Global Forum 30

8/8/2019 Global Forum 30

1/20

Chapter 6

Monitoring financial flows

Section 1Background

Section 2Why measure resource flows?

Section 3Methods

Section 4Results

Section 5Discussion and future stra tegies


2/20

For a summary of this chapter, see the Executive Summary, page xviii.


3/20

1016. Monitoring financial flows

The Commission on Health Research forDevelopment drew attention to theimportance of health research as the essentiallink to equity in development1. It proposedthat low- and middle-income countries shouldreview and strengthen the management ofhealth research so as to meet their nationalneeds as well as contributing to the globalfund of knowledge. The Commission also

recommended that governments in low- andmiddle-income countries allocate at least 2%of national health expenditures and 5% ofexternally funded programmes to research andcapacity strengthening. The Commissionhoped that these financial arr a n g e m e n t swould provide a secure foundation for fundingthe priority re s e a rch needs in low- andmiddle-income countries, based on the newconcept of Essential National HealthResearch.2 The expectation was that low- and

middle-income countries would review theirc u rrent spending on health re s e arch andwould strive to meet the stated goals.

Rather disappointingly, neither the low- andmiddle-income countries nor the donorcommunity enthusiastically followed up theC o m m i s s i o n s recommendations, although

there were a few exceptions. Furthermore,since most low- and middle-income countrieswere not actively tracking the pattern ofspending on health research, it was difficult toknow how close they were to the target andwhat trends were occurring over time. Onemajor obstacle was the lack of testedmethodologies for monitoring spending onhealth research at the country level.

In an attempt to fill this gap, the GlobalForum for Health Research has tackled theproblem through its support of a network ofinvestigators. This chapter synthesizes themain points of a recently published report onthe first three years of the project.3 The aim ofthe publication is to stimulate interest in thisi m p o rtant issue in the hope that otherinvestigators will critically review themethodology that this team has developed

and perhaps offer refinements. Furthermore,the tentative results from a few countriesshould stimulate others to follow the exampleand provide data from many more countries.Ideally, other studies will adopt the coredefinitions so as to facilitate comparisonsamong countries and also to examine trendsover time.

Sec tion 1

Background

1 Commission on Health Research for Development, 1990. Health Research: Essential Link to Equity in Development. New York,

Oxford University Press

2 Task Force on Health Research for Development, 1991. Essential National Health Research. A Strategy for Action in Health and

Human Development. c/o United Nations Development Programme, Geneva, Switzerland.

3 Global Forum for Health Research, Monitoring financial flows for health research. October 2001.


4/20

102

Sec tion 2

Why measure resource flows?

1. Fragmented dataKnowledge of re s o u rce flows for healthresearch is an important input into prioritysetting. Although funding agencies andcompanies in the public and private sectorsmay have internal mechanisms to trackhealth re s e a rch and development (R&D)

e x p e n d i t u res, the available data is veryfragmented. The Organisation for EconomicCooperation and Development (OECD) is theonly institution with a mandate to regularlycollect and disseminate standardized nationalstatistics on aggregated health-related R&Dfor its member States. R&D funds arereported as part of Science and Technology(S&T) information. While no equivalentinstitutional mechanism exists in low- andmiddle-income countries, information onresource flows has recently begun to emerge.

The challenge now is to develop and applyhealth R&D indicators which can be collectedin low- and middle-income countries,countries in transition and high-incomecountries. Wherever possible, such indicatorsshould draw on existing intern a t i o n a lstatistical standards. Consistency will facilitatecomparisons between countries while alsomeeting national and regional needs.

A more detailed mapping of global resourceflows will help decision-makers in allcountries to target, and there f o re betterallocate, funds supporting health R&D.Mapping will also help monitor shifts in R&Dfunding allocations towards the mostprevalent health conditions and determinants,identify the areas which do not attract enough

funding, and avoid unnecessary duplicationof research efforts. These measures, in turn,are expected to have a significant impact onreduction of the burden of disease and injuryin low- and middle-income countries,particularly among the poor.

2. Progress in resource flows measurementSince the Commission re p o rt, the 1996Report of the WHO Ad Hoc Committee onHealth Research reiterated the importance ofestablishing an institutional mechanism forthe systematic tracking of investments inhealth R&D. Although that report provideds u m m a ry data on public and privateinvestments in health research and estimatedglobal health research investments at US$56billion, the authors acknowledged thecomplexity of developing a useful system tomonitor resource flows.

Beginning in 1999, the Global Forum forHealth Research supported efforts to developand implement a system for tracking andre p o rting investments in health re s e a rc h .Monitoring focused on investments made bylow- and middle-income countries, high-income country agencies providing funds tolow- and middle-income countries, and forproblems relevant to low- and middle-income

countries.

