Funding for HIV and non-communicable diseases: Implications for priority-setting in the Pacific region

7/31/2019 Funding for HIV and non-communicable diseases: Implications for priority-setting in the Pacific region

1/20

KNOWLEDGE HUBS FOR HEALTHStrengthening health systems through evidence in Asia and the Pacic

Funding or HIV andNon-Communicable Diseases:

Implications or Priority Setting in

the Pacifc Region

Joel NeginSydney School o Public Health and Menzies Centre or Health Policy, University o Sydney, Australia.

Helen M. RobinsonConsultant, Nossal Institute or Global Health, University o Melbourne, Australia.

HEALTH POLICY AND HEALTH FINANCEKNOWLEDGE HUB

WORKING PAPER SERIES NUMBER 1 | MARCH 2010

The Nossal Institute

for Global Health

www.ni.unimelb.edu.au


2/20

WORKING PAPER SERIES HEALTH POLICY AND HEALTH FINANCE KNOWLEDGE HUB

Funding or HIV and Non-Communicable Diseases NUMBER 1 | MARCH 2010

ABOUT THIS SERIESThis Working Paper is produced by the Nossal Institute or Global Health at the University o Melbourne,Australia.

The Australian Agency or International Development (AusAID) has established our Knowledge Hubs or

health, each addressing dierent dimensions o the health system: Health Policy and Health Finance; HealthInormation Systems; Human Resources or Health; and Womens and Childrens Health.

Based at the Nossal Institute, the Health Policy and Health Finance Knowledge Hub aims to support regional,

national and international partners to develop eective evidence-inormed policy making, particularly in the eldo health nance and health systems.

The Working Paper series is not a peer-reviewed journal; papers in this series are works-in-progress. The aim is

to stimulate discussion and comment among policy makers and researchers.

The Nossal Institute invites and encourages eedback. We would like to hear both where corrections

are needed to published papers and where additional work would be useul. We also would like to hear

suggestions or new papers or the investigation o any topics that health planners or policy makers wouldnd helpul. To provide comment or get urther inormation about the Working Paper series please contact;[email protected] with Working Papers as the subject.

For updated Working Papers, the title page includes the date o the latest revision.

Funding or HIV and Non-Communicable Diseases: Implications or Priority Setting in the Pacifc

Region

First drat March 2010

Lead author: Joel Negin

Sydney School o Public Health and Menzies Centre or Health Policy, University o Sydney,

[email protected]

Other contributors: Helen M. Robinson, Consultant, The Nossal Institute or Global Health, University oMelbourne

This Working Paper represents the views o its author/s and does not represent any ocial position o The

University o Melbourne, AusAID or the Australian Government.

ACKNOWLEDGEMENTSThe HIV unding inormation was collected in collaboration with the George Institute or International Health(http://www.thegeorgeinstitute.org/) and with Associate Proessor Heather Worth o the University o New South

Wales, as part o work commissioned by the Commission on AIDS in the Pacic.


3/20

HEALTH POLICY AND HEALTH FINANCE KNOWLEDGE HUB WORKING PAPER SERIES

NUMBER 1 | MARCH 2010 Funding or HIV and Non-Communicable Diseases: Implications or Priority Setting in the Pacic Region i

ABSTRACTObjectives: There has been increasing global interest in documenting unding fows or health, but none o thatwork has ocused on the Pacic region. This paper outlines external unding or two specic areas o overseas

development assistance (ODA) or health in the regionHIV/AIDS and non-communicable diseases (NCDs)during 2002-09. These are compared to the comparative disease burdens, and some initial thoughts arepresented on the dynamics o setting donor health priorities in the Pacic.

Methods: Empirical data on development partner aid unding were accessed through a review o web sites,

annual reports, published data, unding proposals and other publicly available documentation o donor countryaid agencies, multilateral agencies and programs and that o recipient governments. The document review was

supplemented by 27 key inormant interviews to veriy and clariy the available data. Interviewees were drawn

mainly rom bilateral and multilateral agencies active in the Pacic and researchers working in the eld. The HIV

component was commissioned work or the Commission on AIDS in the Pacic.

Results: Despite much higher mortality rates rom NCDs, external unding or HIV is higher than or NCDs. From

2002 to 2009, unding totalled US$68,481,730 or HIV and US$32,910,778 or NCDs. External assistance or

HIV activities in the Pacic in 2009 was more than US$18 million, while unding or NCDs in the same year wasalmost US$12 million.

Conclusions: Despite cooperation rom many agencies, the unding data were dicult to gather, highlighting

the need or greater transparency o unding inormation and more thorough record keeping. The externalunding does not align with the disease and mortality gures, and urther interviews suggested that donor

unding decisions in the region are driven not by local priorities but by actors including a strong global HIV

community, the commitment to the Millennium Development Goals (MDGs) and the lack o coherence in the

way NCDs are presented to policy makers.


4/20


ii Funding or HIV and Non-Communicable Diseases: Implications or Priority Setting in the Pacic Region NUMBER 1 | MARCH 2010


5/20


NUMBER 1 | MARCH 2010 Funding or HIV and Non-Communicable Diseases: Implications or Priority Setting in the Pacic Region 1

INTRODUCTIONThe three or our decades ollowing the Second World War brought dramatic improvements in the quality ohealth o most people living in Pacic island countries (PICs). Data on estimated lie expectancy o both men and

women indicate that these countries have relatively low rates o adult mortality. Declining adult mortality rom1970 until 2000 also refects a generally positive situation, particularly or women (WHO 2003).

