State of Inequality

STATE OF INEQUALITY
Reproductive, maternal,
newborn and child health
I N T E R A C T I V E V I S U A L I Z A T I O N O F H E A LT H D A T A
EVERYONE EVERYWHERE ALWAYS
STATE OF INEQUALITY
Reproductive, maternal,
I N T E R A C T I V E V I S U A L I Z A T I O N O F H E A LT H D A T A
WHO Library Cataloguing-in-Publication Data
State of inequality: reproductive, maternal, newborn and child health
I.World Health Organization.
ISBN 978 92 4 156490 8
Subject headings are available from the WHO institutional repository
© World Health Organization 2015
All rights reserved. Publications of the World Health Organization are available on the WHO website (www.who.int) or can be purchased from WHO Press,
World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: [email protected]).
Requests for permission to reproduce or translate WHO publications — whether for sale or for non-commercial distribution — should be addressed to WHO
Press through the WHO website (www.who.int/about/licensing/copyright_form/en/index.html).
The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the
World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or
boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement.
The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the World Health
Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are
distinguished by initial capital letters.
All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published
material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with
the reader. In no event shall the World Health Organization be liable for damages arising from its use.
Design and layout by L’IV Com Sàrl, Villars-sous-Yens, Switzerland.
Printed in Luxembourg.
Table of contents
Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
Acknowledgements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
Executive summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii
1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.1 Inequality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Health inequality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.3 Reproductive, maternal, newborn and child health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3. Monitoring the state of inequality in RMNCH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

3.1 Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1.1 Data sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1.2 Health indicator data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1.3 Dimension of inequality data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1.4 Country selection.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 Analysis.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2.1 Data disaggregation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2.2 Summary measures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3 Reporting.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.3.1 Data visualization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.3.2 Feature stories. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4. The state of inequality in RMNCH: stories from low- and

middle-income countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.1 Reproductive health interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2 Maternal health interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.3 Care-seeking for sick children. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.4 Childhood immunization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.5 Child malnutrition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.6 Child mortality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.7 RMNCH interventions, combined. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.8 Potential for improvement in RMNCH interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
iii
STATE OF INEQUALITY: REPRODUCTIVE, MATERNAL, NEWBORN AND CHILD HEALTH
5. Reporting the state of inequality: taking stock. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

5.1 The importance of data disaggregation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.2 Equity orientation of policies, programmes and practices. . . . . . . . . . . . . . . . . . . . . . . . . 57
5.3 Equity-oriented health information systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.4 Reducing inequality across health topics and the post-2015 sustainable
development agenda. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Appendices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Appendix 1. Data and analysis methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Appendix 2. Assessing health inequality: methodological considerations. . . . . . . . . . . . . . . . . . 72
Appendix 3. Visualizing disaggregated data using maps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Appendix 4. Guide to interpreting the visuals used in this report.. . . . . . . . . . . . . . . . . . . . . . . . 78
Appendix 5. Interactive visualization of health data.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Appendix 6. Additional interactive visuals: references for further data exploration. . . . . . . . . . . 83
Appendix 7. Patterns of inequality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Supplementary tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
iv
TABLE OF CONTENTS
Figures
Figure 4.1 Contraceptive prevalence (modern methods) by woman’s education
in 71 low- and middle-income countries: latest situation (DHS and MICS
2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Figure 4.2 Contraceptive prevalence (modern methods) by woman’s education in
38 low- and middle-income countries: change over time (DHS and MICS
1995–2004 and 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Figure 4.3 Births attended by skilled health personnel by economic status in
30 low-income and 53 middle-income countries: latest situation
(DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Figure 4.4 Births attended by skilled health personnel in 83 low- and middle-income
countries (national average and within-country economic-related inequality):
latest situation (DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Figure 4.5 Births attended by skilled health personnel: change over time in
national average (absolute change) and in the poorest compared with
the richest quintile (absolute excess change) in 42 low- and middle-income
countries (DHS and MICS 1995–2004 and 2005–2013). . . . . . . . . . . . . . . . . . . . . 27
Figure 4.6 Children aged less than five years with pneumonia symptoms taken
to a health facility by place of residence in 72 low- and middle-income
countries: latest situation (DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . 30
Figure 4.7 Children aged less than five years with pneumonia symptoms taken
to a health facility: change over time in national average (absolute change)
and in rural compared with urban areas (absolute excess change) in
33 low- and middle-income countries (DHS and MICS 1995–2004
and 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Figure 4.8 DTP3 immunization coverage among one-year-olds by economic
status in 78 low- and middle-income countries: latest situation
(DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Figure 4.9 DTP3 immunization coverage among one-year-olds: change over
time in national average (absolute change) and in the poorest compared
with the richest quintile (absolute excess change) in 41 low- and
middle-income countries (DHS and MICS 1995–2004 and 2005–2013). . . . . . . . . 35
Figure 4.10 Stunting prevalence in children aged less than five years by mother’s
education in 30 low-income and 36 middle-income countries: latest
situation (DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Figure 4.11 Stunting prevalence in children aged less than five years: change over
time in national average (absolute change) and in the least-educated
compared with the most-educated subgroup (absolute excess change)
in 29 low- and middle-income countries (DHS and MICS 1995–2004
and 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Figure 4.12 Under-five mortality rates by place of residence in 54 low- and
middle-income countries: latest situation (DHS 2005–2013). . . . . . . . . . . . . . . . . . 42
Figure 4.13 Under-five mortality rates: change over time in national average
(absolute change) and in rural compared with urban areas (absolute
excess change) in 37 low- and middle-income countries (DHS 1995–
2004 and 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
v
Figure 4.14 RMNCH composite coverage index by multiple dimensions of

inequality in low- and middle-income countries: latest situation
(DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Figure 4.15 Potential for improvement in RMNCH intervention coverage by
eliminating within-country economic-related inequality in Egypt
(DHS 2008) and Niger (DHS 2012). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Figure 5.1 Under-five mortality rate doubly disaggregated by place of residence
and economic status in Benin (DHS 2006). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Figure A2.1 Antenatal care coverage (at least four visits) by woman’s education
in Indonesia (DHS 2002 and 2012).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Figure A2.2 Proportion of women aged 15–49 years with a live birth within three
years preceding the survey by education in Indonesia (DHS 2002
and 2012). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Figure A3.1 Stunting prevalence in children aged less than five years by mother’s
education in 74 low- and middle-income countries: latest situation
(DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Figure A7.1 Patterns of inequality in health intervention coverage by economic status. . . . . . . . 85
Tables
Table 2.1 Development initiatives and RMNCH: a recent history. . . . . . . . . . . . . . . . . . . . . . . . 6
Table 3.1 Selected RMNCH indicators used in this report. . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Table 3.2 Health indicator categories and relevant dimensions of inequality. . . . . . . . . . . . . . 11
Table 3.3 The calculation of selected summary measures and their application
in this report. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Table 4.1 Potential for improvement in national RMNCH intervention coverage by
eliminating within-country economic-related inequality (population
attributable risk) in low- and middle-income countries with available data
(DHS and MICS 2005–2013). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Table A1.1 Health determinants included in correlation analyses with RMNCH
indicators, grouped according to the EQuity-oriented Analysis of
Linkages between health and other sectors (EQuAL) Framework. . . . . . . . . . . . . . . 68
Table A1.2 Difference and ratio summary measure calculations by dimension
of inequality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Table A2.1 Guide to interpreting annual absolute excess change estimates. . . . . . . . . . . . . . . . 75
Table A4.1 Interpreting the visuals used in this report.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Table A5.1 Best practices in interactive visualization of data.. . . . . . . . . . . . . . . . . . . . . . . . . . . 82
vi
TABLE OF CONTENTS
Interactive visuals
Interactive visual 1. Reproductive health interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Interactive visual 2. Maternal health interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Interactive visual 3. Care-seeking for sick children.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Interactive visual 4. Childhood immunization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Interactive visual 5. Child malnutrition.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Interactive visual 6. Child mortality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Interactive visual 7. RMNCH composite coverage index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Interactive visual 8. Potential for improvement in RMNCH interventions. . . . . . . . . . . . . . . . . 53
Interactive visual 9. Determinants of RMNCH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Interactive visual A1. Equity country profiles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Interactive visual A2. Maps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Interactive visual A3. Reference tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Interactive visual A4. RMNCH interventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Video clips
Video clip 1. Health inequality is multidimensional . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Video clip 2. Benchmarking puts inequality in context. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Video clip 3. Health inequalities are widespread. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Video clip 4. Health inequalities inform policies, programmes and practices. . . . . . . . . . . . . . . . 86
Frequently used abbreviations

BCG one dose of Bacille Calmette-Guérin vaccine
DHS Demographic and Health Survey
DTP3 three doses of the combined diphtheria, tetanus toxoid and pertussis vaccine
MICS Multiple Indicator Cluster Survey
RMNCH reproductive, maternal, newborn and child health
vii
Foreword
2015 is the target year for the Millennium Development Goals, marking
both the end of an ambitious 15-year effort to improve the lives of the
world’s poorest, but also a time for new beginnings, with opportunities
to refocus, renew and revitalize the approach of successive global
development initiatives.
The post-2015 sustainable development agenda aims to build upon and expand the unfinished work of
the Millennium Development Goals, and to adopt a broader perspective that includes all countries. The
report of the Open Working Group on Sustainable Development Goals, endorsed by the United Nations
General Assembly in 2014, emphasizes the importance of poverty eradication, environmental sustainability,
inclusive growth, equality and a people-centred agenda for sustainable development. The Secretary-
General’s synthesis report on the post-2015 sustainable development agenda, The road to dignity by 2030:
ending poverty, transforming all lives and protecting the planet, describes a renewed paradigm for sustainable
development based on six elements: dignity, people, prosperity, planet, justice and partnership.
A sense of optimism prevails – now is the time to take inequity more seriously and create a world where
every person enjoys a basic standard of well-being.
At this important juncture, and under the banner of the post-2015 sustainable development agenda, the
World Health Organization supports universal health coverage as the means to ensure that high-quality,
essential health services are available and affordable to all. When universal health coverage is pursued
through progressive realization it upholds principles of fairness and equity, setting the course for realizing
the right to health. In the words of World Health Organization Director-General Margaret Chan, “universal
health coverage is the single most powerful concept that public health has to offer”.
Achieving equity in health requires a commitment to monitoring health inequalities which, in turn,
necessitates strong, equity-oriented health information systems. High-quality data and robust monitoring
systems ensure that efforts can be targeted appropriately and that progress can be tracked. Countries
must strengthen health information systems to generate better data and evidence to measure progress.
Integral to the health inequality monitoring process is the task of reporting data in a meaningful way.
This State of inequality report exemplifies effective reporting practices, featuring the topic of reproductive,
maternal, newborn and child health. The report addresses the challenge of how to best communicate a large
and complex body of data in a manner that is comprehensible, flexible and appealing to a wide readership.
viii
FOREWORD
Feature stories provide an in-depth look at the state of inequality for selected indicators and highlight key
observations in reproductive health interventions, maternal health interventions, care-seeking for sick
children, childhood immunization, child malnutrition and child mortality. Perhaps most notable, however,
is the innovative use of electronic visualization technology. Story-point dashboards, for instance, guide the
reader through a succession of visuals where readers can use interactive tools to further explore, sort and
filter the data. Similarly, interactive maps and tables engage readers in customizing how data are viewed.
The report reveals that significant inequalities exist in low- and middle-income countries in the area of
reproductive, maternal, newborn and child health. The good health of women, infants and children is
essential for sustainable development, and there is still much work to be done. Discussions will increasingly
call into question how efforts to improve reproductive, maternal, newborn and child health can achieve
early and accelerated progress among those who are falling behind.
This State of inequality report helps to focus the monitoring and reporting of health inequalities, and provides
comprehensive information on the state of inequality in reproductive, maternal, newborn and child health
in low- and middle-income countries.
Flavia Bustreo Marie-Paule Kieny

Assistant Director-General Assistant Director-General
Family, Women’s and Children’s Health Cluster Health Systems and Innovation Cluster
World Health Organization World Health Organization
ix
Acknowledgements
Ahmad Reza Hosseinpoor, Nicole Bergen and Anne Schlotheuber (World Health Organization, Geneva,
Switzerland) conceptualized the report, wrote the first draft and prepared the electronic interactive
components. Ahmad Reza Hosseinpoor coordinated the overall development of the report.
The disaggregated data used in the report are the product of a reanalysis of survey micro-data by Aluisio JD
Barros, Cesar Victora, Giovanny Araújo, Maria Clara Restrepo and Kerry Wong at the International Center
for Equity in Health based in the Federal University of Pelotas, Pelotas, Brazil. Further data analyses, including
the calculation of summary measures and data preparation for the interactive visuals, were conducted by
Anne Schlotheuber and Ahmad Reza Hosseinpoor.
A number of individuals reviewed the interactive dashboards and provided comments, including, in
alphabetical order, Stephen Few (Perceptual Edge, Berkeley, United States of America), Katherine Rowell
(HealthDataViz, Newton, United States of America) and Michael Wong (Partnership for Maternal,
Newborn & Child Health, Geneva, Switzerland), as well as several World Health Organization staff
members: Fiona Gore, Jessica Ho, Doris Ma Fat, Annet Mahanani, Liliana Pievaroli, Gretchen Stevens,
Hazim Timimi (World Health Organization, Geneva, Switzerland) and Ramon Martinez (World Health
Organization, Washington DC, United States of America). Florence Rusciano (World Health Organization,
Geneva, Switzerland) provided advice on the preparation of interactive maps and developed the maps in
Figure A3.1. Humna Amjad (intern, World Health Organization, Geneva, Switzerland) contributed to the
development of the video clips.
The report benefited from the contributions of a number of World Health Organization programme staff.
The following individuals reviewed sections of the report and provided comments according to their area
of expertise in reproductive, maternal, newborn and/or child health (acknowledged in alphabetical order):
Monika Barbara Bloessner, Elaine Borghi, Cynthia Boschi Pinto, Doris Chou, Marta Gacic-Dobo, Anna
Gruending, Rajat Kholsa, Matthews Mathai, Ann-Beth Moller, Shamim Ahmad Qazi, Lale Say, Kamel
Senouci and Nicole Valentine (World Health Organization, Geneva, Switzerland).
Aluisio JD Barros, Cesar Victora (Federal University of Pelotas, Pelotas, Brazil), Ties Boerma, Somnath
Chatterji, Theadora Koller, Veronica Magar (World Health Organization, Geneva, Switzerland) and Amit
Prasad (World Health Organization, Kobe, Japan) reviewed the report in full.
x
ACKNOWLEDGEMENTS
Petra Schuster provided valuable administrative support.
Ann Morgan provided copy-editing support, AvisAnne Julien proofread the document and Liza Furnival
prepared the index.
Funding for the project was provided in part by the Rockefeller Foundation and the Norwegian Agency for
Development Cooperation (Norad).
This work is the product of a collaboration between the World Health Organization Department of Health
Statistics and Information Systems; the World Health Organization Gender, Equity and Human Rights Team;
and the International Center for Equity in Health, Federal University of Pelotas, Pelotas, Brazil.
xi
Executive summary
The State of inequality: reproductive, maternal, newborn and child health report delivers both promising and
disappointing messages about the situation in low- and middle-income countries. On the one hand,
within-country inequalities have narrowed, with a tendency for national improvements driven by faster
improvements in disadvantaged subgroups. In certain indicators and countries, these improvements have
been substantial. On the other hand, however, inequalities still persist in most reproductive, maternal,
newborn and child health (RMNCH) indicators. The extent of within-country inequality differed by
dimension of inequality and by country, country income group and geographical region. There is still much
progress to be made in reducing inequalities in RMNCH.
REPORTING ON INEQUALITIES IN RMNCH
The State of inequality: reproductive, maternal, newborn and child health report looks at the state of inequality in health,
answering key questions: according to the latest available data, what is the status of inequality across and within countries?
How have levels of health changed in population subgroups over time?
The objective of this report is to showcase best practices in reporting the state of inequality in low- and middle-income
countries using high-quality data, sound and transparent analysis methods, and user-oriented, comprehensive reporting.
This report encompasses the latest status of inequality and changes over time across 23 RMNCH indicators, disaggregated
by four dimensions of inequality (economic status, education, place of residence and sex). It draws on data from 86 low-
and middle-income countries from all world regions. In a subset of 42 low- and middle-income countries (where data
availability permitted), it was also possible to assess how the extent of inequality had changed over time.
The use of effective reporting practices helps to convey clear, salient messages about the state of inequality. Visualization
technology facilitates the presentation and interpretation of large amounts of data, as results can be displayed using
interactive, customizable views.
Overall, inequalities were to the detriment of women, infants and children in disadvantaged population
subgroups; that is, the poorest, the least educated and those residing in rural areas had lower health
intervention coverage and worse health outcomes than the more advantaged. In a minority of cases, child
health interventions or outcomes were unequal between boys and girls.
xii
EXECUTIVE SUMMARY
Latest situation of inequality

The latest situation of inequality in RMNCH revealed inequalities across low- and middle-income countries
in terms of national figures. Within-country inequality differed across health indicators. Maternal health
intervention indicators demonstrated pronounced within-country inequalities. The largest gaps in coverage
– between the richest and poorest, the most and least educated, and urban and rural areas – were reported
for births attended by skilled health personnel, followed by antenatal care coverage (at least four visits).
Inequalities were also reported in antenatal care coverage (at least one visit), though to a lesser extent
than the two above-mentioned maternal health interventions.
• The proportion of births attended by skilled health personnel differed by up to 80 percentage points
between the richest and poorest subgroups; this difference was 37 percentage points or higher in half
of countries.
• In half of countries, antenatal care coverage (at least four visits) differed by at least 25 percentage
points between both the most and least educated, and the richest and poorest.
• Antenatal care coverage (at least one visit) was at least 10 percentage points higher among women
in the richest subgroup than those in the poorest subgroup in half of countries.
Reproductive health intervention indicators also indicated a situation of inequality.

• The use of modern contraception was at least twice as high among women with secondary schooling
or higher than among women with no education in nearly half of countries.
Immunization indicators demonstrated low to moderate coverage gaps across different dimensions of
inequality.
• Countries demonstrated no – or very low levels of – sex-related inequality in immunization coverage.
The difference in immunization coverage between boys and girls did not exceed 10 percentage points
in any study country.
• Looking at BCG, polio, measles and DTP3 immunization among one-year-olds, in each case there
was a difference of less than 5 percentage points between coverage in rural and urban areas in half of
countries.
• Over one third of countries reported a gap of less than 5 percentage points between BCG immunization
coverage in the richest and poorest subgroups.
Indicators related to care-seeking for sick children showed higher inequality in care-seeking for pneumonia
symptoms than for diarrhoea. (Note that estimates were subject to small sample sizes, and results were
highly variable across countries.) There were divergent patterns across countries in the level of inequality
in the early initiation of breastfeeding.
• In half of countries, there was at least an 18 percentage point gap in care-seeking for children with
pneumonia symptoms between the poorest and richest subgroups.
• About the same number of countries reported pro-poor inequality in early initiation of breastfeeding
(higher prevalence of breastfeeding in the poorest than in the richest subgroup) as reported pro-rich
inequality (higher prevalence in the richest than in the poorest subgroup). Overall, there was no prevailing
pattern in economic-related inequality in breastfeeding practices across countries.
xiii
Inequalities were also reported in child health outcomes. Under-five mortality rate and stunting prevalence
in children aged less than five years demonstrated particularly high levels of inequality by economic status,
education, place of residence and, to a lesser extent, sex.
• A large majority of countries reported a higher under-five mortality rate in rural than in urban areas. In
half of countries, the difference between rural and urban areas exceeded 16 deaths per 1000 live births.
• Stunting prevalence in children aged less than five years was elevated by as much as 39 percentage
points in the children of mothers with no education compared with those children whose mothers had
attended secondary school or higher. In half of countries, the education-related difference between these
two subgroups was 15 percentage points or more.
Change in inequality over time

Inequalities in health are not static, but change over time. Looking at changes over a period of about 10
years, global figures indicated improvements at the national level in many areas of RMNCH. Also, countries
tended to report gains that were faster in disadvantaged subgroups than in advantaged subgroups, which
is desirable for the reduction of inequalities. The patterns of change in inequality over time varied by health
indicator, and according to country and dimension of inequality.
For example, among the immunization indicators, improvements at the national level tended to be
accompanied by gains in the disadvantaged subgroups that outpaced those in the advantaged subgroups.
• In half of countries, the changes in polio and DTP3 immunization coverage among one-year-olds
indicated situations that were pro-poor, favouring children in the poorest subgroup over the richest by
a margin of at least 9 percentage points over 10 years.
For a given indicator, the change in inequality over time sometimes varied across the dimensions of inequality.
• Demand for family planning satisfied, for example, showed substantial progress in narrowing education-
related inequality over the past 10 years, with increases in the no education subgroup exceeding increases
among those who attended secondary school or higher by at least 9 percentage points in half of countries.
The gains in reducing place-of-residence inequality, however, were slower, with progress in rural areas
outpacing that in urban areas by at least 3 percentage points over the 10-year period in half of countries.
Overall, the change over time in child mortality indicators indicated improved national averages and narrowing
inequalities, particularly for under-five mortality. Child malnutrition indicators reported a similar tendency
towards decreasing national averages; however, there was little change in the level of existing inequality.
• The under-five mortality rate decreased more rapidly in the poorest than in the richest subgroup, by a
margin of at least 26 deaths per 1000 live births over a 10-year period.
• Comparing the pace of change in stunting prevalence among children aged less than five years in the
poorest and richest subgroups revealed divergent patterns across study countries. Several countries
reported a strong pro-poor situation (changes in prevalence favoured the poorest subgroup) whereas
several other countries reported a pro-rich situation (changes in prevalence favoured the richest
subgroup). Overall, there was little indication that economic-related inequality in stunting prevalence
had decreased globally.
xiv
EXECUTIVE SUMMARY
THE COMPOSITE COVERAGE INDEX

The composite coverage index is a single indicator that summarizes the level of coverage across the spectrum of RMNCH
interventions. It includes eight indicators: demand for family planning satisfied; antenatal care coverage (at least one visit);
births attended by skilled health personnel; BCG immunization coverage among one-year-olds; measles immunization
coverage among one-year-olds; DTP3 immunization coverage among one-year-olds; children aged less than five years
with diarrhoea receiving oral rehydration therapy and continued feeding; and children aged less than five years with
pneumonia symptoms taken to a health facility. Overall, more than half of countries reported composite coverage index
values of 70% or more. The level of RMNCH interventions coverage varied substantially across countries, ranging from
under 40% to nearly 90%.
Within-country inequality existed according to different dimensions of inequality, and variations were observed by country.
• There was a poorest-to-richest difference of at least 20 percentage points in half of countries; the maximum economic-
related difference in combined RMNCH interventions coverage was over 60 percentage points.
• Those with secondary schooling or higher education reported composite coverage index levels of up to 46 percentage
points greater than those with no education.
• The rural-to-urban gap in coverage was over 10 percentage points in half of countries.
Nationally, the coverage levels of RMNCH interventions increased over the past decade; this was usually accompanied by
faster improvements in the most-disadvantaged subgroups, though there was variation by country.
• Half of countries reported an increase in coverage that was at least 6 percentage points higher in the poorest than in
the richest subgroup over a period of 10 years.
• Countries reported coverage increases in all education subgroups, with the no education subgroup outpacing those with
secondary schooling or higher by up to a maximum of 18 percentage points over 10 years.
• In most countries, the rural-to-urban gap in coverage narrowed, with faster improvements in rural than in urban areas
by a margin of 4 percentage points or higher over 10 years in half of countries.
Implications of health inequality monitoring

While current national averages and improvements over time are important indications of progress on
a global level, reporting inequalities within countries reveals the different experiences of rural and urban
residents, the poor and the rich, the educated and the non-educated, and females and males. Monitoring the
state of inequality, which includes tracking the change over time, unravels how progress in national averages
is realized by population subgroups. Establishing goals and targets that specify a reduction in inequality
encourages the orientation of policies, programmes and practices to promote health in disadvantaged
subgroups. Without a dedicated focus on equity, efforts to improve health risk perpetuating or intensifying
within-country inequality, even as increases in national coverage are achieved.
Equity-oriented health information systems are the foundation for monitoring health inequality. When
health information systems are equity oriented they have the tools available to collect, analyse and report
data about health inequality. Building capacity for health inequality monitoring requires developing,
strengthening and/or expanding equity-oriented health information systems at the national level.
Health inequality monitoring is an essential step towards achieving health equity. It has broad applications
and can be conducted across diverse health topics. Applying the best practices in health inequality
monitoring presents an opportunity to share the state of inequality with stakeholders, indicate areas in
need of improvement and track progress over time.
xv
1
Introduction
This report, State of inequality: reproductive, maternal, newborn and child health, was developed to demonstrate
best practices in reporting the results of health inequality monitoring, and to introduce innovative, interactive
ways for audiences to explore inequality data. The report draws on data about reproductive, maternal,
newborn and child health (RMNCH) in low- and middle-income countries – an important topic in global
health – but the approach and underlying concepts can be applied to any health topic.
Using comparable and publically available data from 86 low- and middle-income countries, this report
showcases sound and transparent analytical methods and user-oriented, comprehensive reporting
practices. The report is novel because it adopts an expansive scope, presenting harmonized data for a wide
selection of RMNCH indicators and allowing comparisons to be made across countries and over time. In
addition, electronic visualization components provide readers with a unique opportunity to explore the
data in ways appropriate to their needs and interests; customized data views and outputs can be created
for setting-specific benchmarking and reporting purposes.
The report was primarily developed for those who work with health information systems and have basic
skills in interpreting health-related data. This encompasses a broad audience of technical staff (for example,
in ministries of health), public health professionals, policy-makers, researchers, students and others. The
content and principles contained within this report have relevance to those interested in health inequality
monitoring, health data communication, novel applications of interactive technologies and the state of
inequality in RMNCH. Readers are not required to have specialized knowledge about health inequality nor
experience with interactive visualization technologies to engage with this report.
The report is comprised of five chapters with accompanying appendices, supplementary tables and
electronic visualization components.
1
Readers are first introduced to the concept of inequality and the importance of monitoring inequalities in
health. A brief overview of current issues in RMNCH is presented to familiarize readers with the relevance
of the topic at hand (Chapter 2).
Next, the approach to monitoring the state of inequality is discussed, covering specific matters such as
data, analysis and reporting methods (Chapter 3). The State of inequality: reproductive, maternal, newborn
and child health report uses the latest available disaggregated data about RMNCH from the World Health
Organization (WHO) Health Equity Monitor database to derive and report estimates for 23 RMNCH
indicators. Data are disaggregated by four dimensions of inequality: economic status, education level,
place of residence and sex (1).
The main body of the report is devoted to a series of feature stories about inequalities in RMNCH
(Chapter 4). These have been selected to provide readers with an insight into some of the key messages
about inequalities in health among women, mothers, newborns and children.
FEATURE STORIES ABOUT THE STATE OF INEQUALITY IN RMNCH
Eight selected feature stories illustrate the state of inequality in RMNCH.