The five-year Resource Flows Projects goal isto improve priority setting through developinga database of internationally comparablestatistics on global resource flows for healthresearch. To reach this goal, the Global Forumand its partners intend to:


5/20


define the inputs for the database develop institutional mechanisms for

providing the inputs report health R&D expenditures

ensure that decision-makers have access tothe database link these activities with priority-setting

e x e rcises in order to maximize thee ffectiveness of investments in healthresearch.

This chapter describes the first three years ofproject work. An Advisory Group (Annex 6.1)met with Global Forum staff four times

between January 1999 and February 2000 toassist in the development and assessment ofthe methodology used for obtaining data forthe project, including the conceptual

framework.

Collecting and reporting data on funding forhealth research are challenging tasks and thisreport represents only the first step towardsthat end. The Global Forum is activelysupporting the work carried out by others,facilitating standardization where feasible,helping to fill in gaps to disseminate theinformation.

Sec tion 3

Methods

1. Definition of health research anddevelopment

The following definitions of research andhealth res ea rch, used by the OECD andUNESCO, were adopted for this study:4

R e s e a rch and experimental development

comprises creative work undertaken on a

systematic basis in order to increase the stock of

knowledge, including knowledge of man, culture

and society, and the use of this knowledge to

devise new applications.

Thus, health re s e a rch is a process forgenerating systematic knowledge, and to testhypotheses, within the domain of medicaland natural sciences as well as social sciences,including economics and behavioural science.The information resulting from this processcan be used to improve the health ofindividuals or groups.

2. Conceptual modelOne objective of the project was to measuretotal funding of health R&D worldwide, with

4 OECD. The Measurement of Scientific and Technological Activities, Proposed Standard Practice for Surveys of Research and

Experimental Development, Frascati Manual 1993, Paris, 1994.


6/20

104

* Proportions for surfaces A, B and A/B are indicative only.

Insert 6.1Graphic representation of health research funding*

A = R&D by high-income countries

B = R&D by low- and middle-income countries

A/B = R&D efforts converge or overlap

(see text for details)

A BA/B

particular emphasis on R&D for or by low-and middle-income countries. Insert 6.1illustrates the main components. Area Ac orresponds to the health R&D eff orts of

high-income countries. Area B rep res en t sthe health R&D effo rts of low- and middle-income coun tries. The overlapping Are aA/B depicts where these eff orts converge oroverlap. These three areas could be furth erdefined in several ways. For the purpose offinancial flows in the present study, Area Adescribes all health R&D fund e d by high-income countries; Area B, all health R&Dfinanced by and carried out in low- andmiddle-income countries. Area A/Bco rresponds to R&D funded by high-incomecountries and carried out in and for thepr i m a ry benefit of low- and middle-incomecountries. The area should also incorporateR&D carried out in high-income countrieswhich is for, or relevant to, the needs of low-and middle-income countries, and R&Dc a rried out in low- and middle-incomecountries which is for, or relevant to, theneeds of high-income countries. The three

a reas constitute the framework for projec tdata collection.

Data on health R&D expenditures can be

collected from the unit providing the funds(the funder) or from the unit actuallycarrying out the research (the performer).The data compiled within areas A and A/Bw e re generally collected from funders,whereas the data for area B were collectedfrom both performers and funders. Becausethe three categories of data were compiledusing different approaches and from differentsources, it was challenging to aggregate theminto the global total, and especially to avoiddouble counting of area A/B.

The countries undergoing transition fro mcentralized to market economies do not fiteasily into the model. They are examined in aseparate section but are also treated in thediscussion of area A/B, as they are eligible forsome of the types of support for health R&Dtraditionally oriented towards low- andmiddle-income countries.


7/20

105

3. Classification frameworkThe major product developed under theguidance of the Advisory Group was aclassification system to cover total health R&D.

The aim was to produce a set of categories thatwould be useful for decision-makers especiallyin low- and middle-income countries. Itwould, in addition, serve as a framework forspecial surveys and for documenting datacompiled from other sources.

The main categories of the classification arelisted in Insert 6.2.

There are other dimensions by which R&Dre s o u rce flows are commonly classified.These may include activity, discipline, topic,

location, beneficiary and developmentoutcome. The Advisory Group andconsultants endorsed the development of ac o m p rehensive framework that includedmultiple levels of disaggregated data andthoroughly discussed the details.

Insert 6.2Classification of resource flows for health research

Levels of aggregation of R&D funds

A.1 Non-oriented, fundamental research

No further disaggregation

A.2 Health conditions, diseases or injuries

A.2.1 Group I (communicable, maternal, perinatal and nutritional conditions) *

A.2.2 Group II (noncommunicable diseases) *

A.2.3 Group III (injuries) *

A.3 Exposures, risk factors that impact on health (determinants)

A.3.1 Risk factors within the health system

A.3.2 Risk factors outside the health system

A.4 Health systems research

A.4.1 Policy and p lanning research

A.4.2 Health services delivery research

A.4.3 Surveillance

A.5 Research capacity building

A.5.1 Recurrent expenses

A.5.2 Capital expenditures

* Groups I, II and III follow the Global Burden of Disease classification (C.J. Murray & A. Lopez, Global Burden of Diseases and Injuries. Volume1. WHO, 1996)



8/20

106

The funding classification tested in previousexercises had been used for health R&Dfinanced by high-income countries (Areas Aand A/B in Insert 6.1).