Today, the health situation in the region does not look quite as rosy. Most PICs seem unlikely to achieve the

targets or 2015 established by regional MDGs (AusAID 2009). Despite the historic successes and the availability

o health aid, it seems there is a risk that the health gains o last century will be eroded, with the possibility oreversals or uture generations in some communities. Given this background, we eel it is important to examine

the relationships between national policy makers and their development partners to gain insights into how

better to align development eorts and so improve aid eectiveness through more sustainable processes and

partnerships.

The Paris Declaration on Aid Eectiveness and the Accra Agenda or Action (OECD 2008) agreed that countries

should take charge o setting priorities or themselves and drive their own policy agendas. But is this happening?

Are fows o overseas development assistance making it too dicult or small countries to set their own healthpriorities? ODA or health makes up a relatively large component o health unding in PICs. Much o this aid

is directed towards disease-specic initiatives, and it is not clear whether the health priorities implied by ODA

priorities match the health priorities o the countries.

Recently, there has been greater global recognition o unding imbalances. Gostin and Mok (2009) argue that the

problem o skewed priorities in international assistance mean a signicant amount o unding is directed towards

specic diseases or narrowly perceived national security interests that have been placed high on the global

health agenda by a small number o wealthy donors (such as OECD countries, the Gates Foundation and theGlobal Fund).

Gostin (2007) argues that rich countries and philanthropists have oten set priorities that do not refect local needs

and preerences. The Paris Declaration and Accra Agenda both highlighted the need or disease control programs

to be better integrated into eorts to strengthen health systems; Cometto, Ooms et al (2009) believe the emphasison vertical disease control programs can exacerbate problems or developing country health systems.

This paper takes a rst step towards understanding this situation by examining external unding or twospecic areas o health ODA in PICsHIV/AIDS and non-communicable diseases (NCDs)1during 2002-

09. The unding situation is assessed in conjunction with comparative disease burdens in order to develop a

better understanding o the dynamics o donor health priorities in the region. Internationally, there has beenconsiderable recent work in this area (McCoy, Chand et al 2009; Shiman 2009), but little work has been done

specically on the Pacic region.

The paper is an initial review o available sources on levels o unding in the region and what they suggest about

priorities or health. By ocusing the review on external unding, we can more easily see the extent to which high-level promises and pledgesby PICs and by development partnersare translated into action by providing

resources. This empirical analysis aims to assess unding or HIV/AIDS and NCDs, comment on what theserelative unding levels suggest about whose policies more strongly infuence priorities and set a baseline oruture comparisons o development partner priorities or HIV and NCDs in the Pacic.

HIV and NCDs were selected or this study because:

botharerelativelyrecentphenomenainburdenofdiseaseforPICs;

bothrequirelong-termandsustainabletreatmentsthathaveprofoundimpactsonhealthpolicymakingand

resourcing (Bischo, Ekoe et al 2009);

bothhaveconsiderableimpactontheeconomiesandcommunitylifeofPICsduetotheprematurelossof

productive adults;

botharelargelypreventable.

1 Non-communicable diseases are principal ly cardiovascular diseases, cancers, chronic respiratory diseases and diabetes, as classied by the Internationa lStatistical Classication o Diseases and Related Health Problems rules used by WHO to produce internationally comparable health metrics (WHO 2008).


6/20


2 Funding or HIV and Non-Communicable Diseases: Implications or Priority Setting in the Pacic Region NUMBER 1 | MARCH 2010

In the paper, we rst describe how we collected and analysed the data. We describe the disease burden

associated with each condition. We then briefy describe the characteristics o development assistance in PICs

and measure unding available or the HIV and NCD response in the region. We then discuss the ndings andtheir implications or health policy making in PICs and, nally, the implications o these ndings or uture work.

METHODOLOGYThe methodology is in two parts. First, the empirical data on development partner aid unding were accessed

through a review o web sites, annual reports, published data, unding proposals and other publicly available

documentation o donor country aid agencies, multilateral agencies and programs and that o recipientgovernments. As a result, the unding data come largely rom publicly available primary sources, including

the Global Fund to Fight AIDS, Tuberculosis and Malaria, the Secretariat o the Pacic Community (SPC), US

Centres or Disease Control and Prevention (CDC), UNAIDS, World Health Organization (WHO) and AusAID.The academic literature was reviewed through reerence lists o relevant articles identied. Searches ocused on

the period rom 2000 to 2009. Wherever possible, all data have been cross-reerenced against other sources,

including published data rom other researchers working in this area. All nancial data are presented in currentUS dollars unless otherwise noted.

Secondly, the results o this review were supplemented by eld interviews to veriy and clariy the meaning o

the publicly available data. Interviewees were mainly rom bilateral and multilateral agencies active in the Pacic

and researchers working in the eld. They were involved in unding decisions regarding HIV and NCDs. Theinterviews were conducted either in person or by telephone. Interview topics included how the unding priorities

were set and what process was used to determine allocations. Email was used or ollow-up questions,

clarications and requests or conrmation o specic points.

