research-for-universal-health-coverage(9789240690837_eng).pdf

8/20/2019 research-for-universal-health-coverage(9789240690837_eng).pdf

1/162

Research for Universal

Health Coverage

T h e W o r l d H e a l t h R e p o r t 2 0 1 3


2/162

WHO Library Cataloguing-in-Publication Data

The world health repor t 2013: research for univers al health coverage.

1.World health - trends. 2.Universal coverage. 3.Health services accessibility. 4.Research. 5.Insurance, Health.I.World Health Organization.

ISBN 978 92 4 156459 5 (NLM classification: W 84.6)ISBN 978 92 4 069081 3 (ePub)ISBN 978 92 4 069082 0 (Daisy)ISBN 978 92 4 069083 7 (PDF)ISSN 1020-3311

Printed in Luxembourg

© World Health Organization 2013

All rights reserved. Publications of the World Health Organization are available on the WHO web site (www.who.int) or can be purchased fromWHO 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 web site (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 onthe part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning thedelimitation of its frontiers or boundaries. Dot ted 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 WorldHealth Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of propr ietaryproducts 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 interpretationand use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use.

Acknowledgements

Under the aegis of Assistant Directors-General Hiroki Nakatani and Marie-Paule Kieny, the following people wrote and produced this report:

Lead authors

Christopher Dye, Ties Boerma, David Evans, Anthony Harries, Christian Lienhardt, Joanne McManus, Tikki Pang, Rober t Terry, Rony Zachariah.

WHO staff in Geneva

Caroline Allsopp, Najeeb Al-Shorbaji, John Beard, Douglas Bettcher, Diarmid Campbell-Lendrum, Andrew Cassels, A’Isha Commar, Luis DeFrancisco Serpa, Carlos Dora, Gerald Dziekan, Christy Feig, Fiona Fleck, Haileyesus Getahun, Abdul Ghaffar, Laragh Gollogly, Andre Griekspoor,Sophie Guetaneh Aguettant, Metin Gülmezoglu, Ali Hamandi, Asli Kalin, Ghassan Karam, Edward Kelley, Richard Laing, Melanie Lauckner, KnutLönroth, Mary MacLennan, Clarisse Mason, Elizabeth Mason, Mike Mbizvo, Shanti Mendis, Thierry Mertens, Zafar Mirza, Maria Neira, UlyssesPanisset, Kimberly Parker, Michaela Pfeiffer, Kent Ranson, Mario Raviglione, John Reeder, Alex Ross, Cathy Roth, Sarah Russell, Ritu Sadana,Abha Saxena, Trish Saywell, Thomas Shakespeare, Isobel Sleeman, Johannes Sommerfeld, Marleen Temmerman, Diana Weil, Karin Weyer.

WHO staff in regional and country offices

Naeema Al-Gasseer, Luis Cuervo Amore, Govin Permanand, Manju Rani, Issa Sanou, Gunawan Setiadi, Claudia Stein, Edouard Tursan d’Espaignet,Adik Wibowo.

Members of the Scientific Advisory Panel

Andy Haines (chair), Fred Binka, Somsak Chunharas, Maimunah Hamid, Richard Horton, John Lavis, Hassan Mshinda, Pierre Ongolo-Zogo,Silvina Ramos, Francisco Songane.

Other individuals who contributed to or reviewed the content

Claire Allen, Thomas Bombelles, David Bramley, Martin Buxton, Anne Candau, Michael Clarke, Sylvia de Haan, David Durrheim, Toker Ergüder,Mahmoud Fathalla, Stephen Hanney, Mark Harrington, Sue Hobbs, Carel IJsselmuiden, Nasreen Jessani, Anatole Krattiger, Gina Lagomarsino,Guillermo Lemarchand, David Mabey, Dermot Maher, Cristina Ortiz, Adolfo Mart inez Palomo, Charlotte Masiello-Riome, Peter Massey, MartinMckee, Opena Merlita, Amanda Milligan, Peter Ndumbe, Thomson Prentice, Bernd Rechel, Jan Ross, Sabine Schott, Peter Small, Hanna Steinbach,Sheri Strite, Yot Teerawattananon, Göran Tomson, Ian Viney, Laetitia Voneche, Shaw Voon Wong, Judith Whitworth, Suwit Wibulpolprasert,Catherine Wintrich.

http://www.who.int/mailto:[email protected]%20http://www.who.int/about/licensing/copyright_form/en/index.htmlhttp://www.who.int/about/licensing/copyright_form/en/index.htmlmailto:[email protected]%20http://www.who.int/


3/162

iii

“Universal health coverage is the single most powerful concept

that public health has to offer”

Dr Margaret Chan, Address to the Sixty-fth

World Health Assembly, May 2012

“Another lesson is the importance of long-term investment in

the research institutions that generate evidence for policy ...”

Lancet , 2012, 380:1259,

on the approach to universal health coverage in Mexico


4/162

As we approach the 2015 deadline or meeting the United Nations MillenniumDevelopment Goals (MDGs), it is time to take stock o the progress that has beenmade since 2000. It is also time to reflect on how we made progress, and on how wecould do better.

All eight o the MDGs have conse-quences or health, but three put healthat ront and centre – they concernchild health (MDG 4), maternal health(MDG 5), and the control o HIV/AIDS,malaria, tuberculosis and other majorcommunicable diseases (MDG 6). ohighlight just one o these, MDG 4 callsor a reduction in the number o child

deaths rom 12 million in 1990 to ewerthan 4 million by 2015. Although greatstrides have been taken since the turn othe millennium, especially in reducingdeaths afer the neonatal period, the bestmeasurements indicate that nearly 7 mil-lion children under five years o age died in 2011. From experience in high-incomecountries, we know that almost all o these deaths can be prevented. But how can thatbe done everywhere?

One idea is to make greater use o community-based interventions. But do theywork? Experiments in the orm o randomized controlled trials provide the most per-

suasive evidence or action in public health. By 2010, 18 such trials in Arica, Asia andEurope had shown that the participation o outreach workers, lay health workers, com-munity midwives, community and village health workers, and trained birth attend-ants collectively reduced neonatal deaths by an average o 24%, stillbirths by 16% andperinatal mortality by 20%. Maternal illness was also reduced by a quarter (1). Tese

Message from the Director-General

iv

1. Lassi ZS, Haider BA, Bhutta ZA. Community-based intervention packages for reducing maternal andneonatal morbidity and mortality and improving neonatal outcomes. Cochrane Database of SystematicReviews (Online), 2010,11:CD007754. PMID:21069697

http://www.ncbi.nlm.nih.gov/pubmed/21069697http://www.ncbi.nlm.nih.gov/pubmed/21069697


5/162

v

trials clearly do not give all the answers – or instance, the benefits o these interven-tions in reducing maternal mortality, as distinct rom morbidity, are still unclear – butthey are a powerul argument or involving community health workers in the care omothers and newborn children.

Tese rigorous investigations have the potential to benefit millions around theworld. Tey conront the challenge presented by just one o the MDGs, but they cap-ture the general spirit o this report – to promote investigations in which creativity isharnessed by the highest-quality science in order to deliver affordable, quality healthservices and better health or everyone. More than that, the process o discovery is asource o inspiration and motivation, stirring ambitions to deeat the biggest problemsin public health. Tis is the purpose o Research for universal health coverage.

Tis report is or everyone concerned with understanding how to reach the goalo universal health coverage – those who und the necessary research, those who doresearch and who would like to do research, and those who use the evidence romresearch. It shows how research or health in general underpins research or universalhealth coverage in particular.

Understanding how to make progress towards achieving the MDGs is central tothis report. But its scope is wider. As the 2015 deadline draws closer, we are looking orways to improve all aspects o health, working within and beyond the MDG ramework.And we are investigating how better health can contribute to the larger goal o humandevelopment. In this broad context, I invite you to read Research for universal healthcoverage. I invite you to assess the report’s arguments, review its evidence, and helpsupport the research that will bring us closer to the goal o universal health coverage.