• Contraceptive use is compared among women with differing levels of education.
• The percentage of babies delivered by skilled health personnel is presented by wealth quintile.
• Care-seeking for sick children with pneumonia symptoms is compared between rural and urban areas.
• The coverage of three doses of the combined diphtheria, tetanus toxoid and pertussis vaccine (DTP3) among one-year-
olds is explored in subgroups of differing economic status.
• The prevalence of stunting in children under five years of age is presented according to mother’s education level.
• The rate of under-five mortality is compared between children living in rural and urban areas.
• A composite index of health interventions is contrasted across education levels, wealth quintiles and places of residence.
• The potential for improvement in the coverage of health interventions that could be achieved by eliminating within-
country inequality is explored.
Conventional means of data visualization, such as static tables, graphs and maps, are presented in the
report to illustrate the feature stories, while a multitude of other findings can be uncovered in the interactive
visualization components that accompany each feature story. Readers can engage with these interactive
visuals to animate the data behind feature stories and construct displays relevant to their own interests
and needs.
The report concludes with a reflection on the state of inequality and reiterates the importance of health
inequality monitoring and effective reporting as a means to inform health policies, programmes and
practices (Chapter 5). Capacity-building for health inequality monitoring requires the equity orientation
of health information systems, with improved collection, analysis and reporting of health data that are
disaggregated by population subgroups. Health inequality monitoring is a growing priority on the global
health agenda, particularly with the movement towards the progressive realization of universal health
coverage. Reports about the state of inequality are warranted across all health topics.
2
2
Background
2.1 Inequality
Inequalities exist in many areas and can be measured using various indicators. Often, inequalities are
quantified by comparing the national average value of an indicator across countries. Such national figures,
however, do not account for inequalities that exist within countries, that is, between the different subgroups
that comprise the national population. In addition to cross-country inequality, it is important to also consider
within-country inequality, which captures the different experiences of men and women, boys and girls, rural
and urban residents, the rich and the poor, the young and the old, the educated and the non-educated, etc.
Until recently, development goals and agendas have lacked a systematic focus on the reduction of within-
country inequality. Emphasis has tended to be placed on improving the overall national situation (that is,
the national average), with too little attention devoted to narrowing the gaps that exist between subgroups
of the population. The Millennium Development Goals (MDGs), adopted in the year 2000, called for
improvements in national averages, but did not address how efforts to achieve the goals might affect
within-country inequality. While a country may be on track to achieve national MDG targets, the situation
with respect to some subgroups of the population may have stagnated or even worsened over time. In
the absence of inequality monitoring, it remains unknown whether countries have narrowed or widened
the gap between the advantaged and the disadvantaged. Acknowledging this deficit, several subsequent
initiatives have emerged that promote the practice of monitoring the state of inequality alongside actions
to attain the targets specified in the MDGs.
The emerging post-2015 sustainable development agenda, which will be adopted at the United Nations
summit in September 2015, cites equity as a central principle of the renewed global development goals
and targets. Learning from the oversights of the past, the post-2015 development agenda recognizes the
monitoring and reduction of inequalities as a global priority (2). Identifying and understanding inequalities
helps to pinpoint the key drivers of inequity and, in turn, informs targeted action to improve the situation
of the disadvantaged.
3
Inequalities are perpetuated when certain subgroups are routinely subject to discrimination, human rights
violations and other structural barriers related to cultural, economic, environmental, political and social
domains. These effects cumulate in stalled progress or deteriorating situations, which reinforce vulnerability,
powerlessness and disadvantage. Unless actions are taken to benefit the most disadvantaged, they will
continue to fall further behind. Efforts to achieve equity and reduce inequalities are warranted not only
for the sake of social justice and human rights, but also because of the concomitant social, political and
economic benefits.
2.2 Health inequality

The health of the world’s population is in a state of inequality. That is to say, there are vastly different stories
to tell about a person’s health depending on where they live, their level of education, and whether they
are rich or poor, etc. Describing the state of inequality in health compares the experiences of population
subgroups of different social classes, ages and sex. It sheds light on questions such as: how do mortality
rates differ between rural and urban areas? Do the richest members of a population have better access to
skilled health personnel than the poorest? Is there a difference between the levels of malnutrition among
children born to women with higher versus lower levels of education? Finding answers to these – and
similar – questions helps to identify those differences in health that are unjust, and is an important first
step towards promoting health equity and the right to health.
Calling attention to the importance of health equity is neither a new, nor novel practice. The right to health
is a fundamental human right, as affirmed in the WHO 1946 constitution and in numerous legally binding
human rights conventions (3). Ensuring that all individuals of a population have the opportunity to realize
the right to health sets nations on a course to develop and thrive. The Declaration of Alma-Ata, adopted
in 1978, was among the first major international proclamations that identified the need for urgent action
“to protect and promote the health of all people of the world” and recognize the inequalities in health that
exist, both between countries and within them (4). The Global Strategy for Health for All was subsequently
adopted by the World Health Assembly in 1981, prioritizing the achievement of equity in the way that
health resources and health care are distributed and accessed (5).
A HUMAN RIGHTS APPROACH TO PROMOTING HEALTH
A human rights based approach to promoting health is guided by the key principles of availability; accessibility; acceptability
and quality of facilities and services; participation; equality and non-discrimination; and accountability (6).
More recently, this call to promote health among disadvantaged populations has been echoed through
other important global initiatives, notably the Commission on Social Determinants of Health (7), and the
Rio Political Declaration on Social Determinants of Health (8). Increasingly, global initiatives are orienting
towards establishing health inequality monitoring practices and recommending tangible actions to reduce
health inequalities, with a focus on accountability and results.
4
2. BACKGROUND
The inclusion of universal health coverage as part of the health-related post-2015 sustainable development
agenda puts equity at the forefront of a major global movement (9). The concept of universal health
coverage encompasses two components: all people should be able to obtain high-quality, essential
health interventions, which they should be able to access without experiencing undue financial hardship.
Proposed targets for universal health coverage are that all populations achieve a minimum of 80%
coverage of essential health interventions and 100% financial protection (9). The reduction of inequalities
in both components – coverage of health interventions and financial protection – is key for the progressive
realization of universal health coverage (10, 11). From the initial implementation of universal health coverage
through to its realization, ongoing monitoring of the state of inequality is vital to ensure that disadvantaged
populations are identified and prioritized.
INEQUALITY IN HEALTH VERSUS INEQUITY IN HEALTH
Health inequalities are observable differences in health between subgroups of a population. Subgroups can be defined
by demographic, geographic or socioeconomic factors such as age, economic status, education, place of residence and
sex. Inequalities exist wherever there are differences in health indicators between subgroups. When health data are
disaggregated – broken down by subgroups – they reveal differences between social groups that might have otherwise
remained hidden behind the overall average.
Health inequity is a normative concept, defined as the avoidable and/or unjust differences in health between population
subgroups. Statements about health equity involve a judgement about what is deemed to be right, fair or acceptable in
a society. Measuring and monitoring health inequalities is a starting point from which health equity can be evaluated.
To illustrate, socioeconomic differences in under-five mortality rates – based on, for example, economic status or mother’s
education level – suggest a situation where inequality represents an inequity. The interpretation of sex-related inequality
in under-five mortality, however, is more complex because under-five mortality rates tend to be higher in boys due to
biological reasons that are not related to gender discrimination (12). Thus, an observed inequality may not constitute a
situation of inequity.
2.3 Reproductive, maternal, newborn and child health

Infancy, childhood and women’s childbearing years are widely recognized as critical junctures for lifelong
health, and by extension, thriving and productive populations. Any level of preventable maternal or child
mortality is unacceptable, and inequities associated with RMNCH interventions and outcomes warrant
action. Improving RMNCH is the explicit focus of countless initiatives at regional, national and subnational
levels, and is a priority area for international health and development organizations, such as WHO, World
Bank, the United Nations Children’s Fund and the United Nations Population Fund (Table 2.1).
5
TABLE 2.1 Development initiatives and RMNCH: a recent history
Initiative Key publications

The MDGs directly address RMNCH in goal four (to A series of annual reports and progress charts provide
ESTABLISHED 2000
reduce child mortality) and goal five (to improve detailed assessments of global progress towards the MDGs
maternal health). Specific targets for these goals aim (13).
to reduce the 1990 under-five child mortality rate by
two thirds, reduce the 1990 maternal mortality ratio
➜ The Millennium Development Goals report 2014 demonstrates
inequality in selected child and maternal health outcomes
by three quarters, and achieve universal access to
by region and/or country income group (15).
reproductive health by the year 2015 (13, 14).
Building on the momentum of the MDGs, other initiatives have emerged that monitor, report and promote advancements in
RMNCH, emphasizing equity as a key component.
The Partnership for Maternal, Newborn & Child Beginning in 2011, annual publications by the Partnership
ESTABLISHED 2005
Health brings together and supports key players in the for Maternal, Newborn & Child Health report stakeholders’
global health community to achieve progress and spur progress on commitments to the Global Strategy for
the implementation of actions to advance MDGs four ➜ Women’s and Children’s Health (16–19).
and five.
Countdown to 2015 was the first initiative to Fulfilling the health agenda for women and children: the
systematically report inequalities in RMNCH, 2014 report features summarized accountability profiles
ESTABLISHED 2005
tracking progress towards the MDGs within and for 75 countries with country-specific data about health
across the 75 countries where 95% of maternal indicators and information about coverage, trend over time,
and child deaths occurs. Comprised of academic socioeconomic inequity and demographic indicators (20).
institutions, governments, international agencies, ➜ This report was preceded by other biannual reports (21–24).
professional organizations, donor organizations and
nongovernmental organizations, Countdown to 2015
disseminates country-specific data on topics related to
RMNCH.
The Every Woman, Every Child movement was The 2014 report, Every newborn: an action plan to end
ESTABLISHED 2010
launched at the 2010 Millennium Development Goals preventable deaths, identifies specific global and national
Summit, and aims to galvanize action by governments, targets and milestones with the goals of ending preventable
multilaterals, the private sector and civil society to ➜ newborn deaths and still births (25).
address health challenges of women and children
around the world.
In 2010, the United Nations General Assembly, aiming

to save the lives of 16 million women and children
ESTABLISHED 2010
by the year 2015, launched the Global Strategy for

Women’s and Children’s Health. The Commission’s final report, Keeping promises, measuring
Subsequently, the Commission on Information and ➜ results, sets forth a framework to guide oversight,
accountability and reporting in RMNCH (26).
Accountability for Women’s and Children’s Health
was established to make recommendations for tracking
and monitoring progress.
6
2. BACKGROUND
Initiative Key publications

The independent Expert Review Group was The 2014 progress report, Every woman, every child: a
ESTABLISHED 2011
established to provide global oversight on the results, post-2015 vision, provides recommendations for a renewed
resources and progress related to the United Nations commitment to strengthen women’s and children’s health,
Global Strategy and the Commission on Accountability. ➜ including a focus on accountability (27).
The UN Commission on Life-Saving Commodities The 2012 Commissioners’ Report identified and endorsed 13
ESTABLISHED 2012
for Women’s and Children’s Health took up a life-saving commodities, outlined key barriers that prevent
challenge by the Global Strategy for Women’s and their access and use, and recommended 10 time-bound
Children’s Health to increase access to and appropriate
use of life-saving commodities that effectively address
➜ actions to address them (28).
the leading preventable causes of death during

pregnancy, childbirth and childhood.
Committing to Child Survival: a Promise Yearly reports on child survival track progress and promote
ESTABLISHED 2012
Renewed is a global movement that brings together accountability for global commitments (29–31).
public, private and civil society actors to advocate for
action to end preventable child deaths by accelerating ➜
progress on maternal, newborn and child survival.
MDGs: Millennium Development Goals; RMNCH: reproductive, maternal, newborn and child health.
The post-2015 sustainable development agenda calls for a continued focus on the health-related MDGs
where there is still progress to be made. Furthermore, RMNCH remains an important priority for the health-
related aspects of sustainable development. RMNCH was chosen as the theme for this report, given its
relevance and global importance, and the existence of good quality, comparable data from a number of
low- and middle-income countries (1).
7
3
Monitoring the state of inequality
in RMNCH
An understanding of the state of inequality reveals gaps in population health and lends insight into how
policies, programmes and practices can be aligned to promote the ideal of health for all. This report presents
selected examples of the state of inequality in low- and middle-income countries, highlighting important
and relevant stories in RMNCH.
A complete assessment of the state of inequality should detail both the latest situation and change over
time. Together, data about the current and past state of inequality in a country indicate how a country has
progressed and is performing.
• Descriptions of the latest situation of inequality answer the question: according to the latest available
data, what is the status of inequality within a country?
• Investigations of change over time use the latest available data and comparable data from the past to
explore the question: how has the state of inequality changed?
First and foremost, reporting the state of inequality in RMNCH requires that relevant and comparable
data be available from reliable sources for a number of countries. Then, these data must be analysed
appropriately and presented in a meaningful way.
Comprehensive and transparent reporting provides the target audience with all the information necessary
to understand the strengths, limitations and assumptions of the data and analytical methods that underlie
the validity of the conclusions. In presenting this information, this report gives readers an appreciation of
the steps and complexities involved in conducting health inequality monitoring.
9
3.1 Data
Two types of data are required for measuring within-country health inequality: data about “health
indicators” that describe an individual’s experience of health (in terms of health intervention coverage and
health outcomes) and data about “dimensions of inequality” that allow populations to be organized into
subgroups according to their demographic, geographic and/or socioeconomic characteristics. Ideally, data
sources should provide health indicator and dimension of inequality data that are reliable, of high quality
and comparable across settings and over time.
3.1.1 Data sources

The health indicator and dimension of inequality data used in this report were sourced from Demographic
and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS). DHS and MICS are large-scale,
nationally representative household health surveys that are routinely conducted in low- and middle-income
countries. Standardized questionnaires are used to collect information through face-to-face interviews with
women aged 15–49 years. These surveys provide all the data required for health inequality monitoring –
data about multiple health indicators, as well as data that allow disaggregation of health data according
to a given dimension of inequality. By virtue of their design, DHS and MICS data have high comparability
between settings and over time.
Further details about the data used here are available in Appendix 1.
3.1.2 Health indicator data

Twenty-three health indicators were selected to explore the state of inequality in RMNCH in this report,
covering reproductive health interventions, maternal health interventions, newborn and child health
interventions, child malnutrition and child mortality. These health indicators represent an array of diverse
indicators within the topic, and were selected on the basis of data availability and relevancy, as evidenced
by their inclusion in previous RMNCH initiatives (1, 20). The selected indicators are listed in Table 3.1. A
detailed description of each indicator is available in Supplementary table S1 (appended to this report).
3.1.3 Dimension of inequality data

Data on economic status, education level, place of residence and sex are used to categorize populations
according to dimensions of inequality. These four dimensions of inequality represent common sources of
discrimination, and can be widely applied to populations in low- and middle-income countries.
Economic status is described in terms of a household wealth index, which accounts for ownership of
certain household items and access to specific services. On the basis of the wealth index, populations are
categorized into five subgroups, the quintile with the lowest scores representing the poorest members of
the population and the quintile with the highest, the richest (32). Education as a dimension of inequality
reflects the level of education attained by a woman (in the case of the reproductive and maternal health
interventions) or by a child’s mother (in the case of the newborn and child health indicators). Three
subgroups are specified: no education, primary school education and secondary school or higher education.
Place of residence (rural or urban) and child sex (female or male) each consist of two subgroups.
With the exception of the composite coverage index and the reproductive and maternal health indicators,
analyses of health inequalities may be conducted according to all four dimensions of inequality (Table 3.2).
10
3.1. DATA
TABLE 3.1 Selected RMNCH indicators used in this report
Category Indicator
Reproductive health interventions Contraceptive prevalence – modern and traditional methods (%)
Contraceptive prevalence – modern methods (%)
Demand for family planning satisfied (%)*
Maternal health interventions Antenatal care coverage – at least one visit (%)*
Antenatal care coverage – at least four visits (%)
Births attended by skilled health personnel (%)*
Newborn and child health interventions Early initiation of breastfeeding (%)
Children aged 6–59 months who received vitamin A supplementation (%)
BCG immunization coverage among one-year-olds (%)*
Measles immunization coverage among one-year-olds (%)*
Polio immunization coverage among one-year-olds (%)
DTP3 immunization coverage among one-year-olds (%)*
Full immunization coverage among one-year-olds (%)
Children aged less than five years with diarrhoea receiving oral rehydration salts (%)
Children aged less than five years with diarrhoea receiving oral rehydration therapy
and continued feeding (%)*
Children aged less than five years with pneumonia symptoms taken to a health
facility (%)*
RMNCH interventions, combined Composite coverage index (%)
Child malnutrition Stunting prevalence in children aged less than five years (%)
Underweight prevalence in children aged less than five years (%)
Wasting prevalence in children aged less than five years (%)
Child mortality Neonatal mortality rate (deaths per 1000 live births)
Infant mortality rate (deaths per 1000 live births)
Under-five mortality rate (deaths per 1000 live births)
BCG: one dose of Bacille Calmette-Guérin vaccine; DTP3: three doses of the combined diphtheria, tetanus toxoid and pertussis vaccine; RMNCH:
reproductive, maternal, newborn and child health.
* Indicator is part of the composite coverage index.
TABLE 3.2 Health indicator categories and relevant dimensions of inequality

Health indicator category Place of
(number of indicators) Economic status Education residence Sex
Reproductive health interventions (3) ✔ ✔ ✔
Maternal health interventions (3) ✔ ✔ ✔
Newborn and child health interventions (10) ✔ ✔ ✔ ✔
RMNCH interventions, combined (1) ✔ ✔ ✔
Child malnutrition (3) ✔ ✔ ✔ ✔
Child mortality (3) ✔ ✔ ✔ ✔
RMNCH: reproductive, maternal, newborn and child health.
11
VIDEO CLIP 1. HEALTH INEQUALITY IS MULTIDIMENSIONAL
Watch a short video clip to explore the question: how do AVAILABLE ON CD/USB SCAN HERE
subgroup variations in under-five mortality rates differ
between low-income countries and middle-income
state of InequalIty
reproductive, maternal,
I n t e r a c t I v e v I s u a l I z at I o n o f H e a lt H D ata
countries? World HealtH organization

20, avenue appia
CH-1211 geneva 27
SWitzerland
http://www.who.int/gho/health_equity/report_2015/
ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
ht is tr
wor e d
ld h or fre
e a lt h o r g a n iz at i o n , 2 0 1 5 . f
www.who.int/gho/health_equity/videos/en/
3.1.4 Country selection

Overall, this report covers 86 low- and middle-income countries that have comparable data from recent
surveys (either DHS or MICS) conducted between 2005 and 2013. These countries span all WHO regions.
For 42 of these 86 countries, survey data are also available for a previous time point (that is, between
1995 and 2004), allowing for an assessment of the change within countries over a period of about 10
years. A full list of study countries, with details of survey type and year is given in Supplementary table
S2 (appended to this report).
STUDY COUNTRIES INCLUDED IN THIS REPORT
Afghanistan, Albania, Armenia,* Azerbaijan, Bangladesh,* Belarus, Belize, Benin,* Bhutan, Bolivia (Plurinational State of),*
Bosnia and Herzegovina, Burkina Faso,* Burundi, Cambodia,* Cameroon,* Central African Republic, Colombia,* Comoros,
Congo, Costa Rica, Côte d’Ivoire,* Cuba, Democratic Republic of the Congo, Djibouti, Dominican Republic,* Egypt,* Ethiopia,*
Gabon,* Gambia, Georgia, Ghana,* Guinea,* Guinea-Bissau, Guyana, Haiti,* Honduras, India,* Indonesia,* Iraq, Jamaica,
Jordan,* Kazakhstan,* Kenya,* Kyrgyzstan,* Lao People’s Democratic Republic, Lesotho,* Liberia, Madagascar,* Malawi,*
Maldives, Mali,* Mauritania, Mongolia, Montenegro, Mozambique,* Namibia,* Nepal,* Niger,* Nigeria,* Pakistan, Peru,*
Philippines,* Republic of Moldova, Rwanda,* Sao Tome and Principe, Senegal,* Serbia, Sierra Leone, Somalia, Suriname,
Swaziland, Syrian Arab Republic, Tajikistan, Thailand, The former Yugoslav Republic of Macedonia, Timor-Leste, Togo,*
Uganda,* Ukraine, United Republic of Tanzania,* Uzbekistan,* Vanuatu, Viet Nam,* Yemen, Zambia,* Zimbabwe*
*Denotes the 42 countries included in analyses of change over time.
12
3.2. ANALYSIS
3.2 Analysis
The starting point for analysing health inequalities is data disaggregation. Disaggregated data show the
level of health in each subgroup of a given dimension of inequality across each country, and can be used
to explore the latest status of health inequality or change in inequality over time. In preparing this report,
analyses of disaggregated data were done to capture the latest situation of inequality in RMNCH indicators,
and then to determine how inequalities have changed over time.
For convenience and ease of understanding, health inequalities may be quantified in terms of summary
measures; such measures build on disaggregated data, combining estimates of a given health indicator
for two or more subgroups into a single numerical figure. Summary measures of inequality may be applied
to assess the latest situation or change over time.
More details about the analysis methods used in this report are available in Appendix 1, and other
methodological considerations are discussed more fully in Appendix 2.
3.2.1 Data disaggregation

Drawing from household health surveys, disaggregated data for each of the health indicators can be
generated. For example, data about stunting prevalence in children aged less than five years, an indicator of
child malnutrition, may be broken down by the mother’s level of education. Assessing disaggregated data
helps to answer questions such as: how did stunting vary between the education subgroups in a particular
country? How do levels of stunting vary across countries for a given education subgroup?
Looking at disaggregated data over time helps to assess progress within subgroups and to compare
progress between subgroups. Again, considering child stunting prevalence by mother’s education level,
an analysis of disaggregated data from two or more time points helps to address questions such as: for a
given country, did stunting prevalence increase or decrease in the least educated? The most educated?
Were decreases in child stunting prevalence faster in the no education subgroup than in the secondary
school or higher subgroup?
3.2.2 Summary measures

Summary measures are used to represent the degree of inequality in a health indicator across population
subgroups in a single numerical figure. A number of summary measures are used throughout this report,
namely difference, ratio, population attributable risk and absolute excess change (Table 3.3).
Difference and ratio are simple measures of inequality, expressing inequalities between two population
subgroups. For example, difference and ratio can be applied to show inequality in under-five mortality
rates by place of residence. For a given country, difference can be quantified by subtracting the urban
from the rural rate; this provides an estimate of the absolute place-of-residence inequality. Relative place-
of-residence inequality may be quantified by calculating the ratio of the urban to rural rates. Note that
for dimensions of inequality that have more than two subgroups, such as economic status or education,
difference and ratio measures often make comparisons between the most-advantaged and the most-
disadvantaged subgroups.
13
TABLE 3.3 The calculation of selected summary measures and their application in this report
Summary measure Calculation Application

Difference The level of health intervention All dimensions of inequality: economic
(absolute inequality)* coverage in the most-disadvantaged status, education, place of residence, sex
subgroup is subtracted from health All health indicators
intervention coverage in the most-
advantaged subgroup (or vice versa for
child malnutrition and child mortality
indicators).
Ratio The level of health intervention All dimensions of inequality: economic
(relative inequality)* coverage in the most-advantaged status, education, place of residence, sex
subgroup is divided by health All health indicators
intervention coverage in the most-
disadvantaged subgroup (or vice
versa for child malnutrition and child
mortality indicators).
Population attributable risk (potential for The calculation shows the possible Economic status, education and place of
improvement in national coverage) improvement in coverage by residence
eliminating within-country inequality Health intervention indicators
related to economic status, education or
place of residence.
The national level of coverage is
subtracted from the level of coverage in
the most-advantaged subgroup.
Absolute excess change** The absolute change in the most- All dimensions of inequality: economic
advantaged subgroup is subtracted status, education, place of residence, sex
from the absolute change in the most- All health indicators
disadvantaged subgroup.
* Further details about these calculations are available in Appendix table A1.2.
** Information about how to interpret excess change is provided in Appendix 2.
Reference subgroups for difference and ratio were selected based on convenience of data interpretation (that is, providing positive values for difference
calculations and values above one for ratio calculations). For example, the poorest/no education/rural/males subgroups tended to have higher child
mortality or higher prevalence of child malnutrition than the richest/secondary school or higher/urban/females subgroups, respectively. In the case of
sex, this selection does not represent an assumed advantage of one sex over the other.
ABSOLUTE VERSUS RELATIVE MEASURES OF INEQUALITY
Absolute measures of inequality, such as difference, reflect the magnitude of the difference in health status between two
subgroups. For example, DTP3 immunization coverage of 100% in one subgroup and 50% in another subgroup would
mean that there is an absolute difference in coverage of 50 percentage points.
Relative measures of inequality, such as ratio, are calculated as the quotient between two subgroups. For example, DTP3
immunization coverage of 100% in one subgroup and 50% in another subgroup would generate a ratio of 2, implying
that coverage in one group is twice that in the other.
14
3.2. ANALYSIS
The population attributable risk takes into account the situation in all subgroups (the whole population). For
health intervention indicators, this measure is based on the principle that each subgroup has the potential
to achieve the same coverage as the most-advantaged subgroup. The measure can be interpreted as the
potential for improvement in the national average of health intervention coverage that could be achieved
by eliminating within-country inequality related to economic status, education or place of residence. In
this report, population attributable risk is applied to assess the latest situation of inequality.
The absolute excess change summary measure can be used to answer questions such as: how much faster
(or slower) was the change in the health indicator in the most-disadvantaged group compared with the
most-advantaged group? This summary measure is applied to compare the change in a health indicator over
time between two subgroups, building on the absolute change experienced by each of the two subgroups.
Absolute change shows how coverage in a single subgroup has changed over time. For example, the annual
absolute change in contraceptive use among women in a given education subgroup can be calculated as
the prevalence in the most recent survey minus the prevalence in an older survey, divided by the number
of years between the two surveys. In this example, the annual absolute change is expressed in units of
percentage points per year. From here, annual absolute excess change is calculated as the annual absolute
change in the no education subgroup (the most disadvantaged) minus the annual absolute change in the
secondary school or higher subgroup (the most advantaged). A positive excess change value indicates that
the pace of change in contraceptive use over time was more favourable among women with no education
compared with women who had attended secondary school or higher. In most cases, this means that
contraceptive use increased faster in the disadvantaged subgroup (women with no education) than in the
advantaged subgroup (women with secondary school or higher), which is a desired situation that indicates
narrowing inequality. Other possible scenarios are explored in Appendix 2.
MEDIAN VALUES
When assessing the extent of inequality across multiple study countries, the median value (middle point) of disaggregated
data (or summary measures) is a useful statistic for summarizing the situation across a set of countries. For instance, the
median value of multiple country estimates of stunting within the no education subgroup could be compared with the
median value of multiple country estimates within the primary school subgroup and the secondary school or higher
subgroup. This type of analysis helps to answer questions such as: did median stunting prevalence vary between education
levels across countries? Was the median prevalence of stunting in the no education subgroup lower than the median
prevalence in the secondary school or higher subgroup?
15
3.3 Reporting
Reporting on health inequality provides information to a diverse group of stakeholders – policy-makers,
technical staff in ministries of health, health professionals (including public health professionals), academics
and others – and informs the development of equity-oriented policies, programmes and practices. Inequality
data should be reported in a manner that considers not only what content is most relevant to the target
audience, but also how that content can be effectively presented to that audience.
Considerations for assessing and reporting health inequality data are discussed in Appendix 2.
3.3.1 Data visualization

One of the primary aims of this report is to showcase best practices in data visualization. The effective use
of visual displays of health data helps to add meaning to otherwise cumbersome and complex datasets.
Visual displays can communicate large amounts of information efficiently by accelerating and deepening
understanding. Three common tools used to present data are tables, graphs and maps. Tables present
data comprehensively, and are appropriate for situations where the audience requires precise information,
such as exact numerical figures. Graphs highlight key messages, and may draw upon attributes such as
colour, shading, shapes, lines, sizing and/or patterns of arrangement to deliver those messages. Maps
are useful for presenting data with a geographical component, and can be effective for the visualization
of disaggregated data. Appendix 3 provides an example of maps displaying stunting prevalence among
children, disaggregated according to the mother’s level of education.
Throughout the report, both conventional and interactive data visuals have been employed to enhance the
communication of the state of inequality in RMNCH. Conventional data visualization refers to the static
visuals that appear throughout the text and appendices of this report, including tables, maps and graphs.
Interactive data visualization features include story-points, as well as interactive country profiles, maps
and reference tables. Tables, graphs and maps can be prepared as static (conventional) data visuals or as
interactive features.
A guide to interpreting the primary types of visuals that are used in this report is provided in Appendix 4.
INTERACTIVE DATA VISUALIZATION TERMINOLOGY
➜ A dashboard in an interactive visual is the term to describe the view that consolidates and presents multiple types
of related data and information on a single screen. Often the components of a single dashboard are interconnected,
and users may filter or highlight multiple components of the dashboard by clicking on one selection.
➜ Story-point is a feature that links several dashboards together in sequence, allowing users to be guided through data
views and information in a story-like arrangement.
For information about techniques employed in interactive data visualization, please refer to Appendix 5.
16
3.3. REPORTING
3.3.2 Feature stories