Ethical approval or the interview process was given by the University o NSW and the University o Sydney

Human Research Ethics Committees. Condentiality or interviewees was guaranteed to encourage more open

discussion. Where direct quotes are provided, names are not attributed. As is oten the case with interviews okey policy makers, not all those approached were able to complete an interview. Data or the HIV component

o this study were collected as part o work or the report o the Commission on AIDS in the Pacic (2009) and

used with the permission o the Commission.

There have been many calls in recent health research literature or more transparency by aid donors in

reporting aid fows (McCoy, Chand et al 2009; IHME 2009). With the entrance o new players, such as the Gates

Foundation and the Global Fund, to an already crowded stage, it is dicult to piece together a ull and clear

picture o the purposes o various unding projects and programs, and the resources they involve. The workundertaken here has proven no less dicult. Generally, it is easier to distinguish aid programs or HIV because

they are generally delivered through stand-alone projects, while those directed to NCDs are more dicult to

identiy or separate rom overall health budgets. Domestic unding was not systematically collected due todiculties in gathering such data broken down by disease areas.

CURRENT BURDEN OF DISEASE IN PACIFIC ISLANDCOUNTRIESBeore investigating unding levels or the two priority areas selected or this study, we outline here the Pacic

burden o disease.

The Pacic region accounts or just 0.2% o the global burden o HIV and AIDS, the majority o these cases

occurring in Papua New Guinea (PNG), Australia and New Zealand (UNAIDS 2008). According to data provided by

the SPC, as o the end o 2008 there were 1337 cumulative reported cases o HIV and AIDS across the PICs andterritories (excluding PNG) (SPC 2009). Fourteen o 21 Pacic countries and territories have ewer than 25 reported

cases o HIV and AIDS, making it clear that HIV is not currently a large public health burden in many o these

countries. Table 1 shows the numbers o cumulative HIV and AIDS cases in each o the countries o the region.


7/20



Table 1. Pacifc population and reported cases o HIV and AIDS, 2008

Country Population (mid-2008) HIV reported cases:cumulative

Cumulative incidenceper 100,000

MELANESIA (excluding PNG) 1,804,350 590 32.7Fiji 831,600 259 31.1

New Caledonia 241,700 316 130.7

Solomon Islands 503,900 10 2.0

Vanuatu 227,150 5 2.2

MICRONESIA 525,309 331 63.0

Federated States o Micronesia 110,600 35 31.6

Guam 172,300 187 108.5

Kiribati 95,500 54 56.5

Republic o Marshall Islands 52,700 13 24.7

Nauru 9,900 2 20.2

Northern Mariana Islands 64,109 32 49.9

Palau 20,200 8 39.6

POLYNESIA 649,650 334 51.4

American Samoa 65,000 3 4.6

Cook Islands 13,500 2 14.8

French Polynesia 261,400 286 109.4

Niue 1,600 0 0.0

Pitcairn Islands 50 0 0.0

Samoa 179,500 16 8.9

Tokelau Islands 1,200 0 0.0

Tonga 102,300 15 14.7

Tuvalu 9,700 10 103.1

Wallis and Futuna 15,400 2 13.0

ALL PICs (excluding PNG) 2,979,309 1,255 42.1

Note: Kiribati reporting period Dec 2004; Tuvalu reporting period Dec 2005.Source: SPC 2009.

At the same time, WHO data (WHOSIS 2009) indicate that NCDs are the leading cause o death in most PICs,

contributing approximately 75% o deaths in 2007; or example, 82% o deaths in Fiji during 2007 are attributed

to NCDs (WHOSIS 2009). NCDs also contribute a very large proportion o morbidity or people living in thePacic region, and indications are that NCD-related mortality and morbidity are rising (WHO 2007, 2009).

Rates o diabetes in the Pacic are among the highest in the world, with a prevalence o more than 40% among

adults aged 25 to 64 years in American Samoa, Tokelau and the Marshall Islands and rates o 13% or higherin Tonga, Tuvalu, Samoa and French Polynesia (Colagiuri, Palu et al 2008; Buckley and Colagiuri 2007). Levelso overweight and obesity in the region are extremely troubling and range rom 47% to 93% in countries that

have completed STEPS surveys, being over 80% in Samoa and Tonga (WHO 2009).2 Data on body mass

index, obesity and risk actors rom a number o sources highlight the threat o chronic NCD in the Pacic(Schultz, Vatucawaqa et al 2007; Rasanathan and Tukuitonga 2007; Carlot-Tary, Hughes et al 1998; Colagiuri,

Muimuiheata et al 2002; Keke, Phongsavan et al 2007).

The impact o these high rates o NCDs on health budgets is immense, as much as 60% o health budgetsin some Pacic island countries being allocated to expensive overseas care or those aected (WHO 2007).

A World Bank document estimated that in Samoa in 2000, NCDs accounted or over 43% o total health care

expenditure (World Bank 2008). Overall, while NCDs are a major contributor to mortality in the region, WHO

data show that AIDS-related mortality is extremely low or does not exist in the countries o the region (Figure 1).

2 World Health Organizations STEPwise approach to survei llance (STEPS), http://www.who.int /chp/steps/en/.


8/20



Figure 1. Deaths due to NCDs and AIDS per 100,000 population, by country, 2002 and 2005

Source: WHOSIS 2009.