Dr Margaret ChanDirector-GeneralWorld Health Organization


6/162


7/162

vii

Message from the Director-General iv

Executive summary xi

1. The role of research for universal health coverage 5

Developing the concept of universal health coverage 6

Investigating financial risk protection 11

Investigating the coverage of health services 15

Equity and universal health coverage 19

Coverage of health services: quality as well as quantity 20

Conclusions: research needed for universal health coverage 21

2. The growth of research for universal health coverage 31

Creativity everywhere 35

Research ascending 35

Growing unevenly 42

The value of health research 46

Conclusions: building on the foundations 47

3. How research contributes to universal health coverage 57

Case-study 1 61

Insecticide-treated mosquito nets to reduce childhood mortality

Case-study 2 63Antiretroviral therapy to prevent sexual transmission of HIV

Case-study 3 65

Zinc supplements to reduce pneumonia and diarrhoea in young children

Contents


8/162

viiiviii

Case-study 4 67

Telemedicine to improve the quality of paediatric care

Case-study 5 69

New diagnostics for tuberculosis

Case-study 6 71

The “polypill” to reduce deaths from cardiovascular disease

Case-study 7 73

Combination treatment with sodium stibogluconate (SSG) and paromomycincompared to SSG monotherapy for visceral leishmaniasis

Case-study 8 75

Task shifting in the scale-up of interventions to improve child survival

Case-study 9 77Improving access to emergency obstetric care

Case-study 10 79

Conditional cash transfers to improve the use of health services andhealth outcomes

Case-study 11 81

Insurance in the provision of accessible and affordable health services

Case-study 12 82

Affordable health care in ageing populations

Conclusions: general lessons drawn from specific examples 84

4. Building research systems for universal health coverage 95

Setting research priorities 96

Strengthening research capacity 98

A framework for strengthening capacity 99

Creating and retaining a skilled research workforce 103

Ensuring transparency and accountability in research funding 105

Building research institutions and networks 107

Defining and implementing norms and standards 110

Ethics and ethical review 110

Reporting and sharing research data, tools and materials 110

Registering clinical trials 110

Using evidence to develop policy, practice and products 113


9/162

ix

Translating evidence into policy and practice 113

Monitoring and coordinating research, nationally and internationally 116

Financing research for universal health coverage 117

National and international governance of health research 118

Conclusions: building effective research systems 118

5. Action on research for universal health coverage 129

Research – essential for universal coverage and a source of inspirationfor public health 129

Defining and measuring progress towards universal health coverage 131

The path to universal health coverage, and the path to better health 132

Research for universal health coverage in every country 133Supporting the people who do research 134

Translating research evidence into health policy and practice 135

Supporting research for universal health coverage, nationallyand internationally 136

WHO’s role in research for universal health coverage 137

Index 139


10/162


11/162

xi

Three key messages from The world health report

■ Universal health coverage, with full access to high-quality services for health promotion,prevention, treatment, rehabilitation, palliation and financial risk protection, cannot be

achieved without evidence from research. Research has the power to address a widerange of questions about how we can reach universal coverage, providing answers toimprove human health, well-being and development.

■ All nations should be producers of research as well as consumers. The creativity andskills of researchers should be used to strengthen investigations not only in academiccentres but also in public health programmes, close to the supply of and demand forhealth services.

■ Research for universal health coverage requires national and international backing. Tomake the best use of limited resources, systems are needed to develop national researchagendas, to raise funds, to strengthen research capacity, and to make appropriate andeffective use of research findings.

Why universal health coverage?

In 2005, all WHO Member States made the commitment to achieve universalhealth coverage. Te commitment was a collective expression o the belie thatall people should have access to the health services they need without risk ofinancial ruin or impoverishment. Working towards universal health coverage isa powerul mechanism or achieving better health and well-being, and or pro-moting human development.

Chapter 1 explains how the resolution adopted by all WHO Member Statesembraces the two acets o universal health coverage: the provision o, and accessto, high-quality health services; and financial risk protection or people who needto use these services. “Health services” in this report mean methods or promo-tion, prevention, treatment, rehabilitation and palliation, encompassing healthcare in communities, health centres and hospitals. Te term includes ways otaking action on social and environmental determinants both within and beyondthe health sector. Financial risk protection is part o the package o measures thatprovides overall social protection.

Executive summary


12/162

xii

Research for universal health coverage

Why research?

Scientific research has been undamental to the improvement o human health.Research is vital in developing the technology, systems and services needed to achieveuniversal health coverage. On the road to universal coverage, taking a methodicalapproach to ormulating and answering questions is not a luxury but a necessity.

When WHO Member States made the pledge to achieve universal coveragethey took a significant step orward or public health. As described in Chapter 1,taking that step effectively launched an agenda or research. In this report, researchis the set o ormal methods that turns promising ideas into practical solutions orimproving health services, and consequently or improving health. Te goal o thereport is to identiy the research questions that open the way to universal healthcoverage and to discuss how these questions can be answered.

Many recent advances have been made in health service coverage and in finan-cial risk protection as shown, or example, by progress towards the United NationsMillennium Development Goals (MDGs). Despite this progress, the gap betweenthe present coverage o health services and universal health coverage remains largeor many conditions o ill-health in many settings. For instance, nearly hal o allHIV-inected people eligible or antiretroviral therapy were still not receiving it in2011, and an estimated 150 million people suffer financial catastrophe each yearbecause they have to pay cash out-o-pocket or the health care they need. Teocus o this report is on the research needed to provide wider access to essentialservices o this kind, and how to create the environment in which this research canbe carried out.

What questions need to be answered by research?

Chapter 1 identifies research questions o two kinds. Te causes o ill-health differrom one setting to another and so too must the necessary health services, includ-ing mechanisms or financial risk protection. Te first group o questions thereoreasks how to choose the health services needed in each setting, how to improveservice coverage and financial protection, and consequently how to protect andimprove health and well-being.

Tese questions throw up a wide range o topics or research. Research is neededto find out how to improve the coverage o existing interventions and how to select

and introduce new ones. Research must explore the development and use o both“sofware” (such as schemes or financial protection and simplified approaches totreatment) and “hardware” (research and development or commodities and tech-nology). And research is needed to investigate ways o improving health romwithin and outside the health sector.

Te most pressing research questions have been identified or many specifichealth topics, such as maternal and child health, communicable diseases, andhealth systems and services. Although there are notable exceptions, less effort has


13/162

xiii

Executive summary

generally been given worldwide to establishing and publicizing national researchpriorities, to assessing the strengths and weaknesses o national research pro-grammes, and to evaluating the health, social and economic benefits o research.

Te second group o questions asks how to measure progress towards univer-sal coverage in each setting or each population, in terms o the services that areneeded and the indicators and data that measure the coverage o these services.Te answer to this group o questions is a measure o the gap between the presentcoverage o services and universal coverage. Te challenge or research is to fillthat gap.

Many specific indicators, targets and data sources are already used to measurethe coverage o specific health interventions. Te metrics used to monitor progresstowards the MDGs track, or example, access to antiretroviral therapy, births attendedby skilled health personnel, and immunization coverage. However, the measuremento other aspects o coverage needs urther development; interventions to prevent andcontrol noncommunicable diseases, or to track healthy ageing, are two examples.

It is not usually possible to measure the coverage o the hundreds o interven-tions and services that make up a national health system. However, it is possible tochoose a subset o services, with their associated indicators, that are representativeo the overall quantity, quality, equity and financing o services. Ten a practicaldefinition o universal health coverage is that all persons who are eligible haveaccess to the services they need. o choose the essential health services that shouldbe monitored, and a set o indicators to track progress towards universal coverage,is a research task or health programmes in each country. Out o these investiga-tions will emerge a common set o indicators that can be used to measure andcompare progress towards universal health coverage across al l countries.

With its ocus on research, the goal o this report is not to measure definitivelythe gap between the present coverage o health services and universal coverage but,instead, to identiy the questions that arise as we move towards universal coverageand to discuss how these questions can be answered.

Should all countries have the capacity to do research?