Chapter 4 of this report contains a series of feature stories that describe current patterns and trends in
the state of inequality in RMNCH. The stories highlight the latest situation and change over time with
respect to selected RMNCH indicators and dimensions of inequality, illustrating how health inequality data
can be interpreted and reported for a particular topic. Alongside each of the feature stories, conventional
visuals (figures) illustrate key findings; additionally, interactive visuals can be accessed that expand upon
the information presented in the feature stories, and allow further exploration RMNCH indicator data.
Thus, readers can explore the underlying data, customizing the dashboards and engaging in benchmarking
according to their interests. The feature stories presented here represent only a small sample of stories
that can be told about the state of inequality in RMNCH.
BENCHMARKING
Benchmarking considers one country’s status in a multinational context, drawing comparisons with other countries of the
same geographical region and/or economic situation. Benchmarking helps to answer the questions: how does the latest
situation of inequality in one country compare among a group of countries with similar characteristics? Did inequality in
one country increase or decrease in the same direction as other countries? And was the pace faster or slower?
VIDEO CLIP 2. BENCHMARKING PUTS INEQUALITY IN CONTEXT
Watch a short video clip to explore the question: which AVAILABLE ON CD/USB SCAN HERE
countries increased modern contraceptive use at the
national level and also decreased the place-of-residence
state of InequalIty
inequality in this health intervention? World HealtH organization

20, avenue appia
CH-1211 geneva 27
SWitzerland
ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
ht is tr
wor e d
ld h or fre
ACCESSING INTERACTIVE VISUALS
In Chapter 4, each feature story is linked to an interactive visual; these contain a large bank of data specific to the indicators
of the featured RMNCH subtopic. Links to additional interactive components, including country profiles, maps, reference
tables and a comprehensive interactive visual of all RMNCH interventions, are available in Appendix 6.
Links to interactive visuals can be found in the green boxes throughout Chapters 4 and 5 and Appendix 6. All interactive
visuals are contained on the CD or USB that accompanies this report, and can be viewed offline. Interactive visuals can be
accessed online by scanning the QR code or following the URL provided.
17
4
The state of inequality in RMNCH:
stories from low- and middle-
income countries
Inequality in RMNCH is pervasive. Women, mothers, infants and children from low- and middle-income
countries have a wide range of health experiences that are often correlated with underlying demographic,
geographic and socioeconomic characteristics.
Inequalities were present in most RMNCH indicators across countries (as evidenced by the wide range
in national estimates for various health indicators) and within countries (as evidenced by the range in
subgroup estimates for various health indicators). There was also evidence of large variations in the level
of within-country inequalities across countries.
Certain aspects of the state of inequality in RMNCH, however, are more optimistic. Change-over-time
analyses demonstrated that improvements were frequently realized both nationally and in population
subgroups. A comparison of the pace of change in disadvantaged subgroups versus advantaged subgroups
often revealed faster improvements among the most disadvantaged. This suggests that, although
inequalities in RMNCH still persist, they have narrowed over time.
19
4.1 Reproductive health interventions

Voluntary and effective use of contraception enables women to better make family planning decisions
and helps to reduce maternal and infant deaths. Other benefits of contraceptive use include gains in
women’s empowerment and equality, and increased opportunities for women to participate in educational,
economic and development activities that benefit society and themselves (33). In 2012, contraceptive use
was estimated to have prevented 218 million unintended pregnancies in developing countries; additionally,
contraceptive use averted 55 million unplanned births, 138 million abortions, 25 million miscarriages and
118 000 maternal deaths (34).
Acknowledging that the unmet need for contraception is highest among the marginalized and the
disadvantaged, reproductive health initiatives must appropriately support gender equality and prioritize
the poor and other disadvantaged subgroups (35).
FEATURE STORY
EDUCATION-RELATED INEQUALITY IN MODERN CONTRACEPTIVE USE
MODERN CONTRACEPTIVE USE

84 low- and middle-income study countries, DHS and MICS 2005–2013
In half of study countries, at least one in three women reported using modern methods of contraception;
nearly one quarter of study countries reported levels of modern contraceptive use of 50% or higher. Use
was much lower (10% or less) in one tenth of study countries. Modern contraceptive use across study
countries ranged from 1.2% in Somalia to 75.1% in Thailand.
LATEST SITUATION: BY EDUCATION LEVEL

The use of modern methods of contraception tended to be lowest in women with no education, and
generally increased across education levels. This pattern was evident in many of the study countries. The
median values of contraceptive prevalence for each of the three education subgroups are compared in
Figure 4.1, which illustrates greater prevalence of modern contraceptive use in subgroups with higher levels
of education. Ethiopia is highlighted as one example of a country that demonstrated a gradient pattern of
contraceptive use across education subgroups.
The gap in modern contraceptive use between the most- and least-educated women was substantial in
many study countries. For example, in half of study countries, the prevalence of modern contraceptive
use in the secondary school or higher subgroup exceeded the prevalence in the no education subgroup by
over 14 percentage points. In the Central African Republic, Guinea-Bissau and Nigeria, contraceptive use
among women in the highest education category was at least 10 times greater than that among women
with no education.
20
4.1. REPRODUCTIVE HEALTH INTERVENTIONS
FIGURE 4.1 Contraceptive prevalence (modern methods) by woman’s education in 71 low- and middle-income countries: latest
situation (DHS and MICS 2005–2013)
No education Primary school Secondary school +
100
90
80
70
60
Coverage (%)
50
40
35.3
29.9
30
20 18.9
10
Circles indicate countries – each study country is represented on the graph by three circles. Horizontal lines indicate the median value (middle point)
for each subgroup. Light grey bands indicate the interquartile range (middle 50% of study country estimates). Highlighted country: Ethiopia.
Large education-related inequalities in modern contraceptive use were not reported by all countries. Some
countries reported relatively low levels of education-related inequality: Cambodia and the Dominican
Republic, for example, had a difference of less than 1 percentage point between contraceptive use in
the secondary school or higher subgroup and the no education subgroup. Bhutan, Maldives, Nepal and
Viet Nam demonstrated higher usage among women with no education, with contraceptive use at least
7 percentage points higher in the no education subgroup than in the secondary school or higher subgroup.
The magnitude of inequality in modern contraceptive use did not appear to be associated with the national
level of usage. For example, the overall prevalence of modern contraceptive use averaged around 35% in
both the Plurinational State of Bolivia and Cambodia; whereas disaggregated data for Cambodia showed
little education-related inequality, the Plurinational State of Bolivia reported a marked education-related
difference in contraceptive use (usage was 21.0 percentage points greater among women with secondary
schooling or higher than women with no education).
21
CHANGE OVER TIME: BY EDUCATION LEVEL

38 low- and middle-income study countries, DHS and MICS 1995–2004 and 2005–2013
The majority of low- and middle-income study countries included in change-over-time analyses reported
national increases in modern contraceptive use between the two survey periods. Rwanda reported an
especially elevated increase in the national prevalence of modern contraceptive use of 4.1 percentage
points per year.
In most countries, the pace of increase in contraceptive use tended to be faster – or otherwise favourable
– among women with no education than in women with secondary schooling or higher. In Zambia, for
instance, the annual absolute increase was faster in the no education subgroup (1.9 percentage points per
year) than the secondary school or higher subgroup (1.3 percentage points per year), resulting in a positive
annual absolute excess change of 0.6 percentage points (Figure 4.2: highlighted circles).
The median absolute increase in contraceptive use across all study countries was 0.7 percentage points per
year in the no education subgroup, and 0.2 percentage points per year in the secondary school or higher
subgroup; the median absolute excess change was positive, reflecting a tendency for faster increases in
contraceptive use among women in the no education subgroup (Figure 4.2).
FIGURE 4.2 Contraceptive prevalence (modern methods) by woman’s education in 38 low- and middle-income countries: change
over time (DHS and MICS 1995–2004 and 2005–2013)
Change over time Excess change over time
Comparing change in no education
No education Secondary school + with change in secondary school +
4 4
Annual absolute excess change (percentage points)
3 3
Annual absolute change (percentage points)
2 1.9 2
1.3
1 1
0.6
0 0
-1 -1
-2 -2
Circles indicate countries – each study country is represented by two circles in the change over time graph on the left (one for no education and one
for secondary school or higher), and one circle on the excess change over time graph on the right. Horizontal lines indicate the median value (middle
point) for each subgroup. Light grey bands indicate the interquartile range (middle 50% of study country estimates). Highlighted country: Zambia.
22
4.1. REPRODUCTIVE HEALTH INTERVENTIONS
OTHER KEY FINDINGS: INEQUALITY IN REPRODUCTIVE HEALTH INTERVENTIONS

• In over half of study countries, the demand for family planning satisfied was at least 17 percentage points
higher among women with secondary school or higher education than in women with no education.
• Nearly one third of study countries reported considerable economic-related inequality in the use of
modern and traditional contraception, with prevalence values at least twice as high in the richest
compared with the poorest quintile of households.
• Over a 10-year period, most study countries demonstrated increasing national levels of demand for
family planning satisfied. In nearly half of study countries, increased national prevalence was realized
alongside faster improvements in the poorest subgroups compared with the richest subgroups.
INTERACTIVE VISUAL 1. REPRODUCTIVE HEALTH INTERVENTIONS
Electronic visualization components accompany this AVAILABLE ON CD/USB SCAN HERE

report to enable independent data exploration and
benchmarking.
state of InequalIty
This story-point interactive visual guides you through

World HealtH organization
20, avenue appia
CH-1211 geneva 27
SWitzerland
the state of inequality in this feature story and other

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
reproductive health intervention indicators.

ht is tr
wor e d
ld h or fre
23
4.2 Maternal health interventions

Each year hundreds of thousands of women die for reasons related to pregnancy and child birth (36). Given
that the risk of maternal death is highest immediately postpartum and in the following 24–48 hours, the
presence of skilled health personnel during childbirth is a key intervention for preventing maternal and
newborn deaths. In 2012, about 40 million births in developing regions were not attended by skilled health
personnel (15).
Inequalities in maternal health have been widely acknowledged, both across countries (15, 37, 38) and
within countries (1, 39). The proportion of births attended by skilled health personnel has been identified
as the maternal health intervention indicator with the most pronounced economic-related inequality (39).
Thus, improving on the coverage of this health intervention is a priority for initiatives that aim to promote
maternal health.
FEATURE STORY
ECONOMIC-RELATED INEQUALITY IN BIRTHS ATTENDED BY SKILLED HEALTH

PERSONNEL
BIRTHS ATTENDED BY SKILLED HEALTH PERSONNEL

Almost half of low- and middle-income study countries reported that at least 80% of live births were
attended by skilled health personnel; however, there was wide variation in the proportion of attended births
across countries. Generally, the proportion of births attended by skilled health personnel was much lower
across low-income countries than in middle-income countries. Study countries in the WHO European
Region demonstrated little cross-country inequality in the proportion of births attended by skilled health
personnel, with complete – or very high – levels of coverage for this intervention across all countries. The
proportion of births attended by skilled health personnel in study countries of other regions, however,
ranged from less than 12% to nearly 100%.
LATEST SITUATION: BY ECONOMIC STATUS

Overall, the proportion of births attended by skilled health personnel increased with rising economic status:
poorer subgroups typically experienced lower levels of skilled birth attendance than richer subgroups. This
relationship was more pronounced in low-income study countries than in middle-income study countries.
In low-income countries, the median coverage for this intervention ranged from 33.7% in the poorest to
89.0% in the richest quintile, whereas in the middle-income countries it ranged from 76.5% in the poorest
to 98.1% in the richest quintile (Figure 4.3). Most study countries (90%) reported coverage of over 80%
in the richest quintile, whereas only 30% of study countries reported this level of coverage in the poorest
quintile.
24
4.2. MATERNAL HEALTH INTERVENTIONS
The variation in the proportion of births attended by skilled personnel was larger among the poorer quintiles
of countries. Taking the group of middle-income study countries as an example, the interquartile range
(middle 50% of study country estimates) was 58.9 percentage points in the poorest quintile of households,
26.3 percentage points in the middle quintile and 4.7 percentage points in the richest quintile (Figure 4.3).
FIGURE 4.3 Births attended by skilled health personnel by economic status in 30 low-income and 53 middle-income countries:
latest situation (DHS and MICS 2005–2013)
Low-income study countries Middle-income study countries
Quintile 1 Quintile 5 Quintile 1 Quintile 5
Quintile 2 Quintile 3 Quintile 4 Quintile 2 Quintile 3 Quintile 4
(poorest) (richest) (poorest) (richest)
100 100
90 90
80 80
70 70
60 60
Coverage (%)
Coverage (%)
50 50
40 40
30 30
20 20
10 10
0 0
Within each subgroup, the top and bottom lines indicate maximum and minimum values, the centre line indicates the median value (middle point)
and the light grey box indicates the interquartile range (middle 50% of study country estimates).
Figure 4.4 plots economic-related inequality in skilled birth attendance (calculated as the difference in
coverage between the richest and poorest quintiles, in percentage points, for each country) alongside
national coverage. A clustering pattern of countries by WHO region can be observed. For example, most of
the study countries from the WHO European Region demonstrated low or no economic-related inequality
in skilled birth attendance alongside high or complete national coverage (Figure 4.4: grey triangles).
Conversely, many countries from the WHO African Region reported high levels of economic-related
inequality while national levels of skilled birth attendance spanned a wide range (Figure 4.4: brown circles).
25
FIGURE 4.4 Births attended by skilled health personnel in 83 low- and middle-income countries (national average and within-
country economic-related inequality): latest situation (DHS and MICS 2005–2013)
80
Difference between the richest and the poorest quintile (percentage points)
70
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
National average (%)
 African Region;  Region of the Americas;  South-East Asia Region;  European Region;  Eastern Mediterranean Region;  Western Pacific Region
Coloured shapes indicate countries – each study country is represented on the graph by one coloured shape. Each type of shape represents one WHO
region. Dashed orange lines indicate the median values (middle points).
CHANGE OVER TIME: BY ECONOMIC STATUS

The national proportion of births attended by skilled health personnel increased over the 10-year period
between surveys in the majority of study countries (Figure 4.5: all countries to the right of the vertical grey
zero line). Half of study countries reported an absolute increase in national coverage for this intervention
of at least 1 percentage point per year, which translates into a 10 (or more) percentage point increase over
10 years (Figure 4.5: all countries to the right of the vertical orange median line).
Study countries reported variation in the pace of change between the poorest and richest subgroups in the
proportion of births attended by skilled health personnel. In more than half of study countries, the annual
absolute excess change was positive, reflecting a pro-poor trend of increasing coverage favouring the most
disadvantaged (Figure 4.5: countries above the horizontal grey zero line).
Considering the pace of change in subgroups alongside change in national averages, about half of study
countries reported a desirable situation: improved national average with increases in the poorest quintile
outpacing the change in the richest quintile (Figure 4.5: countries in the top right quadrant – above and to the
right of the grey zero lines). Several countries reported no change in national coverage and/or in economic-
related inequality. For example, in Jordan and Kazakhstan there was no change in either national coverage
or in economic-related inequality because these countries reported complete coverage at both time points.
26
4.2. MATERNAL HEALTH INTERVENTIONS
FIGURE 4.5 Births attended by skilled health personnel: change over time in national average (absolute change) and in the poorest
compared with the richest quintile (absolute excess change) in 42 low- and middle-income countries (DHS and MICS 1995–2004 and
2005–2013)
Decrease in national average Increase in national average

4 Pro-poor change Pro-poor change
Annual absolute change in the poorest compared with the richest quintile
1
(percentage points)
Jordan
Kazakhstan
-1
-2
-3
-4 Decrease in national average Increase in national average

Pro-rich change Pro-rich change
-3 -2 -1 0 1 2 3 4 5
Annual absolute change in national average (percentage points)
OTHER KEY FINDINGS: INEQUALITY IN MATERNAL HEALTH INTERVENTIONS

• For all three dimensions of inequality (economic status, education and place of residence), the lowest
levels of inequalities were reported for antenatal care coverage (at least one visit), followed by antenatal
care coverage (at least four visits), and then births attended by skilled health personnel.
• Half of study countries reported the prevalence of births attended by skilled health personnel to be at
least 20 percentage points higher in urban than in rural areas.
• One quarter of study countries reported that antenatal care coverage (at least four visits) was at least
twice as high in women with secondary schooling or higher than in women with no education.
• In most study countries, maternal health interventions demonstrated faster improvements – or more
favourable changes – in the most-disadvantaged subgroups (the poorest, the least educated and rural
residents) over a 10-year period.
27
INTERACTIVE VISUAL 2. MATERNAL HEALTH INTERVENTIONS

benchmarking.
state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
maternal health intervention indicators.

ht is
wor e d
ld h or fre
28
4.3. CARE-SEEKING FOR SICK CHILDREN
4.3 Care-seeking for sick children

Many of the deaths in children under the age of five years could be averted by early, low-cost, appropriate
interventions in the home or community. Unfortunately, many countries fall short in making these early
interventions widely available, and large numbers of sick children do not have contact with health facilities.
Pneumonia is a major cause of death in children under five years of age, and the leading infectious cause
of death in children under five years (40). It accounts for 15% of child deaths, and resulted in an estimated
935 000 deaths in children under five years in 2013 (41). Treatment for pneumonia with antibiotics is well
established, safe and effective, yet many children who need these drugs fail to receive them.
Initiatives to end preventable childhood deaths due to pneumonia and diarrhoea build on the principles
of protecting children by promoting good health practices from birth, preventing ill health and intervening
early with appropriate treatments.
FEATURE STORY
PLACE-OF-RESIDENCE INEQUALITY IN CARE-SEEKING FOR CHILDREN WITH

PNEUMONIA SYMPTOMS
CHILDREN AGED LESS THAN FIVE YEARS WITH PNEUMONIA SYMPTOMS TAKEN TO A
HEALTH FACILITY
In about 1 in 10 countries, at least 80% of children under five years of age with pneumonia symptoms was
taken to a health facility. However, nearly one third of study countries had low rates of care-seeking, around
50% or less. The national prevalence of care-seeking for children with pneumonia symptoms ranged from
13.0% in Somalia to 96.5% in Cuba.
LATEST SITUATION: BY PLACE OF RESIDENCE

When disaggregated by place of residence, the prevalence of care-seeking for children with pneumonia
symptoms differed greatly between rural and urban areas. Three quarters of low- and middle-income study
countries reported a higher level of care-seeking in urban than in rural areas. While 14% of study countries
reported the prevalence of care-seeking to be at least 80% in urban areas, only 6% of study countries
achieved this level of coverage in rural areas.
The variation in care-seeking behaviour across study countries was less extreme among urban than among
rural residents, as indicated by the smaller size of the interquartile range (middle 50% of study country
estimates). The interquartile range for urban areas was 19.8 percentage points, whereas for rural areas,
the interquartile range was 30.7 percentage points (Figure 4.6: light grey boxes).
29
FIGURE 4.6 Children aged less than five years with pneumonia symptoms taken to a health facility by place of residence in 72 low-
and middle-income countries: latest situation (DHS and MICS 2005–2013)
Rural Urban
100
90
80
70
60
Coverage (%)
50
40
30
20
10
Almost one in five study countries reported the prevalence of care-seeking for children with pneumonia
symptoms to be at least 20 percentage points higher in urban than in rural areas. The largest absolute
inequality was reported by Guinea, where the level of care-seeking was 40.1 percentage points higher in
the urban than in the rural subgroup.
Pro-urban inequality, however, was not always the case. In a minority of study countries, levels of care-
seeking were higher in rural than in urban areas (pro-rural). Countries such as Bosnia and Herzegovina,
Cuba, Serbia and Thailand reported an optimal overall situation, with low place-of-residence absolute
inequality (5 percentage points or less) and national coverage exceeding 80%.
CHANGE OVER TIME: BY PLACE OF RESIDENCE

Many countries reported improved national levels of care-seeking for children with pneumonia symptoms,
with increases in both rural and urban areas. Overall, half of study countries saw care-seeking increase
by at least 10 percentage points in rural areas over 10 years. Of all study countries, the fastest rural
improvement was a 46.0 percentage point increase over 10 years, achieved in Malawi. Across urban areas,
the median improvement in the prevalence of care-seeking for children with pneumonia symptoms was
about 6 percentage points over 10 years. The most pronounced urban gain was reported in Rwanda, with
an increase of 51.3 percentage points over 10 years.
30
4.3. CARE-SEEKING FOR SICK CHILDREN
FIGURE 4.7 Children aged less than five years with pneumonia symptoms taken to a health facility: change over time in national
average (absolute change) and in rural compared with urban areas (absolute excess change) in 33 low- and middle-income countries
(DHS and MICS 1995–2004 and 2005–2013)

Annual absolute change in rural compared with urban areas (percentage points)
Pro-rural change Pro-rural change

3 Malawi
Burkina Faso
Cambodia
0
Nepal Namibia
-1
Rwanda
-2
-3
Pro-urban change Pro-urban change
-3 -2 -1 0 1 2 3 4 5
Comparing the pace of change in care-seeking for children with pneumonia symptoms in rural and urban
areas, two thirds of study countries indicated a pro-rural situation (that is, a faster increase or slower decrease
in rural areas compared with urban areas) (Figure 4.7: countries above the horizontal grey zero line). Burkina
Faso and Malawi, for example, reported considerable national-level gains, driven by faster progress in rural
than in urban areas. The optimal situation – an increasing national average coupled with a pro-rural change
– was achieved by about half of study countries (Figure 4.7: countries in the top right quadrant – above and
to the right of the grey zero lines).
Cambodia reported an increase in national average with an equal pace of improvement in rural and urban
areas.
A minority of study countries reported changes that were pro-urban (Figure 4.7: countries below the
horizontal grey zero line). In Rwanda, a considerable increase in the national prevalence of care-seeking
was reported; however, this was achieved through faster gains in urban areas than in rural areas by a margin
of 20.6 percentage points over 10 years. Namibia and Nepal also reported large improvements in national
averages driven by pro-urban changes of around 5 percentage points over the 10-year period.
31
OTHER KEY FINDINGS: INEQUALITY IN CARE-SEEKING FOR SICK CHILDREN

• In half of study countries, there was at least a 20 percentage point gap in care-seeking for children with
pneumonia symptoms between the poorest and richest subgroups.
• In over one third of study countries, the difference between rural and urban areas in the prevalence of
treating diarrhoea with oral rehydration therapy and continued feeding was low, at less than 5 percentage
points; the maximum place-of-residence difference, however, approached 40 percentage points.
• Over a 10-year period, the use of oral rehydration salts to treat diarrhoea in children under the age of
five years increased in two thirds of study countries; in the majority of these countries with increasing
national figures, this change was pro-rural (that is, the proportion of children with diarrhoea who were
treated increased more rapidly in rural than in urban areas).
INTERACTIVE VISUAL 3. CARE-SEEKING FOR SICK CHILDREN

benchmarking.
state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
care-seeking for sick children indicators.

ht is
wor e d
ld h or fre
32
4.4. CHILDHOOD IMMUNIZATION
4.4 Childhood immunization

The coverage of childhood immunization has risen substantially over the past decades, which has
contributed significantly to falling child mortality rates. However, diseases that are preventable by vaccines
currently recommended by WHO still claim the lives of 1.5 million children every year. Approximately 17%
of deaths in children under five years (0–59 months) and 29% of deaths in children aged 1–59 months
are vaccine preventable (42).
In 2013, 84% of infants worldwide received DTP3 immunization, an increase from 76% of infants in 1990.
Similar gains have been seen in polio and measles immunization, which now reach at least 80% of infants
in the majority of countries worldwide. Despite the recent progress in immunization coverage, in 2013, 21.8
million children under the age of one year did not receive the DTP3 vaccine; most of these unprotected
children (70%) live in just 10 underserved countries (42).
Improving immunization coverage is widely recognized as a successful and cost-effective intervention to

improve child health and reduce child mortality.
FEATURE STORY
ECONOMIC-RELATED INEQUALITY IN DTP3 IMMUNIZATION COVERAGE
DTP3 IMMUNIZATION COVERAGE AMONG ONE-YEAR-OLDS

Overall, half of low- and middle-income study countries reported DTP3 immunization coverage of about
80% or higher among one-year-olds. DTP3 immunization coverage was above 90% in more than one
quarter of study countries.
LATEST SITUATION: BY ECONOMIC STATUS

Overall, DTP3 immunization coverage was lower in children from poorer households. Across study
countries, DTP3 immunization coverage tended to increase with rising economic status. Half of study
countries achieved coverage of at least 73% in the poorest quintile, whereas for the richest quintile, half
of study countries reported coverage of over 86% (Figure 4.8).
Study countries were more likely to report DTP3 immunization coverage of over 80% in richer than in
poorer quintiles: 73% of countries achieved this level of coverage among the richest quintile, while only
38% of countries achieved this level of coverage among the poorest quintile.
The degree of within-country economic-related inequality varied from country to country. In more than
one quarter of study countries, there was no – or very little – difference in DTP3 immunization coverage
between the richest and poorest quintiles (less than 5 percentage points). On the other hand, pronounced
levels of inequality (absolute differences of at least 25 percentage points between the richest and poorest
quintiles) were apparent in one quarter of study countries.
33
FIGURE 4.8 DTP3 immunization coverage among one-year-olds by economic status in 78 low- and middle-income countries: latest
Quintile 1 (poorest) Quintile 2 Quintile 3 Quintile 4 Quintile 5 (richest)
100
90
85.9
82.5
80 79.1
76.3
72.5
70
60
Coverage (%)
50
40
30
20
10
Circles indicate countries – each study country is represented on the graph by five circles. Horizontal lines indicate the median value (middle point) for
each subgroup. Light grey bands indicate the interquartile range (middle 50% of study country estimates).
CHANGE OVER TIME: BY ECONOMIC STATUS

In general, DTP3 immunization coverage increased with increasing economic status in most low- and
middle-income study countries. In most study countries, coverage increased substantially faster in the
poorest than in the richest quintile. The increase in DTP3 immunization coverage in the poorest quintile
exceeded that in the richest quintile by a margin of at least 0.9 percentage points or more per year over
the period between surveys in half of study countries (Figure 4.9: countries above the horizontal orange
median line).
Overall, national levels of DTP3 immunization coverage showed a median increase of 0.7 percentage points
per year (Figure 4.9: the vertical orange median line). In the majority of countries, an increase in national
average was reported alongside an excess change that favoured the poorest over the richest quintiles
(Figure 4.9: countries in the top right quadrant – above and to the right of the grey zero lines).
34
4.4. CHILDHOOD IMMUNIZATION
FIGURE 4.9 DTP3 immunization coverage among one-year-olds: change over time in national average (absolute change) and in
the poorest compared with the richest quintile (absolute excess change) in 41 low- and middle-income countries (DHS and MICS
1995–2004 and 2005–2013)
Pro-poor change Pro-poor change
Annual absolute change in the poorest compared with the richest quintile
Plurinational State of Bolivia
Gabon Burkina Faso

Niger
2 Cambodia
(percentage points)
-1
Namibia
-2
Cameroon
-3 Decrease in national average Increase in national average