The user/perf o rmer classification wasdeveloped during the experimental healthR&D surveys in selected low- and middle-income countries (Area B in Insert 6.1).

4. Sources of data5

Previous global resource flow studies have, byand large, focused on data from existingdatabases and estimated the data from low-and middle-income countries. The presentp roject extends that work by developing

special surveys based on the newclassification; by making more extensive useof recently published data sets; and byundertaking institution-specific case studiesinvolving personal contacts with fundingagencies and low- and middle-incomecountry institutions. The following strategieswere used:

(a) Funder questionnaires(b) Special survey for low- and middle-

income countries(c) Funder surveys/databases(d) Government S&T surveys(e) Evaluations, annual reports, websites(f) Interviews/personal contacts.

5 For further details, see Global Forum for Health Research, Monitoring financial flows for health research. October 2001

It was also necessary to identify someinstitutional categories for the main types ofhealth R&D funders and performers. The

following groups of funders and users/performers were identified (Insert 6.3):

Insert 6.3Classification of funders and performers

Funders Performers in low- and middle-income countr ies

Public sector Government departments Government departments

(national aid agencies) Academic/research institutes

Hospitals

Others

Private sector Pharmaceutical firms Pharmaceutical firmsPrivate non-profit organizations Academic/research institutes

Hospitals/laboratories

NGOs

Others

International Multilateral Foreign institutions

Bilateral Government departments

Others


9/20


1. Global health R&D and main aggregatesfunded by high-income and transitioncountriesBased on partial estimates, public and privatesources worldwide invested a minimum ofUS$73.5 billion in health R&D in 1998 (orabout 2.7 % of total health expenditures

worldwide). Governments in high-incomecountries, countries in transition, and low-and middle-income countries invested at leastUS$ 37 billion (50%), and the pharmaceuticalindustry US$30.5 billion (42%). Private, non-p rofit and university funds provided theremaining US$6 billion (8 %). See Insert 6.4.

Governments of countries having establishedmarket economies (high-income countries)spent US$34.2 billion on health R&D, in

addition to an estimated US$350 million indevelopment assistance for health R&D.

G o ve rnments of the Central and EasternEuropean countries in transition for whichestimates are available (Czech Republic,H u n g a ry, Poland , Romania, the RussianFederation, Slovak Republic and Slovenia)spent an estimated US$200 million out of atotal health R&D expenditure of about

US$360 million in these countries.

For low- and middle-income countries, it isestimated that Argentina, Brazil, Mexico andother Latin American countries, in additionto India, Malaysia, the Philippines, Thailand,Turkey and Chinese Taipei, spent aminimum of US$2.5 billion in 1998 onhealth R&D. Data for other low- and middle-income countries which spent significantamounts on health res e arch, such as theP e o p l e s Republic of China, are not availableat this stage.

Sec tion 4

Results

Insert 6.4Estimated global health R&D funding 1998 (in current US$)

Total US$73.5 billion

Total Percent(billion US$)

Public funding: high-income and transition countries 34.5 47

Private funding: pharmaceutical industry 30.5 42

Private not-for-profit funding 6.0 8

Public funding: low- and middle-income countries 2.5 3

Total 73.5 100


10/20

108

Overall investments in health R&D fro mpublic, industrial and non-profit sourc e si n c reased in real terms in high-incomecountries during the 1990s, in contrast to a

general decrease in the countries int r a n s i ti o n .

The figure of US$73.5 billion contrasts withthat of US$56 billion in 1992 (in currentterms). It is estimated that up to one-third ofthe increase between 1992 and the presentstudy is in real terms. Data from low- andmiddle-income countries, when available,indicate considerably larger R&D investmentsin health from national sources than earlierstudies had estimated.6 While this increasereflects real growth in overall investments inhealth R&D, it probably also reflects betterreporting for these countries.

2. Funding health R&D in high-incomecountries(a) Public funding of health R&DG o v e rnments in high-income countriesinvested US$34.2 billion in health R&D in1998. The United States provided over half ofthis amount, investing US$19.5 billion. Japan

contributed US$2.9 billion, Germany US$2.4billion, France US$2.2 billion, the UnitedKingdom US$1.8 billion and CanadaUS$0.75 billion. Together, the G7 countries(including a rough estimate for Italy)accounted for 90% of total publicly fundedhealth R&D in the high-income countries. Allother high-income country govern m e n t stogether contributed US$3.5 billion.

For the United States, public funds spent for

health R&D are estimated as corresponding to0.22% of GDP, the highest figure among high-income countries. This is followed bySweden, Austria and Finland, whose R&D

funds correspond to more than 2% of nationalhealth expenditure.