Te results o some research studies are widely applicable, but many questionsabout universal health coverage require local answers. All countries thereore needto be producers o research as well as consumers o it. An abundance o data,

presented in Chapter 2, shows that most low- and middle-income countries nowhave, at least, the oundations on which to build effective national health researchsystems. Some countries have much more than the oundations; they have thrivingresearch communities with a growing number o “south–south” as well as “north–south” international links. By strengthening these systems, countries will be ableto capitalize more effectively on the supply o ideas, using ormal research methodsto turn them into useul products and strategies or better health.


14/162

xiv


Which kinds of research studies have shownhow to improve the coverage of healthservices and how to improve health?

Te case or investing in research is made, in part, by demonstrating that scientificinvestigations really do produce results that can be translated into accessible andaffordable health services that provide benefits or health. Chapter 3 presents 12examples o studies that show how research can address some o the major ques-tions about achieving universal health coverage, and can deliver results that haveinfluenced, or could influence, policy and health outcomes.

Tree examples make the point. In one, a systematic review o survey datarom 22 Arican countries showed how the use o insecticide-treated mosquito netswas associated with ewer malaria inections and lower mortality in young chil-dren. Tis evidence underlines the value o scaling up and maintaining coverage oinsecticide-treated nets in malaria-endemic areas. In a second set o experimentaltrials in Ethiopia, Kenya, Sudan and Uganda, a combination o the drugs sodiumstibogluconate and paromomycin was ound to be an effective treatment or visceralleishmaniasis. reatment with the drug combination is shorter than with sodiumstibogluconate alone and is less likely to lead to drug resistance. On the basis o thesefindings, WHO recommended the drug combination as a first-line treatment or vis-ceral leishmaniasis in East Arica. A third systematic review o evidence rom Brazil,Colombia, Honduras, Malawi, Mexico and Nicaragua showed how conditional cashtransers – cash payments made in return or using health services – encourage theuse o these services and lead to better health outcomes.

Te successes o these investigations, and the others described in Chapter 3,should be a stimulus to invest in urther research. Not all investigations will findthat ideas or improving health services are successul, or that the provision o newservices actually improves health. In mapping the route to universal coverage, thenegative results o research studies are just as valuable as the positive ones.

Which research methods are used to answerquestions about universal health coverage?

Te examples described in Chapter 3 expose the diversity o questions about uni-

versal health coverage, and also the variety o research methods used to investigatethem. Methods include quantitative and qualitative evaluations, observational andcase-control studies, intervention studies, randomized controlled trials, and sys-tematic reviews and meta-analyses. Te report shows the benefits o having evidencerom multiple sources, explores the link between experimental design and strengtho inerence, and highlights the compromises in study design (better evidence isofen more costly, but not always) that must be made by all investigators. Te survey


15/162

xv

Executive summary

o research methods reveals the nature o the research cycle, where questions leadto answers that lead to yet more questions. Te chapter illustrates some o the waysin which research is linked with health policy and practice.

What can be done to strengthen nationalhealth research systems?

Research is likely to be most productive when it is conducted within a supportivenational research system. Chapter 4 is an introduction to the essential unctionso national health research systems, namely: to set research priorities, to developresearch capacity, to define norms and standards or research, and to translateevidence into practice.

Standard methods have been developed to set research priorities. Tese meth-ods should be used more widely by governments to set national priorities across allaspects o health and to determine how best to spend limited unds on research.

With regard to strengthening capacity, effective research needs transparent andaccountable methods or allocating unds, in addition to well-equipped researchinstitutions and networks. However, it is the people who do research – with theircuriosity, imagination, motivation, technical skills, experience and connections –that are most critical to the success o the research enterprise.

Codes o practice, which are the cornerstone o any research system, are alreadyin use in many countries. Te task ahead is to ensure that such codes o practice arecomprehensive and apply in all countries, and to encourage adherence everywhere.

Achieving universal health coverage depends on research ranging rom studies

o causation to studies o how health systems unction. However, because manyexisting cost-effective interventions are not widely used, there is a particular needto close the gap between exist ing knowledge and action. Areas o research that needspecial attention concern the implementation o new and existing technologies,health service operations, and the design o effective health systems. o help bridgethe gap between science and practice, research should be strengthened not only inacademic centres but also in public health programmes, close to the supply o anddemand or health services.

How can research for universal health coverage

be supported nationally and internationally?In the wake o many previous reports, Chapter 4 presents three mechanisms tostimulate and acilitate research or universal health coverage – monitoring, coor-dination and financing. Provided there is a commitment to share data, nationaland global observatories could be established to monitor research activities.Observatories could serve a variety o unctions, acting as repositories o data on


16/162

xvi


the process o doing research and presenting and sharing the findings o researchstudies. Such data would help in tracking progress towards universal health cover-age, country by country.

Monitoring supports the second unction, coordination, on various levels – bysharing inormation, by jointly setting research priorities, or by acilitating col-laboration on research projects.

Regarding the third unction, financing, health research is more effective andproductive i there is a guaranteed, regular income. Sustained financing guaranteesthat research projects are not interrupted or otherwise compromised by a suddenlack o resources. Various mechanisms or raising and disbursing additionalresearch unds have been proposed and are under discussion. Whatever mecha-nism is adopted, international donors and national governments should measureprogress against their own commitments to investing in health research.

How will WHO support research foruniversal health coverage?

Chapter 5 draws out the dominant themes o the report, and proposes a set oactions by which the research community, national governments, donors, civilsociety and international organizations, including WHO, can support the researchthat is needed i we are to reach universal health coverage.

Although the debate about universal health coverage has added to the vocabu-lary o public health in recent years, “to promote and conduct research in the fieldo health” has always been central to WHO’s goal o achieving “the highest attain-

able standard o health”. Chapter 5 briefly explains how WHO plays a role in bothdoing and supporting research through the Organization’s Strategy on Researchor Health. Tis report is closely aligned with the aims o the WHO strategy, whichencourages the highest-quality research in order to deliver the greatest health ben-efits to the maximum number o people.


17/162

Chapter 1

The role of research foruniversal health coverage


18/162


19/162

Key points 4

Developing the concept ofuniversal health coverage

6

Investigating financialrisk protection

11

Investigating the coverage ofhealth services

15

Equity and universal health coverage 19

Coverage of health services:quality as well as quantity

20

Conclusions: research needed foruniversal health coverage

21

Chapter 1

A field worker interviewing a young child near Bala Kot, Pakistan (WHO).


20/162

Key points

■ he goal of universal health coverage is to ensure that all people obtain the healthservices they need – prevention, promotion, treatment, rehabilitation and palliation– without risk of financial ruin or impoverishment, now and in the future.

■ Since 2005, when all WHO Member States made the commitment to universalhealth coverage, many advances have been made in the provision of health servicesand in financial risk protection. his is i llustrated by progress towards the health-related Millennium Development Goals (MDGs), and in the widespread fal l in cashpayments made for using health services.

■ Despite this progress, the coverage of health services and financial risk protectioncurrently fall far short of universal coverage. hus nearly half of all HIV-infectedpeople eligible for antiretroviral therapy were still not receiving it in 2011; and anestimated 150 mill ion people suffer f inancial catastrophe each year because theyhave to pay out-of-pocket for health services.

■ he conditions causing ill-health, and the f inancial capacity to protect people fromill-health, vary among countries. Consequently, given limited resources, each nationmust determine its own priorities for improving health, the services that are needed,and the appropriate mechanisms for f inancial risk protection.

■ hese observations lead to research questions of two kinds. First, and mostimportant, are questions about improving health and well-being – questions thathelp us to define the interventions and services that are needed, including financialrisk protection, discover how to expand the coverage of these services, including the

reduction of inequities in coverage, and investigate the effects of improved coverageon health. he second set of questions is about measurement – of the indicatorsand data needed to monitor service coverage, financial risk protection, and healthimpact. One task for research is to help define a set of common indicators forcomparing progress towards universal coverage across al l countries.

■ Neither of these areas of questioning has permanent answers. hrough the cycle ofresearch – questions yield answers which provoke yet more questions – there willalways be new opportunities to improve health. oday’s targets for universal healthcoverage will inevitably be superseded in tomorrow’s world of greater expectations.