Pro-rich change Pro-rich change
-3 -2 -1 0 1 2 3 4
The Plurinational State of Bolivia, Burkina Faso, Cambodia, Gabon and Niger are among the countries that
reported substantial increases in national DTP3 immunization coverage coupled with pro-poor changes
in economic-related inequality (a faster pace of improvement in the poorest than in the richest quintile).
Other countries, such as Cameroon and Namibia, reported an increase in national coverage that was
achieved through improvements in the richest subgroup but no improvements in the poorest subgroup
(a pro-rich change).
35
OTHER KEY FINDINGS: INEQUALITY IN CHILDHOOD IMMUNIZATION

• In most study countries, there was no – or very little – difference in immunization coverage between
boys and girls.
• Economic-related absolute inequalities were more pronounced in DTP3 and measles immunization
coverage than in Bacille Calmette-Guérin (BCG) and polio immunization coverage.
• A considerable proportion of study countries (about 40%) reported large economic-related inequalities
in full immunization coverage, with coverage at least 20 percentage points higher in the richest than in
the poorest quintiles.
• Over a 10-year period, over 85% of study countries reported increases in measles immunization coverage.
In all but a few countries, rural areas showed similar or faster improvements than urban areas.
• Over a 10-year period, over 80% of study countries achieved national-level increases in full immunization
coverage; in the majority of these countries with national improvements, coverage increased faster in
the poorest than in the richest quintile.
INTERACTIVE VISUAL 4. CHILDHOOD IMMUNIZATION

benchmarking.
state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
childhood immunization indicators.

ht is
wor e d
ld h or fre
36
4.5. CHILD MALNUTRITION
4.5 Child malnutrition

Children who are undernourished at a young age may experience poor cognitive development, and miss out
on a critical window for optimizing adult height, lean body mass, intelligence and educational achievement,
economic productivity, and reproductive performance (20). Stunting refers to inadequate length or height for
age, and affects about 161 million young children globally (43). It is commonly the outcome of insufficient
or low-quality diets, poor childcare and frequent infections. Child stunting is the most sensitive indicator
of a child’s quality of life (20), and is widely employed as a marker for malnutrition among children.
Effective action to reduce and eliminate child malnutrition – and thereby alleviate short- and long-term
adverse health consequences – must be comprehensive and multifaceted, building upon efforts to address
immediate concerns about hunger and food availability. This means taking action to improve both the
root causes of malnutrition (such as household and family factors, inadequate complementary feeding,
early cessation of breastfeeding and frequent infections) and its broader underlying social, economic and
environmental determinants (44).
FEATURE STORY
EDUCATION-RELATED INEQUALITY IN STUNTING PREVALENCE IN CHILDREN
STUNTING PREVALENCE IN CHILDREN AGED LESS THAN FIVE YEARS

In half of study countries, nearly one in every three children under the age of five years was classified
as stunted. Over one quarter of countries had a national stunting prevalence of under 20% but another
quarter had a prevalence of 40% or higher. Overall, stunting in children under five years of age was more
prevalent in low-income study countries than in middle-income study countries.
LATEST SITUATION: BY EDUCATION LEVEL

In most countries, stunting prevalence in children was lower among the subgroup with the highest level of
maternal education. That is, the no education subgroup tended to have the highest prevalence of stunting,
the primary school subgroup had the second highest prevalence, and the secondary school or higher
subgroup had the lowest prevalence. In both low-income and middle-income study countries, the median
prevalence of stunting decreased most sharply between mothers who received primary-level schooling
and those who attended secondary school or higher (relative to the decrease between no education and
primary school subgroups) (Figure 4.10)
In half of study countries, the absolute difference in stunting prevalence between the secondary school
or higher subgroup and the no education subgroup was 15 percentage points or higher; this education-
related difference in stunting prevalence reached a maximum of 38.6 percentage points in Honduras, and
the minimum inequality was 0.9 percentage points in Gabon.
37
FIGURE 4.10 Stunting prevalence in children aged less than five years by mother’s education in 30 low-income and 36 middle-
income countries: latest situation (DHS and MICS 2005–2013)
Low-income study countries Middle-income study countries
No education Primary school Secondary school + No education Primary school Secondary school +
70 70
60 60
50 50
Prevalence (%)
Prevalence (%)
40 40
30 30
20 20
10 10
0 0
Interesting patterns emerge if the degree of education-related inequality in stunting prevalence is viewed
alongside national levels of stunting. A low national prevalence of stunting did not necessarily indicate
a favourable situation in terms of inequality. Study countries from the WHO Region of the Americas, for
example, demonstrated relatively low stunting prevalence (compared with the median of study countries
from all WHO regions), but a wide range of levels of inequality. Belize, Guyana, Haiti, Honduras and Peru
all reported a national stunting prevalence of around 20%, but absolute inequality (the difference between
the most and least educated) ranged from 8.1 percentage points in Guyana to 38.6 percentage points in
Honduras.
CHANGE OVER TIME: BY EDUCATION LEVEL

Overall, the prevalence of stunting tended to decrease over time in all education subgroups. However,
annual absolute excess change values revealed considerable variation across countries in terms of the
pace of change in stunting prevalence in education subgroups. The Dominican Republic, Malawi, Mali,
Mozambique, Zambia and Zimbabwe, for instance, all reported improvements in the least-educated
subgroup but no improvement or worsening in the most-educated subgroup; for these countries, excess
change indicated a situation that was more favourable among the least educated (that is, pro less-
38
4.5. CHILD MALNUTRITION
educated). Jordan realized improvements in all subgroups, with substantially faster reductions in stunting
prevalence in the no education subgroup relative to the secondary school or higher subgroup; improvements
among the least educated outpaced those in the most educated by 11 percentage points over 10 years.
Cambodia had similar rates of improvement in all education subgroups, and thus education-related
inequality remained unchanged.
All but a few study countries reported an improvement (or no change) in national stunting prevalence
over time (Figure 4.11: countries to the left of the vertical grey line). In half of study countries, the national
prevalence of stunting decreased by at least 6 percentage points over 10 years (Figure 4.11: countries to
the left of the vertical orange median line). More than half of study countries fell into the most desirable
quadrant, having achieved a decrease in national stunting prevalence and a faster pace of improvement
in the least-educated subgroup (Figure 4.11: countries in the bottom left quadrant – below and to the left
of the grey zero lines).
FIGURE 4.11 Stunting prevalence in children aged less than five years: change over time in national average (absolute change)
and in the least-educated compared with the most-educated subgroup (absolute excess change) in 29 low- and middle-income
countries (DHS and MICS 1995–2004 and 2005–2013)
Pro more-educated change Pro more-educated change
2
Annual absolute change in the no education compared with the
secondary school + subgroup (percentage points)
0
-0.3
-1
-2
Pro less-educated change -0.6 Pro less-educated change
-2 -1 0 1 2
39
OTHER KEY FINDINGS: INEQUALITY IN CHILD MALNUTRITION

• In most study countries, the prevalence of underweight in children under the age of five years was highest
in disadvantaged populations (the poorest, the least educated and those residing in rural areas).
• In over two thirds of study countries, underweight prevalence in children under five years of age was at
least twice as high in the poorest relative to the richest quintiles.
• In the majority of study countries, stunting prevalence among all children under five years decreased
over time; the pace of change in the poorest and richest subgroups tended to reveal divergent patterns
across study countries, thus there was no global indication of a decrease in economic-related inequality
in stunting prevalence.
INTERACTIVE VISUAL 5. CHILD MALNUTRITION

benchmarking.
state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
indicators of child malnutrition.

ht is
wor e d
ld h or fre
40
4.6. CHILD MORTALITY
4.6 Child mortality

In 2013, child mortality accounted for 6.3 million deaths globally, translating into 17 000 deaths of children
aged five years or younger each day (45, 46). The leading causes of deaths in the post-neonatal and early
childhood period include preterm birth complications, pneumonia, and intrapartum-related complications
such as birth asphyxia (40, 47). Malnutrition contributes to a sizable proportion of deaths, especially in low-
income countries (15). Approximately four out of five deaths in children under the age of five years occurred
in sub-Saharan African or Southern Asia (15). It is estimated that more than half of all child deaths are due to
preventable or treatable conditions that could be averted with access to simple, affordable interventions (48).
Since 1990, the global rate of child mortality has fallen by nearly 50%, with accelerated reductions in recent
years. These reductions in child mortality are largely attributable to declines in infectious diseases such as
pneumonia, diarrhoea and measles (40). Despite these recent gains, there is still significant progress to be
made and the continued reduction of child mortality remains a global priority (46).
FEATURE STORY
PLACE-OF-RESIDENCE INEQUALITY IN UNDER-FIVE MORTALITY RATE
UNDER-FIVE MORTALITY RATE

54 low- and middle-income study countries, DHS 2005–2013
In over half of low- and middle-income study countries, at least 75 children out of every 1000 live births
died before reaching their fifth birthday. The countries reporting the highest under-five mortality rates
were in the WHO African Region.
Under-five mortality rates varied greatly by country income group: whereas half of the middle-income study
countries reported child mortality rates in excess of 50 deaths per 1000 live births, half of the low-income
study countries had rates of over 100 deaths per 1000 live births.
LATEST SITUATION: BY PLACE OF RESIDENCE

54 low- and middle-income study countries, DHS 2005–2013
The under-five mortality rate was higher in rural areas than in urban areas in most of the low- and middle-
income study countries. In half of countries, the under-five mortality rate in rural areas was 84 deaths per
1000 live births or higher. In urban areas, the under-five mortality rate was about 61 deaths per 1000 live
births or higher in half of study countries.
The magnitude of the difference in under-five mortality rates between rural and urban areas varied by
country. Half of study countries reported a difference of about 16 deaths per 1000 live births or less (Figure
4.12). In Burkina Faso, Burundi, Cameroon, Guinea, Niger and Nigeria, however, the difference in under-five
mortality rates between rural and urban areas was at least 50 deaths per 1000 live births (Figure 4.12:
countries highlighted in red).
41
FIGURE 4.12 Under-five mortality rates by place of residence in 54 low- and middle-income countries: latest situation
(DHS 2005–2013)
Niger
Niger
Nigeria
Nigeria
Cameroon
Cameroon
80
Guinea
Guinea
Burkina Faso
Burkina Faso
Burundi
Burundi
Mali
Mali
Cambodia
Cambodia
Bolivia(Plurinational
Bolivia (Plurinational State
State of)
of) 70
Uganda
Uganda
India
India
Pakistan
Pakistan
Ethiopia
Ethiopia
Senegal
Senegal 60
Comoros
Comoros
Difference between rural and urban areas (deaths per 1000 live births)
Timor-Leste
Timor-Leste
Côted'Ivoire
Côte d’Ivoire
Sierra
Sierra Leone
Leone
Rwanda
Rwanda
Lesotho
Lesotho 50
Democratic
DemocraticRepublic
Republicofofthe
the Congo
Congo
Benin
Benin
Madagascar
Madagascar
Nepal
Nepal
Indonesia
Indonesia 40
Malawi
Malawi
Gabon
Gabon
Albania
Albania
Namibia
Namibia
Liberia
Liberia
Philippines
Philippines 30
Peru
Peru
Azerbaijan
Azerbaijan
Mozambique
Mozambique
Congo
Congo
Kenya
Kenya 20
Bangladesh
Bangladesh
Republic
Republicof of Moldova
Moldova
Tajikistan
Tajikistan
Egypt
Egypt
Armenia
Armenia
Zambia 10
Zambia
Maldives
Maldives
Colombia
Colombia
Ukraine
Ukraine
Zimbabwe
Zimbabwe
Dominican
DominicanRepublic
Republic 0
Honduras
Honduras
Kyrgyzstan
Kyrgyzstan
Jordan
Jordan
United
UnitedRepublic
RepublicofofTanzania
Tanzania
Sao
SaoTome
TomeandandPrincipe
Principe
-10
Guyana
Guyana
Haiti
Haiti
0 50 100 150 200
Mortality rate (deaths per 1000 live births)
 Study countries with no (or very low) place-of-residence inequality  Study countries with high place-of-residence inequality
Circles indicate countries – each study country is represented by two circles in the left graph and one circle in the right graph. In the graph on the left,
the light blue bands indicate the absolute difference in mortality rate between rural and urban areas for each country. In the graph on the right, the
horizontal line indicates the median value (middle point), and the light grey band indicates the interquartile range (middle 50% of study country
estimates).
42
4.6. CHILD MORTALITY
Certain study countries reported no – or very low – place-of-residence inequality in child mortality rates. For
example, the Dominican Republic, Honduras, Jordan, Kyrgyzstan, Ukraine, the United Republic of Tanzania
and Zimbabwe reported an under-five mortality rate difference of less than 3 deaths per 1000 live births
between rural and urban areas (Figure 4.12: countries highlighted in blue). In some cases, such as Jordan
and Ukraine, low inequality was achieved alongside low national under-five mortality rates (around 20
deaths per 1000 live births). The United Republic of Tanzania and Zimbabwe, however, had moderately
high national rates of under-five mortality (around 80–90 deaths per 1000 live births) and low inequality,
indicating that the situation was equally unfavourable in both rural and urban areas.
CHANGE OVER TIME: BY PLACE OF RESIDENCE

37 low- and middle-income study countries, DHS 1995–2004 and 2005–2013
Change-over-time analyses revealed improvements in national average child mortality rates in nearly all
study countries (Figure 4.13: countries to the left of the vertical grey zero line). Notably, Mail, Niger and
Rwanda reported an average reduction in under-five mortality of more than 10 deaths per 1000 live births
per year.
Reductions in under-five mortality rates across both rural and urban areas were seen in most study
countries. Faster decreases tended to be reported for rural areas. For instance, the pace of improvement
in rural areas outpaced that in urban areas by more than 45 deaths per 1000 live births over 10 years in
Burkina Faso, Malawi, Mozambique, Niger and Rwanda. In contrast, Cambodia and Cameroon reported
faster improvements in urban areas, but the pro-urban advantage did not exceed 15 deaths per 1000 live
births over 10 years in any of these countries (Figure 4.13).
A clear majority of study countries reported decreasing national under-five mortality rates alongside pro-
rural changes in the degree of inequality over time (Figure 4.13: countries in the bottom left quadrant).
Furthermore, in nearly one quarter of study countries, the decrease in national under-five mortality rates
was particularly fast (at least 60 deaths per 1000 live births over 10 years) and achieved through pro-rural
reductions.
OTHER KEY FINDINGS: INEQUALITY IN CHILD MORTALITY

• In half of study countries, the infant mortality rate was at least 8 deaths per 1000 live births higher in
rural than in urban areas.
• In about one quarter of study countries, the gap in neonatal mortality rates between the most- and
least-educated subgroups was at least 15 deaths per 1000 live births.
• In half of study countries, the under-five mortality rate decreased more rapidly in the poorest than in the
richest subgroup, by a margin of at least 26 deaths per 1000 live births over a 10-year period.
43
FIGURE 4.13 Under-five mortality rates: change over time in national average (absolute change) and in rural compared with urban
areas (absolute excess change) in 37 low- and middle-income countries (DHS 1995–2004 and 2005–2013)
Pro-urban change Pro-urban change
6
Annual absolute change in rural compared with urban areas
4
(deaths per 1000 live births)
2 Cambodia Cameroon
Mali
-2
-4 Burkina Faso
Mozambique
Niger
Malawi
Rwanda
-6
Pro-rural change Pro-rural change
-12 -10 -8 -6 -4 -2 0 2 4 6 8
Annual absolute change in national average (deaths per 1000 live births)
INTERACTIVE VISUAL 6. CHILD MORTALITY

benchmarking. state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland

ISBN 978 92 4 156491 5
© n.
co tio
child mortality indicators.

py ibu
rig tr
ht is
wor e d
ld h or fre
44
4.7. RMNCH INTERVENTIONS, COMBINED
4.7 RMNCH interventions, combined

The composite coverage index is a weighted score that incorporates the following eight RMNCH
intervention indicators:
• demand for family planning satisfied
• antenatal care coverage (at least one visit)
• births attended by skilled health personnel
• BCG immunization coverage among one-year-olds
• measles immunization coverage among one-year-olds
• DTP3 immunization coverage among one-year-olds
• children aged less than five years with diarrhoea receiving oral rehydration therapy and continued feeding
• children aged less than five years with pneumonia symptoms taken to a health facility.
The composite coverage index captures both the provision and use of key RMNCH interventions, expressing
the number of people receiving a specified intervention as a percentage of those who require that
intervention. The composite coverage index can be used to indicate a country’s overall progress towards
achieving universal coverage for RMNCH. A composite index tends to be more stable and representative
of the overall situation than any one of the component RMNCH intervention indicators, which are more
sensitive to factors such as small sample sizes.
The Countdown to 2015 initiative has applied the composite coverage index as a summary measure to
track key interventions in RMNCH, noting a tendency towards lower child mortality in countries with
higher values of the composite coverage index and, conversely, higher child mortality in countries with
lower values (20). The expansion and improvement of RMNCH interventions is embedded in all major
global initiatives to promote RMNCH.
FEATURE STORY
INEQUALITY IN RMNCH COMPOSITE COVERAGE INDEX
RMNCH COMPOSITE COVERAGE INDEX

Overall, almost one quarter of study countries reported composite coverage index values of 80% or more;
however, coverage index values varied substantially across countries, ranging from 37.4% in Ethiopia to
89.7% in Costa Rica.
45
LATEST SITUATION: BY MULTIPLE DIMENSIONS OF INEQUALITY

48 low- and middle-income study countries (by economic status)
45 low- and middle-income study countries (by education level)
62 low- and middle-income study countries (by place of residence)
DHS and MICS 2005–2013
Disaggregation of the composite coverage index score across three dimensions of inequality revealed a
common pattern of higher coverage among advantaged subgroups. The coverage of RMNCH interventions
tended to be greatest in the economically advantaged subgroups, the secondary school or higher subgroup
and in urban areas. Figure 4.14 illustrates the scale of within-country inequalities and the cross-country
variation within subgroups.
FIGURE 4.14 RMNCH composite coverage index by multiple dimensions of inequality in low- and middle-income countries: latest
Economic status Education* Place of residence
Quintile 1 Quintile 5 Secondary Rural Urban
Quintile 2 Quintile 3 Quintile 4 No education Primary school
(poorest) (richest) school +
100
90
80 79.1
76.3 76.3
71.8
70
64.9 65.9 65.1
61.4
60
Coverage (%)
56.7 56.4
50
40
30
20
10
* Education refers to the highest level of schooling attained by the woman or, in the case of newborn and child health interventions, the child’s mother.
Circles indicate countries – each study country is represented on each graph by one circle per subgroup. Horizontal lines indicate the median value
(middle point) for each subgroup. Light grey bands indicate the interquartile range (middle 50% of study country estimates).
46
Economic-related inequality
Median values of the composite coverage index increased in a linear fashion across economic status
subgroups, moving from poorest to richest. Economic-related inequality (expressed in absolute terms as
a difference between the richest and the poorest quintiles) varied among study countries, ranging from
3.1 percentage points in Jordan to 60.9 percentage points in Nigeria.
In nearly one quarter of study countries, the difference between the richest and poorest quintiles was
particularly large – 30 percentage points or higher. Cameroon, the Central African Republic, Ethiopia, Guinea
and Nigeria reported the highest levels of within-country economic-related relative inequality, with a level
of coverage in the richest subgroup at least two times that of the poorest subgroup.
Low levels of economic-related inequality were reported by Burundi, Colombia, the Dominican Republic,
Honduras, Jordan, Kyrgyzstan, Sierra Leone and Swaziland, where the difference in the value of the
composite coverage index between the richest and poorest quintiles was 10 percentage points or less.
Education-related inequality
All low- and middle-income study countries reported composite coverage index values that were greater
in the secondary school or higher subgroup than in the no education subgroup. Over one third of study
countries reported a difference of more than 20 percentage points between the most- and least-educated
subgroups. In Egypt and Swaziland, this difference was less than 8 percentage points. In contrast, the
difference was 40 percentage points or more in Cameroon and Nigeria.
Place-of-residence inequality
Composite coverage index values tended to be higher in urban than in rural areas. In nearly half of study
countries, however, the difference between urban and rural areas was less than 10 percentage points.
One fifth of study countries showed minimal place-of-residence inequality with differences of less than 5
percentage points. In Bosnia and Herzegovina, the Dominican Republic, Jordan, Republic of Moldova and
Uzbekistan, this difference was less than 2 percentage points.
Ethiopia and Nigeria had the largest level of within-country relative inequality, with coverage in urban areas
exceeding that of rural areas by a factor of two.
CHANGE OVER TIME: BY MULTIPLE DIMENSIONS OF INEQUALITY

28 low- and middle-income study countries (by economic status)
25 low- and middle-income study countries (by education level)
34 low- and middle-income study countries (by place of residence)
DHS and MICS 1995–2004 and 2005–2013
Overall, countries tended to demonstrate improvements in composite coverage index over time, with
gains nationally and within subgroups of the population. The pace of these improvements (or changes)
tended to favour the disadvantaged subgroups; that is, the changes were for the most part pro-poor, pro
less-educated and pro-rural. The highest rates of increase in coverage (at the national level) were reported
by Cambodia (30.9 percentage points over 10 years) and Rwanda (23.9 percentage points over 10 years).
47
Economic-related inequality
Composite coverage index scores increased faster in the poorest population subgroup compared with
the richest subgroup in three quarters of study countries. The Plurinational State of Bolivia and Cambodia
reported the fastest pro-poor increases (narrowing of inequality), with the change over time in the poorest
subgroup outpacing change in the richest subgroup by at least 20 percentage points over 10 years.
While Ethiopia and Uganda reported a substantial increase in national RMNCH composite coverage index
(by more than 10 percentage points over 10 years), improvements in the poorest quintile were outpaced
by those in the richest (a pro-rich change).
Education-related inequality
In the majority of cases, RMNCH composite coverage index increased over time in all three education
subgroups. Most countries reported faster improvements in the no education subgroup than in the
secondary school or higher subgroup. The Plurinational State of Bolivia, Cambodia and Rwanda reported
substantial increases in RMNCH composite coverage index of at least 24 percentage points over 10 years
in the no education subgroup.
In many study countries, national-level increases in RMNCH composite coverage index were accomplished
through rapid and sizeable gains in the no education subgroup. However, this was not true for all countries.
Namibia, for example, achieved a substantial overall increase in RMNCH composite coverage index (by
12.3 percentage points over 10 years), but this was driven largely by improvements in the primary school
and secondary school or higher subgroups.
Place-of-residence inequality
Most low- and middle-income study countries reported an increase in the value of the composite coverage
index over time in both rural and urban areas. In the majority of cases, the changes were pro-rural; that
is, changes in RMNCH composite coverage index occurred faster in, or otherwise favoured, rural areas.
Notably, the Plurinational State of Bolivia, Malawi, Mali and Niger reported a pro-rural change by a margin
of at least 10 percentage points over 10 years. One in every six countries reported the pace of change to
be the same in rural and urban areas (excess change was negligible).
OTHER KEY FINDINGS: INEQUALITY IN RMNCH COMPOSITE COVERAGE INDEX

• While half of countries had composite coverage index values of over 80% for the richest quintile, only
one country achieved this level of coverage in the poorest quintile.
• A considerable proportion of countries – about one quarter – reported both a low national average
composite coverage index value (under 60%) and a high level of education-related inequality (a
difference between the most- and least-educated of at least 20 percentage points).
• Over a 10-year period, the changes in composite coverage index tended to favour the most-disadvantaged
subgroups to a greater extent than the most advantaged for all dimensions of inequality (economic
status, education and place of residence); that is, coverage increased more rapidly (or decreased more
slowly) in the most-disadvantaged subgroups.
48
INTERACTIVE VISUAL 7. RMNCH COMPOSITE COVERAGE INDEX

benchmarking.
state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland
the state of inequality in the RMNCH composite

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
coverage index.
ht is
wor e d
ld h or fre
49
4.8 Potential for improvement in RMNCH interventions

Population attributable risk is a summary measure that expresses the magnitude of within-country
inequality in a single number. It combines the health indicator estimate of the whole population and that
of the most-advantaged subgroup, showing the improvement possible if the whole population in a country
had the same level of coverage as the most-advantaged subgroup. Population attributable risk can thus
be interpreted as the improvement in coverage that would be realized at the national level if the whole
population was able to experience the same level of coverage as the most-advantaged subgroup.
FEATURE STORY
POTENTIAL FOR IMPROVEMENT IN THE COVERAGE OF RMNCH

INTERVENTIONS BY ELIMINATING WITHIN-COUNTRY ECONOMIC-RELATED
INEQUALITY
Population attributable risk estimates for the composite coverage index and its eight constituent health
intervention indicators are summarized in Table 4.1. Note that because composite coverage index is
comprised of eight components, these estimates show less variation from country to country than estimates
for each of the individual indicators.
TABLE 4.1 Potential for improvement in national RMNCH intervention coverage by eliminating within-country economic-related
inequality (population attributable risk) in low- and middle-income countries with available data (DHS and MICS 2005–2013)
Maximum
Median population population
Number of countries attributable risk attributable risk
RMNCH indicator with available data (percentage points) (percentage points)
Composite coverage index 50 9.6 32.2
Demand for family planning satisfied 60 8.0 25.0
Antenatal care – at least one visit 83 4.1 41.9
Births attended by skilled health personnel 83 16.6 49.2
BCG immunization coverage among one-year-olds 78 2.7 41.1
Measles immunization coverage among one-year-olds 78 5.3 35.8
DTP3 immunization coverage among one-year-olds 78 5.0 41.1
Children aged less than five years with diarrhoea receiving 24.7
70 3.9
oral rehydration therapy and continued feeding
Children aged less than five years with pneumonia 40.4
57 8.8
symptoms taken to a health facility
BCG: one dose of Bacille Calmette-Guérin vaccine; DTP3: three doses of the combined diphtheria, tetanus toxoid and pertussis vaccine;
RMNCH: reproductive, maternal, newborn and child health.
50
4.8. POTENTIAL FOR IMPROVEMENT IN RMNCH INTERVENTIONS
POTENTIAL FOR IMPROVEMENT IN SELECTED RMNCH INTERVENTION INDICATORS

Composite coverage index and births attended by skilled health personnel, DHS and MICS 2005–2013
By eliminating economic-related inequality in eight RMNCH interventions and increasing coverage to the
level of the richest quintile, around half of study countries could potentially achieve an increase in their
composite coverage index of about 10 percentage points from current levels (median population attributable
risk, 9.6 percentage points). Given that Jordan reported a population attributable risk of 1.5 percentage
points, it has the least room for improvement (and the lowest level of within-country economic-related
inequality), whereas Nigeria has the most potential for improvement with a population attributable risk
of 32.2 percentage points (indicating the highest level of within-country economic-related inequality).
According to this analysis, if economic-related inequalities were eliminated, almost half of study countries
would have a national composite coverage index value of about 80% or higher. Currently, only one in every
seven study countries reported national index values of over 80%.
Of the indicators that comprise the composite coverage index, the births attended by skilled health
personnel indicator demonstrated the most room for improvement according to population attributable
risk analyses. Overall, the difference between coverage at the national level and that in the richest quintile
was more than 16 percentage points in half of low- and middle-income study countries. This median value,
however, masks cross-country variations in population attributable risk. Several countries had a population
attributable risk of less than 1 percentage point, while others reported values in excess of 40 percentage
points. The maximum population attributable risk for births attended by skilled health personnel was
reported in the Lao People’s Democratic Republic (49.2 percentage points).
If economic-related inequalities were eliminated (by increasing the proportion of attended births in the
whole population to that in the richest quintile), more than half of study countries would achieve levels of
skilled birth attendance of over 95%. Only about one quarter of study countries reported current coverage
of 95% or higher.
POTENTIAL FOR IMPROVEMENT IN SELECTED COUNTRY EXAMPLES