Public funding of health R&D grew in the

high-income countries both as a group, and inv i rtually all of the countries studied,individually. This was partly due to improvedcoverage and reporting of the data series. Forexample, the category funding of hospitalR&D was added during the project period inFrance, the United Kingdom and Finland.

(b) Indu str y funding of health -re l a t e dR&DThe pharmaceutical industry is the dominantindustrial funder of health-related R&D. Themajority of pharmaceutical research is fundedby multinational companies, which areo fficially headquart ered in high-incomecountries. There is of course somepharmaceutical R&D carried out in transitionaland low- and middle-income countries.

The pharmaceutical industry, includingbiotechnology companies, spent an estimatedUS$30.5 billion in 1998, corresponding to42% of all health R&D funding (Inserts 6.4

and 6.5). Reported investment in R&D as ashare of sales in the pharmaceutical industryis very high. It ranged between 12% and 21%of turnover in the 15 companies having thel a rgest R&D investment. The share washigher still in the 10 biotechnologycompanies making the largest R&Dinvestments, corresponding to allocations of26% to 67% of revenues to R&D (Insert 6.5)7.

It has not been possible to provide a

breakdown of the global total by country.From national sources it is estimated thatresearch-based pharmaceutical companies inthe United States invested US$20.3 billion in

6 C. Michaud, C.J.L. Murray, 1996. Resources for health research and development, 1992: a global overview. Annex 5 ofInvesting

in Health Research and Development. Report of the Ad Hoc Committee on Health Research relating to future intervention

options. Geneva, World Health Organization, 1996.

7 For further details, see Global Forum for Health Research, Monitoring financial flows for health research. October 2001


11/20


R&D in human-use pharmaceuticals, ofwhich US$16.9 billion were spent at homeand US$3.4 billion abroad.8

(c) Private foundations and other not-for-profit organizationsPrivate foundations and other not-for-profit

Source: SCRIP 1999, Pharmaceutical Company League Tables; Ernst & Young: European Life Sciences 99, Sixth Annual Repor t

8 Pharmaceutical Research and Manufacturers of America, Annual Survey 2000.

Insert 6.5R&D expenditures by major pharmaceutical and biotechnology companies, 1998

(US$ million)

Pharmaceutical companies

15 leading companies with largest R&D R&D expenditures Per US$ of totalpharmaceutical sales

AstraZeneca 2,183.0 0.17

Glaxo Wellcome 1,927.5 0.15

Roche 1,893.1 0.19

Merck & Co 1,821.1 0.12

Novartis 1,801.3 0.16

Bristol-Myers Squibb 1,559.0 0.12

Hoechst Marion Roussel 1,426.2 0.18

Johnson & Johnson 1,400.0 0.16

SmithKline Beecham 1,394.0 0.18

American Home Products 1,389.9 0.16

Rhne-Poulenc Rorer 1,010.5 0.17

Boehringer Ingelheim 866.0 0.19

Bayer 852.3 0.18

Novo Nordisk 420.1 0.21

Yamanouchi 415.1 0.17

Biotechnology comp anies

10 companies with largest R&D

Amgen 663.3 0.26

Chiron 108.0 NA

Genentech 396.2 0.55

Biogen 177.2 0.45

ALZA 156.8 0.67Immunex 92.0 NA

Genzyme 63.0 NA

British Biotech 20.8 NA

Chiroscience 51.3 NA

Genset 10.1 NA


12/20

o rganizations spent an estimated US$3.4billion on health research in 1998 of whichUS$1.9 billion came from the United States,US$700 million from the United Kingdom,

US$240 million from Japan, US$200 millionfrom Canada and US$120 million came fromFrance. An estimated US$200 million camef rom all other high-income countr iescombined.

The two largest private sponsors of research in1998 were the Wellcome Trust in the UnitedKingdom, which spent US$650 million onbiomedical research, and the Howard HughesMedical Institute (HHMI) in the UnitedStates, which spent US$389 million.9

In addition to these sources, at least US$2.5billion was contributed to health researchthrough the private funds of universities andcolleges in Canada, Japan and the UnitedStates.

3. Funding medical research in Central andEastern European countries in transitionCountries in transition do not fit neatly intothe model of country groups used for this

study. Like the high-income countries, mosthad fully developed science and technology aswell as health care systems. However, thesesystems suffered greatly during their difficultinitial period of adjustment to marketeconomies. And like the low- and middle-income countries, they have been recipientsof aid from high-income countries, mostly toimprove economic performance rather thanfor social objectives.

In 1998, the Czech Republic, Hungary,Poland, Romania, the Russian Federation, theSlovak Republic and Slovenia spent theequivalent of approximately US$360 millionon health R&D. Government financing

accounted for just over US$200 million. Themagnitude of R&D efforts are not adequatelyreflected in these dollar figures, however, as aresult of these countries weak currencies.