21/162

Te goal o universal health coverage is to ensure that everyone can use thehealth services they need without risk o financial ruin or impoverishment (1).As a descendant o the “Health or All” movement (Box 1.1), universal healthcoverage takes a broad view o the services that are needed or good health andwell-being. Tese services range rom clinical care or individual patients tothe public services that protect the health o whole populations. Tey includeservices that come rom both within and beyond the health sector. Financialrisk protection is one element in the package o measures that provides overallsocial protection (7 ). And protection against severe financial difficulties in theevent o illness gives the peace o mind that is an integral part o well-being.

o support the goal o universal health coverage is also to express concern orequity and or honouring everyone’s right to health (8). Tese are personal and moral

choices regarding the kind o society that people wish to live in, taking universal cov-erage beyond the technicalities o health financing, public health and clinical care.With a greater understanding o the scope o universal health coverage, many

national governments now view progress towards that goal as a guiding principleor the development o health systems, and or human development generally. It isclear that healthier environments mean healthier people (9). Preventive and curativeservices protect health and protect incomes (10, 11). Healthy children are better ableto learn, and healthy adults are better able to contribute socially and economically.

Te path to universal health coverage has been dubbed “the third globalhealth transition”, afer the demographic and epidemiological transitions (12).Universal coverage is now an ambition or all nations at all stages o develop-

ment. Te timetable and priorities or action clearly differ between countries,but the higher aim o ensuring that all people can use the health services theyneed without risk o financial hardship is the same everywhere.

1Te role of research foruniversal health coverage

5


22/162


6

Developing the concept ofuniversal health coverage

Te world health report 2010 represented theconcept o universal health coverage in three

dimensions: the health services that are needed,the number o people that need them, and thecosts to whoever must pay – users and third-party unders (Fig. 1.1) (1, 13).

Te health services include approaches toprevention, promotion, treatment, rehabilitation

and palliative care, and these services must besufficient to meet health needs, both in quantityand in quality. Services must also be prepared orthe unexpected – environmental disasters, chem-ical or nuclear accidents, pandemics, and so on.

Te need or financial risk protection isdetermined by the proportion o costs thatindividuals must themselves cover by makingdirect and immediate cash payments.a Underuniversal coverage, there would be no out-o-pocket payments that exceed a given threshold

a Indirect costs, due for example to lost earnings, are not considered to be part of financial risk protection, but are part of thelarger goal of social protection.

Box 1.1. From “Health for All” to universal health coverage

Universal health coverage is an aspiration that underpins “the enjoyment of the highest attainable standard of health”which, as stated in WHO’s constitution, is “one of the fundamental r ights of every human being without distinction ofrace, religion, political belief, economic or social condition” ( 2). To reach the highest attainable standard of health isan objective that has guided health policy nationally and internationally for 65 years, f inding voice in WHO’s “Healthfor All” programme which began in the 1970s and was enshrined in the Alma Ata Declaration of 1978.

The Alma Ata Declaration is best known for promoting primary health care as a means to address the main healthproblems in communities, fostering equitable access to promotive, preventive, curative, palliative and rehabilitativehealth services.

The idea that everyone should have access to the health services they need underpinned a resolution of the 2005World Health Assembly, which urged Member States “to plan the transition to universal coverage of their citizensso as to contribute to meeting the needs of the population for health care and improving its quality, to reducingpoverty, and to attaining internationally agreed development goals” (3).

The central role of primary care within health systems was reiterated in The world health report 2008 which was

devoted to that topic (4). The world health repo rt 2010 on health systems financing built on this heritage by proposingthat health financing systems – which countries of all income levels constantly seek to modify and adapt – shouldbe developed with the specific goal of universal health coverage in mind.

The twin goals of ensuring access to health services, plus financial risk protection, were reaffirmed in 2012 by aresolution of the United Nations General Assembly which promotes universal health coverage, including socialprotection and sustainable financing (5). The 2012 resolution goes even further; it highlights the importance ofuniversal health coverage in reaching the MDGs, in alleviating poverty and in achieving sustainable development(6). It recognizes, as did the “Health for All” movement and the Alma Ata Declaration, that health depends not onlyon having access to medical services and a means of paying for these services, but also on understanding the linksbetween social factors, the environment, natural disasters and health.

This brief history sets the scene for this report. The world health report 2013: research for universal health coverage addresses questions about prevention and treatment, about how services can be paid for by individuals and govern-

ments, about their impact on the health of populations and the health of individuals, and about how to improvehealth through interventions both within and beyond the health sector. Although the focus of universal health cover-age is on interventions whose primary objective is to improve health, interventions in other sectors – agriculture,education, finance, industry, housing and others – may bring substantial health benefits.


23/162

Chapter 1 The role of research for universal health coverage

7

o affordability – usually set at zero or the poor-est and most disadvantaged people. Te total

volume o the large box in Fig. 1.1 is the cost oall services or everyone at a particular pointin time (1). Te volume o the smaller blue box

shows the health services and costs that are cov-ered rom pre-paid, pooled unds. Te goal ouniversal coverage is or everyone to obtain theservices they need at a cost that is affordable tothemselves and to the nation as a whole.

All governments should thereore decidewhat health services are needed, and how tomake sure they are universally available, aord-able, eicient, and o good quality (14, 15). heservices that are needed dier rom one settingto another because the causes o il l-health also

vary. he balance o services inevitably changesover time, as new technologies and proceduresemerge as a result o research and innovation,ollowing the changes in the causes o il l-health.In deciding which services to provide, institu-tions such as the National Institute or Healthand Clinical Excellence (NICE) in Englandand Wales and the Health Intervention andechnology Assessment Programme (HIAP)

in hailand (Box 1.2) have a vital role in evalu-ating whether interventions are eective andaordable.

In every country, there are people who areunable to pay directly, out-o-pocket, or theservices they need, or who may be seriously dis-advantaged by doing so. When people on lowincomes with no financial risk protection allill they ace a dilemma: i a local health serviceexists, they can decide to use the service andsuffer urther impoverishment in paying or it,or they can decide not to use the service, remainill and risk being unable to work (20). Te generalsolution or achieving wide coverage o financialrisk protection is through various orms o pre-payment or services. Prepayments allow undsto be pooled so that they can be redistributed toreduce financial barriers or those who need touse services they could not otherwise afford. Tisspreads the financial risks o ill-health acrosswhole populations. Prepayment can be derivedrom taxation, other government charges orhealth insurance, and usually comes rom a mix-ture o sources (1).

Financial risk protection o this kind is an

instrument o social protection applied to health(7 ). It works alongside other mechanisms osocial protection – unemployment and sicknessbenefits, pensions, child support, housing assis-tance, job-creation schemes, agricultural insur-ance and so on – many o which have indirectconsequences or health.

Governments, especially in low-incomecountries, cannot usually raise suicient undsby prepayment to eliminate excess out-o-pocket expenditures or all the health services

that people need (1). It is thereore a challengeto decide how best to support health withinbudgetary limits. Fig. 1.1 oers three optionsor spending: maximize the proportion o thepopulation covered by existing services, diver-siy health services by oering more types ointervention, or use the money or inancialcompensation, thereby reducing cash paymentsor health care.

Fig. 1.1. Measuring progress towards

universal health coverage in three

dimensions

Direct costs:proportion of the

costs covered

Population: who is covered?

Includeother

services

Extend tonon-covered

Services:which servicesare covered?

Current pooledfunds

Reduce cost-sharingand fees

Source: World Health Organization (1) and Busse, Schreyögg& Gericke (13).


24/162


8

Box 1.2. How Thailand assesses the costs and benefits of health interventions and

technologies

In 2001 the Government of Thailand introduced universal health coverage financed from general taxation. Economic

recession underlined the need for rigorous evaluation of health technologies that would be eligible for funding inorder to prevent costs from escalating. At the time, no organization had the capacity to carry out the volume of healthtechnology assessments (HTAs) demanded by the government. Therefore the Health Intervention and TechnologyAssessment Programme (HITAP, www.hitap.net) was set up to assess the costs, effectiveness and cost–effectivenessof health technologies – not only medications and medical procedures but also social interventions, public healthmeasures and changes to the health system itself (16, 17 ).