Egypt and Niger, DHS 2008 and 2012
Within countries, population attributable risk varied according to the health indicator. Egypt, for
example, reported little economic-related inequality in the child immunization indicators. Maternal health
intervention indicators, on the other hand, each indicated a population attributable risk value of over
16 percentage points. Reproductive health interventions indicators in Egypt had intermediate levels of
population attributable risk, of around 5 percentage points (Figure 4.15).
Niger demonstrated different levels of population attributable risk in two maternal health interventions with
similar levels of current coverage. Both antenatal care coverage (at least four visits) and births attended
by skilled health personnel had national coverage of just over 30%. The potential for improvement by
eliminating economic-related inequality, however, was much higher for births attended by skilled health
personnel (41.0 percentage points) than antenatal care coverage (at least four visits) (13.5 percentage
points) (Figure 4.15).
51
FIGURE 4.15 Potential for improvement in RMNCH intervention coverage by eliminating within-country economic-related
inequality in Egypt (DHS 2008) and Niger (DHS 2012)
Egypt Niger
Contraceptive prevalence – modern and

Reproductive
traditional methods
health
interventions
Contraceptive prevalence – modern methods
Demand for family planning satisfied
Maternal Antenatal care coverage – at least one visit

health
interventions
Antenatal care coverage – at least four visits
Births attended by skilled health personnel
Newborn and Early initiation of breastfeeding

child health
interventions
Children aged 6–59 months who received
vitamin A supplementation
BCG immunization coverage among

one-year-olds
Measles immunization coverage among

one-year-olds
Polio immunization coverage among

one-year-olds
DTP3 immunization coverage among

one-year-olds
Full immunization coverage among

one-year-olds
Children aged < 5 years with diarrhoea

receiving oral rehydration salts
Children aged < 5 years with diarrhoea

receiving oral rehydration therapy and
continued feeding
Children aged < 5 years with pneumonia

symptoms taken to a health facility
RMNCH
interventions, Composite coverage index
combined
0 20 40 60 80 100 0 20 40 60 80 100
National average (%) National average (%)
 Current situation  Potential for improvement
The potential for improvement (pale blue shaded area) represents the improvement possible if the whole population had the same level of coverage as the
richest subgroup.
52
4.8. POTENTIAL FOR IMPROVEMENT IN RMNCH INTERVENTIONS
INTERACTIVE VISUAL 8. POTENTIAL FOR IMPROVEMENT IN RMNCH INTERVENTIONS

benchmarking.
state of InequalIty
These interactive dashboards guide you through the

20, avenue appia
CH-1211 geneva 27
SWitzerland
potential for improvement in RMNCH intervention

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
indicators, by multiple dimensions of inequality.

ht is
wor e d
ld h or fre
53
5
Reporting the state of inequality:
taking stock
Overall, health inequalities were identified across low- and middle-income countries in RMNCH. While
national averages demonstrated improvements over the past decade – often as a result of rapid gains in
disadvantaged subgroups – this progress was not sufficiently equity oriented to close the gap. Inequalities
were still pervasive between those of different economic status subgroups, education levels, places of
residence and, in a few cases, sex, though there were variations based on the country and the indicator.
Supplementary tables S3, S4 and S5 (appended to this report) present a summary of state-of-inequality
data for the 23 RMNCH indicators.
55
UNDERSTANDING WHY INEQUALITIES PERSIST
A study of the drivers of inequality attempts to explain the root causes of inequality, looking at a broad scope of
determinants of health (demographic, environmental and socioeconomic factors) within and outside of the health sector.
Comprehensive investigations of the causes of inequality draw on many types of research and diverse perspectives to
provide context and frame the issue. There are many possible causes of inequality, which may manifest in different ways
depending on the setting and health topic (and indicator) of interest, as well as the timing of the study.
This report examines both cross-country inequalities, comparing national averages across low- and middle-income
countries, as well as within-country inequality by economic status, education level, place of residence and sex. As an
extension of these findings, the associations between selected determinants of health and RMNCH indicators may be
explored to gain preliminary insights into possible country-level factors that may be driving cross-country inequalities in
RMNCH. (See Appendix 1 for more details about the possible determinants of health featured in the interactive visual.)
INTERACTIVE VISUAL 9. DETERMINANTS OF RMNCH

benchmarking.
state of InequalIty
These interactive dashboards guide you through data

20, avenue appia
CH-1211 geneva 27
SWitzerland
about the associations between selected country-level

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
determinants and health indicators.

ht is
wor e d
ld h or fre
5.1 The importance of data disaggregation

Disaggregated data are at the heart of monitoring inequalities. The practice of disaggregating data reveals
where inequalities exist; appreciating how health levels vary across subgroups can help to identify the best
approach to address inequalities in health.
In addition to the four dimensions of inequality discussed in this report, health data may be disaggregated
by other relevant dimensions of inequality such as age, race/ethnicity, subnational region and religion,
as appropriate and per data availability. For example, age may be considered an important stratifier for
reproductive health indicators that are subject to age discrimination (this may occur if adolescent and
adult-aged women experience different access to reproductive health interventions, for instance).
In this report, inequalities were analysed by a single dimension of inequality at a time. In some cases,
however, it is more instructive to doubly disaggregate, that is, to apply two dimensions of inequality
simultaneously. For example, socioeconomic inequalities in health may exist within urban and rural settings.
In Benin, the under-five mortality rate was higher in rural areas than in urban areas. However, a closer
inspection reveals that, when disaggregated again by economic status, the rate of under-five mortality
among the urban poor was even higher than the rate in rural areas (Figure 5.1).
56
5. REPORTING THE STATE OF INEQUALITY: TAKING STOCK
FIGURE 5.1 Under-five mortality rate doubly disaggregated by place of residence and economic status in Benin (DHS 2006)
180
153
150
Mortality rate (deaths per 1000 live births)
134
120
106
90
73
60
30
0
Rural Urban (overall) Urban poorest quintile Urban richest quintile
Source: adapted from Country profiles on urban health: Benin. Kobe: World Health Organization Centre for Health Development
(http://www.who.int/kobe_centre/measuring/urbanheart/benin.pdf?ua=1, accessed 11 March 2015).
Such a situation may have important implications for policies, programmes and practices. If double-
disaggregation analyses were not conducted for Benin, for example, policy-makers might concentrate
efforts in rural areas and neglect the disadvantaged urban poor. Thus, when monitoring health inequalities,
it may be appropriate to divide the population by multiple dimensions of inequality at the same time. One
limitation of double disaggregation using household health survey data, however, is that the sample size
in each subgroup diminishes when data are disaggregated. Smaller sample sizes mean that the estimates
become more uncertain and the ability to make valid comparisons between subgroups may be hindered.
5.2 Equity orientation of policies, programmes and practices

Monitoring the state of inequality, both across and within countries, serves as a warning system; it draws
attention to the presence of inequality in different areas of health and identifies priority areas for further
investigation. Policies, programmes and practices should be equity oriented to promote improvements
among the most-disadvantaged subgroup(s) that are at least as fast as the most advantaged. Without a
dedicated focus on equity, such actions may achieve increases in national coverage but risk intensifying
within-country inequality.
Many study countries achieved improved health intervention coverage and outcomes at the national level;
however, in some cases this was fuelled by more rapid gains in the advantaged subgroups than in the
disadvantaged subgroups. In these cases, health policies, programmes and practices should be re-oriented
to uphold the principle of equity and promote faster improvements among the disadvantaged. For instance,
in Ethiopia, the national prevalence of modern contraceptive use increased but disproportionately more
so among the richer and better-educated subgroups. Similarly, while the percentage of births attended
by skilled health personnel in Bangladesh, Ethiopia, Nepal and Uganda increased overall, the pace of the
increase favoured the rich over the poor.
57
In other cases, improvements at the national level were accompanied by pro-disadvantaged changes at the
population subgroup level, suggesting that interventions may already be equity oriented. In Bangladesh,
for example, the increase in measles immunization coverage in rural areas outpaced that in urban areas;
there were also faster coverage gains in the poorest (relative to the richest) and among children born
to mothers with no education (relative to children whose mothers had received secondary schooling or
higher). Likewise, in Malawi, Mozambique, Niger and Rwanda, national infant mortality rates declined,
with faster decreases in the least-educated subgroups than in the most-educated subgroups.
Characteristic patterns of inequality across multiple, ordered subgroups – and suggestions for corresponding
policy, programme or practice responses – are described further in Appendix 7.
5.3 Equity-oriented health information systems

The purpose of health information systems is to collect, analyse and report data. When health information
systems are equity oriented they have the necessary tools to conduct health inequality monitoring and
to generate useful inputs to inform equity-oriented policies, programmes and practices. Developing and
expanding equity-oriented health information systems at a national level requires capacity-building to
support the improved collection, analysis and reporting of health data by population subgroups. For some
countries, significant investment may be required to build and strengthen capacity for health inequality
monitoring. Countries that already have strong, equity-oriented health information systems should regularly
review and update their standardized data collection, analysis and reporting practices.
One central consideration for the expansion of health inequality monitoring is data availability. The data
used for health inequality monitoring should be of high quality and comparable across settings and over
time. The two types of data required for health inequality monitoring – data about health and data about
the dimensions of inequality – can either be collected from a single source or linked together from different
sources. Currently, national assessments of health inequality in low- and middle-income countries – such
as the current report – usually draw from household health survey data. Household health surveys such
as DHS and MICS are repeated at regular intervals, and have high-quality, comparable data on a specific
health topic, as well as on living standards. Health facility data are increasingly recognized as a valuable
source of readily available data to assess geographical inequality at the subnational level, especially when
data collection practices are harmonized across subnational regions (49).
Dedicated resources are also required to develop and support the technical expertise and skills required to
perform health inequality analyses. Inequality is a complex concept that can be conveyed using a variety of
analysis techniques. Proficiency in analysing health inequality data demands not only technical knowledge
of summary measures, but also an awareness of best practices in data analysis and interpretation.
Effective reporting and dissemination practices take into account the unique needs and abilities of the
target audiences. Clear and salient reporting about the state of inequality should aim to achieve a balance
between presenting comprehensible messages, while maintaining sufficient technical accuracy and rigour.
Sometimes, the clear and effective communication of multiple dimensions of inequality may necessitate a
reduction in the amount of data that are presented (49). The use of data visuals – both in the conventional
static and novel interactive forms – can greatly enhance the presentation and interpretation of large and
complex inequality datasets.
58
5. REPORTING THE STATE OF INEQUALITY: TAKING STOCK
PROMOTING EQUITY IN RMNCH
Promoting equity within the health sector begins with monitoring health inequalities. Actions to address inequities include:
recognizing that health services often contribute to increasing inequities; prioritizing diseases of the poor; deploying and
improving health interventions where the poor live; employing appropriate health service delivery channels; removing or
reducing financial barriers to health care; and setting goals and monitoring progress through an equity lens (49).
Achieving improvements in RMNCH, however, requires intersectoral action that addresses the broader determinants of
health.
Outside of the health sector, actions to promote RMNCH encompass factors such as education, population dynamics,
environmental management, poverty reduction and income inequality, women’s political and socioeconomic participation,
good governance, and economic growth. The use of reliable, timely evidence for decision-making and accountability is a
key strategy to improve health outcomes (50, 51).
A renewed Global Strategy for Women’s, Children’s and Adolescent’s Health will have a stronger focus on equity and rights,
and call attention to the integration of environmental, political and socioeconomic determinants in garnering intersectoral
action for improved RMNCH (52). The renewed Global Strategy, to be launched in September 2015, will be a roadmap for
ending all preventable deaths of women, children and adolescents by 2030, and will align with the targets and indicators
developed for the Sustainable Development Goal Framework. Partners in this process have made a commitment to focus on
critical and underserved population groups such as newborns, adolescents and those living in fragile and conflict settings.
Relative to other health topics, inequality monitoring in RMNCH has been accelerated by the availability of
relevant household health survey data. Other areas stand to benefit by learning from the data collection,
analysis and reporting practices that have been established by initiatives to monitor inequalities in RMNCH.
Better cooperation between governments, academia and community stakeholders can help to expand and
enhance the quality of health information systems for inequality monitoring, not only in RMNCH, but also
in all areas of health. Moving forward, the strengthening of equity-oriented health information systems is a
necessary and lasting investment that will enable improved health inequality monitoring across all settings.
5.4 Reducing inequality across health topics and the post-2015

sustainable development agenda
Inequalities are not exclusive to RMNCH. There are countless stories about the state of inequality in
other areas of health. These stories also need to be explored and shared, drawing upon reliable data,
sound analyses and good reporting practices. Monitoring inequalities across the spectrum of health
and by multiple dimensions of inequality represents a necessary step in helping all populations achieve
their potential for good health. The practice of data disaggregation is acknowledged as a key principle for
sustainable development (53). Building capacity for inequality monitoring across a diversity of health topics
is necessary, relevant and important, especially as global movements call attention to promoting equity
through initiatives such as universal health coverage (10).
59
In this spirit, the renewed sustainable development goals and targets specifically call for reductions in
health inequalities. Indeed, there is a heightened requirement for equity orientation across health-related
components of the post-2015 sustainable development agenda, which – in addition to the inclusion of
universal health coverage – seek improvements in areas such as environmental health and noncommunicable
diseases (54). The overarching goal of the health-related post-2015 sustainable development agenda, to
“ensure healthy lives and promote well-being for all at all ages”, represents a commitment to promoting
equity in health, and necessitates a focus on monitoring and reporting inequalities (55).
The post-2015 sustainable development agenda focuses on “leaving no one behind”, calling for efforts
that prioritize marginalized groups. The reduction of inequality within and between countries is included
on the list of proposed sustainable development goals to be attained by 2030. Other proposed goals call
for gender equality, poverty elimination and education for all, addressing social factors that contribute to
inequalities in health (55).
Monitoring the state of inequality in health is a key step towards identifying where action is needed, and
determining how health-related policies, programmes and practices can best be implemented to benefit the
people who need them the most. Only by comparisons of data disaggregated by population subgroups is it
possible to delve deeper into how various facets of health are experienced throughout the entire population.
When health inequalities are taken into consideration, decisions about policies, programmes and practices
can align with broader efforts to reduce social inequalities and discrimination, promote gender equality,
and realize the right to health for all.
60
References
1. Global Health Observatory (GHO) data: Health Equity Monitor [online database]. Geneva: World Health
Organization; 2013 (www.who.int/gho/health_equity/en/, accessed 11 March 2015).
2. Global Thematic Consultation on the Post-2015 Development Agenda. Addressing inequalities: synthesis
report of Global Public Consultation. New York, NY: United Nations Children’s Fund and UN Women; 2013.
3. The core international human rights instruments and their monitoring bodies [website]. New York,
NY: Office of the United Nations High Commissioner for Human Rights; 2015 (www.ohchr.org/EN/
ProfessionalInterest/Pages/CoreInstruments.aspx, accessed 11 March 2015).
4. Declaration of Alma-Ata. Alma-Ata, USSR, 6–12 September 1978 (http://www.who.int/publications/
almaata_declaration_en.pdf, accessed 11 March 2015).
5. Global Strategy for Health for All by the Year 2000. Geneva: World Health Organization; 1981 (http://
whqlibdoc.who.int/publications/9241800038.pdf, accessed 11 March 2015).
6. Office of the United Nations High Commissioner for Human Rights. Human rights indicators: a guide
to measurement and implementation. New York, NY: United Nations; 2012 (http://www.ohchr.org/
Documents/Publications/Human_rights_indicators_en.pdf, accessed 11 March 2015).
7. Commission on Social Determinants of Health. Closing the gap in a generation: health equity through action
on the social determinants of health: final report of the Commission on Social Determinants of Health.
Geneva: World Health Organization; 2008 (http://www.who.int/social_determinants/thecommission/
finalreport/en/, accessed 11 March 2015).
8. Rio Political Declaration on Social Determinants of Health. Rio de Janeiro, Brazil, 21 October 2011 (http://
www.who.int/sdhconference/declaration/Rio_political_declaration.pdf, accessed 11 March 2015).
9. Boerma T, Eozenou P, Evans D, Evans T, Kieny MP, Wagstaff A. Monitoring progress towards universal health
coverage at country and global levels. PLoS Med. 2014;11(9):e1001731.
10. Hosseinpoor AR, Bergen N, Koller T, Prasad A, Schlotheuber A, Valentine N et al. Equity-oriented monitoring
in the context of universal health coverage. PLoS Med. 2014;11(9):e1001727.
11. Saksena P, Hsu J, Evans DB. Financial risk protection and universal health coverage: evidence and
measurement challenges. PLoS Med. 2014;11(9):e1001701.
12. Hill K, Upchurch D. Gender differences in child health: evidence from the Demographic and Health Surveys.
Population and Development Review 1995;21(1):127–51.
13. Reports. In: United Nations Millennium Development Goals and beyond 2015 [website]. New York, NY:
United Nations; 2014 (http://www.un.org/millenniumgoals/reports.shtml, accessed 11 March 2015).
14. Statistics Division of the Department of Economic and Social Affairs. Millennium Development Goals
indicators. The official United Nations site for the MGD indicators [website]. New York, NY: United Nations;
2013 (http://mdgs.un.org/unsd/mdg/Host.aspx?Content=indicators/officiallist.htm, accessed 11 March
2015).
15. The Millennium Development Goals report 2014. New York, NY: United Nations; 2014 (http://www.un.org/
millenniumgoals/2014%20MDG%20report/MDG%202014%20English%20web.pdf, accessed 11 March
2015).
61
16. The Partnership for Maternal, Newborn & Child Health. The PMNCH 2011 report: analysing commitments
to advance the Global Strategy for Women’s and Children’s Health. Geneva: World Health Organization;
2011 (http://www.who.int/pmnch/media/press/2011/2011_pmnch_report/en/, accessed 11 March 2015).
17. The Partnership for Maternal, Newborn & Child Health. The PMNCH 2012 report: analysing progress
on commitments to the Global Strategy for Women’s and Children’s Health. Geneva: World Health
Organization; 2012 (http://www.who.int/pmnch/media/press/2012/2012_pmnch_report/en/, accessed 11
March 2015).
18. The Partnership for Maternal, Newborn & Child Health. The PMNCH 2013 report: analysing progress
on commitments to the Global Strategy for Women’s and Children’s Health; Geneva: World Health
Organization; 2013 (http://www.who.int/pmnch/knowledge/publications/2013_pmnch_report/en/,
accessed 11 March 2015).
19. The Partnership for Maternal, Newborn & Child Health. The PMNCH 2014 accountability report: tracking
financial commitments to the Global Strategy for Women’s and Children’s Health. Geneva: World Health
Organization; 2014 (http://www.who.int/pmnch/knowledge/publications/2014_pmnch_report/en/,
20. Countdown to 2015. Fulfilling the health agenda for women and children: the 2014 report. Geneva: World
Health Organization; 2014 (http://www.countdown2015mnch.org/reports-and-articles/2014-report,
21. Countdown to 2015. Tracking progress in child survival: the 2005 report. New York, NY: United Nations
Children’s Fund; 2005 (http://www.countdown2015mnch.org/reports-and-articles/previous-reports/2005,
22. Countdown to 2015. Tracking progress in maternal, newborn and child survival: the 2008 report. New York,
NY: United Nations Children’s Fund; 2008 (http://www.countdown2015mnch.org/reports-and-articles/
previous-reports/2008, accessed 11 March 2015).
23. Countdown to 2015. Countdown to 2015 decade report (2000–2010). Geneva: World Health Organization;
2010 (http://www.countdown2015mnch.org/reports-and-articles/previous-reports/2010-decade-report,
24. Countdown to 2015. Building a future for women and children: the 2012 report. Geneva: World Health
Organization; 2012 (http://www.countdown2015mnch.org/reports-and-articles/previous-reports/2012-
report, accessed 11 March 2015).
25. World Health Organization, United Nations Children’s Fund. Every newborn: an action plan to end
preventable deaths. Geneva: World Health Organization; 2014 (http://www.everynewborn.org/Documents/
Full-action-plan-EN.pdf, accessed 11 March 2015).
26. Commission on Information and Accountability for Women’s and Children’s Health. Keeping promises,
measuring results. Geneva: World Health Organization; 2011 (http://www.who.int/topics/millennium_
development_goals/accountability_commission/Commission_Report_advance_copy.pdf, accessed 11 March
2015).
27. Independent Expert Review Group. Every woman, every child: a post-2015 vision. The third
report of the independent Expert Review Group on Information and Accountability for Women’s
and Children’s Health. Geneva: World Health Organization; 2014 (http://apps.who.int/iris/
bitstream/10665/132673/1/9789241507523_eng.pdf, accessed 11 March 2015).
28. UN Commission on Life-Saving Commodities for Women and Children. Commissioners’ Report: September
2012, New York, NY: United Nations; 2012 (http://www.everywomaneverychild.org/images/UN_
Commission_Report_September_2012_Final.pdf, accessed 11 March 2015).
29. Committing to child survival: a promise renewed – Progress report 2012. New York, NY: United Nations
Children’s Fund; 2012 (http://www.unicef.org/publications/index_65820.html, accessed 11 March 2015).
62
REFERENCES
32. The DHS Program: wealth index [website]. Rockville, MD: United States Agency for International
Development; 2014 (http://dhsprogram.com/topics/wealth-index/Index.cfm, accessed 11 March 2015).
33. Reproductive health strategy to accelerate progress towards the attainment of international development
goals and targets. Global strategy adopted at the 57th World Health Assembly. Geneva: World Health
Organization; 2004 WHO/RHR/04.8 (http://www.who.int/reproductivehealth/publications/general/
RHR_04_8/en/, accessed 11 March 2015).
34. Singh S, Darroch JE. Adding it up: costs and benefits of contraceptive services – estimates for 2012. New
York, NY: Guttmacher Institute and New York, NY: United Nations Population Fund; 2012.
35. Ensuring human rights in the provision of contraceptive information and services: guidance and
recommendations. Geneva: World Health Organization; 2014 (http://www.who.int/reproductivehealth/
publications/family_planning/human-rights-contraception/en/, accessed 11 March 2015).
36. World Health Organization, United Nations Children’s Fund, United Nations Population Fund, World
Bank, United Nations Population Division. Trends in maternal mortality: 1990 to 2013. Geneva: World
Health Organization; 2014 (http://www.who.int/reproductivehealth/publications/monitoring/maternal-
mortality-2013/en/, accessed 11 March 2015).
37. Task Force on Child Health and Maternal Health. Who’s got the power? Transforming health systems for
women and children. London: Earthscan; 2005.
38. Progress report – reproductive health strategy. To accelerate progress towards the attainment of international
development goals and targets. Geneva: World Health Organization; 2010 (http://www.who.int/
reproductivehealth/publications/general/rhr_10_14/en/, accessed 11 March 2015).
39. Gwatkin D, Rutstein S, Johnson K, Suliman E, Wagstaff A, Amouzou A. Socio-economic differences in health,
nutrition, and population within developing countries. Washington (DC): World Bank; 2007.
40. Liu L, Oza S, Hogan D, Perin J, Rudan I, Lawn JE et al. Global, regional, and national causes of child mortality
in 2000–13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet.
2014;385(9966):430–40.
41. Pneumonia. Fact sheet 331. Geneva: World Health Organization; 2014 (http://www.who.int/mediacentre/
factsheets/fs331/en/, accessed 11 March 2015).
42. World Health Organization, United Nations Children’s Fund. Global immunization data. July 2014. Geneva:
World Health Organization; 2014 (http://www.who.int/immunization/monitoring_surveillance/global_
immunization_data.pdf, accessed 11 March 2015).
43. Levels & trends in child malnutrition. Joint child malnutrition estimates. New York, NY: United Nations
Children’s Fund, Geneva: World Health Organization and Washington (DC): World Bank; 2013 (http://www.
who.int/nutgrowthdb/summary_jme_2013.pdf, accessed 11 March 2015).
44. Stewart CP, Iannotti L, Dewey KG, Michaelsen KF, Onyango AW. Contextualising complementary feeding in a
broader framework for stunting prevention. Matern Child Nutr. 2013;9 (Suppl. 2):27–45.
45. Global Health Observatory (GHO) data: Under-five mortality [online database]. Geneva: World Health
Organization; 2014 (http://www.who.int/gho/child_health/mortality/mortality_under_five/en/, accessed 11
March 2015).
46. United Nations Inter-agency Group for Child Mortality Estimation. Levels & trends in child mortality:
Report 2014. New York, NY: United Nations Children’s Fund; 2014 (http://www.data.unicef.org/fckimages/
uploads/1410869227_Child_Mortality_Report_2014.pdf, accessed 11 March 2015).
63
47. Global Health Observatory (GHO) data: Causes of child mortality [online database]. Geneva: World Health
Organization; 2014 (http://www.who.int/gho/child_health/mortality/causes/en/, accessed 11 March 2015).
48. Children: reducing mortality. Fact sheet 178. Geneva: World Health Organization; 2014 (http://www.who.int/
mediacentre/factsheets/fs178/en/, accessed 11 March 2015).
49. Handbook on health inequality monitoring with a special focus on low- and middle-
income countries. Geneva: World Health Organization; 2013 (http://apps.who.int/iris/
bitstream/10665/85345/1/9789241548632_eng.pdf?ua=1, accessed 11 March 2015).
50. Kuruvilla S, Schweitzer J, Bishai D, Chowdhury S, Caramani D, Frost L et al. Success factors for reducing
maternal and child mortality. Bull World Health Organ. 2014;92(7):533–44B.
51. Success factors for women’s and children’s health: how some countries are accelerating progress towards
MDGs 4 and 5. Geneva: World Health Organization; 2014 (http://www.who.int/pmnch/knowledge/
publications/success_factors_overview.pdf, accessed 11 March 2015).
52. Shaping the future for health women, children & adolescents: learn more about the process to update the
Global Strategy [website]. Every Woman Every Child; 2014 (http://www.everywomaneverychild.org/global-
strategy-2, accessed 11 March 2015).
53. A world that counts: mobilizing the data revolution for sustainable development [edocument]. Independent
Expert Advisory Group on a Data Revolution for Sustainable Development (http://www.undatarevolution.
org/wp-content/uploads/2014/12/A-World-That-Counts2.pdf, accessed 11 March 2015).
54. Monitoring framework for the post-2015 health goals of the SDGs: targets and indicators [draft]. Geneva:
World Health Organization; 2014 (http://unstats.un.org/unsd/post-2015/activities/egm-on-indicator-
framework/docs/Background%20note%20by%20WHO-%20Health%20SDG%20Targets%20and%20
Indicators_EGM-Feb2015.pdf, accessed 30 March 2015).
55. Open Working Group of the General Assembly on Sustainable Development Goals. Open Working Group
proposal for Sustainable Development Goals; 2014 (https://sustainabledevelopment.un.org/content/
documents/1579SDGs%20Proposal.pdf, accessed 30 March 2015).
64
Appendices
Appendix 1. Data and analysis methods
Data
Data sources
Health indicator and dimension of inequality data were sourced from publicly available Demographic and
Health Surveys (DHS) – rounds three, four, five and six – and Multiple Indicator Cluster Surveys (MICS) –
rounds three and four. DHS and MICS are large-scale, nationally representative household health surveys
that collect data through standardized, face-to-face interviews with women aged 15–49 years in low- and
middle-income countries (1, 2). Country income group was determined using the World Bank classification
as of July 2014 (3).
The disaggregated data used in this report are the product of a reanalysis of DHS and MICS micro-data by
the International Center for Equity in Health based in the Federal University of Pelotas, Brazil, and can be
freely accessed from the World Health Organization (WHO) Health Equity Monitor database (4).
The survey tools used by DHS and MICS permit direct comparisons between surveys, and the analyses
in this report assume that the survey design and implementation quality are sufficiently similar between
DHS and MICS, across countries and over time (5, 6). The data were taken from rounds of DHS and MICS
that were not conducted in the same year in all countries. In a few cases, there may be minor differences
between the data reported here and in previous DHS or MICS country reports due to small discrepancies
in the time span, definition and/or calculation of some indicators.
Health indicator data