Comparison of purchasing power parities,reflecting the average cost of goods andservices in each country, raises total healthR&D funding to US$800 million, of which anestimated US$450 million was financed bypublic sources.

4. Funding for health R&D by low- andmiddle-income countriesThe study did not attempt to be acomprehensive review of all low- and middle-income countries investing in health research.Research focused on a few, selected countriesin which teams conducted special surveys onhealth R&D, in addition to countries forwhich published information already existed.As such, this section is not meant to provide acomprehensive analysis of investments.

It is estimated that Argentina, Brazil, India,Malaysia, Mexico, Panama, Peru, thePhilippines, Thailand and Turkey spent aminimum of US$2.3 billion in 1998 on health

R&D. Data for other low- and middle-incomecountries, among them countries which spendimportant amounts on health research such asthe Peoples Republic of China, are notavailable at this stage. These gaps in knowledgewill be addressed during Phase 2 of the project.

(a) Special surveys of health R&DA thre e - co u n try study conducted for theCouncil on Health Research for Development(COHRED) in Malaysia, the Philippines and

Thailand traced flows of funds for health R&Dfrom the funding sources to the performers ofthe research projects concerned. As a fullreport has been published, only the mainaspects will be described here.10

110

9 Global Forum for Health Research, Monitoring financial flows for health research. October 2001.

10 B.A. Alano Jr and E.S. Almeria, Tracking country resource flows for health research development (R&D). The Philippines, Centre

for Economic Policy Research, 2000.


13/20

The survey concluded that these thre ecountries spent over US$33 million in 1997and US$30 million in 1998 (total expendituresby public and private sectors), with Thailand

spending about 50% of the total.

Government is the main source of funds forhealth R&D. In Malaysia these funds come

largely from the Department of Science andTechnology whereas the Department ofHealth is the main source in Thailand. In thePhilippines, both ministries contribute.

Multilateral and bilateral funding arerelatively much higher (28%) in thePhilippines than in the other two countries(see Insert 6.6).

(b) Health R&D data from ongoing R&DsurveysTotal annual investment in Scientific andTechnological Activities (S&T) in LatinAmerica amounted to US$15.3 billion in1998, of which R&D accounted for nearlyUS$11 billion.11 Three countries (Argentina,Brazil and Mexico) accounted for 86% of theR&D spending. The percentage of GDPdevoted to R&D ranged from about 1% in

Brazil and Costa Rica to about 0.1% inEcuador, El Salvador and Trinidad with aregional average of 0.58%. The public sector(go vernment and higher education) tendsto play the major role in both funding

and carrying out national R&D efforts inthe region, though this share is declining.Total health research (R&D) spending in LatinAmerica in 1998 is estimated as US$1.4 billion(about 12.7% of total investments in R&D).Of this figure, Argentina (about US$240million), Brazil (about US$850 million) andMexico (about US$200 million) accounted forall but US$100 million (estimated for all otherLatin American countries). The proportion of

health research to total R&D investments inLatin America varies between more than 20%in Panama to less than 5% in Chile andUruguay. It is not possible to identify the sharefunded from public sources.


Insert 6.6Funding of health R&D in three Asian countries, 1998

Thailand Philippines Malaysia

US$ million 15.7 7.4 6.9

% total government budget 0.06 0.11 0.04

% health budget 0.90 0.61 0.60

% GDP 0.012 0.049 0.010

11 RICYT (Red Iberoamericana de Ciencia y Tecnologia). El estado de la ciencia: principales indicadores de ciencia y tecnologia

Iberoamericanos/Interamericanos. Quilmes, 2000.


14/20

112

According to these estimates on health R&D,funding increased between 1992 and 1998 forall three major countries. The increases wereabout 40% (in current US dollars) in

Argentina and Mexico, and may have doubledin Brazil.

(c) Investments in health re s e a rch asrecommended by the Commission forHealth Research and DevelopmentThe Commission on Health Researc h

for Development, convened in 1990,recommended that at least 2% of nationalhealth expenditures in low- and middle-incomecountries be allocated to health research and

capacity building. Of the countries included inthis study, Brazil and Cuba approached the 2%mark (Insert 6.7). Turkey was not included inInsert 6.7 as higher education subsidies in thatcountry, particularly in state universities formedical education, influenced the highpercentage reported.

Insert 6.7

Selected low- and middle-income countries: estimated health R&D as % of totalhealth expenditure

Brazil

Cuba

Panama

Costa Rica

Argentina

India

Mexico

VenezuelaBolivia

Colombia

Malaysia

Phillipines

Thailand

Peru

Chile

Uruguay

Ecuador

El Salvador

Trinidad

Pale countries are particularly rough estimates.

Sources: Health R&D data: as above

GDP: World Development Report 2000-2001: Attacking Poverty. World Bank, Washington DC, 2000; RICYT, El estado de la ciencia: principales

indicadores de ciencia y technologia Iberoamericanos/Interamericanos. Quilmes, 2000 .