Unlike the National Institute for Health and Clinical Excellence (NICE) in England and Wales, which evaluates existinginterventions only, HITAP does primary research, including observational studies and randomized controlled trials,as well as systematic reviews and meta-analyses based on secondary literature analysis. Its output takes the form offormal presentations, discussion with technical and policy forums and academic publications.

One example of HITAP’s work is in devising a screening strategy for cervical cancer which is caused by infection withthe human papillomavirus (HPV) and is a major cause of morbidity and mortality among Thai women. Despite the intro-duction of Papanicolaou (Pap) screening at every hospital over 40 years ago, only 5% of women were screened. Visualinspection of the cervix with the naked eye after application with acetic acid (VIA) was introduced as an alternative in2001 because it did not require c ytologists. When HITAP’s study began, both VIA and Pap smears were being offered towomen in parallel and there was pressure from vaccine companies, international health agencies and nongovernmentalorganizations (NGOs) to introduce the new HPV vaccine (18).

The options considered by HITAP were conventional Pap screening, VIA, vaccination or a combination of Pap screeningand VIA. Costs were calculated on the basis of estimated levels of participation and included costs to the health-careprovider, costs for women attending screening and costs for those who were treated for cervical cancer. Potentialbenefits were analysed by using a model that estimated the number of women who would go on to develop cervicalcancer in each scenario, and the impact on quality-adjusted life years (QALYs) was calculated by using data from acohort of Thai patients.

The study concluded that the most cost-effective strategy was to offer VIA to women every five years between theages of 30 and 45, followed by a Pap smear every five years for women aged between 50 and 60 years. The strategywould offer an additional 0.01 QALYs and a total cost saving of 800 Baht, when compared to doing nothing. Universalintroduction of vaccination for 15-year-old girls without screening would result in a gain of 0.06 QALYs at a cost of 8000Baht, and either VIA and Pap screening alone would have costs and benefits somewhere between the two amounts (19).

The approach recommended by HITAP was piloted in several provinces starting in 2009, and this has now been imple-mented nationally. The actual impact is currently being assessed.

HITAP attributes its success to several factors:

■ the strong research environment in Thailand which, for instance, provides staff for HITAP and supports peerreview of their recommendations;

■ collegiate relationships with similar institutions in other countries, such as NICE in England and Wales;

■

working with peers (HITAP meets with other Asian HTA institutions, and has formed an association with Japan,Malaysia and the Republic of Korea); ■ transparency in research methods, so that difficult or unpopular decisions can be understood; ■ a code of conduct (HITAP adheres to a strict code of behaviour which, for instance, precludes acceptance of

gifts or money from pharmaceutical companies); ■ political support from government, fostered by opening doors to, and discussing methods with,

decision-makers; ■ popular support, generated by lectures at universities and dissemination of recommendations to the

general public; ■ external review (HITAP commissioned an external review of its methods and work in 2009).

http://www.hitap.net/http://www.hitap.net/


25/162


9

Financial investments are made in medi-cines and other commodities, as well as in inra-structure, in order to generate the services that

have an impact on health. Fig. 1.2 is one way toportray this chain o events. Consider, or exam-ple, the links between tobacco smoking andhealth. Te proportion o people who smokein a population (outcome), which representsa risk actor or lung, heart and other diseases(impact), is affected by various services and poli-cies that prevent ill-health and promote goodhealth (outputs). Among these services and poli-cies are ace-to-ace counselling, anti-smokingcampaigns, bans on smoking in public places,

and taxes on tobacco products. Te populationcoverage achieved by these interventions, whichare ofen used in combination, influences thenumber o smokers in the population (21).

In act, the problem o tobacco smoking inrelation to health goes beyond the results chainin Fig. 1.2. Smoking, like many other risk ac-tors, tends to be more requent among those whohave had less ormal education and who have

lower incomes. When seeking health care orsmoking-related illnesses, people educated to ahigher level are typically more aware o the ser-

vices available and more disposed to use them.Tese “social determinants”, which influenceprevention and treatment o illness, are a reasonor taking a broad view o research or health;they highlight the value o combining investiga-tions both within and outside the health sectorwith the aim o achieving policies or “heath inall sectors” (Box 1.3 and Chapter 2).

Even with an understanding o the deter-minants and consequences o service coverage,the balancing o investments in health services is

more than a technical matter. Te allocation opublic money to health also has ethical, moral andpolitical implications. Public debate, based on evi-dence rom research, is the mechanism or obtain-ing consensus on, or instance, who should beentitled to health care paid rom the public purse,under what conditions, and or what range o ser-

vices. Decisions on these issues, which involve acombination o ethical imperatives and political

Fig. 1.2. A representation of the results chain for universal health coverage, focusing

on the outcomes

Impact

Improved health status

Improved financialwell-being

Increased responsiveness

Increased health security

Inputs and processes

Health financing

Health workforce

Medicines, health productsand infrastructure

Information

Governance and legislation

Outputs

Service access andreadiness, including

medicines

Service quality and safety

Service utilization

Financial resources pooled

Crisis readiness

Outcomes

Coverage ofinterventions

Financial riskprotection

Risk factor mitigation

Quantity, quality and equity of services

Social determinants

Note: Each of these outcomes depends on inputs, processes and outputs (to the lef t), and eventually makes an impact on health(to the right). Access to financial risk protection can also be considered an output. All measurements must reflec t not only thequantity of services, but also quality and equity of access (first cross panel). Equity of coverage is influenced by “social determi-nants” (second cross panel), so it is vital to measure the spectrum from inputs to impact by income, occupation, disability, etc.


26/162


10

continues ...

Box 1.3. What do universal health coverage and social protection mean for people affected

by tuberculosis?

Tuberculosis (TB) is a disease of poverty that drives people deeper into poverty ( 22). In recognition of this fact, TB

diagnosis and treatment are free of charge for patients in most countries. The cost of TB treatment, provided as apublic service, is covered by domestic health-care budgets, often supplemented by international grants or loans( 23). This helps to reduce the financial barriers to accessing and adhering to treatment. However, free public healthservices are often not entirely free, and patients always face other expenses. Payments are made for medical tests,medicines, consultation fees and transport, and there are indirect costs of illness due to lost earnings.

For patients, therefore, the total cost of an episode of TB is often large in relation to their income ( 24). The aver-age total cost incurred by TB patients in low- and middle-income countries has been estimated at between 20%and 40% of annual family income, and the relative cost is higher in the lower socioeconomic groups ( 25–32). Thepoorest patients become indebted: 40−70% of them according to three studies carried out in Africa and Asia ( 26,

28, 29). A large part of the cost of TB treatment is incurred during the diagnostic phase before treatment starts ina subsidized TB programme. Costs are especially high for diagnosis and treatment by private doctors, with whommany of the very poorest seek care first ( 28, 29, 33, 34). Financial costs are commonly compounded by adverse social

consequences – such as rejection by family and friends, divorce, expulsion from school and loss of employment –which affect women in particular (35–37 ).

The research behind these findings has been essential for documenting the obstacles to the use of health servicesand the financial vulnerability of families affected by TB. It has helped to pinpoint where improved services, healthinsurance coverage and social protection can safeguard against the consequences of potentially fatal and financiallycatastrophic illness (38).

To estimate patients’ costs and identify barriers to access, WHO and partners have developed a toolkit which hasrecently been field-tested in surveys in several countries. The results have begun to inform national policy on socialprotection for people with TB (39, 40). Beyond free diagnosis and treatment, a full package of measures for socialprotection requires the following:

■ Universal health care, free of cost, or heavily subsidized. People do not enter the health-care system as TB

patients eligible for free treatment; they typically enter as patients with a respiratory illness. The journey tocorrect diagnosis and the start of treatment often takes weeks or months. Out-of-pocket expenses need to beminimized across the health system ( 23).

■ Specific social or financial risk protection schemes, compensating for the adverse financial or socialeffects of TB. For example, these may include travel vouchers, food packages, or cash transfers, as well aspsychosocial support.

■ Legislation to protect workers, ensuring that people with TB are not expelled from employment due to adisease that is normally rendered non-infectious after two weeks of correct treatment, and from which mostpatients fully recover.