Reproductive, maternal, newborn and child health (RMNCH) indicators were selected for inclusion in this
report based on the global importance of the indicators and data availability. The RMNCH intervention
indicators included in this report are similar to those used by the Countdown to 2015 initiative and the
Commission on Information and Accountability for Women’s and Children’s Health, and overlap with the
indicators for the Millennium Development Goals (MDGs) targets. The selected intervention indicators
represent the continuum of care from reproductive to maternal health, and from newborn to child health;
the child malnutrition and mortality indicators also have direct relevancy to the topic.
A description of each of the health indicators used in the report is provided in Supplementary table S1.
Detailed information about the criteria used to calculate the numerator and denominator values for each
indicator is available in the WHO Indicator and Measurement Registry, under the topic “Health Equity
Monitor” (7). Note that the definitions for two of the maternal health intervention indicators used in this
65
report (births attended by skilled health personnel and antenatal care coverage (at least one visit)) differ
from the official WHO definitions, which employ a uniform definition of skilled health personnel across
all countries (only doctors, nurses and midwives). In this report, the definition of skilled health personnel
is country specific, and relevant health professionals were determined for each country by reviewing all
DHS and MICS country reports. For child immunization indicators, the reference age group used in the
denominator (12–23 months) was adjusted to align with alternative immunization schedules adopted in
certain countries (18–29 months or 15–26 months).
The year associated with the survey reflects the year of data collection, but is not necessarily the status
of the indicator during that year (Supplementary table S1). Mortality indicators, for example, refer to the
10-year period prior to the survey. We also note that there are minor discrepancies for indicators related
to maternal health interventions and early initiation of breastfeeding, as DHS data for these indicators are
based on live births in the three years prior to the survey and MICS data are based on live births in the
two years prior to the survey (6).
Dimensions of inequality data

Health data were disaggregated by four dimensions of inequality: economic status, education level, place
of residence and sex. Economic status was determined at the household level, using a wealth index.
Country-specific indices were based on owning selected assets and having access to certain services,
and constructed using principal component analysis. Within each country the index was used to create
quintiles, thereby identifying five equal subgroups that each account for 20% of the population (8, 9). Note
that certain indicators have denominator criteria that do not include all households and/or are more likely
to include households from a specific quintile; thus the share of the population for a given indicator may
not equal 20%. For example, the birth rate in the poorest quintile is often higher than in the richest quintile,
resulting in the poorest quintile representing a larger share of the affected population (number of live births)
for indicators such as births attended by skilled health personnel.
Education level refers to the highest level of schooling attained by the woman or, in the case of newborn and
child indicators, the mother. Three subgroups were defined: no education, primary school and secondary
school or higher.
For place of residence classifications (that is, urban or rural), country specific criteria were applied.
Country selection
Countries were selected for inclusion in our analyses based on data availability and survey year. When a
survey was conducted over more than one calendar year, the year of survey was assigned based on the
initial year of data collection. Low- and middle-income countries with surveys from the past 10 years were
chosen to illustrate the “latest situation” of inequality. This included 86 countries (30 low-income countries
and 56 middle-income countries) with the year of their most recent survey falling between 2005 and 2013
(54 DHS and 32 MICS). Countries encompassed all WHO regions: 36 countries from the African Region,
12 countries from the Region of the Americas, 8 countries from the South-East Asia Region, 15 countries
from the European Region, 9 countries from the Eastern Mediterranean Region and 6 countries from the
Western Pacific Region (Supplementary table S2).
66
APPENDICES
For the sake of consistency, the most recent DHS or MICS conducted during the period 2005–2013 was
selected for each country, and then that survey was evaluated for health indicator data availability. If
indicator data were not available from the most recent survey, the country was not included in analyses for
that indicator, regardless of whether that country had relevant data from an older survey conducted during
the period 2005–2013. Thus, the number of study countries included in analyses for each indicator was
variable. For example, data for the indicators antenatal care coverage (at least four visits) and demand for
family planning satisfied were only available from DHS and round four of MICS. Mortality indicator data
were taken from DHS but not MICS, excluding from mortality-related analyses in two countries that had
a post-2005 DHS but a more recent MICS: Ghana (DHS 2008) and Swaziland (DHS 2006).
“Change over time” was analysed for each study country that had surveys from two time points (a recent
survey falling between 2005 and 2013 and an older survey falling between 1995 and 2004), and reflects
the change in national averages and inequalities within countries over a period of about 10 years. Change-
over-time analyses were possible for 42 countries (19 low-income countries and 23 middle-income
countries) from all WHO regions (24 countries from the African Region, 5 countries from the Region of the
Americas, 4 countries from the South-East Asia Region, 4 countries from the European Region, 2 countries
from the Eastern Mediterranean Region and 3 countries from the Western Pacific Region). The number
of years between surveys within countries ranged from 5 to 15 years, as per data availability, though for
the majority of countries this gap was 10 or 11 years. When more than one older survey was available, the
survey closest to 10 years prior to the most recent survey was selected.
With the exception of reference tables and maps, study countries were excluded on a case-by-case basis if
data about the relevant health indicator and/or dimension of inequality were not available, or if the sample
size was too low to report a valid estimate for one or more of the relevant subgroups (that is, less than
25 cases, or in the case of mortality indicators, less than 250 unweighted person-years of exposure to the
risk of death). For example, there were 54 countries included in the latest-situation analysis of demand
for family planning satisfied by education level: from the pool of 86 countries, 23 countries were excluded
because the country’s most recent survey did not have data about the indicator, 5 countries were excluded
because the survey did not have data about women’s education levels according to the classification used
in this report, and 4 countries were excluded because the sample size was less than 25 cases in any one
education subgroup.
In all visuals, situations of low sample size were noted. This included estimates based on 25–49 cases, or
in the case of mortality indicators, 250–499 unweighted person-years of exposure to the risk of death. In
reference tables and maps, all countries were included (even if data from one or more subgroups were not
listed); missing data were flagged as not available, not reported or not calculated.
In 13 MICS, education was classified according to different criteria than those applied in other surveys,
and subgroup data could not be reasonably compared with those from other study countries. Data from
these 13 surveys were thus excluded from subsequent disaggregation and analyses by education. These
surveys were conducted in Albania (MICS 2005), Belize MICS (2006), Cuba (MICS 2006 and MICS 2010),
Georgia (MICS 2005), Guyana (MICS 2006), Kazakhstan (MICS 2006 and MICS 2010), Kyrgyzstan
(MICS 2005), Montenegro (MICS 2005), Serbia (MICS 2005), Ukraine (MICS 2005) and Uzbekistan
(MICS 2006). In five DHS, mortality data could not be calculated for all education subgroups; these
data were also excluded from further disaggregation and analyses by education. This applied to surveys
conducted in Kazakhstan (DHS 1999), Kyrgyzstan (DHS 1997), Republic of Moldova (DHS 2005), Ukraine
(DHS 2007) and Uzbekistan (DHS 1996).
67
The composite coverage index for a given subgroup was not calculated when any of the eight component
indicators had a sample size of less than 25 cases.
Determinants of health
Two databases were searched for potentially relevant determinants of health: the World Bank DataBank (10)
and the WHO Global Health Observatory (11). Following consultation with WHO experts, 23 relevant
factors were selected to showcase the associations between health indicators and their determinants
(Table A1.1). Correlation analyses were also done to show the relationship between health outcomes and
selected RMNCH interventions: contraceptive prevalence (modern methods); demand for family planning
satisfied; antenatal care coverage (at least one visit); births attended by skilled health personnel; DTP3
immunization coverage among one-year-olds; full immunization coverage among one-year-olds; and
composite coverage index.
TABLE A1.1 Health determinants included in correlation analyses with RMNCH indicators, grouped according to the EQuity-oriented
Analysis of Linkages between health and other sectors (EQuAL) Framework*
Health determinant Source

Environmental quality World Bank DataBank
• Access to electricity (% of population)
• Population using improved drinking-water sources (%)** WHO Global Health
• Population using improved sanitation facilities (%)** Observatory
• Road density (kilometre of road per 100 square kilometres of land area)
Accountability and inclusion World Bank DataBank
• Labour force participation rate for ages 15–24 years, female (%)
• Literacy rate for ages 15–24 years, female (%)
• Primary school completion rate, female (% of relevant age group)
• Ratio of female to male labour force participation rate
Livelihoods and skills World Bank DataBank
• GDP per capita, PPP (current international dollars)
• GNI per capita, PPP (current international dollars)
• Gini index
• Labour force participation rate for ages 15–24 years, total (%)
• Literacy rate for ages 15–24 years, total (%)
• Mobile cellular subscriptions (per 100 people)
• Poverty headcount ratio at national poverty line (% of population)
Demography World Bank DataBank
• Age dependency ratio (% of working-age population)
• Population (total)
• Population aged 0–14 years (% of total)
• Population growth (% per year)
• Urban population (% of total)
Health system inputs WHO Global Health
• Per capita government expenditure on health, PPP (international dollars) Observatory
• Per capita total expenditure on health, PPP (international dollars)
• Physicians density (per 1000 population)
GDP: gross domestic product; GNI: gross national income; PPP: purchasing power parity.
* More information about the EQuAL Framework will be available in a forthcoming WHO report Monitoring health determinants for equity (in
preparation).
** Determinant was only analysed for health outcomes (and not health interventions).
68
APPENDICES
For each country, data about the determinants of health were matched to the year of the most recent
survey used to derive the health indicator estimates. The criterion for matching was as follows: the date of
data collection for the determinant must fall within the five years prior to the collection of the most recent
RMNCH indicator data (the observation was dropped if the determinant of health data were collected
after the RMNCH indicator data, or if the time difference between the two was greater than five years).
Analysis
Data disaggregation
Micro-level DHS and MICS data were used to generate national average and disaggregated estimates for
each indicator. Survey design specifications were taken into consideration in the estimation. The same
methods of calculation were applied across all surveys to generate comparable estimates across countries
and over time.
Health data were disaggregated at the country level according to four dimensions of inequality. In the
interactive visuals, point estimates of disaggregated data are presented alongside 95% confidence intervals
(CIs) and the population share of the subgroup. The population share for each indicator is the percentage
of the affected population – the indicator denominator – represented by the subgroup in a given country.
RMNCH indicator estimates are presented separately by subgroup within each country. Median values
of these country-level estimates in each subgroup show the global level of the indicator (or by country
income group). Comparing the median values across subgroups indicates the extent of health inequality
at the global level (or by country income group).
Summary measures
Two measures, difference and ratio, show absolute and relative inequality, respectively, between two
subgroups within the same country (Table A1.2).
TABLE A1.2 Difference and ratio summary measure calculations by dimension of inequality
Indicator type Dimension of inequality Difference calculation Ratio calculation

Desirable Economic status richest quintile – poorest quintile richest quintile / poorest quintile
(health interventions Education secondary school or higher – secondary school or higher / no
such as antenatal care) no education education
Place of residence urban – rural urban / rural
Sex females – males females / males
Undesirable Economic status poorest quintile – richest quintile poorest quintile / richest quintile
(health outcomes such Education no education – secondary no education / secondary school
as child mortality) school or higher or higher
Place of residence rural – urban rural / urban
Sex males – females males / females
Reference subgroups for difference and ratio were selected based on convenience of data interpretation (that is, providing positive values for difference
calculations and values above one for ratio calculations). For example, the poorest/no education/rural/males subgroups tended to have higher child
mortality or higher prevalence of child malnutrition than the richest/secondary school or higher/urban/females subgroups, respectively. In the case of
sex, this selection does not represent an assumed advantage of one sex over the other.
69
Population attributable risk was calculated as the difference between the level of RMNCH intervention
coverage in the most-advantaged subgroup (richest quintile, secondary school or higher, or urban) and
the national average. Generally, the subtraction yielded a positive value; however, in exceptional cases the
result was a negative value (indicating that coverage in the most-advantaged subgroup was lower than
the national average). For the purposes of this report, negative population attributable risk values were
reassigned to zero to convey that there would be no improvement in the national average if coverage in
the total population reached the level of the most advantaged.
For change over time in a given health indicator, the annual change calculation indicates the average change
per year within a subgroup of a given country. Annual change was calculated using the number of years
between survey one (falling between 1995 and 2004) and survey two (falling between 2005 and 2013),
a value that ranged from 5 to 15 years. Annual absolute change was calculated as the difference in health
indicator level at the two survey points (most recent survey data minus older survey data), divided by the
gap between survey years. A positive value indicates increasing level of the health indicator; that is, an
increase in health intervention coverage (a favourable event) or an increase in child mortality or malnutrition
prevalence (a non-favourable event). A negative value indicates decreasing level of the health indicator; that
is, a decrease in health intervention coverage or a decrease in child mortality or malnutrition prevalence.
Annual absolute excess change compares the pace of change in the most-disadvantaged subgroup (poorest
quintile, no education or rural) with that in the most-advantaged subgroup (richest quintile, secondary
school or higher, or urban). It is calculated as the annual absolute change in the disadvantaged subgroup
minus the annual absolute change in the advantaged subgroup. In the case of sex, excess change compares
the pace of change in males with that in females, and is calculated as the annual absolute change in males
minus the annual absolute change in females. This selection does not represent an assumed advantage
of one sex over the other. More information about the interpretation of annual absolute excess change is
available in Appendix 2.
For all summary measures, values between -0.1 and +0.1 (percentage points or deaths per 1000 live births)
were interpreted as no inequality (in case of latest status analyses) or no change over time. The same logic
was applied to evaluate change in national average over time.
References
1. The DHS Program [website]. Rockville, MD: United States Agency for International Development; 2015 (http://
dhsprogram.com, accessed 11 March 2015).
2. Statistics and monitoring: Multiple Indicator Cluster Survey (MICS) [website]. New York, NY: United Nations
Children’s Fund; 2015 (http://www.unicef.org/statistics/index_24302.html, accessed 11 March 2015).
3. Data: Country and lending groups [website]. Washington (DC): World Bank; 2014 (http://data.worldbank.org/
about/country-and-lending-groups#OECD_members, accessed 11 March 2015).
4. Global Health Observatory (GHO) data: Health Equity Monitor [online database]. Geneva: World Health
Organization; 2013 (http://www.who.int/gho/health_equity/en/, accessed 11 March 2015).
5. Requejo J, Victora C, Bryce J. Data resource profile: countdown to 2015: maternal, newborn and child survival.
Int J Epidemiol. 2014;43(2):586–96.
6. Hancioglu A, Arnold F. Measuring coverage in MNCH: tracking progress in health for women and children using
DHS and MICS household surveys. PLoS Med. 2013;10(5):e1001391.
70
APPENDICES
7. Global Health Observatory (GHO) data: WHO indicator registry [online database]. Geneva: World Health
Organization; 2015 (http://www.who.int/gho/indicator_registry/en/, accessed 11 March 2015).
8. Rutstein S, Johnson K. DHS comparative reports no. 6: the DHS wealth index. Calverton, MD: ORC Macro;
2004.
9. The DHS Program: wealth index [website]. Rockville, MD: United States Agency for International Development;
2014 (http://dhsprogram.com/topics/wealth-index/Index.cfm, accessed 11 March 2015)
10. Data: World Bank open data [website]. Washington (DC): World Bank; 2015 (http://data.worldbank.org/,
11. Global Health Observatory (GHO) data: Data repository [online database]. Geneva: World Health Organization;
2015 (http://www.who.int/gho/database/en/, accessed 11 March 2015).
71
Appendix 2. Assessing health inequality: methodological

considerations
General guidelines for assessing and reporting the state of inequality include (1):
• Disaggregated data and summary measures should be reported together. Disaggregated data give a
sense of the underlying level of health, and summary measures enhance interpretation and reporting
by expressing inequality in a single number.
• Both relative and absolute summary measures should be considered and/or reported to give a sense
of the magnitude of the difference between two subgroups and also to provide a relative comparison
between two subgroups.
• National average should be provided alongside inequality estimates to provide a more complete
assessment of the situation.
• Accounting for population share when reporting inequality provides a more nuanced indication of how
inequalities exist within populations.
• When assessing change over time, it is important to consider the baseline level of health, as there is
greater room for improvement in situations of poor performance at baseline. This is often the case when
comparing the progress in disadvantaged and advantaged subgroups, as the disadvantaged often have
lower levels of health at baseline and therefore a larger margin of improvement is possible.
• When interpreting estimates that approach lower and upper limits it is important to consider:
- as the overall coverage of a health intervention approaches 100%, the difference and ratio values
typically decrease;
- if the coverage in the disadvantaged group is very low, the resulting ratio (between disadvantaged and
advantaged subgroups) may be very high.
Confidence intervals
An important feature of these analyses is the inclusion of 95% confidence intervals (CIs) for point estimates
of disaggregated data. CIs can be accessed in the interactive visualization components that accompany
this report.
Reporting CIs can help users of the data to judge whether there are statistical differences between
subgroups. Some caution is required, however, when using confidence measurements to draw conclusions
about health inequality data. Estimates that are derived from large samples may prove to be statistically
different mathematically, but in the realm of public health this difference may not be meaningful. For
example, there was a statistical difference in the prevalence of demand for family planning satisfied in
Peru between rural areas (88.8%) and urban areas (90.9%). However, in terms of public health policies,
programmes and practices, this 2.1% difference bears little importance.
Nevertheless, this does not mean that CIs should be ignored when reporting data. Rather, there is a need
to ensure that point estimates do not lead to false conclusions and misinformed policy. This includes
considering whether the CIs of the point estimates are narrow enough to allow valid conclusions about
inequality to be drawn. For example, CIs for estimates of the treatment of sick children are important to
help indicate the underlying precision of the data. The sample size of the affected population for these
indicators is typically much larger among poorer households than richer households. The CI for the richest
quintile may be wider, and therefore estimates for the richest quintile may be less precise.
72
APPENDICES
Population share
When reporting disaggregated data, providing information about the population share in each subgroup
allows for a more thorough interpretation of whether inequalities and changes in inequality over time are
meaningful.
An example from Indonesia demonstrates the importance of accounting for population share. Looking at
the disaggregated data presented in Figure A2.1, the education-related inequality in antenatal care coverage
(at least four visits) appears to have increased between 2002 and 2012 due to hastened increases in the
primary school and secondary school or higher subgroups, and decreasing coverage in the no education
subgroup. Based on this information, a logical conclusion would be that education-related inequality has
increased over time.
FIGURE A2.1 Antenatal care coverage (at least four visits) by woman’s education in Indonesia (DHS 2002 and 2012)
100
90
80
70
60
Coverage (%)
50
40
30
20
10
0
2002 2012
n No education n Primary school n Secondary school +
Considering the population share, however, brings another layer of meaning to the data. The data presented
in Figure A2.2 show the proportion of women with a live birth within three years preceding the survey
in Indonesia, by education level, in DHS 2002 and 2012. Information about population share reveal that
there was a substantial increase in the proportion of women belonging to the secondary school or higher
subgroup, and a decrease in the proportion of women in the no education subgroup. Evidently, there was
a population shift between education subgroups over the 10-year period in Indonesia.
Inequality monitoring should draw upon summary measures that take into account disaggregated data and
population share across subgroups. For example, summary measures such as the slope index of inequality
and the concentration index provide a more sophisticated estimation of inequality than simple measures of
inequality (1). By taking into account population shift in the above example, these measures may indicate
a decrease in inequality over time, and thus are appropriate measures to capture the impact of upstream
policies, such as those promoting education among women. However, such measures require more effort
to understand and were not used in this report.
73
FIGURE A2.2 Proportion of women aged 15–49 years with a live birth within three years preceding the survey by education in
Indonesia (DHS 2002 and 2012)
100
90
80
70
Population share (%) 60
50
40
30
20
10
0
2002 2012
n Secondary school + 48.5 67.2
n Primary school 46.9 30.9
n No education 4.6 1.9
INTERPRETING 95% CONFIDENCE INTERVALS AND POPULATION SHARE: AN EXAMPLE
Maldives reported a stunting prevalence of 15.4% in children less than five years of age who were born to mothers with a
secondary school or higher level of education, with a 95% CI ranging from 13.1 to 18.0%. The share of the population in this
subgroup (mother’s education: secondary school or higher) was 49.7%, meaning that about half of the children under five
years of age in Maldives represented by the DHS 2009 had a mother with a secondary school or higher level of education.
Maldives, DHS 2009

Stunting prevalence in children aged < 5 years (%)
By mother’s education: Secondary school + (49.7% of affected population)
Estimate: 15.4%; 95%CI: 13.1–18.0
National average: 18.0%; 95%CI: 16.0–20.1
Interpreting absolute excess change

Excess change calculations compare the pace of change in two population subgroups over a number
of years. In this report, the annual absolute excess change value is interpreted differently for health
interventions (where increased intervention coverage is desired) than for health outcomes (where
decreased malnutrition and mortality are desired). Several possible scenarios are possible, as detailed in
Table A2.1.
74
APPENDICES
TABLE A2.1 Guide to interpreting annual absolute excess change estimates

Direction of annual
Sign of annual absolute change
absolute excess in advantaged and
change (nature of disadvantaged
Indicator type change) subgroups Pace of annual absolute change in subgroups
Desirable Positive value Increasing in both* Increase occurred faster in the disadvantaged subgroup
(health (pro-disadvantaged than in the advantaged subgroup
interventions such change) Decreasing in both Decrease occurred slower in the disadvantaged subgroup
as antenatal care) than in the advantaged subgroup
Mixed directions Increase (or no change) occurred in the disadvantaged
subgroup and decrease (or no change) occurred in the
advantaged subgroup
Negative value Increasing in both Increase occurred slower in the disadvantaged subgroup
(pro-advantaged than in the advantaged subgroup
change) Decreasing in both** Decrease occurred faster in the disadvantaged subgroup
than in the advantaged subgroup
Mixed directions Decrease (or no change) occurred in the disadvantaged
subgroup and increase (or no change) occurred in the
advantaged subgroup
Zero value*** No change in either
Same direction of change Subgroups reported the same pace of change
Undesirable Negative value Decreasing in both* Decrease occurred faster in the disadvantaged subgroup
(health outcomes (pro-disadvantaged than in the advantaged subgroup
such as child change) Increasing in both Increase occurred slower in the disadvantaged subgroup
mortality) than in the advantaged subgroup
Mixed directions Decrease (or no change) occurred in the disadvantaged
subgroup and increase (or no change) occurred in the
advantaged subgroup
Positive value Decreasing in both Decrease occurred slower in the disadvantaged subgroup
(pro-advantaged than in the advantaged subgroup
change) Increasing in both** Increase occurred faster in the disadvantaged subgroup
than in the advantaged subgroup
Mixed directions Increase (or no change) occurred in the disadvantaged
subgroup and decrease (or no change) occurred in the
advantaged subgroup
Zero value*** No change in either
Same direction of change Subgroups reported the same pace of change
* Indicates the best-case scenario of improved national average and narrowing inequality.
** Indicates the worst-case scenario of worsening national average and widening inequality.
*** Absolute excess change values in the range of -0.1 to +0.1 percentage points or deaths per 1000 live births were interpreted as no change.
Reference
1. Handbook on health inequality monitoring with a special focus on low- and middle-income countries. Geneva:
World Health Organization; 2013 (http://apps.who.int/iris/bitstream/10665/85345/1/9789241548632_eng.
pdf?ua=1, accessed 11 March 2015).
75
Appendix 3. Visualizing disaggregated data using maps

Figure A3.1 displays data about the prevalence of child malnutrition in 74 low- and middle-income countries,
disaggregated by the mother’s level of education. Data about the health indicator, stunting prevalence in
children under the age of five years, are presented according to three educational subgroups: no education,
primary school and secondary school or higher.
The prevalence of stunting not only varied between the three subgroups (comparing between the three map
views), but it also varied between countries for a given subgroup (shown on a single map view). Overall,
stunting prevalence tended to be lower in subgroups with higher levels of education.
FIGURE A3.1 Stunting prevalence in children aged less than five years by mother’s education in 74 low- and middle-income
countries: latest situation (DHS and MICS 2005–2013)
Prevalence (%)
<20.0
20.0–29.9
30.0–39.9
≥40.0 Data not available
NO EDUCATION
0 850 1,700 3,400 Kilometers
Not reported* Not applicable
Prevalence (%)
<20.0
20.0–29.9
30.0–39.9
≥40.0 Data not available
PRIMARY SCHOOL
0 850 1,700 3,400 Kilometers
Not reported* Not applicable
76
APPENDICES
Prevalence (%)
<20.0
20.0–29.9 SECONDARY SCHOOL +
30.0–39.9 Data not available 0 850 1,700 3,400 Kilometers
≥40.0 Not applicable
*Estimate was based on fewer than 25 cases.
VIDEO CLIP 3. HEALTH INEQUALITIES ARE WIDESPREAD
Watch this short video clip to explore the question: how AVAILABLE ON CD/USB SCAN HERE
much does child malnutrition vary across education
subgroups within countries?
state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland
ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
ht is tr
wor e d
ld h or fre
77
Appendix 4. Guide to interpreting the visuals used in this report

Effective conventional and interactive data visuals contribute to the enhanced communication of health
inequality data. A brief guide to interpreting the visuals used in this report is provided in Table A4.1.
TABLE A4.1 Interpreting the visuals used in this report

Description of visual
(type of data) Snapshot Interpretation
Horizontal bar chart Thailand
Costa Rica
Median=30.2
Horizontal bars show the point

(national data) estimate of the national value for
Cuba
Colombia
Dominican Republic
each country.
Bhutan
Honduras
Swaziland
Viet Nam
The vertical line indicates the

Uzbekistan
Indonesia
Egypt
median value (middle point of all

Zimbabwe
Namibia
Bangladesh
values).
Peru
Belize
Belarus
Mongolia
Kazakhstan
India
Ukraine
Suriname
Lesotho
Rwanda
Republic of Moldova
Nepal
0 10 20 30 40 50 60 70 80 90 100
National average (%)
Horizontal circle plot, by Thailand

Costa Rica Disaggregated data (point
country (disaggregated estimates) for each population
Colombia
Dominican Republic
Bhutan
data) subgroup are presented by country.

Honduras
Swaziland
Viet Nam
Indonesia
Shading indicates the difference

Egypt
Zimbabwe
Namibia
(absolute inequality) between the

Bangladesh
Peru
Belize
most extreme subgroup values.