Health expenditure: World Health Report 2000. Health Systems: Improving Performance. WHO, 2000; OECD Health DATA 2000; A Comparative

Analysis of 29 Countr ies. OECD, Paris, 2000 .

0 2%


15/20


Health research is essential to improve thedesign and implementation of healthi n t e rventions, policies and health serv i c edelivery. It is evident that the 1990s have seena worldwide increase in funds for healthresearch and a transition of donors. Yet, inorder to improve the health of the majority of

the worlds population, res e a rch must betargeted to solving the problems of greatestimportance worldwide now and in the future.Thus, re s e a rch funds must be rationallyallocated in order to:

Develop new and improved technologies toadd ress the diseases and conditions ofgreatest magnitude;

Improve the delivery of and accessibility tohealth care, includ ing pre v e n t i v einterventions;

Address the cross-sectoral issues relevant toimproved health.

Access to research findings not only by theresearch and biomedical community, but bythe global population is critical. Hence theimportance of their application at the policyand programme levels. It is therefore essentialthat information on health research fundingon a disaggregated basis be collected anddisseminated.

1. Demand for dataThe demand for data on resource flows ishighly segmented. Various constituenciesrequire different types of information. Someconstituencies want resource flows data toi n fo rm policy and, ultimately, to prov i deguidance for action. Other constituencies

want re s o u rce flows data for advocacypurposes; for example, to point out thatinadequate resources are being allocated forhealth re s e a rch by a government ororganization. At disease or research topiclevel, constituencies need data to show thatimp ortant areas are being neglected. The

diversity of the demand for resource flowsdata is reflected in the diversity of the datatracked by funders and performers.

2. Supply of data(a) Total health R&D dataData is readily available for advancedcountries from existing data collectionsystems. Improvements in quality andstandardizations are already underway. Aspart of this process, potential as well as realdouble counting are being reviewed. Areasconstituting gaps, such as re s e a rch inhospitals, are included. While it is stilldifficult to obtain reliable health R&D totalsfor some low- and middle-income countriesand countries in transition, data collectionsystems are evolving: for example, the LatinAmerican region. The best inform a t i o nobtained to date has been through specialstudies and surveys. While the initial studymay take as long as two years to complete,such a study can form the countries basis for

a more systematic approach to monitoringresource flows in the future. In addition, bybuilding such systems in a manner that iscompatible with existing global datacollection systems, it will be easier in thefuture to obtain a more accurate overview oftotal health R&D funding worldwide.

Sec tion 5

Discussion and future strategies


16/20

114

(b) Disaggregated health R&D flows dataFunding flows from high-income countries to low-and middle-income countries, or countries intransition, are usually very difficult to trace.

For example, funding may be passed laterallyfrom one advanced country agency to anotherbefore it is provided to a low- and middle-income country agency. Furthermore, thesefunds may be provided through multilateralchannels, bilateral channels or via secondaryfunders, such as advanced country universitiesor non-governmental organizations whoadminister the funds on behalf of agovernment agency. In addition, many fundingagencies are highly decentralized withdecisions on allocations made in low- andmiddle-income countries and re port i n grequirements based on the overall goals andobjectives developed within the bilateralrelationship. Many advanced countriesfunding agencies, especially those disbursingODA, do not collect disease-specific data;therefore, this is unavailable or not easilyaccessible.

Funding flows within low- and middle-income

c o u n t r i e s a re also complex. Researc h

institutions receive public funds bilaterally,multilaterally and from their owngovernments and may concurrently receivefunds from external and internal non-governmental entities. As tracking these fundsis usually very difficult and time-consuming,a mapping of institutions and fundingstructures must be done first.

Private investments by pharm a c e u t i c a l

companies account for almost half of the total

investments into research worldwide. Onlyaggregated information is released in thisgroup. Information on the cost of researchand clinical trials for discovery anddevelopment of medicines was not consideredin this study. The widely quoted figure ofUS$500 million required to develop a newdrug was not addressed in this study andshould be studied and discussed in future.

(c) Usefulness of data sources for healthR&D informationThe following summary (Insert 6.8) examinedthe utility of available data sources and the

quality of the information:

Estimates of total R&D in high-incomecountries

Results obtained mainly from S&Td a t a base s /surveys and supplemented bydata from published reports were good.

Estimates on health R&D in low- and middle-income countries and countries in transition

Results obtained from the methodologydeveloped for three-country studies weregood. Improvements are needed intracking and obtaining disaggregated dataat the country level. Results from scienceand technology surveys and databases gavei n f o rmation on total funds for healthresearch and development. They provideduseful information on both performers andfunders. Information on countries notresearched in this first phase (for example,the Peoples Republic of China) will becarried out in the second phase.