■ Sickness insurance, compensating income loss during illness. ■ Instruments to protect human rights, minimizing stigma and discrimination, with special attention to

gender, ethnicity and protection of the vulnerable groups that are at particularly high risk of TB.

■ Whole-of-government approaches to address social determinants of health, and policies based on“heath in all sectors”, taking a broad view of the drivers of TB epidemics (Chapter 2). Poverty-reduction strate-gies and financial safety nets help prevent TB on many levels. Most important for TB prevention are goodliving and working conditions and good nutrition. Basic education supports universal health coverage byenabling healthy lifestyle choices and informing health-care decisions.


27/162


11

possibilities, place constraints on the analysis ohow to maximize health impact or the money

spent.In summary, the first challenge in movingtowards universal health coverage is to define theservices and supporting policies needed in anysetting, including financial risk protection, thepopulation that needs to use these services, andthe cost. Tis requires an understanding o thecauses o ill-health, the possible interventions,who currently has access to these services andwho does not, and the extent o financial hard-ship incurred by paying out-o-pocket. Acting

on behal o their populations, governmentsmust decide how to move closer to universalcoverage with limited financial resources. Tesecond challenge is to measure progress towardsuniversal coverage, using valid indicators and

appropriate data. Te two challenges go together,and research provides the evidence to address

them both.o highlight the role o research, the con-cepts o financial risk protection and healthservice coverage are expanded below, and thestrengths and weaknesses o methods or track-ing progress in each area are considered.

Investigating financialrisk protection

It is significant that, at a time o widespreadeconomic austerity, even high-income countriesare struggling to maintain current health ser-

vices and to make sure that everyone can affordto use them (41, 42). Te question o how to

None of the above is specific to TB, but TB control programmes are among those affected by the presence or absenceof health services and mechanisms for social protection. While disease-specific solutions can help partly and tempo-

rarily, universal health coverage, including social protection, is vital for sustained and effective TB control. Diseasecontrol programmes need to ensure that the patients they ser ve are eligible for, and actually receive, support fromthe general health services and not only from TB control programmes.

TB has close links with poverty and social vulnerability, and is one of the conditions that can function as a tracer foruniversal coverage. However, national TB control programmes need to add measures of financial risk protection toexisting indicators of service coverage. Among the measurable indicators are the following:

Outcome

■ For coverage of health services: TB diagnosis and treatment coverage (percentage of TB cases receivingproper care, and percentage successfully treated; see Fig. 1.5) and equity in coverage.

■ For financial risk protection: Access to financial risk protection schemes (percentage of patients using exist-ing schemes) and equity of access.

Impact ■ For financial risk protection: Cost of TB illness to patients (percentage with catastrophic expenditure, data

from surveys, using the tool to estimate patients’ costs). ■ Combined for universal coverage, financial risk pro tection and addressing soc ial determinants: TB

incidence, prevalence and death rates (from programme surveillance data, vital registration and population-based surveys).

... continued


28/162


12

provide and maintain financial risk protectionis relevant everywhere.

Access to financial risk protection couldbe expressed as the number o people enrolledin some type o insurance scheme or coveredby a tax-unded health service ree at the pointo use (43). In act, financial risk protection isofen more accurately judged by the adverseconsequences or people who are not protected(Box 1.4). As an example, survey data or 92countries (inhabited by 89% o the world’spopulation) show that the annual incidence ocatastrophic health expenditure is close to zeroin countries with well-established social protec-tion systems, but up to 11% in countries withoutsuch systems. In 37 o the 92 countries surveyed,the annual incidence o financial catastropheexceeded 2%, and in 15 it was above 4%.

An indirect measure o (the lack o) finan-cial risk protection is the ratio o out-o-pocketpayments to total health expenditure (table inBox 1.4; Fig. 1.3). In 63 countries, most o themlow-income countries where many people needfinancial risk protection, more than 40% o allhealth expenditure took the orm o direct out-o-

pocket payments. At the other end o the scale, in62 countries less than 20% o health expenditurewas out-o-pocket. Although the majority o the62 are high-income countries, among them areAlgeria, Bhutan, Cuba, Lesotho and Tailand.Te governments o these countries have shownhow, despite low average incomes, the poorestpeople can be protected rom having to makedisastrously large cash payments or health.

Tese surveys are also being used to track theprogress being made in financial risk protection

over time. Between 2005 and 2010 the proportiono health spending made through out-o-pocketpayments ell, on average, in all but one WHOregion (46 ). Te exception was Arica, where thelevel remained stable. wenty-three countriesacross all regions and income levels achieved areduction o at least 25% in the proportion o

health spending made through out-o-pocketpayments. Nevertheless, an estimated 150 mil-lion people suffered financial catastrophe in2010, and 100 million people were pushed belowthe poverty line (poverty is defined in Box 1.4)because they had to pay out-o-pocket or healthcare (46 ).

Tese conclusions derive rom two differentways o expressing financial risk protection; oneuses a direct measure rom primary survey data,the other uses an indirect measure derived romtwo different sets o surveys. Although the indi-cators differ, the results are similar. Te data sug-gest, as a rule o thumb, that when out-o-pocketpayments al l to or below 15–20% o total healthexpenditure, the incidence o financial catastro-phe will be negligible (47 , 48).

While these surveys give useul insightsinto financial risk protection, they raise urtherquestions about the different ideas that underpinfinancial risk protection, and about the sources odata and methods o measurement. For instance,should the incidence o catastrophic expenditureand impoverishment be given equal weight indescribing the extent o financial risk protection

in a country? Is it better to improve financial riskprotection on average, or to set a minimum levelo protection or everyone? How does financialrisk protection reflect the broader goal o socialprotection? What targets or milestones should beset or measures o financial risk protection untiluniversal coverage is ully achieved? Which con-ditions o ill-health, perhaps with costly treat-ments, tend to all outside national financial riskprotection mechanisms and thereore result infinancial impoverishment or households? Do

any o these measures capture the value associ-ated with peace-o-mind – the assurance thatis conerred by accessible, affordable, and reli-able health services? Tese are topics or urtherresearch, and in some cases public debate, on themechanisms o financial risk protection, and onthe methods o measurement.


29/162


13

Box 1.4. Measuring financial risk protection

The measurement of financial risk protection should ideally capture the number of people enrolled in some kind ofhealth insurance scheme and the number of people who are eligible to use – and able to afford – health services

provided by government, private sector or civil society.

Direct and indirect indicators of financial risk protection

Direct indicators Explanation

Incidence of catastrophic healthexpenditure due to out-of-pocketpayments

The number of people or the proportion of the population at all income levelswho spend a disproportionate share of their incomes on out-of-pocket pay-ments each year. Financial catastrophe is defined as out-of-pocket expenditureexceeding 40% of household income net of subsistence needs.

Mean positive overshoot ofcatastrophic payments

Shows the average amount by which households affected by catastrophicexpenditures pay more than the threshold used to define catastrophic healthspending.

Incidence of impoverishment due toout-of-pocket payments

The number of people or proportion of the population pushed below the pov-erty line because of out-of-pocket payments. The poverty line is crossed whendaily income falls below a locally-defined threshold, typically around US$1–2 perday. For people who are living near the poverty threshold, even small paymentspush them below the threshold.

Poverty gap due to out-of-pocketpayments

The extent to which out-of-pocket health payments worsen a households’pre-existing level of poverty.

Indirect indicators

Out-of-pocket payments as a share oftotal health expenditure

There is a high correlation between this indicator and the incidence of financialcatastrophe.

Government health expenditure as ashare of GDP

This recognizes that in all countries the poor need to be covered by financial riskprotection from general government revenues; they are rarely all covered whenthis proportion is less than 5%.

GDP, gross domestic produc t; US$, United States dollars.

There are, however, some difficulties in determining who is actually financially protected and to what extent, as twoexamples will make clear. First, health insurance as such does not guarantee full financial risk protection. Many formsof insurance cover only a minimum set of services, so that those insured are still required to make out-of-pocketpayments of different types, including informal cash payments (1). Second, government-financed services may beinadequate. For instance, they may not be available close to where they are needed, there may be too few healthworkers or no medicines, or the services may be perceived to be unsafe. In India, for example, everyone is eligibleto use government health services, but direct out-of-pocket payments are still among the highest in the world (44).