Funding or Disease Control in the PacifcFor the size o their economies and populations, Pacic countries receive some o the largest ODA amounts in the

world (AusAID 2009; Hughes 2003). The three million people in the Pacic (excluding PNG) received US$706.5million in ODA in 2007 or US$235.51 per capita (AusAID 2009). Australia, New Zealand, Japan and the USA arethe major bilateral development partners, along with the multilateral World Bank and Asian Development Bank

(ADB). Although detail is dicult to conrm, Hanson (2009) reports that Chinas aid to the region is estimated at

between US$100 million and US$150 million per year, including both loans and direct grants.

Development assistance or health in the Pacic has been increasing over the past decade (Figure 2). Most

PICs received increasing amounts o aid, particularly ater 2000. Micronesia and the Marshall Islands receive

levels o unding that place them among the largest recipients o per capita health assistance in the world(IHME 2009). The Solomon Islands, Palau, Tonga and the Cook Islands (as well as the aorementioned

Micronesia and Marshall Islands) all received more than US$20 per capita per year (annualised over the past

three years) in health assistance.3 IHME (2009) estimates that Australian health aid increased rom US$28

million in 1990 to US$220 million in 2007 (in constant 2007 dollars). While not all o this aid went to the Pacic, aconsiderable proportion did.

At the same time as there have been increases in ODA or health, many PIC governments made modest

increases to their health expenditures. In most o the larger Pacic countries, per capita government spendingon health increased less than the global average (AusAID 2009). Government health spending in Fiji, the

Solomon Islands, Vanuatu, Marshall Islands, Palau and Tonga all increased by less than 25% between 2000

and 2006 in purchasing power parity terms, against a lower middle income country average o 86% and a lowincome country average o 62% (AusAID 2009).

3 In comparison, in 2007 per capita health aid was US$4.38 or Ethiopia and US$7.68 or Cambodia. However, two o the three Arican islands that have comparablechallenges o remoteness and lack o economy o scale are also recipients o high levels o per capita development assistance or health: Sao Tome and Principe(US$23.26) and Cape Verde (US$21.22). The Comoros, on the other hand, receive only US$2.51less than any Pacic Island or which data are available.


9/20



Figures 2A, 2B and 2C. ODA per capita or health, by recipient, 1990-2007, in constant 2007 US$

Note that the vertical scale is dierent in each gure.Source: IHME 2009

Funding or HIV ResponseIn determining the extent o unding going to HIV activities in PICs, we included unding or all activities related

to prevention, care and treatment. Activities directed towards HIV together with other related goals, including orexample tuberculosis or maternal care, were included only i the response to HIV was clearly identiable and

could be isolated or, in the case where the aims were clearly integrated with HIV, they were included as a whole.

In cases where HIV and reproductive health were part o the same initiative, as is the case with many UNFPA-

unded projects and programs, these were included in their entirety.

The amount o external unding available or HIV activities in the Pacic has been increasing or the last decade

and reached US$18 million in 2009 (Figure 3). Notable issues include the large amount o US government

undingexclusively to the six US-aliated countries and territoriesand large amounts provided by AusAIDand the Global Fund.

This is in the context o total global unding or HIV and AIDS increasing dramatically over the past decade,

rom less than US$300 million in 1996 to US$10 billion in 2007 (UNAIDS 2008). Domestic expenditure was not

included, but, based on interviews and document review, this was determined to be either very small or evennon-existent in many countries. Data on HIV-related household expenditure were not available, nor does the


10/20



gure include any domestic private sector spendingor example, spending by employers or corporations on

HIV prevention or treatment. Where possible, actual expenditure has been included rather than budgeted ororecast gures, but this was not available rom all agencies.

Analysis o unding by country reveals that ve o the eight largest recipients o HIV unds in the Pacic orwhich country data are available are the US-aliated countries and territories (Figure 4). These countries

and territories receive levels o US government unding that dwar that disbursed by the Global Fund and the

AusAID-unded Pacic Regional HIV/AIDS Program (PRHP). Only two countries, Palau and the Federated

States o Micronesia, receive unding rom all three sources.

US government spending on HIV represents very high per capita unding in some o the countries and

territories, particularly in Palau (see Figure 5). Other small Pacic island countries, such as Tuvalu, Niue,

the Cook Islands and Tokelau, also receive large amounts per capita. Some o the countries with largerpopulations, such as Fiji, Vanuatu, Samoa and the Solomon Islands, receive smaller amounts per capita. Fiji,

with a relatively larger number o cases, receives only US$0.40 per person per year rom these three majorunding sources. However, per capita spending is not a completely accurate measure. A large xed-costcomponent in HIV responsesparticularly or treatment elementsmakes per capita gures disproportionately

large or small island states.

The various unding sources or HIV activities in the region unction dierently. While the ADB and Global Fund

contributions are channelled through the SPC and managed regionally, US unding is provided directly tocountry governments and is earmarked or HIV activities. The UN programs and WHO work largely through

their regional technical networks, and, as o 2009, AusAID and NZAID have developed a pooled unding

mechanism called the Pacic HIV Response Fund. More discussion o the manner in which these undingsources operate is available in Commission on AIDS in the Pacic (2009).