Mongolia
India
Suriname
Lesotho
Rwanda
Nepal
Syrian Arab Republic
Jordan
Malawi
Lao People's Democratic Republic
Guyana
Kenya
0 10 20 30 40 50 60 70 80 90 100
Coverage (%)
Box-and-whisker plot Box-and-whisker plots give

(disaggregated data) No education Primary school Secondary school +
information about the distribution
of a set of data (for example,
100
multiple country estimates for a

90
80
70
subgroup) without listing all data
points.
Coverage (%)
60
50
The top and bottom lines indicate
maximum and minimum values,
40
the centre line indicates the median

30
20
10
value and the shaded box indicates
0 the interquartile range (middle 50%
of study country estimates).
78
APPENDICES
Vertical circle graph No education Primary school Secondary school + Circles represent the point
(disaggregated data) 100
estimate for each country within
90
each subgroup: each country
is represented by one circle per
80
70
60
subgroup.
Coverage (%)
50
Horizontal grey lines indicate the
median values and the grey band
40
30
20
indicates the interquartile range.
10
Vertical circle graph, Circles represent the value of the

by country (summary specified summary measure for each
measures) 60 country.
50 Summary measures include:
difference or ratio value (calculated
40
using the most-advantaged and
30
most-disadvantaged subgroups);
change over time (annual absolute
20
change in the most-advantaged or
10
most-disadvantaged subgroup);
and annual absolute excess change
0 value (comparing pace of change
-10
in the most-advantaged and most-
disadvantaged subgroup).
-20
The horizontal grey line indicates
the median value and the grey band
indicates the interquartile range.
Scatterplot – four Each country is represented by one
quadrant view shape.
(national data and 60
For latest situation, the national
summary measures) 50 average (x-axis) is plotted against
40 the difference (absolute inequality)
30 between the most extreme
20
subgroups (y-axis) for each country.
10
For change over time, the annual
0
change in the national average
-10 (x-axis) is plotted against the annual
-20 absolute excess change (y-axis) for
0 10 20 30 40 50 60 70 80 90 100
each country.
The horizontal and vertical orange
lines indicate the median values.
79
Change-over-time line No education Primary school Secondary school + Each country is represented by one
plot (disaggregated 100
line per subgroup.
data) 90
80 Lines indicate the change over

70
time for each country within each
subgroup.
Coverage (%)
60
50
40
30
20
10
0
Survey 1 Survey 2 Survey 1 Survey 2 Survey 1 Survey 2
Map with vertical bar Maps show data for a selected

charts (disaggregated subgroup across countries.
data)
The bar chart at the bottom
left displays data for a selected
subgroup across countries.
The bar chart at the bottom right
displays disaggregated data for a
selected dimension of inequality
across all subgroups of a selected
country.
80
APPENDICES
Appendix 5. Interactive visualization of health data

Interactive data visualization technology can help to enhance the accessibility of health inequality data. This
report links to several types of interactive components: map-based dashboards, story-points (consisting
of a variety of interactive graphs), reference data tables and country profiles. These components reinforce
the content of conventional forms of data presentation employed in this report (text and static tables,
graphs and maps). Interactive components also provide users with tools to explore a large database of
health data for other stories about health inequalities.
Like the more conventional forms of data presentation used in this report, the accompanying interactive
visuals aim to motivate users to engage with the key messages that emerge from data and ultimately take
appropriate steps to monitor and address health inequalities. The effective use of interactive visualization
technology confers certain benefits over conventional forms of data presentation by:
• presenting large amounts of data in an approachable and modifiable format;
• allowing users to select and customize data views, such as choosing a health indicator of interest;
• making it possible to drill down through levels of data views, from the global perspective to a selected
local level;
• enabling users to selectively benchmark in order to assess how countries are performing in relation to
one another; and
• providing a novel and efficient way to explore patterns in health data.
For interactive visuals to be effective, users need to know how to use their features and how to interpret the
underlying messages. Optimally, little effort should be spent on decoding the surface of what one is looking
at – if a visual is well designed with the target audience in mind, this should be almost intuitive. Instead,
the focus should remain on exploring the substantive content of the visual and the health stories within.
Many of the considerations that surround creating effective data visuals depend on the purpose of the
communication product, the needs and abilities of the target audience, and the characteristics of the
data that are being communicated. The possibilities may be limited by the methods and tools that are
available to generate the visual. However, in a general sense, there are certain best practices that have
wide applicability to visualizing health data and are exemplified in the interactive visuals that accompany
this report (Table A5.1).
81
TABLE A5.1 Best practices in interactive visualization of data
Best practice Application

Make the visualization(s) cohesive. Story-points feature comprehensive information about the
Ensure that all parts of the dashboard are relevant to the state of inequality in a selected category of RMNCH.
central theme and messages that are being communicated. ➜ Interactive maps are built around each dimension of
inequality.
Make navigation straightforward. Dashboards are designed with common visual cues to
Provide concise directions to support new users. When guiding signal navigation, such as arrows, tabs and buttons.
readers through multiple dashboards, maintain a consistent
Yellow stickers on the first few story-point views introduce
design and placement of navigation tools.
➜ users to dashboard features, and a help button is available
for more detailed navigation assistance.
The selection menu is placed on the left throughout all
visuals.
Introduce data in a logical manner. Inequalities in health are explained through the use of
Avoid overwhelming users with too much information at once. story-points. Stories feature data and information, building
A good approach is to progress from simple to more complex from national average to within-country inequality and
concepts. ➜ from the latest situation to change over time. Complex
views, such as scatterplots, are introduced first as separate
components, and then combined into one view.
Use interactive features in a meaningful way. Selection, filtering and highlighting options were chosen
Be cognizant of the types of customization that are relevant based on their logical application to aid interpretation.
to the user. ➜ For example, users can click on legend entries to highlight
data.
Use non-interactive (static) aspects of visualization The use of the same background colours, title font/size/
software effectively. placement and layout of dashboards creates a unified look.
When developing interactive dashboards consider how to best ➜
apply formatting, settings and design options.
Emphasize graphics over text. Buttons and tooltips (pop-ups) provide access to additional
In most cases, the graphics of a dashboard should be text without cluttering the main view of the dashboard.
the prominent feature; text should be used sparingly in
dashboards, only when necessary to augment the information
➜
in the graphic.
Use colours purposefully. Data points and labels are colour-coded when appropriate.
Colours can be a valuable feature for conveying meaning and For example, red is used to denote unfavourable situations
patterns in data. All colour assignments should be deliberate.
Note that green–red colour combinations may be difficult for
➜ and blue is used to denote favourable situations.
those who are colour blind.

Make additional information available. Buttons and tooltips contain information about the content
Users should be able to access sufficient information about the of the visual, and technical notes are available on the
underlying data to assess its strengths and limitations. Details dashboards.
about the data sources, data selection criteria and method of
➜
Users are prompted to refer to the text of the report for
analysis should be available.
more detailed information.
RMNCH: Reproductive, maternal, newborn and child health.
82
APPENDICES
Appendix 6. Additional interactive visuals: references for further

data exploration
Audiences have diverse data needs, and may benefit from additional and alternative approaches to data
presentation and reporting. The following interactive visuals present the state of inequality in reproductive,
maternal, newborn and child health (RMNCH) through country profiles, maps, reference tables and a
comprehensive interactive visual of all RMNCH interventions.
INTERACTIVE VISUAL A1. EQUITY COUNTRY PROFILES

benchmarking. state of InequalIty
Country profiles contain all available disaggregated

20, avenue appia
CH-1211 geneva 27
SWitzerland
data related to RMNCH interventions and outcomes

ISBN 978 92 4 156491 5
© n.
co tio
indicators for a selected country.

py ibu
rig tr
ht is
wor e d
ld h or fre
INTERACTIVE VISUAL A2. MAPS

benchmarking.
state of InequalIty
Interactive maps display disaggregated data for RMNCH

20, avenue appia
CH-1211 geneva 27
SWitzerland
indicators by different dimensions of inequality:

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
economic status, education, place of residence and sex.

ht is
wor e d
ld h or fre
INTERACTIVE VISUAL A3. REFERENCE TABLES

benchmarking.
state of InequalIty
Interactive reference tables contain complete data from

20, avenue appia
CH-1211 geneva 27
SWitzerland
all available surveys from 86 low- and middle-income

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
countries.
ht is
wor e d
ld h or fre
83
INTERACTIVE VISUAL A4. RMNCH INTERVENTIONS

benchmarking.
state of InequalIty

20, avenue appia
CH-1211 geneva 27
SWitzerland
the state of inequality in a selected RMNCH intervention

ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
tr
indicator and dimension of inequality.

ht is
wor e d
ld h or fre
84
APPENDICES
Appendix 7. Patterns of inequality

For indicators that have more than two ordered subgroups (that is, multiple subgroups that can be ranked
based on logical criteria, such as wealth quintiles or multiple levels of education), characteristic patterns
of inequality across disaggregated data may be identified. To illustrate, Figure A7.1 displays four patterns
in health intervention coverage data, disaggregated by economic status.
FIGURE A7.1 Patterns of inequality in health intervention coverage by economic status
100
80
60
Coverage (%)
40
20
0
Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5
(poorest) (richest)
n Complete coverage shows a need for continued monitoring

n Marginal exclusion requires action targeted to the underserved
n An incremental linear pattern requires a combined universal and targeted approach
n Mass deprivation requires universal action oriented to the whole population
Each of the four distinct patterns of inequality prompts a different general policy response (1, 2).
• A complete coverage pattern is shown by the green line. All quintiles report nearly 100% coverage, and
thus universal coverage has been achieved. Ongoing monitoring may be warranted to ensure that the
situation remains favourable for all.
• A marginal exclusion pattern – also called top inequality – is represented by the blue line. The indicator
demonstrated much lower coverage in the poorest quintile relative to the other four quintiles. This
scenario calls for a targeted approach, whereby resources are directed at the most disadvantaged.
• An incremental linear (or queuing) pattern is apparent in the orange line. A linear gradient indicates
equal increases across quintiles, moving from the poorest to the richest. This pattern requires an approach
that combines population-wide and targeted interventions.
• A mass deprivation pattern – also called bottom inequality – is indicated by the purple line. Health service
coverage is low or very low in all but the richest quintile. Interventions to address mass deprivation should
target the whole population, investing resources in all (or most) subgroups.
85
The patterns of inequality and their corresponding policy responses are intended to be a general guide for
consideration in policy-making. In isolation, this interpretation of data does not constitute evidence for a
definitive course of action. Other considerations, such as context-specific factors and national priorities,
help to inform decisions about where resources in a country should be focused.
VIDEO CLIP 4. HEALTH INEQUALITIES INFORM POLICIES, PROGRAMMES AND PRACTICES
Watch this short video clip to explore the question: AVAILABLE ON CD/USB SCAN HERE
which countries demonstrate characteristic patterns
of inequality in births attended by skilled health state of InequalIty
personnel, disaggregated by economic status?


20, avenue appia
CH-1211 geneva 27
SWitzerland
ISBN 978 92 4 156491 5
© n.
co
py tio
rig ibu
ht is tr
wor e d
ld h or fre
References
1. Handbook on health inequality monitoring with a special focus on low- and middle-income countries. Geneva:
World Health Organization; 2013 (http://apps.who.int/iris/bitstream/10665/85345/1/9789241548632_eng.
pdf?ua=1, accessed 11 March 2015).
2. Victora CG, Fenn B, Bryce J, Kirkwood BR. Co-coverage of preventive interventions and implications for child-
survival strategies: evidence from national surveys. Lancet. 2005;366(9495):1460–6.
86
Supplementary tables
TABLE S1 Descriptions of reproductive, maternal, newborn and child health indicators
Affected population Time period reflected

Indicator Definition* (denominator) in indicator data
Reproductive health interventions
Contraceptive Percentage of women aged 15–49 years, married Women aged 15–49 Current use at time of
prevalence – modern or in union, who are currently using (or whose years who are married survey
and traditional sexual partner is using) at least one method of or in union
methods contraception, regardless of the method used.
Contraceptive Percentage of women aged 15–49 years, married Women aged 15–49 Current use at time of
prevalence – modern or in union, who are currently using (or whose years who are married survey
methods sexual partner is using) at least one modern or in union
method of contraception. Modern methods
of contraception include female and male
sterilization, oral hormonal pills, intrauterine
device (IUD), male condom, injectables, implant
(including Norplant), vaginal barrier methods, the
female condom and emergency contraception.
Demand for family Percentage of women aged 15–49 years, married Women aged 15–49 Current use at time of
planning satisfied or in union, who are currently using any method years who are in need survey
of contraception, among those in need of of contraception, are
contraception. Women in need of contraception fecund and are married
include those who are fecund but report wanting or in union
to space their next birth or stop child-bearing
altogether.
Maternal health interventions
Antenatal care coverage Percentage of women aged 15–49 years with a Women aged 15–49 Three years preceding
– at least one visit live birth within the period preceding the survey, years with a live birth the survey for DHS, and
attended at least once during pregnancy by within the period two years preceding the
skilled health personnel for reasons related to preceding the survey survey for MICS
the pregnancy. Skilled health personnel include (only the last live birth
doctors, nurses, midwives and other medically- was considered)
trained personnel, as defined according to each
country.
Antenatal care coverage Percentage of women aged 15–49 years with a Women aged 15–49 Three years preceding
– at least four visits live birth within the period preceding the survey, years with a live birth the survey for DHS, and
attended at least four times during pregnancy within the period two years preceding the
by any provider (skilled or unskilled) for reasons preceding the survey survey for MICS
related to the pregnancy. (only the last live birth
was considered)
87

Births attended by Percentage of live births attended during delivery Live births within the Three years preceding
skilled health personnel by skilled health personnel, within the period period preceding the the survey for DHS, and
preceding the survey. Skilled health personnel survey two years preceding the
include doctors, nurses, midwives and other survey for MICS
medically-trained personnel, as defined according
to each country.
Newborn and child health interventions
Early initiation of Percentage of children who were put to the breast Live births within the Three years preceding
breastfeeding within one hour of birth. period preceding the the survey for DHS, and
survey two years preceding the
survey for MICS
Children aged Percentage of children aged 6–59 months who Children aged 6–59 Six months preceding
6–59 months who received a high dose vitamin A supplement months the survey
received vitamin A within the six months prior to the survey. A
supplementation high-dose of vitamin A supplement, according to
the International Vitamin A Consultative Group
definition, refers to doses equal to or greater than
25 000 IU.
BCG immunization Percentage of one-year-olds who have received Children aged 12–23 Two years preceding
coverage among one- one dose of BCG vaccine. months the survey
year-olds Note: some countries
use a different reference
age group of 15–26 or
18–29 months
Measles immunization Percentage of one-year-olds who have received at Children aged 12–23 Two years preceding
coverage among one- least one dose of measles-containing vaccine. months the survey
18–29 months
Polio immunization Percentage of one-year-olds who have received Children aged 12–23 Two years preceding
coverage among one- three doses of polio vaccine. months the survey
18–29 months
DTP3 immunization Percentage of one-year-olds who have received Children aged 12–23 Two years preceding
coverage among one- three doses of DTP3 vaccine. months the survey
18–29 months
Full immunization Percentage of one-year-olds who have received Children aged 12–23 Two years preceding
coverage among one- one dose of BCG vaccine, three doses of polio months the survey
year-olds vaccine, three doses of DTP3 vaccine, and one Note: some countries
dose of measles vaccine. use a different reference
18–29 months
88
SUPPLEMENTARY TABLES

Children aged less Percentage of children aged 0–59 months who Children aged 0–59 Two weeks preceding
than five years with had diarrhoea in the two weeks prior to the months who had the survey
diarrhoea receiving oral survey and received oral rehydration salts. diarrhoea in the two
rehydration salts weeks prior to the
survey
Children aged less Percentage of children aged 0–59 months who Children aged 0–59 Two weeks preceding
than five years with had diarrhoea in the two weeks prior to the survey months who had the survey
diarrhoea receiving oral and were treated with oral rehydration therapy – diarrhoea in the two
rehydration therapy oral rehydration salts or an appropriate household weeks prior to the
and continued feeding solution – and continued feeding. survey
Children aged less Percentage of children aged 0–59 months with Children aged Two weeks preceding
than five years with pneumonia symptoms in the two weeks prior 0–59 months with the survey
pneumonia symptoms to the survey who were taken to an appropriate pneumonia symptoms
taken to a health health provider. in the two weeks prior
facility to the survey
RMNCH interventions, combined
Composite coverage The composite coverage index is a weighted score This indicator is based Not applicable
index reflecting coverage of eight RMNCH interventions on aggregate estimates
along the continuum of care: demand for family
planning satisfied; antenatal care coverage (at
least one visit); births attended by skilled health
personnel; BCG immunization coverage among
one-year-olds; measles immunization coverage
among one-year-olds; DTP3 immunization
coverage among one-year-olds; children aged
less than five years with diarrhoea receiving oral
rehydration therapy and continued feeding; and
children aged less than five years with pneumonia
symptoms taken to a health facility.
Child malnutrition
Stunting prevalence in Percentage of children aged 0–59 months who Children aged 0–59 Current status at time
children aged less than are stunted (defined as more than two standard months of survey
five years deviations below the median height-for-age of
the WHO Child Growth Standards).
Underweight Percentage of children aged 0–59 months who Children aged 0–59 Current status at time
prevalence in children are underweight (defined as more than two months of survey
aged less than five standard deviations below the median weight-
years for-age of the WHO Child Growth Standards).
Wasting prevalence in Percentage of children 0–59 months who are Children aged 0–59 Current status at time
children aged less than wasted (defined as more than two standard months of survey
five years deviations below the median weight-for-height
of the WHO Child Growth Standards).
89

Child mortality
Neonatal mortality rate Probability (expressed as a rate per 1000 live Number of surviving 10 years preceding the
births) of a child born in a specific year or period children at beginning survey
dying in the first 30 days of life,** if subject to of specified age range
age-specific mortality rates of that period. during the specified
time period
Infant mortality rate Probability (expressed as a rate per 1000 live Number of surviving 10 years preceding the
births) of a child born in a specific year or period children at beginning survey
dying before reaching the age of one year, if of specified age range
subject to age-specific mortality rates of that during the specified
period. time period
Under-five mortality Probability (expressed as a rate per 1000 live Number of surviving 10 years preceding the
rate births) of a child born in a specific year or period children at beginning survey
dying before reaching the age of five years, if of specified age range
subject to age-specific mortality rates of that during the specified
period. time period
BCG: one dose of Bacille Calmette-Guérin vaccine; DHS: Demographic and Health Survey; DTP3: three doses of the combined diphtheria, tetanus toxoid
and pertussis vaccine; IU: international unit; MICS: Multiple Indicator Cluster Survey; RMNCH: reproductive, maternal, newborn and child health.
* Detailed information about the criteria used to calculate the numerator and denominator values for each indicator is available from the WHO Indicator
and Measurement Registry, under the topic “Health Equity Monitor” (www.who.int/gho/indicator_registry/en/, accessed 11 March 2015).
** Standard DHS calculations (as applied in this report) specify the first 30 days of life; however, other accepted definitions may specify the first 28 days.
90
TABLE S2 Study countries: survey source(s) and year(s), WHO region and country income group
Country income
Country Survey source(s) and year(s) WHO Region group*
Afghanistan MICS 2010–2011 Eastern Mediterranean Low-income
Albania DHS 2008–2009 European Middle-income
Armenia** DHS 2010, DHS 2000 European Middle-income
Azerbaijan DHS 2006 European Middle-income
Bangladesh** DHS 2011, DHS 1999–2000 South-East Asia Low-income
Belarus MICS 2012 European Middle-income
Belize MICS 2011 Americas Middle-income
Benin** DHS 2011–2012, DHS 2001 African Low-income
Bhutan MICS 2010 South-East Asia Middle-income
Bolivia (Plurinational State of)** DHS 2008, DHS 1998 Americas Middle-income
Bosnia and Herzegovina MICS 2011–2012 European Middle-income
Burkina Faso** DHS 2010, DHS 1998–1999 African Low-income
Burundi DHS 2010 African Low-income
Cambodia** DHS 2010, DHS 2000 Western Pacific Low-income
Cameroon** DHS 2011, DHS 2004 African Middle-income
Central African Republic MICS 2010 African Low-income
Colombia** DHS 2010, DHS 2000 Americas Middle-income
Comoros DHS 2012 African Low-income
Congo DHS 2011–2012 African Middle-income
Costa Rica MICS 2011 Americas Middle-income
Côte d'Ivoire** DHS 2011–2012, DHS 1998–1999 African Middle-income
Cuba MICS 2010–2011 Americas Middle-income
Democratic Republic of the Congo DHS 2013–2014 African Low-income
Djibouti MICS 2006 Eastern Mediterranean Middle-income
Dominican Republic** DHS 2007, DHS 1996 Americas Middle-income
Egypt** DHS 2008, DHS 2000 Eastern Mediterranean Middle-income
Ethiopia** DHS 2011, DHS 2000 African Low-income
Gabon** DHS 2012, DHS 2000 African Middle-income
Gambia MICS 2005–2006 African Low-income
Georgia MICS 2005 European Middle-income
Ghana** MICS 2011, DHS 2003 African Middle-income
Guinea** DHS 2012, DHS 1999 African Low-income
Guinea-Bissau MICS 2006 African Low-income
Guyana DHS 2009 Americas Middle-income
Haiti** DHS 2012, DHS 2000 Americas Low-income
Honduras DHS 2011–2012 Americas Middle-income
India** DHS 2005–2006, DHS 1998–1999 South-East Asia Middle-income
91
Country income
Indonesia** DHS 2012, DHS 2002–2003 South-East Asia Middle-income
Iraq MICS 2011 Eastern Mediterranean Middle-income
Jamaica MICS 2005 Americas Middle-income
Jordan** DHS 2012, DHS 2002 Eastern Mediterranean Middle-income
Kazakhstan** MICS 2010–2011, DHS 1999 European Middle-income
Kenya** DHS 2008–2009, DHS 1998 African Low-income
Kyrgyzstan** DHS 2012, DHS 1997 European Middle-income
Lao People's Democratic Republic MICS 2011–2012 Western Pacific Middle-income
Lesotho** DHS 2009, DHS 2004 African Middle-income
Liberia DHS 2013 African Low-income
Madagascar** DHS 2008–2009, DHS 1997 African Low-income
Malawi** DHS 2010, DHS 2000 African Low-income
Maldives DHS 2009 South-East Asia Middle-income
Mali** DHS 2012–2013, DHS 2001 African Low-income
Mauritania MICS 2007 African Middle-income
Mongolia MICS 2010 Western Pacific Middle-income
Montenegro MICS 2005–2006 European Middle-income
Mozambique** DHS 2011, DHS 2003 African Low-income
Namibia** DHS 2006–2007, DHS 2000 African Middle-income
Nepal** DHS 2011, DHS 2001 South-East Asia Low-income
Niger** DHS 2012, DHS 1998 African Low-income
Nigeria** DHS 2013, DHS 2003 African Middle-income
Pakistan DHS 2012–2013 Eastern Mediterranean Middle-income
Peru** DHS 2012, DHS 2000 Americas Middle-income
Philippines** DHS 2013, DHS 2003 Western Pacific Middle-income
Republic of Moldova DHS 2005 European Middle-income
Rwanda** DHS 2010, DHS 2000 African Low-income
Sao Tome and Principe DHS 2008–2009 African Middle-income
Senegal** DHS 2012–2013, DHS 1997 African Middle-income
Serbia MICS 2010 European Middle-income
Sierra Leone DHS 2013 African Low-income
Somalia MICS 2006 Eastern Mediterranean Low-income
Suriname MICS 2010 Americas Middle-income
Swaziland MICS 2010 African Middle-income
Syrian Arab Republic MICS 2006 Eastern Mediterranean Middle-income
Tajikistan DHS 2012 European Low-income
Thailand MICS 2005–2006 South-East Asia Middle-income
The former Yugoslav Republic of Macedonia MICS 2011 European Middle-income
92
Country income
Timor-Leste DHS 2009–2010 South-East Asia Middle-income
Togo** MICS 2010, DHS 1998 African Low-income
Uganda** DHS 2011, DHS 2000 African Low-income
Ukraine DHS 2007 European Middle-income
United Republic of Tanzania** DHS 2010, DHS 1999 African Low-income
Uzbekistan** MICS 2006, DHS 1996 European Middle-income
Vanuatu MICS 2007 Western Pacific Middle-income
Viet Nam** MICS 2010–2011, DHS 2002 Western Pacific Middle-income
Yemen MICS 2006 Eastern Mediterranean Middle-income
Zambia** DHS 2007, DHS 1996 African Middle-income
Zimbabwe** DHS 2010–2011, DHS 1999 African Low-income
DHS: Demographic and Health Survey; MICS: Multiple Indicator Cluster Survey.
* Country income group was determined using the World Bank classification as of July 2014 (available from: http://data.worldbank.org/about/country-
and-lending-groups#OECD_members, accessed 11 March 2015).
** At least one survey was also conducted during the period 1995–2004 and the country was considered for inclusion in change-over-time analyses.
93
TABLE S3 Summary estimates* for reproductive, maternal, newborn and child health indicators: national average and absolute
inequality by four dimensions of inequality (DHS and MICS 2005–2013)
National average Economic status

(difference)
richest quintile – poorest
quintile
(or vice versa for child
malnutrition and child
mortality indicators)
Number of countries
Number of countries
Maximum
Maximum
Minimum
Minimum
Median
Median
Category Indicator
Reproductive health Contraceptive prevalence – modern and traditional methods 43.6 5.6 79.0 84 13.2 -11.9 39.2 82
interventions
Contraceptive prevalence – modern methods 30.2 1.2 75.1 84 13.0 -12.6 38.8 82
Demand for family planning satisfied 60.6 19.7 92.2 61 15.9 -11.1 63.8 60
Maternal health Antenatal care coverage – at least one visit 93.2 26.1 99.6 85 10.5 -4.2 68.9 83
interventions
Antenatal care coverage – at least four visits 63.6 14.7 99.7 72 25.1 -6.4 73.5 72
Births attended by skilled health personnel 77.1 11.8 100.0 85 36.6 0.0 80.2 83
Newborn and child Early initiation of breastfeeding 52.6 7.6 96.0 86 -0.4 -33.9 22.9 83
health interventions
Children aged 6–59 months who received vitamin A supplementation 59.6 2.9 92.9 65 6.8 -30.6 46.3 64
BCG immunization coverage among one-year-olds 94.1 29.7 100.0 81 7.8 -12.8 78.3 78
Measles immunization coverage among one-year-olds 81.1 28.8 98.3 81 12.7 -11.7 64.7 78
Polio immunization coverage among one-year-olds 77.4 35.9 98.5 80 8.8 -19.8 37.3 77
DTP3 immunization coverage among one-year-olds 78.5 14.1 98.4 81 13.1 -19.3 72.3 78
Full immunization coverage among one-year-olds 65.9 11.6 95.1 79 11.4 -21.9 54.1 76
Children aged less than five years with diarrhoea receiving oral rehydration 39.8 11.3 85.1 83 8.9 -17.3 34.3 66
salts
Children aged less than five years with diarrhoea receiving oral rehydration 47.6 6.8 77.1 83 8.4 -9.7 30.8 66
therapy and continued feeding
Children aged less than five years with pneumonia symptoms taken to a 63.8 13.0 96.5 80 18.2 -21.8 63.2 53
health facility
RMNCH interventions, Composite coverage index 68.8 37.4 89.7 62 20.0 3.1 60.9 48
combined
Child malnutrition Stunting prevalence in children aged less than five years 29.6 4.9 57.9 78 17.7 -0.8 40.9 77
Underweight prevalence in children aged less than five years 13.2 1.3 44.2 77 10.1 -3.1 37.0 76
Wasting prevalence in children aged less than five years 6.0 0.7 25.4 77 2.2 -5.5 12.4 76
Child mortality Neonatal mortality rate 29.0 9.4 57.6 54 6.1 -17.9 33.1 54
Infant mortality rate 53.9 16.4 109.8 54 17.7 -19.1 62.4 54
Under-five mortality rate 76.6 18.7 174.7 54 36.5 -11.7 115.2 54
BCG: one dose of Bacille Calmette-Guérin vaccine; DHS: Demographic and Health Survey; DTP3: three doses of the combined diphtheria, tetanus toxoid and
pertussis vaccine; MICS: Multiple Indicator Cluster Survey; RMNCH: reproductive, maternal, newborn and child health.
* National average estimates are expressed as percentages or, in the case of the child mortality indicators, as deaths per 1000 live births. Difference estimates
are expressed as percentage points or, in the case of the child mortality indicators, as deaths per 1000 live births.
** For reproductive and maternal health interventions, education refers to the woman’s education. For newborn and child health indicators, education refers to
the mother’s education.
94
Education** Place of residence Sex