Estimates of resource flows using high-incomecountry funders as sources of data

Responses to a questionnaire sent to thefunders were disappointing and this datacollection approach should be abandoned.Results obtained using personal interviewsand public documents were useful butrequired time and repeated efforts from theconsultants and staff. Future efforts alongthese lines should be focused and

adequately supported. Disease-specific datawas difficult to obtain as few organizationstrack this information. Funding invested inre s e a rch capacity strengthening wasidentified, along with insights forprogramming of resources by funders. Thiscomponent should be further developed inthe second phase.


17/20


Insert 6.8Usefulness of sources for health R & D data obtained by the resource

flows project

Funder Sp ecial Fu nder Govern ment Evalu ation s/ Interviews/

questionnaire survey surveys/ S&T annual reports/ personal

databases surveys websites contacts

High-income count ries

Government ministries/public ODA/orgs x x xxx

Other public* xxx xx

Pharmaceuticalcompanies xx xx

Non-profit/foundations x x xxx x xxx

EC x x xxx

WHO xx xxx xxWorld Bank x xxx

Low-, middle-income andcountries in transition

Government ministries xx

State government xxx xx

Academic/researchinstitutions x x xx

Hospitals x xx

Multilateral/bilateral x xx

NGOs x x xx

Pharmaceutical companies x xx xx

Academic researchinstitutions x xx xx

Total Global R&DAggregate xx xx xxx

* Other public: public sector funding other than for ODA such as national research institutes, medical research councils,university-based researchBlank = of limited or no use x = of some use xx = very useful xxx = extremely useful

(d) Obstacles encounteredThe following is a list of obstaclesencountered during the process of obtainingfinancial data:

O rganizations surveyed do notsystematically track or monitor healthresearch as per categories defined in this

paper or in the questionnaires. Members ofstaff surveyed were too busy to provideinf ormation beyond the scope of theirrecords.

While most organizations track someaspects of research capacity strengthening such as academic degree pro g r am m etraining, postdoctoral training and


18/20

international projects they generally donot maintain records on the low- andmiddle-income countries components ofi n t e rnational projects with which they

collaborate. This adds to the difficulties indetermining resource flows to low- andmiddle-income countries.

Questionnaires developed as a survey toolfor advanced country funders were toolengthy and detailed, thereby contributingto a poor response rate.

Decentralization of management in ODAand multilateral organizations contributes toproblems in obtaining data on financialresources, especially for purposes that arenot high priorities for those organizations.

Impact level measurements for parameterssuch as research capacity strengthening areinfrequently used. As a result, researchcapacity strengthening is reduced in statusas a priority.

Capturing data for organizations thatfacilitate and convene rather than execute isdifficult.

The importance and the relevance of thedata on re s o u rce flows for investororganizations is unclear when compared to

other priorities. Fluctuations in exchange rates complicate

the interpretation of data, especially long-term funding trends.

Obtaining data from funders in advancedcountries on funds actually used for researchin low- and middle-income countries bylocal researchers is difficult. Ascertaining thepercentage of funds used for administrativeand managerial purposes by advancedcountries and multilateral organizations is of

importance to obtain a better estimation offunds actually expended in low- andmiddle-income countries.

Information from low- and middle-incomecountries was not readily available. Aframework of information about resourceflows for health re se a rch in low- andmiddle-income countries was tested as partof this study.

3. Data gaps identifiedIn the course of this study, no attempt wasmade to gather data in the following areas(these will be addressed in the second phase of

the study): Global allocation of funds to R&D forspecific diseases.

Public funding by advanced countries fornorthern institutions conducting R&D onproblems important to low- and middle-income countries.

Pharmaceutical industry funding in low-and middle-income countries.

Cost of R&D to develop drugs and vaccines,including the costs of clinical trials.

Regular budget allocations by UN agenciessuch as WHO to health re s e a rch, asdifferentiated from voluntary contribution.

Relation between health priorities identifiedin low- and middle-income countries andp rojects funded from national andinternational sources.

Fraction of public funds invested intofundamental re s ea rch which eventuallyleads to a marketed drug.

Funding for social science research and forhealth economics research.

4. Donor transition in the late 1990s andearly 2000sIn the course of this study, it became evidentthat important changes were taking place inthe health donor community havingimplications for health re s e a rch in, andrelevant to, low- and middle-incomecountries. There is clear information on shiftsin funding sources in the late 1990s and early2000s, such as the new Global Fund to

Fight HIV, TB and Malaria, and therecommendations of the Commission onMacroeconomics and Health (see Chapter 1).The private sector foundations, particularlyBill and Melinda Gates Foundation, andphilanthropic institutions have taken a largerrole in funding research. The Bill and MelindaGates Foundation increased its investments inthe health research field to US$189 million in

116


19/20


2001. Investments by US pharm a ce u ti c a lcompanies are increasing in the US but notabroad. The access to the research findings bymost of the worlds population is a crucial

component of health research and should beensured.