Figure 3. External unding or HIV activities by source, 2001-2009, in constant 2007 US$ million

Notes: Funding or New Caledonia, French Polynesia and Wallis and Futuna was not included due to lack o availability and to ensure comparability with NCD data.Double counting is possible in the gure although we have tried to avoid it as much as possible. Although we obtained some data on NGO unding, in many cases thatunding was provided by donor agencies, so NGO unds were not included to avoid possible double counting. We also acknowledge that some donors will include onlyprogrammatic costs and others might include travel, salaries and administrative costs. UNICEF unding gures were not provided despite multiple requests. Fundingdata or 2001-04 are particularly unreliable; many agencies did not have records o past expenditure and had no institutional memory rom which to gather suchinormation. No unding was ound rom the Bill and Melinda Gates Foundation or the Clinton Foundation.


11/20



Figure 4. HIV unding rom PRHP, USA and the Global Fund, by country, 2004-07, in constant 2007 US$

Notes: HIV unding by country was available or only a ew o the sources. The Pacic Regional HIV/AIDS Program (which includes substantial amounts o Australian,New Zealand and French unding), Global Fund and US government data were available by country. ADB unding data are not currently available by country, nor isthe unding provided by UN agencies. Some o these unds are provided at the regional level and thereore cannot be broken down by individual country. The dataavailable are rom 2004 to 2007 or the three sources outlined and cover only the countries that receive unding rom those sources. The amount noted is cumulativeover the our years 2004 to 2007. Figures represent unding disbursed.

Figure 5. Annual HIV unding per capita rom PRHP, US and the Global Fund, 2004-07, in constant 2007 US$

Notes: Australia-US exchange rate is based on an average rom 1 January 2004 to 31 December 2007 (0.7731); population gures or 2005.


12/20



Funding or NCD ActivitiesDespite the high NCD disease burden in the Pacic, the NCD response has received less unding rom

donor partners than HIV. Analysis o unding inormation rom the major development partnersincluding

CDC, NZAID, AusAID, Japan and Francereveals that limited unding was allocated to NCD prevention andtreatment up to 2008 (Figure 6). Prior to 2008, the US was the major bilateral under o NCD activities, with

some contribution rom Australia in the orm o the Pacic Action or Health Project rom 2002 to 2005. Financial

support rom New Zealand was ocused on tobacco control and promotion o physical activity as well asunding medical treatment provided in New Zealand.

As o 2007, external unding or NCD activities in the Pacic amounted to less than US$1 per capita. By 2009,

this had increased to US$3.85. These gures do not include unding that some individual countries receivedirectly rom the World Diabetes Foundation, World Heart Foundation and similar organisations. These data

were not readily available, and amounts are likely to be very small since, when interviewed, the major regional

actors in NCD control were not aware o specic projects or unding. Other prominent global health unders,

such as the Gates Foundation, did not provide unding or NCD activities in the region.

Funding or NCD control has increased since 2008 (mostly rom CDC and the governments o Australia andNew Zealand). As would be expected, US unding is directed to US-aliated countries and territories and is

managed through the same grant program that supports domestic state health agencies. Specic undingstreams include diabetes prevention and control, health promotion or the chronically ill, cancer prevention and

control and tobacco-related projects. Cancer-related projects receive more than US$3 million annually. From

2008, Australia and New Zealand unded the joint WHO and SPC Pacic NCD Framework, a new initiative thatnally brought the NCD response in the region to the ore, including a grants program to und various NCD

activities. One interviewee stated that the French government had initially committed to providing unds to the

NCD ramework but pulled out due to the nancial crisis.

Considerably more donor unding has been provided or HIV responses than or NCD control. From 2002 to2009, unding or HIV in the Pacic totalled US$68,481,730 and or NCDs US$32,910,778. Figure 7 illustrates the

dierent levels o unding.

Figure 6. External unding or NCD programs in the Pacifc, by source, 2002-09, in constant 2007 US$ million

Note: Funding or New Caledonia, French Polynesia and Wallis and Futuna not included.


13/20



KEY FINDINGSWhile the data are somewhat limited, a number o key messages emerge rom this study o unding or HIV andNCD responses:

Evidence-basedpolicyanalysisandpolicymakingrequirereliabledata.Despitecooperationfrommany

agencies, the various unding data were dicult to gather, conrming what others who have attempted togather similar data have noted. There is thereore a need or more thorough record keeping and greater

transparency in access to unding inormation.

Nonetheless,theresultsclearlyshowthatdonorfundingdoesnotalignwithburdenofdiseaseandmortality

gures. Despite the low burden o disease or HIV, external actors provide signicantly more unding or HIV

response than or NCD activities. While there may be good reasons why unding does not align with theburden o disease, the disparities seen in the Pacic are signicant.

Theredoesnotappeartobeaclearrationalesupportingthepatternsofdonorfunding.Somehighly

aected countries receive little compared to those with lower disease burdens; and some countries with

small populations receive more than those with much larger populations. To some extent, unding appears

to ollow historic and geopolitical interests.

DISCUSSIONThe purpose o this paper was not to discuss appropriate levels o unding or HIV or or NCD control in the

PICS at the current time or in the uture. Rather, we set out to look more closely at the overall process o policy

making and priority setting in health. The analysis here has several limitations, mostly related to the availability

Figure 7. External Funding or HIV and NCD Responses in the Pacifc, by Source, 2002-09, in constant 2007 US$


14/20


15/20



the Pacic region. While there has been some increase in external unding or NCDs since 2008, the continuing

imbalance in unding or NCDs compared to HIV/AIDS seems dicult to explain without considering the

likelihood that donor priorities have driven the allocation process.