By contrast, it is more straightforward, and often more precise, to measure the consequences for people who donot have financial risk protection. The table above describes four direct indicators and two indirect indicators of

protection which can be measured by household expenditure surveys that include spending on health, as illustratedin the main text. The techniques used to measure these indicators are well established as a result of investment inrelevant research, and the survey data are readily available (45). To assess inequalities in financial risk protection, theseindicators can also be measured for different population groups, and can be stratified by income (or expenditure orwealth), place of residence, migrant status and so on.

Annual updates on the data and indicators that measure financial risk protection for all countries are reported inWHO’s Global Health Expenditure Database (44).


30/162


14

F i g . 1 . 3 .

O u t - o f - p o c k

e t e x p e n d i t u r e s o n h e a l t h a s a

p e r c e n t a g e o f t o t a l e x p e n d i t u r e o n h e a l t h , 2 0 1 3

N o t e : B a s e d o n W H O d a t a F e b r u a r y 2 0 1 3 .


31/162


15

Investigating the coverageof health services

Te evolution o thinking on universal health cov-erage has also led to a greater understanding o theunctions that health systems should serve. Teseunctions should be concerned with prevention aswell as treatment. Tey should assure: (i) access toessential medicines and health products; (ii) moti-

vated and skilled health workers who are accessible tothe people they serve; (iii) integrated, high-quality,patient-centred services at all levels rom primaryto tertiary care; (iv) a combination o priority pro-grammes or health promotion and disease control,including methods or prevention and treatment,

which are integrated into health systems; (v) inor-mation systems that produce timely and accuratedata or decision-making; and (vi) health financ-ing systems that raise sufficient unds or health,provide financial risk protection, and ensure thatunds are used equitably and efficiently.

In outlining the concept o universal cov-erage, Fig. 1.1 depicts health services along asingle axis. In reality, there is a diversity o ser-

vices delivered on several levels, depending onthe nature o the health condition and the typeo intervention. Te elements o each row inFig. 1.4 are the services that are deemed neces-sary. Preventive services (e.g. vaccines) and cura-tive services (e.g. drug treatments) must address

Fig. 1.4. A framework for measuring and monitoring the coverage of health services

Health system inputs

Outputs: availability, readiness, quality, utilization

Indicators ofservice coverage,

including

promotion,

prevention and

treatment

MNCH

HIV/TB/malaria

NCDs andrisk factors

Injuries

Priority healthconditions

Levels of health system/service delivery

Non-personal

Community-based

Primary (facility)

Secondary(hospital)

Tertiary(hospital)

HIV, human immunodeficiency virus; MNCH, maternal, newborn and child health; NCDs, noncommunicable diseases; TB,tuberculosis.Note: “Non-personal” health services are actions applied either to communities or populations – such as mass health educa-tion, policy development or taxation – or to the nonhuman components of the environment – such as environmental healthmeasures. Community-based health services are defined as individual and community health actions delivered in the commu-nity (e.g. by community health workers) and not through health facilities. They are often considered to be par t of the primaryhealth care service.


32/162


16

the principal causes o ill-health now and in theuture (e.g. the causes addressed by MDGs 4–6,and noncommunicable diseases in low-incomecountries). Te columns in Fig. 1.4 represent the

various levels on which services are delivered: inthe community, to individuals at primary carecentres or at secondary or tertiary hospitals,and to whole populations (nonpersonal) (49). Asillustrated by its position in Fig. 1.4, a strong pri-mary care system is central to an effective healthsystem (4). “Nonpersonal” services are actionsapplied either to communities or to populations;broadly, they are educational, environmental,public health and policy measures in a range osectors that influence health.

he MDGs have been a powerul orce bothor better health and or measuring progresstowards better health with precisely-deinedindicators, data collected in standard ways, andwith internationally-agreed targets (46 , 50). Asan illustration, Fig. 1.5 shows some examples oprogress towards MDG 6 (i.e. “to combat HIV/AIDS, malaria and other diseases”). For HIV/AIDS, “universal access” to antiretroviral ther-apy is currently deined as treatment o at least

80% o the eligible population. By 2010, 47% oeligible people were receiving treatment. husthe target was missed globally, but national datashow that it was reached in 10 countries, includ-ing some countries with a high prevalence oHIV, such as Botswana, Namibia and Rwanda.

MDG 7 is concerned with environmentalsustainability. As a contribution to universalcoverage, it includes the target to reduce by hal,between 1990 and 2015, the proportion o peoplewithout access to sae drinking-water and basic

sanitation. Notwithstanding some methodo-logical limitations in measurement, more thantwo billion people gained access to improveddrinking-water sources between 1990 and 2010,including piped supplies and protected wells. TeMDG target was met by 2010, although access toimproved water supplies was generally lower inrural areas than in urban ones (50, 55).

Tese investigations o progress towardsthe MDGs show, or selected interventions, howar we are rom universal coverage. Ideally, weshould measure the coverage o all the interven-tions that make up health services, but that isnot usually possible even in high-income coun-tries. In Mexico, or example, 472 interventionswere covered by five separate health protectionmechanisms in 2012, mainly under the healthinsurance programme known as Seguro Popular (Chapter 3, case-study 11) (43). It is easible,however, to take a selection o interventionsand indicators, and use them as “tracers” o theoverall progress towards universal coverage. Teinterventions selected should be accessible toeveryone who is eligible to receive them underuniversal health coverage in any setting.

Whether the tracers actually represent accessto all health services needs to be evaluated, andthis is a task or research. Nevertheless, to illus-trate the idea, Box 1.5 shows how tracers o thecoverage o maternal and child health services,combined with measures o financial risk pro-tection, give an overview o service coverage inthe Philippines and Ukraine. Te two countries

are similar with respect to the coverage o healthservices. Te differences are in the incidence ocatastrophic health expenditure and o povertydue to out-o-pocket payments.

An important unction o this kind o anal-ysis is to stimulate national policy dialoguesabout why the coverage o certain interventionsis insufficient. For instance, in the comparisonin Box 1.5, would the addition o other inter-

ventions tell a different story about progresstowards universal coverage? Do the indicators o

catastrophic expenditure and poverty representaspects o financial risk protection that differbetween the two countries? And there is alwaysthe question: “Are the underlying data accurate?”

Te coverage o services depends on how thoseservices are provided. Te inputs can be investi-gated in addition to, or as a proxy or, direct meas-ures o coverage (Fig. 1.2). For instance, WHO


33/162


17

Fig. 1.5. Towards universal health coverage: examples of the growing coverage of

interventions for the control of HIV/AIDS, tuberculosis, malaria and neglected tropical

diseases

ACT, artemisinin-based combination therapies; AIDS, acquired immunodeficiency syndrome; ART, antiretroviral therapy; HIV,human immunodeficiency virus; ITN, insecticide-treated bed nets; PMTCT, prevention of mother-to-child transmission.Note: Between 2003 and 2008, the denominator for ART coverage was all HIV-infected people with CD4 cell counts of ≤ 200

cells/μL, but in 2009 and 2010 the denominator was all people with ≤ 350 CD4 cells/μL. Hence the apparent fall in coveragebetween 2008 and 2009.For PMTCT with ART, the numerator in 2010 excludes treatment with single-dose nevirapine.For malaria, data on household coverage with ITN and on suspec ted cases tested are for the WHO African Region. Data on ACTare for the whole world.The interpretation of universal coverage is 100% coverage for all interventions, except for interim targets of ≥ 80% coverage for ART,≥ 90% for the percentage of tuberculosis patients cured, and variable coverage targets for neglected tropical diseases ( 23, 51–53).Reproduced, by permission of the publisher, from Dye et al. (54).


34/162


18

Box 1.5. Measuring the coverage of health services

It is not usually possible to measure all aspects of service coverage even in high-income countries, but it is feasibleto define a set of “tracer” conditions, with associated indicators and targets for interventions, to track progress

towards universal coverage. The choice of tracer conditions and the associated indicators and data, and the work todemonstrate that these measures are representative and robust, are topics for further research (56).