In conclusion, during the late 1990s and early2000s, there has been greater involvementof foundations, CSOs, national re s e a rc hinstitutions in advanced countries, and thep h a rmaceut ical industry in intern a t i o n a lhealth. This shift is coupled with an increasein investments in health research globally,from governments in both advanced andlow- and middle-income countries. Theimplications of this transition to improve thehealth of the majority of the worldspopulation, a global public good, are not clearand have to be documented in future. Byensuring that re s e a rch is conducted ondiseases and determinants with the highestmagnitude of disease burden, we ensure thatthe limited available re s o u rces have thegreatest possible impact on the health of themajority of the worlds population, inparticular the poorer segments.

5. Conclusions and future stepsAt the global level, there is no coordination ofhealth research funding, and perhaps there willnever be. This study is certainly not intendingto attempt such coordination. In the realworld, there is a constellation of institutionsworking towards similar goals, which may ormay not communicate with each other. Ap l a t f o rm for discussion and inform a ti o nsharing can be useful to help improve resource

allocation for health research.

It is expected that many more organizationswill take part in future exercises on resourceflows. The following strategies could beconsidered relevant for the next phase:

a ) M e a s u re re s o u rce flows in a d d i t i o n a ldeveloping/transition countries using the

methodology developed in this study. Thisshould be implemented at the following levels: Government: improve and expand data on

selected topics, such as financial flows

related to health problems andd e t e rminants of disease burden at thecountry level; cross-check data generatedwith that reported by external donors.

Research institutions: encourage analysis ofresource flows into defined country healthresearch priorities by: building research capacity to measure

resource flows facilitating information exchange on

experiences and strategies disseminating lessons learned.

WHO, governments and medical researchcouncils in low- and middle-income countriesand institutions like COHRED are inadvantageous positions to facilitate this strategy.

b) Improve the amount and internationalcomparability of publicly available data on thelevel and structure of aggregate spending onhealth research by encouraging the entitiesalready compiling health statistics to pay greater

attention to R&D and by encouragingUNESCO and the regional org a n i z a t i o n scollecting R&D data to give higher priority tohealth-related series.

c) Periodically obtain disaggregated datafrom large investors in advanced countriesincluding ODA agencies, foundations andp h a rmaceutical companies. Analyse theinformation to study the 10/90 gap in healthresearch funding.

d) Influence partners with established interestsand expertise in specific disease areas to doperiodic studies of resource flows for theconditions representing the highest burdennow and in the future (e.g. International UnionAgainst TB and Lung Disease, Wellcome Trust,WHO/TDR, NIDI, WHO); assist in theidentification of funding for such studies.


20/20

118

Annex 6.1

Resource Flows Advisory Group

The members of the Advisory Group were acting in their individual capacity and were active forvarying amounts of time. The names of their institutions are mentioned for identificationpurposes only. The position indicated for each member is that held at the beginning of theproject and does not necessarily reflect the current position.

Chair

Louis J. Currat

Executive Secretary, Global Forum for HealthResearch

Convenor

Andres de FranciscoSenior Public Health Specialist, Global Forumfor Health Research

Members

Bienvenido P. AlanoP resident, Centre for Economic PolicyResearch, The PhilippinesWendy BaldwinDeputy Dire c t o r, National Institu tes ofHealth, USAJulio FrenkExecutive Director, Evidence and Informationfor Policy Cluster, World Health Organization,GenevaMyint HtweRegional Advisor on Medical Research, WorldHealth Organization Regional Office for SouthEast Asia (WHO/SEARO), New Delhi

Adnan A. HyderResearch Associate, Johns Hopkins UniversitySchool of Hygiene and Public Health, USACatherine MichaudSenior Research Associate, Harvard Center forPopulation and Development Studies, USACaryn MillerRese a rch Policy Advisor, U.S. Agency forI n t e rnational Development (USAID) and

Associate, Johns Hopkins University School ofHygiene and Public Health, USA

Barend MonsThe Netherlands Organization for ScientificResearch (NWO), The NetherlandsEric NoehrenbergD i rector of Programmes, Intern a t i o n a lFederation of Pharmaceutical ManufacturersAssociations, GenevaYvo NuyensCoordinator, Council on Health Research forDevelopment (COHRED), SwitzerlandTikki PangDirector, Research Policy and Cooperation,World Health Organization, GenevaUlysses B. PanissetRegional Advisor, Pan American HealthOrganization (PAHO), Washington DCElettra RonchiPrincipal Administrator, Organisation forEconomic Cooperation and Development(OECD), ParisBruce A. ScogginsDirector, Health Research Council of NewZealand, New Zealand

David SeemungalPolicy Analyst, Wellcome Trust, Great BritainAdik WibowoWorld Health Organization Regional Officefor South East Asia (WHO/SEARO), NewDelhiAlison YoungOrganisation for Economic Cooperation andDevelopment (OECD), Paris

Documents

Global Forum 30