It could be argued that the recent interest o the governments o the USA, Australia and New Zealand inproviding increased unding or NCD control in some PICs may be the result o concern or establishing new

development partnerships. Indeed, the NZAID Health Strategy 2008-13 (NZAID 2007) includes NCDs as one oits key themes perhaps in response to the repeated calls rom PIC health ministers. But this does not explain, or

justiy, the situation prior to 2008.

The disparities between external unding o NCDs and HIV and the burden o disease in PICs raises questionsabout how unding priorities are set. In their investigation o how issues nd a place on the global health

agenda, Shiman and Smith (2007) developed a ramework or analysing what they call the determinants o

political priority or global initiatives. Table 2 illustrates their ramework.

The ocus o this paper is to investigate issues related to unding priorities set by donors who themselves are

active players on the global health stage. Considering the disparities in unding or NCDs and HIV and the

lack o coherence provided by an examination o regional disease patterns in providing an explanation orthis, it seems reasonable to consider the political determinants o priority setting using the Shiman and Smithramework. The ramework in act has resonance with some o the ideas emerging rom interviewees in our

initial investigation, including within the unding agencies, ministries o health and regional bodies. Using the

Shiman Smith ramework, Table 3 summarizes ndings rom our initial survey.

Table 3. Determinants o unding priorities or HIV and NCD responses in the Pacifc

Determinant o political priority HIV NCDs

Actor power Strong developed global policy communityand institutions by around 2002 with UNAIDS,PEPFAR, Global Fund, including globalchampions

Civil society mobilisation in the Pacic withPacic Islands AIDS Foundation

Long-standing interest o global actors indisease control means that major Pacic donorsplaced HIV high on their agenda

There are many NCDs, each having its ownstakeholders, thus limiting cohesion o thepolicy community

Greater diversity o actors and institutions

makes co-ordination more dicult; hencemobilisation o key actors globally and in thePacic is more dicult

Limited voice o Pacic governments

Ideas Singularity o purpose around one diseasemakes or simplicity o message

Well-developed global rame available thatresonates with audiences

NCDs encompass many issues, which limitscohesion in the way issues are ramed

Myths o NCDs being diseases o rich, whitemales are dicult to counter

Political contexts Inclusion in the MDGs and Global Fund putHIV on the agenda or donors and developingcountries

NCD challenges overshadowed by MDGs

Issue characteristics Being a single disease makes it easier to trackand identiy eorts and inputs

Poor data on severity, but threat o HIV epidemic

Inectious disease threatens donor neighbours

Clearly developed HIV response with globalevidence

More complex interplay o multiple risk actorsand interventions makes it more dicult toisolate inputs and eorts

Poor NCD data limit clear indications o severityo problem

Limited developing country evidence on eectiveinterventions

Note: PEPFAR is the (US) Presidents Plan or AIDS Relie.


16/20



These preliminary ndings suggest a way orward or uture in-depth research and are presented here to help

galvanise discussion in the Pacic policy community. Using this analytical ramework gives rise to a number o

issues that require urther investigation:

Actor PowerShiman (2009) has noted the eectiveness o the global HIV and AIDS lobby in raising support or HIV

initiatives. The number o institutions, agencies and bodies, including UNAIDS, involved in ghting the

epidemic is testament to the eectiveness o this lobby. Some interviewees noted that the NCD policycommunity is less cohesive because the dierent groups are engaged with disease-specic issues. In

the Pacic region, civil society has mobilised in support o HIV prevention and control, or example in the

orm o the Pacic Islands AIDS Foundation. Equally strong lobby groups or NCDs are only now beginning

to emerge globally and in the Pacic region (see IDF, IUAC et al 2009). Since the mid-1990s, PIC healthministers have expressed their commitment to NCD prevention and control on several occasions (WHO

and SPC 2003; WHO and SPC 2005; WHO 2007). Despite this, development partners have been slow to

recognise these eorts and provide appropriate unding, demonstrating both the limited voice o PIC healthministers and the dominance o global donor priorities.

Ideas

It appears that constructing a response to the HIV epidemic over more than 30 years has provided the

opportunity to rame a response clearly around a single identiable issue and to make it accessible to a

wide range o policy makers (Shiman 2009). Conversely, the act that NCDs encompass many diverseissues, causes and solutions perhaps limits cohesion in the way the NCD challenge is ramed. According

to interviewees, HIVs status as an inectious disease played a role in its prominence. It appears that donors

quickly realised the threat o rapid increases in prevalence and rightly acted to prevent such a spread; onedonor interviewee asserted, or example, that the two main priorities o donors in the region over the past

ew years have been HIV and pandemic preparedness.

Political Context

One interviewee thought there was little action on NCDs in the region because they do not appear in theMDGs. While HIV and malaria are highlighted as priority diseases, NCDs were listed only under othermajor diseases in MDG 6. Despite this, some PICs have attempted to add NCDs to their country-specic

MDGs. Once the response to a disease secures signicant unding, it may be dicult to remove it rom the

policy agenda: Stuckler, King et al (2008) assert that imbalances in unding are sel-sustaining becausedonors and international agencies develop expertise in an area and continue to provide unding unless

unanticipated results emerge. Shiman and Smith (2007) note the importance o norms and describe the

shared belies about appropriate behaviour and the institutions that negotiate and enorce these norms.