(difference) (difference) (difference)
secondary school or urban – rural females – males
higher – no education (or vice versa for child (or vice versa for child
(or vice versa for child malnutrition and child malnutrition and child
malnutrition and child mortality indicators) mortality indicators)
mortality indicators)
Number of countries
Number of countries
Number of countries
Maximum
Maximum
Maximum
Minimum
Minimum
Minimum
Median
Median
Median
Indicator
16.6 -15.7 45.7 71 6.2 -10.9 29.1 84 Contraceptive prevalence – modern and traditional methods
14.4 -16.4 40.1 71 6.7 -12.0 27.0 84 Contraceptive prevalence – modern methods
16.6 -16.4 51.1 54 8.9 -13.4 32.4 61 Demand for family planning satisfied
16.1 -0.9 63.1 69 4.6 -3.0 48.9 85 Antenatal care coverage – at least one visit
26.8 -1.2 59.9 60 12.8 -6.5 43.4 72 Antenatal care coverage – at least four visits
36.0 1.5 69.1 69 20.4 -0.4 61.2 85 Births attended by skilled health personnel
0.2 -27.6 21.0 69 -0.4 -18.0 20.5 86 0.9 -6.4 5.3 54 Early initiation of breastfeeding
9.2 -17.8 37.0 63 2.5 -29.3 18.6 65 -0.1 -3.5 5.7 65 Children aged 6–59 months who received vitamin A supplementation
9.7 -0.5 66.2 62 2.8 -9.2 39.2 81 -0.2 -12.8 6.5 81 BCG immunization coverage among one-year-olds
17.7 -3.2 54.4 62 4.2 -7.0 30.8 81 0.0 -9.8 6.5 81 Measles immunization coverage among one-year-olds
10.2 -7.1 53.5 62 2.2 -15.6 25.6 80 -0.3 -11.6 10.3 80 Polio immunization coverage among one-year-olds
14.6 -10.4 61.7 62 3.4 -12.9 37.3 81 -0.7 -10.7 8.3 81 DTP3 immunization coverage among one-year-olds
15.1 -9.2 49.1 61 3.3 -17.1 27.6 79 0.2 -11.4 10.3 79 Full immunization coverage among one-year-olds
Children aged less than five years with diarrhoea receiving oral rehydration
4.5 -25.1 41.7 58 5.4 -43.6 28.3 81 -0.6 -24.1 15.1 83 salts
6.3 -9.8 31.6 58 4.2 -24.8 39.4 81 -1.0 -22.6 25.7 83 therapy and continued feeding
Children aged less than five years with pneumonia symptoms taken to a
14.0 1.9 37.3 46 9.8 -14.3 40.1 72 0.2 -14.9 17.0 80 health facility
17.6 7.3 46.1 45 10.8 -4.8 31.1 62 Composite coverage index
14.8 -3.7 38.6 66 8.2 -2.8 22.1 78 3.1 -3.5 10.7 78 Stunting prevalence in children aged less than five years
9.0 -0.8 26.4 65 4.9 -1.2 17.7 77 1.5 -4.1 5.9 77 Underweight prevalence in children aged less than five years
2.2 -2.6 12.3 65 0.7 -3.7 16.8 77 0.9 -1.5 4.0 77 Wasting prevalence in children aged less than five years
9.3 -10.2 45.2 43 2.9 -11.9 23.6 54 6.7 -0.7 23.4 54 Neonatal mortality rate
20.7 -7.4 70.9 43 7.9 -15.5 41.7 54 8.5 -3.3 33.0 54 Infant mortality rate
40.3 -3.1 98.6 43 15.5 -11.7 79.4 54 10.5 -6.3 35.7 54 Under-five mortality rate
95
TABLE S4 Summary estimates for reproductive, maternal, newborn and child health indicators: relative inequality by four dimensions
of inequality (DHS and MICS 2005–2013)
Economic status Education*

(ratio) (ratio)
richest quintile / poorest secondary school or
quintile higher / no education
(or vice versa for child (or vice versa for child
malnutrition and child malnutrition and child
mortality indicators) mortality indicators)
Number of countries
Number of countries
Maximum
Maximum
Minimum
Minimum
Median
Median
Category Indicator
Reproductive health Contraceptive prevalence – modern and traditional methods 1.5 0.8 22.1 82 1.7 0.6 11.6 71
interventions
Contraceptive prevalence – modern methods 1.7 0.8 31.8 82 1.9 0.5 12.6 71
Demand for family planning satisfied 1.4 0.8 6.6 60 1.4 0.7 4.4 54
Maternal health Antenatal care coverage – at least one visit 1.1 1.0 6.1 83 1.2 1.0 3.6 69
interventions
Antenatal care coverage – at least four visits 1.5 0.9 9.1 72 1.6 1.0 11.6 60
Births attended by skilled health personnel 1.6 1.0 17.7 83 1.8 1.0 13.4 69
Newborn and child Early initiation of breastfeeding 1.0 0.2 1.9 83 1.0 0.6 2.0 69
Children aged 6–59 months who received vitamin A supplementation 1.2 0.1 4.7 64 1.2 0.4 3.3 63
BCG immunization coverage among one-year-olds 1.1 0.8 6.6 78 1.1 1.0 4.2 62
Measles immunization coverage among one-year-olds 1.2 0.8 5.9 78 1.3 1.0 4.0 62
Polio immunization coverage among one-year-olds 1.1 0.6 2.3 77 1.1 0.8 2.6 62
DTP3 immunization coverage among one-year-olds 1.2 0.7 10.8 78 1.3 0.8 6.1 62
Full immunization coverage among one-year-olds 1.2 0.4 15.4 76 1.3 0.7 7.1 61
Children aged less than five years with diarrhoea receiving oral rehydration 1.2 0.6 4.8 66 1.1 0.6 3.8 58
salts
Children aged less than five years with diarrhoea receiving oral rehydration 1.2 0.7 2.9 66 1.1 0.8 2.5 58
Children aged less than five years with pneumonia symptoms taken to a 1.4 0.7 6.0 53 1.3 1.0 2.4 46
health facility
RMNCH interventions, Composite coverage index 1.3 1.0 5.2 48 1.3 1.1 3.2 45
combined
Child malnutrition Stunting prevalence in children aged less than five years 2.2 1.0 11.1 77 1.6 0.9 4.9 66
Underweight prevalence in children aged less than five years 2.4 0.2 13.1 76 1.8 0.9 6.3 65
Wasting prevalence in children aged less than five years 1.5 0.4 6.7 76 1.4 0.3 4.0 65
Child mortality Neonatal mortality rate 1.3 0.5 3.3 54 1.4 0.6 4.8 43
Infant mortality rate 1.6 0.6 3.4 54 1.6 0.6 3.5 43
Under-five mortality rate 1.8 0.7 3.7 54 1.7 0.8 3.9 43
* For reproductive and maternal health interventions, education refers to the woman’s education. For newborn and child health indicators, education refers to
96
Place of residence Sex

(ratio) (ratio)
urban / rural females / males
(or vice versa for child (or vice versa for child
malnutrition and child malnutrition and child
mortality indicators) mortality indicators)
Number of countries
Number of countries
Maximum
Maximum
Minimum
Minimum
Median
Median
Indicator
1.2 0.8 3.9 84 Contraceptive prevalence – modern and traditional methods
1.3 0.7 17.3 84 Contraceptive prevalence – modern methods
1.2 0.8 2.8 61 Demand for family planning satisfied
1.1 1.0 3.0 85 Antenatal care coverage – at least one visit
1.2 0.9 3.3 72 Antenatal care coverage – at least four visits
1.3 1.0 10.9 85 Births attended by skilled health personnel
1.0 0.5 1.6 86 1.0 0.9 1.1 54 Early initiation of breastfeeding
1.1 0.3 3.2 65 1.0 0.8 1.2 65 Children aged 6–59 months who received vitamin A supplementation
1.0 0.9 2.4 81 1.0 0.9 1.1 81 BCG immunization coverage among one-year-olds
1.1 0.9 2.0 81 1.0 0.9 1.1 81 Measles immunization coverage among one-year-olds
1.0 0.7 2.0 80 1.0 0.9 1.2 80 Polio immunization coverage among one-year-olds
1.0 0.8 3.9 81 1.0 0.8 1.1 81 DTP3 immunization coverage among one-year-olds
1.1 0.6 3.8 79 1.0 0.7 1.2 79 Full immunization coverage among one-year-olds
1.1 0.2 2.9 81 1.0 0.5 1.5 83 salts
1.1 0.5 1.9 81 1.0 0.7 1.5 83 therapy and continued feeding
1.2 0.7 3.1 72 1.0 0.7 1.6 80 health facility
1.2 0.9 2.0 62 Composite coverage index
1.4 0.7 3.1 78 1.1 0.8 1.5 78 Stunting prevalence in children aged less than five years
1.5 0.6 3.1 77 1.1 0.8 1.8 77 Underweight prevalence in children aged less than five years
1.1 0.5 2.9 77 1.2 0.5 2.6 77 Wasting prevalence in children aged less than five years
1.1 0.5 3.1 54 1.3 1.0 2.6 54 Neonatal mortality rate
1.2 0.7 2.9 54 1.2 0.8 1.7 54 Infant mortality rate
1.3 0.8 2.6 54 1.1 0.9 1.7 54 Under-five mortality rate
97
TABLE S5 Summary estimates* for reproductive, maternal, newborn and child health indicators: annual absolute change in national
average and annual absolute excess change over time by four dimensions of inequality (DHS and MICS 1995–2004 and 2005–2013)
National average Economic status

(annual absolute (annual absolute
change) excess change)
annual change in the
poorest quintile – annual
change in the richest
quintile
Number of countries
Number of countries
Maximum
Maximum
Minimum
Minimum
Median
Median
Category Indicator
Reproductive health Contraceptive prevalence – modern and traditional methods 0.5 -1.7 3.8 42 0.5 -2.6 2.0 42
interventions
Contraceptive prevalence – modern methods 0.6 -1.2 4.1 42 0.3 -2.1 1.8 42
Demand for family planning satisfied 0.5 -2.9 4.7 38 0.5 -3.4 2.5 38
Maternal health Antenatal care coverage – at least one visit 0.6 -0.9 5.2 42 0.6 -1.6 4.0 42
interventions
Antenatal care coverage – at least four visits 0.7 -2.6 5.3 41 0.2 -2.2 4.1 41
Births attended by skilled health personnel 1.0 -0.7 4.7 42 0.3 -3.5 3.3 42
Newborn and child Early initiation of breastfeeding 0.9 -2.8 5.5 41 0.1 -2.1 3.9 41
Children aged 6–59 months who received vitamin A supplementation 1.2 -4.7 4.0 18 0.3 -1.8 4.1 18
BCG immunization coverage among one-year-olds 0.5 -0.5 6.9 41 0.7 -1.8 3.0 41
Measles immunization coverage among one-year-olds 0.9 -0.9 3.5 41 0.5 -4.2 4.1 41
Polio immunization coverage among one-year-olds 0.6 -1.3 4.6 41 0.9 -2.0 3.2 41
DTP3 immunization coverage among one-year-olds 0.7 -0.7 4.0 41 0.9 -2.5 3.2 41
Full immunization coverage among one-year-olds 1.0 -1.1 6.5 41 0.6 -3.4 3.4 41
Children aged less than five years with diarrhoea receiving oral rehydration 0.7 -1.0 3.8 41 0.1 -2.2 5.3 36
salts
Children aged less than five years with diarrhoea receiving oral rehydration 0.6 -3.1 4.3 40 0.3 -1.7 4.3 35
Children aged less than five years with pneumonia symptoms taken to a 0.9 -1.6 4.4 35 0.7 -2.6 2.3 26
health facility
RMNCH interventions, Composite coverage index 0.7 -0.7 3.1 34 0.6 -1.3 2.2 28
combined
Child malnutrition Stunting prevalence in children aged less than five years -0.5 -1.7 0.9 31 -0.2 -1.7 1.0 31
Underweight prevalence in children aged less than five years -0.2 -1.4 0.4 31 -0.2 -1.0 0.9 31
Wasting prevalence in children aged less than five years 0.0 -0.5 0.7 31 0.1 -0.7 0.6 31
Child mortality Neonatal mortality rate -0.8 -2.9 0.6 37 -0.6 -3.7 1.5 37
Infant mortality rate -2.4 -6.6 1.5 37 -1.5 -6.6 3.0 37
Under-five mortality rate -4.4 -12.2 1.4 37 -2.6 -8.3 5.1 37
* Estimates of annual change in the national average and annual absolute excess change are expressed as percentage points per year, or in the case of the child
mortality interventions, as deaths per 1000 live births per year.
** For reproductive and maternal health interventions, education refers to the woman’s education. For newborn and child health indicators, education refers to
98
Education** Place of residence Sex

(annual absolute (annual absolute (annual absolute
excess change) excess change) excess change)
annual change in no annual change in rural annual change in males –
education – annual areas – annual change in annual change in females
change in secondary urban areas
school or higher
Number of countries
Number of countries
Number of countries
Maximum
Maximum
Maximum
Minimum
Minimum
Minimum
Median
Median
Median
Indicator
0.7 -1.2 3.1 38 0.3 -1.1 1.4 42 Contraceptive prevalence – modern and traditional methods
0.6 -1.0 2.6 38 0.2 -0.8 1.5 42 Contraceptive prevalence – modern methods
0.9 -1.4 6.7 35 0.3 -1.7 1.9 38 Demand for family planning satisfied
0.7 -1.7 3.2 38 0.5 -1.6 3.2 42 Antenatal care coverage – at least one visit
0.4 -3.4 3.0 38 0.4 -1.0 3.3 41 Antenatal care coverage – at least four visits
0.5 -2.2 4.5 38 0.4 -1.6 3.4 42 Births attended by skilled health personnel
0.2 -2.6 3.8 37 0.1 -1.2 2.1 41 0.0 -0.4 1.2 36 Early initiation of breastfeeding
0.7 -1.3 3.7 18 0.4 -1.8 1.9 18 -0.1 -1.2 0.4 18 Children aged 6–59 months who received vitamin A supplementation
0.9 -0.5 2.2 35 0.4 -0.7 2.8 41 0.0 -0.9 0.9 41 BCG immunization coverage among one-year-olds
0.5 -3.3 2.1 35 0.5 -1.0 2.4 41 -0.1 -1.2 0.6 41 Measles immunization coverage among one-year-olds
0.5 -1.8 4.2 35 0.7 -0.5 3.2 41 0.1 -0.7 0.7 41 Polio immunization coverage among one-year-olds
0.5 -2.0 3.5 35 0.4 -1.2 2.8 41 0.0 -0.9 0.7 41 DTP3 immunization coverage among one-year-olds
0.2 -2.3 4.2 35 0.5 -2.3 2.7 41 0.0 -1.1 0.9 41 Full immunization coverage among one-year-olds
0.5 -2.8 2.1 34 0.1 -1.6 1.8 40 0.0 -2.8 3.0 41 salts
0.4 -3.1 1.7 32 0.2 -2.0 1.7 39 0.2 -3.3 3.8 40 therapy and continued feeding
0.5 -2.1 2.4 21 0.5 -2.1 2.7 33 0.0 -3.0 2.0 35 health facility
0.7 -0.8 1.8 25 0.4 -1.0 1.4 34 Composite coverage index
-0.3 -1.7 1.0 29 -0.1 -1.0 1.8 31 0.0 -0.4 0.5 31 Stunting prevalence in children aged less than five years
-0.3 -2.7 0.8 29 -0.2 -0.9 0.6 31 0.0 -0.3 0.6 31 Underweight prevalence in children aged less than five years
-0.1 -0.4 0.5 29 -0.1 -0.3 0.3 31 0.0 -0.5 0.3 31 Wasting prevalence in children aged less than five years
-0.9 -2.6 0.8 31 -0.6 -3.4 1.0 37 -0.1 -1.4 2.9 37 Neonatal mortality rate
-2.2 -7.8 0.5 31 -1.4 -4.6 1.9 37 0.0 -1.7 4.6 37 Infant mortality rate
-3.6 -10.7 1.6 31 -1.9 -5.3 1.4 37 0.3 -1.8 4.3 37 Under-five mortality rate
99
Index
A Change-over-time line plot 80
Absolute excess change 14, 15, 70 Child health interventions, see Newborn and child health
interpreting 74–75 interventions
summary estimates 98–99 Child malnutrition, see Malnutrition, child
Age discrimination 56 Child mortality 41
Alma-Ata Declaration 4 feature story 41–44
Analysis, data 13–15, 58, 69–70 indicators used 11, 90
Antenatal care pneumonia 29
education-related inequality 27, 73 summary estimates 94–99
indicators 11, 87 vaccine-preventable diseases 33
inequalities 27 Circle graphs, vertical 79
potential for improvement 50, 51, 52 Circle plots, horizontal 78
summary estimates 94–99 Commission on Information and Accountability for Women’s
and Children’s Health 6
B Commission on the Social Determinants of Health 4
Bacille Calmette–Guérin (BCG) immunization 36, 88 Committing to Child Survival: A Promise Renewed 7
potential for improvement 50 Complete coverage pattern 85
summary estimates 94–99 Composite coverage index 45, 89
Bar charts change over time 47–48
horizontal 78 feature story 45–49
vertical, maps with 80 indicators included 11, 45
Benchmarking 17 potential for improvement 50, 51, 52
Birth attendance, skilled 24–28, 88 summary estimates 94–99
change over time 26, 27 Concentration index 73
economic-related inequality 24–26 Confidence intervals (CIs) 72
potential for improvement 50, 51, 52 Contraceptive use 20–23
summary estimates 94–99 change over time 22, 23
Bottom inequality 85 economic-related inequality 23
Box-and-whisker plots 78 education-related inequality 20–23
Breastfeeding, early initiation 88, 94–99 indicators 11, 87
potential for improvement 52
C summary estimates 94–99
Care-seeking for sick children 29 Countdown to 2015 initiative 6, 45
feature story 29–32 Countries, study 12, 66–68, 91–93
indicators 11, 89
potential for improvement 50, 52 D
summary estimates 94–99 Dashboard 16
Change in inequality over time 9, 19 Data 10–12, 65–69
analysis methods 13, 15, 70 analysis 13–15, 58, 69–70
countries included 12 availability 58
data availability 67 dimension of inequality 10, 11, 66
summary estimates 98–99 disaggregation 13, 56–57, 69
100
INDEX
health indicator 10, 11, 65–66 F

sources 10, 65 Family planning, demand satisfied 23, 87
visualization 16, 76–77 potential for improvement 50, 52
Demographic and Health Surveys (DHS) 10, 58, 65, 66–67 summary estimates 94–99
Diarrhoea, care-seeking for 29 Feature stories 17, 19–53
indicators used 11, 89 Four quadrant view 79
place-of-residence inequality 32
potential for improvement 50, 52 G
summary estimates 94–99 Gender, see Sex
Difference 13, 14, 69 Global Health Observatory, WHO 68
Dimensions of inequality 10, 11, 56, 66 Global Strategy for Health for All 4
Diphtheria, tetanus and pertussis (DTP3) immunization 33, 88 Global Strategy for Women’s and Children’s (and Adolescent’s)
change over time 34–35 Health 6, 59
economic-related inequality 33–36 Graphs 16
potential for improvement 50
sex-related inequality 36 H
summary estimates 94–99 Health determinants 68–69
Disadvantaged subgroups 3–4 Health indicators 10, 11, 65–66
categories assessed 10, 11 descriptions 87–90
change-over-time analyses 19 summary estimates 94–99
health-promoting initiatives 4–5 Health inequality 4–5 (see also Inequality)
Disaggregation, data 13, 56–57, 69 methodology for assessing 72–75
DTP3, see Diphtheria, tetanus and pertussis multidimensional nature 12
versus health inequity 5
E Health inequity 5
Economic-related inequality Health information systems 58–59
care-seeking for sick children 32 Household health surveys 10, 58, 65
child malnutrition 40 Household wealth index 10
composite coverage index 46, 47, 48 Human rights approach 4
contraceptive use 23
doubly disaggregated data 56–57 I
immunization coverage 33–36 Immunization, childhood 33
patterns 85 changes over time 34–35, 36
potential for improvement 50–53 feature story 33–36
skilled birth attendance 24–26, 27 indicators used 11, 88
summary estimates 94–99 potential for improvement 50, 52
under-five mortality rate 43 summary estimates 94–99
Economic status 10, 11, 66 Incremental linear pattern 85
Education 10, 11, 66 Independent Expert Review Group 7
Education-related inequality Inequality 3–4 (see also Health inequality)
antenatal care 27, 73 absolute measures 13, 14, 69
composite coverage index 46, 47, 48 across health topics, reducing 59–60
contraceptive use 20–23 dimensions of 10, 11, 56, 66
maternal health interventions 27 patterns of 85–86
neonatal mortality rate 43 reasons for persistence 56
stunting prevalence 37–39, 74, 76–77 relative measures 13, 14, 69
summary estimates 94–99 Inequity, health 5
Equity Infant mortality rate 43, 90, 94–99
global initiatives 3, 4–5 Interactive data visuals 16, 17, 78–84
health 5 best practices 82
orientation 57–59, 60 further exploration of data 83–84
promoting, in RMNCH 59 guide to interpreting 78–80
Every Woman, Every Child movement 6 International Center for Equity in Health, Brazil 65
Excess change, absolute, see Absolute excess change
101
M Pneumonia, care-seeking for 29–32, 89

Malnutrition, child 37 economic-related inequality 32
change over time 38–39 place-of-residence inequality 29–31
deaths due to 41 potential for improvement 50, 52
feature story 37–40 summary estimates 94–99
indicators used 11, 89 Polio immunization 33, 36, 88
summary estimates 94–99 summary estimates 94–99
Maps 16, 76–77 Population attributable risk 14, 15, 50, 70
with vertical bar charts 80 feature story 50–53
Marginal exclusion pattern 85 Population share 73–74
Mass deprivation pattern 85 Post-2015 sustainable development agenda 3, 5, 7, 59–60
Maternal health interventions 24
feature story 24–28 Q
indicators used 11, 65–66, 87–88 Queuing pattern 85
potential for improvement 50, 51, 52
summary estimates 94–99 R
Measles immunization 33, 36, 88 Ratio 13, 14, 69
potential for improvement 50 Reporting 9, 16–17, 55–60
summary estimates 94–99 Reproductive, maternal, newborn and child health (RMNCH)
Median values 15, 69 5–7
Millennium Development Goals (MDGs) 3, 6 health indicators 10, 11, 65–66
Monitoring 9–17 interventions combined see Composite coverage index
analysis methods 13–15, 69–70 monitoring inequality 9–17
capacity building 58–59 potential for improvement 50–53
data 10–12, 65–69 promoting equity in 59
Multidimensional nature, health inequality 12 Reproductive health interventions 20
Multiple Indicator Cluster Surveys (MICS) 10, 58, 65, 66–67 feature story 20–23
indicators used 11, 87
N potential for improvement 50, 51, 52
Neonatal mortality rate 43, 90, 94–99 summary estimates 94–99
Newborn and child health interventions Rio Political Declaration on Social Determinants of Health 4
indicators used 11, 88–89 RMNCH, see Reproductive, maternal, newborn and child health
potential for improvement 50, 51, 52 Rural–urban differences, see Place-of-residence inequalities
summary estimates 94–99
S
O Sex (child) 10, 11
Oral rehydration therapy 32, 50, 89 Sex-related inequality
immunization coverage 36
P summary estimates 94–99
Partnership for Maternal, Newborn and Child Health 6 Slope index of inequality 73
Patterns of inequality 85–86 Story-points 16
Place of residence 10, 11, 66 Stunting prevalence 89
Place-of-residence inequalities change over time 38–39, 40
child mortality 41–44 economic-related inequality 40
composite coverage index 46, 47, 48 education-related inequality 37–39, 74, 76–77
diarrhoea care-seeking 32 summary estimates 94–99
doubly disaggregated data 56–57 Summary measures 13–15, 69–70, 73
maternal health interventions 27
measles immunization 36 T
pneumonia care-seeking 29–31 Tables 16
summary estimates 94–99 Top inequality 85
102
INDEX
U V
UN Commission on Life-Saving Commodities for Women’s and Visualization, data 16, 17, 76–77 (see also Interactive data
Children’s Health 7 visuals)
Under-five mortality rate 90 Vitamin A supplementation 88, 94–99
change over time 43, 44
doubly disaggregated data 56–57 W
place-of-residence inequality 41–44 Wasting prevalence, child 89, 94–99
summary estimates 94–99 Within-country inequality 3–4
Underweight prevalence, child 40, 89, 94–99 World Bank DataBank 68
Universal health coverage 5
Urban–rural differences, see Place-of-residence inequalities
103
h t t p : / / w w w. w h o . i n t / g h o / h e a l t h _ e q u i t y / re p o r t _ 2 0 1 5 /
Everyone Everywhere Always
ISBN 978 92 4 156490 8
h t t p : / / w w w. w h o . i n t / g h o / h e a l t h _ e q u i t y / re p o r t _ 2 0 1 5 /
DEPARTMENT OF HEALTH STATISTICS AND INFORMATION SYSTEMS GENDER, EQUITY AND HUMAN RIGHTS TEAM
WORLD HEALTH ORGANIZATION

20, AVENUE APPIA
CH-1211 GENEVA 27
SWITZERLAND

State of Inequality - Reproductive Maternal Newborn and Child Health

Uploaded by

Copyright:

Available Formats

You might also like

State of Inequality - Reproductive Maternal Newborn and Child Health

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

State of Inequality - Reproductive Maternal Newborn and Child Health

Uploaded by

Copyright:

Available Formats

State of inequality: reproductive, maternal, newborn and child health

I.World Health Organization.

ISBN 978 92 4 156490 8

Subject headings are available from the WHO institutional repository

© World Health Organization 2015

Design and layout by L’IV Com Sàrl, Villars-sous-Yens, Switzerland.

Executive summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

3. Monitoring the state of inequality in RMNCH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

4. The state of inequality in RMNCH: stories from low- and

5. Reporting the state of inequality: taking stock. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Figure 4.14 RMNCH composite coverage index by multiple dimensions of

Frequently used abbreviations

Flavia Bustreo Marie-Paule Kieny

Petra Schuster provided valuable administrative support.

REPORTING ON INEQUALITIES IN RMNCH

Latest situation of inequality

Reproductive health intervention indicators also indicated a situation of inequality.

Change in inequality over time

THE COMPOSITE COVERAGE INDEX

Implications of health inequality monitoring

FEATURE STORIES ABOUT THE STATE OF INEQUALITY IN RMNCH

Eight selected feature stories illustrate the state of inequality in RMNCH.

2.2 Health inequality

A HUMAN RIGHTS APPROACH TO PROMOTING HEALTH

INEQUALITY IN HEALTH VERSUS INEQUITY IN HEALTH

2.3 Reproductive, maternal, newborn and child health

TABLE 2.1 Development initiatives and RMNCH: a recent history

Initiative Key publications

In 2010, the United Nations General Assembly, aiming

by the year 2015, launched the Global Strategy for

Initiative Key publications

the leading preventable causes of death during

3.1.1 Data sources

3.1.2 Health indicator data

3.1.3 Dimension of inequality data

TABLE 3.1 Selected RMNCH indicators used in this report

TABLE 3.2 Health indicator categories and relevant dimensions of inequality

VIDEO CLIP 1. HEALTH INEQUALITY IS MULTIDIMENSIONAL

countries? World HealtH organization

ISBN 978 92 4 156491 5

3.1.4 Country selection

STUDY COUNTRIES INCLUDED IN THIS REPORT

*Denotes the 42 countries included in analyses of change over time.

3.2.1 Data disaggregation

3.2.2 Summary measures

Summary measure Calculation Application

ABSOLUTE VERSUS RELATIVE MEASURES OF INEQUALITY

3.3.1 Data visualization

INTERACTIVE DATA VISUALIZATION TERMINOLOGY

3.3.2 Feature stories

VIDEO CLIP 2. BENCHMARKING PUTS INEQUALITY IN CONTEXT

inequality in this health intervention? World HealtH organization

ISBN 978 92 4 156491 5

ACCESSING INTERACTIVE VISUALS

4.1 Reproductive health interventions

EDUCATION-RELATED INEQUALITY IN MODERN CONTRACEPTIVE USE

MODERN CONTRACEPTIVE USE