RAKOTONDRASOA RADONIAINA Rudi RASAMIMANAJANAHARY Landy Saholy RABODOARIMALALA Mireille Lalaina South Anosy team Supervisor ANDRIANARISON MADINIRINA Sandaniaina Quality Controller TOSY RAMAHAFAKANJARA Santonia Interviewer RAHOLIARISOA Veromanitra RANDRIANJANAHARY Sela Andrianirina RAZAFIMANDIMBY Fabrianni Harinirina Prisca Rolande North West team Supervisor RAMAHENINJOHARY Mika Quality Controller RAKOTOARIMANANA Tiana Andrisoa Interviewer RAKOTOARIVONY Ginette Eulalie RABEMANANTSOA VERONALA Sylvia Annick RAZANATOVO Norovololona

RAKOTOARIMANANA Heriniaina Nomenjanahary

P a g e | 57


West Central team Supervisor RANDRIAMANANTENA Tojonirina Quality Controller RAVELOARISOA Andrianjavony Interviewer RABEMIAKATRA Fara Hanitrinivony RAKOTOZAFY Andrianirina Malala RAVOSON Felamiarantsoa ANDRIAMIANDRASOA Hasinjanahary Eastern team Supervisor ANDRIANAONINTSOA FARAMALALA Vololoniaina Quality Controller RAZAKAMANANA Andrianavosoa Heidy Interviewer DAHIZANDRY Adele Marina ANDRIAMIKANTO Santatsoa RANAIVOARISOA Julie Anna RANAIVOARISOA ONILALAINA Prisca RAZAFINDRAKOTO TSIMANDRESY Andriamiarintsoa East Central team Supervisor RAZAFIMIANDRY Honoré Quality Controller RANDRIANATREHANA Nambinintsoa Mampionona Interviewer ANDRIANJAFY Fanantenana Haingotiana RAZAFINDRATSIMBA MIALY Andrianina RAKOTONANDRASANA Sedraniaina ANDRIANAIVOSON NAMBININA Jonhson Northern team Supervisor RASOLOFOJAONA Heriniaina Manohisoa Quality Controller NAHAFENOHARAVOANA Madio Sarobidy Interviewer BETOTO Jean Tolherino BESOA Zaramanana ANDRIAMBELOSON Jean Fidelice Richad RAKOTONIRINJANAHARY Miantra South Central team Supervisor RANDRIANOMENJANAHARY Samuelson Jeannot Quality Controller RAKOTOBE Sitrakamampianina Interviewer MIHAROMANA Zafy Adorame ANDRIANOMENJANAHARY Bako Andolalaina RAVELOMANJAKA TANTELY Harisoa

RAKOTOARISON Nivoarisoa

P a g e | 58


6.3 Questionnaire

The questionnaire used for this survey can be downloaded from the ACTwatch website: www.actwatch.info/research/questionnaires.php

http://www.actwatch.info/research/questionnaires.php

Evidence for Malaria Medicines Policy

MINSTERE DE LA SANTE PUBLIQUE

household survey madagascar 2012 survey report - … · p a g e | ii household survey report...

Documents