Stuckler, King et al (2008) assert that WHO plays an important normative role and infuence global undingdecisions. In the Pacic region, AusAID has also played a central role in its prioritisation o the response to

HIV and the relative neglect o NCDs.

Issue Characteristics

Walt, Pavignani et al (1999) argue that strategic data collection, interpretation, analysis and dissemination

are crucial to avoid discussions that can be erratic and ideologically based. Despite the signicance oNCDs in the Pacic region, inormation on morbidity and (to a lesser extent) mortality has only recently

become a ocus o systematic data collection and reporting. WHO has supported countries in gathering

inormation on NCD risk-actors, and, as o 2007, 15 countries had completed the WHO STEPS surveyin their communities. The global HIV community has developed considerable evidence on eective

interventions, but evidence on eective NCD interventions in low and middle-income countries has only

recently been the ocus o systematic research (Gaziano, Galea et al 2007; Lim, Gaziano et al 2007).


17/20



CONCLUSIONSThe analysis provided in this paper is preliminary and incomplete, and readers should be aware o thelimitations o the data and the limited amount o time and resources available or this initial survey. As other

authors have noted (Sridhar and Batniji 2008; IHME 2009), inormation about nancial allocations or varioushealth related activities is commonly ragmented and oten inaccessible. The entrance o many new players,particularly those rom the private or philanthropic sector, into the health assistance arena has made the

tracking o unding fows more complex (IHME 2009), and weak monitoring and evaluation o project and

program outcomes, particularly at the country level, makes the situation even more dicult. The lack ocomprehensive data does not help to increase the quality o the debate.

Data on government unding or national HIV and NCD responses were not collected or this study. It remains

possible that, i large amounts o government unding or the HIV and/or NCD responses were identied, this

would lead to dierent ndings or some countries. But government unding or the HIV and NCD responsesin the Pacic is unlikely to be large because overall government allocations or health care are limited. Annual

per capita government expenditure on health in 2005 was, or example, only US$105 in Fiji, US$109 in Kiribati,

US$91 in Samoa, US$79 in Tonga, US$44 in Vanuatu and US$26 in the Solomon Islands (WHOSIS 2009).Furthermore, discussions with policy makers indicated that government spending on NCDs and HIV was low,

especially relative to need. While some stated that some unds were spent, or example, on treatment or NCD-

related conditions, domestic spending on NCD prevention was limited, and interviewees conrmed that mostunding or these responses came rom external sources.

In the Pacic region, external unding or HIV is greater and more diverse than unding or NCDs, despite

the signicantly larger NCD disease burden. Using the available comparative data, this paper has begun an

exploration o the dynamics o setting health policy priorities. This initial investigation o how the responseto HIV emerged and persisted as a unding priority or Pacic countries and why NCDs have until recently

been relatively neglected by donor partners brings to the ore issues o power, infuence, use o evidence and

agenda-setting processes. There may be, in reality, stronger synergies between NCD and HIV control than theshared chronicity suggested by Bischo, Echo et al (2009). More signicant lessons may be drawn perhaps

rom donor-unded programs that address HIV treatment and control which would be o value to expandingdonor support in the so ar relatively neglected area o NCDs. Analysis o these issues provides a ruitul areaor urther research.


18/20



REFERENCESAusAID. 2009. Tracking development and governance in the Pacifc. Canberra: Commonwealth o Australia.

Bischo, A., T. Ekoe, N. Perone, S. Slama and L. Loutan. 2009. Chronic disease management in sub-Saharan

Arica: whose business is it? International Journal o Environmental Research and Public Health 6: 2258-2270.

Buckley, A. and R. Colagiuri. 2007. Building better diabetes care in Vanuatu. Report on the World DiabetesFoundation Project National Stakeholder Forum, 5-6 June, Port Vila, Vanuatu. Sydney: Diabetes Unit, Australian

Health Policy Institute, University o Sydney.

Carlot-Tary, M., M. Hughes and M.C. Hughes. 1998. Vanuatu non-communicable disease survey report.Noumea, New Caledonia: Multipress.

Colagiuri, S., T. Palu, V. Satupaitea, Z. Hussain and R. Colagiuri. 2008. The epidemiology o diabetes in Pacic

island populations. In The epidemiology o diabetes mellitus, eds. Jean-Marie Eko, Paul Zimmet and RhysWilliams.

Colagiuri, S., S. Muimuiheata, R. Colagiuri, Z. Hussain, S. Naati and T. Palu. 2002. The prevalence o diabetes

in the Kingdom o Tonga. Diabetes Care 25: 1378-1383.

Cometto, G., G. Ooms, A. Starrs and P. Zeitz. 2009. A global und or the health MDGs? Lancet373: 1500-1502.

Commission on AIDS in the Pacic. 2009. Turning the tide: an open strategy or a response to AIDS in the

Pacic. Report o the Commission on AIDS in the Pacifc. Suva, Fiji: UNAIDS Pacic Region.