Using tracers to track progress towards universal coverage in the Philippines and Ukraine

Philippines Incidence of catastrophic health expenditure

Poverty incidencedue to

out-of-pocketpayments

Normalizedpoverty gap

due toout-of-pocket

payments

Antenatal care (4 visits)

Birthsattended by

skilled healthpersonnel

DTP3immunization

UkraineIncidence of catastrophic health expenditure

0

20

40

60

80

100

0

20

40

60

80

100 Poverty incidencedue to

out-of-pocketpayments

Normalizedpoverty gap

due toout-of-pocket

payments

Antenatal care (4 visits)

Birthsattended by

skilled healthpersonnel

DTP3immunization

DTP3, diphtheria–tetanus–pertussis.

As an example, three tracers of the coverage of maternal and child health services, together with three measures offinancial risk protection, give an overview of service coverage in the Philippines and Ukraine (see figure). The threeservice coverage indicators are: skilled birth attendants during delivery, three-dose diphtheria–tetanus–pertussis(DTP3) immunization and four antenatal visits (%). The three indicators of financial risk protection are: incidenceof financial catastrophe due to direct out-of-pocket payments, incidence of impoverishment due to out-of-pocketpayments, and the widening of the poverty gap due to out-of-pocket payments. For impoverishment, the worstpossible outcome was estimated to be 5%, which is higher than measured impoverishment due to out-of-pocketpayments in any country. In the figure, 100% service coverage and financial risk protection lie at the outer edgeof the radar diagram, so a fully-filled polygon represents universal coverage. However, financial risk protection ismeasured as the consequences of its absence (Box 1.3), so the percentage scale is reversed for these three indicators.

With respect to the coverage of health services, the Philippines and Ukraine are similar. The differences are in theincidence of catastrophic health expenditure (higher in the Philippines) and the incidence of poverty due to out-of-pocket payments (higher in Ukraine). These observations, based on this particular set of indicators, raise questionsabout how to make fur ther progress towards universal coverage (see main text).

These six tracers could be supplemented with others. For instance, standard indicators of progress exist for HIV/AIDS, tuberculosis, malaria, and some noncommunicable conditions (Fig. 1.5) (57 ). As more indicators are added, thepolygon in the figure approaches a circle. Ideally, all indicators would be disaggregated by wealth quintile, place ofresidence, disability and gender, and by other important characteristics of population groups.


35/162


19

compiles data rom surveys o the availability andprice o essential medicines (Fig. 1.6) (58). Surveyscarried out between 2007 and 2011 ound that 14generic essential medicines were available on aver-age in 52% o public health acilities and in 69% oprivate health acilities. Te averages differed little

between lower-middle-income countries andupper-middle-income countries, and there werelarge variations among countries within each cat-egory. Among upper-middle-income countries,the availability o the 14 generic medicines variedrom zero in the State o Rio Grande do Sul, Brazil,to 97% in the Islamic Republic o Iran.

One advantage o monitoring essential medi-cines as one way o tracking service coverage is that

comparable data are increasingly available, andthe quality o these data, collected through regularhealth acility assessments, is also improving. Morethan 130 countries had an essential medicines listby 2007, and 81% o the low-income countries hadupdated their lists in the previous five years.

Equity and universalhealth coverage

Systems or monitoring the coverage o servicesshould record not only the total number o peoplewho have or do not have access, but also some soci-odemographic details about them. When coverage

Fig. 1.6. Availability of selected generic medicines in public and private health facilities during

the period 2007–2011

100

80

60

40

20

0

Public sector8

Private sector10

Public sector7

Private sector7

Low-income and lower-middle-income countries Upper-middle-income countries

87.1

50.1

21.2

90.7

67.0

22.2

96.7

44.4

0

96.7

71.1

44.4

68.5

51.8

Mean Maximum Minimum

Average in the private sector Average in the public sector

H e a t h f a c i

l i t i e s w i t h m e d i c i n e s a v a i l a b l e ( % )

Reproduced, by permission of the publisher, from United Nations (58).


36/162


20

is truly universal everyone has access, but partialcoverage may benefit certain groups over others. omonitor equity in the supply of, and demand for,health services, indicators should be disaggregatedby income or wealth, sex, age, disability, place ofresidence (e.g. rural/urban, province or district),migrant status and ethnic origin (e.g. indigenousgroups). For instance, gains in access to safe drink-ing-water have been uneven: 19% of people livingin rural areas did not have improved water sourcesby 2010, in contrast with only 4% in urban areas(50). Tis analysis shows where to target furtherefforts to improve coverage.

Another example o the uneven distributiono services – or maternal, newborn and childhealth – is portrayed in Fig. 1.7. Te summarymeasure o service coverage includes amily plan-ning, maternal and newborn care, childhood vac-cination, and treatment o childhood illness. Temean coverage in 46 low- and middle-incomecountries varied by wealth quintile as expected,but there was also great variation within eachquintile. o achieve universal health coverage,it is necessary to eliminate the gap between thepoorest and richest both within and between

quintiles, and to raise the levels in all quintiles.As a general rule, the countries that make thegreatest progress in maternal and child health arethose that successully narrow the gap betweenthe poorest and richest quintiles (59, 60). Tis is aorm o “progressive universalism” in which thepoorest individuals gain at least as much as therichest on the way to universal coverage (61).

Coverage of health services:

quality as well as quantityIt is not just the quantity of health services pro-

vided that is important, but also the quality ofthem. Following a long tradition of researchon the quality of care, the Organisation forEconomic Co-operation and Development(OECD) has developed measures of quality forselected interventions: for cancer and mental

health, for aspects of prevention and health pro-motion, and for patient safety and patient experi-ences (15, 62–64).

Fig. 1.8 illustrates one aspect o the qualityo care, namely the risk o death in hospital ol-lowing ischaemic stroke. Te risk is measured

as the proportion o people who die within 30days o admission (Fig. 1.8) (65). As with manymeasures o quantity, national statistics on thequality o care are ofen not precisely compara-ble. In this instance, case-atality rates shouldideally be based on individual patients, butsome national databases do not track patients inand out o hospitals, between hospitals or evenwithin the same hospital, because they do not

Fig. 1.7. A summary measure of service

coverage for maternal and

child health, in which inequity

is reflected in the differencesbetween wealth quintiles

H e a l t h s e r v i c e c o v e r

a g e ( c o m p o s i t e i n d e x % )

Q1 Q2 Q3 Q4 Q50

10

20

30

40

50

60

70

80

90

100

Wealth quintile (Q1 = poorest, Q5 = richest)

49%

58%

63%

70%

77%

Median Interquartile range

Source: Demographic and Health Surveys or Multiple Indicator

Cluster Surveys in 46 low- and middle-income countries.


37/162


21

use unique patient identifiers. Te data in Fig. 1.8 are thereore based on single hospital admissionsand are restricted to mortality within the samehospital. Tere are big differences in case-atalityrates between countries, but some o the vari-ation might be explained by local practices odischarging patients rom hospitals, and trans-erring patients to other hospitals. o select andagree on internationally comparable indicatorso quality is another task or research.

Conclusions: research neededfor universal health coverage

When all WHO Member States made a commit-ment to achieving universal health coverage in2005 they took a major step orward or publichealth. aking that step launched an agenda orresearch. We do not yet know how to ensure thateveryone has access to all the health services theyneed in all settings, and there are many gaps inunderstanding the links between service cover-age and health (66 , 67 ). Research is the means ofilling these gaps.

With a ocus on research, the goal o thisreport is not to measure definitively the gapbetween the present coverage o health servicesand universal coverage, but rather to identiy thequestions that arise as we try to achieve universalcoverage and to discuss how these questions canbe answered in order to accelerate progress.

In this chapter research questions o twokinds have been identified. Te first and mostimp

Date post:	07-Aug-2018
Category:	Documents
Upload:	anonymous-im9mma5
View:	215 times
Download:	0 times

research-for-universal-health-coverage(9789240690837_eng).pdf

Documents