Gaziano, T.A., G. Galea and K.S. Reddy. 2007. Scaling up interventions or chronic disease prevention: the

evidence. Lancet370 (9603): 1939-1946.

Gostin, L.O. 2007. Meeting the survival needs o the worlds least healthy people: a proposed model or Global

Health Governance.Journal o the American Medical Association 298, 2: 225-228.

Gostin, L.O. and E.A. Mok. 2009. Grand challenges in global health governance. British Medical Bulletin 90:

7-18.Hanson, F. 2009. China: Stumbling through the Pacic. Lowy Institute or International Policy. Policy Brie, July.

Hughes, H. 2003. Aid has ailed the Pacic. Issue Analysis May: 33.

IHME. 2009. Financing global health 2009: tracking development assistance or health. Seattle, WA: Institute or

Health Metrics and Evaluation.

IDF, IUAC and WHF. 2009. Time to act: The Global emergency o non-communicable diseases. Report on

Health and development: Held back by non-communicable diseases: International Diabetes Federation,

International Union Against Cancer and World Heart Foundation.

Keke, K., P. Phongsavan, L. Dan, M. Bacigalupo, B. Smith, R. Thoma, L. Riley, G. Waidubu, G. Galea, J/ Pryor,

S. Raj and T. Chey. 2007. Nauru NCD risk actors STEPS report. Suva, Fiji: WHO.

Lim, S.S., T.A. Gaziano, E. Gakidou, K.S. Reddy, F. Farzadar, R. Lozano and A. Rodgers. 2007. Prevention ocardiovascular disease in high-risk individuals in low-income and middle-income countries: health eects and

costs. Lancet370 (9604): 2054-2062.

McCoy, D., S. Chand and D. Sridhar. 2009. Global health unding: how much, where it comes rom and where itgoes. Health Policy & Planning 24, 6: 407-417.

NZAID. 2007. The NZAID health strategy (2008-2013).

http://www.nzaid.govt.nz/library/docs/nzaid-health-strategy-2008-2013-1389365.pd. (Accessed March 2010)

OECD. 2008. The Paris Declaration on Aid Eectiveness and Accra Agenda or Action.

http://www.oecd.org/dataoecd/30/63/43911948.pd. (Accessed March 2010)


19/20



Rasanathan, K. and C.F. Tukuitonga. 2007. Tobacco smoking prevalence in Pacic island countries and

territories: a review. New Zealand Medical Journal 12 October, Vol. 120 No. 1263.

Schultz, J.T., P. Vatucawaqa and J. Tuivaga. 2007. 2004 Fiji national nutrition survey. Suva: Fiji National Food and

Nutrition Centre.SPC. 2009. 2007 HIV, AIDS and AIDS related deaths cases and incidence, updated May 2009 (200PICT HIV

case data table 2008 v5[1].pd): Secretariat o the Pacic Community.http://www.spc.int/hiv/index.php?option=com_docman&task=cat_view&gid=108&Itemid=148. (Accessed

December 2009

Shiman, J. and S. Smith. 2007. Generation o political priority or global health initiatives: a ramework andcase study o maternal mortality. Lancet370: 1370-1379.

Shiman, J. 2009. A social explanation or the rise and all o global health issues. Bulletin o the World Health

Organization 87: 608-613.

Sridhar, D. and R. Batniji. 2008. Misnancing global health: a case or transparency in disbursements and

decision making. Lancet372 (9644): 1185-1191.Stuckler, D., L. King, H. Robinson and M. McKee. 2008. WHOs budgetary allocations and burden o disease: acomparative analysis. Lancet372: 1563-1569.

UNAIDS. 2008. 2008 Global report on the global AIDS epidemic. Geneva: UNAIDS.

Walt, G., E. Pavignani, L. Gilson and K. Buse. 1999. Health sector development: rom aid coordination toresource management. Health Policy Plan 14, 3: 207-218.

World Bank. 2008. Project Appraisal Document o the World Bank on a Proposed Credit in the amount o SDR

1.9M to the Independent State o Samoa in Support o Health Sector Management Project: World Bank.

WHO. 2003. The world health report 2003Shaping the Future. Geneva: WHO.

WHO. 2007. Pacifc ramework or the prevention and control o non-communicable diseases. Geneva: WHO.WHO. 2008. 2008-2013 Action plan or the global strategy or the prevention and control o non-communicablediseases. Geneva: WHO.

WHO. 2009. Impact o chronic disease in countries. http://www.who.int/chp/chronic_disease_report/media/

impact/en/index.html. (Accessed March 2010)

WHO and SPC. 2003. Tonga commitment to promote healthy liestyles and supportive environment. Meeting oministers o health or the Pacic island countries. 9-13 March, Nukualoa, Tonga.

WHO and SPC. 2005. Samoa commitment: Achieving healthy islands. Geneva: WHO.

WHOSIS. 2009. WHO statistical inormation system. http://apps.who.int/whosis/data/Search.jsp?indicators=[Indicator].[MBD].Members. (Accessed March 2010)


20/20

KNOWLEDGE HUBS FOR HEALTHStrengthening health systems through evidence in Asia and the Pacic

A strateg ic partnerships ini tiative funded by the Aust ralian Agency for Inte rnational Development

The Nossal Institute

for Global Health

Funding for HIV and non-communicable diseases: Implications for priority-setting in the Pacific region

Documents