The Global Fund's resource allocation decisions for HIV programmes: addressing those in need

The Global Fund to Fight AIDS, Tuberculosis and Malaria is a public-private partnership dedicated to attracting and disbursing resources to address HIV, tuberculosis (TB) and malaria pandemics. As of the end of 2010, the Global Fund had allocated US$12 billion and disbursed $7.4 billion for HIV programmes, making it one of the leading sources of funding for HIV programmes worldwide. The resources from the Global Fund, along with resources from key partners, such as the US President's Emergency Plan for AIDS Relief (PEPFAR) and the World Bank Multi-Country HIV/AIDS Program, have made a major contribution to efforts to achieve universal access to prevention, treatment and care services for HIV and AIDS.

By 2009, the joint efforts in this significant expansion in resources had resulted in the reduction of new infections by 19% from the levels in 1999 [1]. However, the global population of people living with HIV continues to be large, numbering an estimated 33.3 million at the end of 2009 [1]. Sub-Saharan Africa remains the region most heavily affected by HIV, accounting for 68% of HIV infections worldwide. The Asian region is home to 4.9 million people living with HIV [2]. The Asian epidemic is still concentrated within specific high-risk populations. Nevertheless, with such a large population, just a small increase could have catastrophic effects [3].

The three regions of the Middle East and North Africa, Latin America and the Caribbean, and Eastern Europe and Central Asia also experience concentrated epidemics. HIV has more heavily affected the Caribbean Region than any other region outside sub-Saharan Africa, with the second highest adult prevalence in the world. In the Eastern Europe and Central Asia region, where injecting drug use is the primary mode of transmission, treatment levels are lower than in sub-Saharan Africa [2], and most people are unaware of their status.

The global economic recession is straining budgets in many low- and middle-income countries, with a decline in health overseas development aid, including commitments to the Global Fund [3]. The Third Voluntary Replenishment of the Global Fund, which led to pledges of US$11.7 billion, will enable further scale up of Global Fund investments for the 2011 to 2013 period, but not at the same pace as in recent years and it is insufficient to meet the anticipated demand. Therefore, not only is there a need to mobilize domestic resources and external aid for HIV programmes, but it is also necessary to ensure that available resources are used as efficiently as possible, and that allocation for HIV prevention, treatment, care and support services matches epidemiological patterns in order to maximize positive outcomes.

This study reviews the Global Fund HIV portfolio in 2002-2010 (funding rounds 1-10). It describes the trends and allocation patterns of the Global Fund investment in HIV programmes and assesses how these investments were allocated in relation to disease burden in the general population and among vulnerable groups, as well as to levels of national income.

By the end of 2010, the Global Fund had approved US$12 billion for HIV programmes in 145 countries. The level of annual HIV investment expanded from $0.3 billion in 2002, when the Global Fund was established, to $1.1 billion in 2003, $2.0 billion in 2008, $2.5 billion in 2009 and $1.2 billion in 2010.

Of the eight Global Fund regions, the three sub-Saharan Africa regions showed the highest absolute gain in investments over time, especially after the high rates of approved funding in Round 8, increasing from US$0.2 billion in 2002 to $1.2 billion in 2008 and $1.1 in 2010), while the Middle East and North Africa region saw the greatest percentage increase. Other regions demonstrated a steady scale up during the reporting period, displaying the highest increases in Rounds 8 and 9.

The Global Fund's guiding principles target investments in line with need for HIV, tuberculosis and malaria, and enable allocation of funding based on country demand.

The key HIV funding provided by the Global Fund was for HIV treatment and care (35%) and prevention activities (29%). There is an emerging consensus that appropriately targeted "know-your-epidemic" prevention efforts need to be expanded and the mix between treatment and prevention interventions need to be adjusted according to the national epidemiological context and assessment of the roots of HIV transmission in the country. In contrast, earlier start points (CD4 cell count of 350 cells/mm³), improved treatment regimens, more effective linkages to care and adherence support and the treatment-as-prevention paradigm [16, 17] would all increase investments needed for HIV treatment and care.

Differences in allocation patterns were observed in relation to the dynamics and severity of the epidemics. The majority of the Global Fund HIV investments (69% of cumulative funds) and the highest per capita funding were channelled to countries in sub-Saharan Africa experiencing generalized epidemics. These countries allocated about 40% of their funding for HIV care and treatment activities. The review of the investment of other key donors in HIV control showed that in 2002-2009, most PEPFAR funds also went to countries with generalized epidemics and mostly for HIV treatment [18], whereas domestic and international funding for prevention remained underfunded [19]. Global investment into HIV treatment and prevention could bring better outcomes if national and international efforts to control HIV epidemics were balanced between the most effective programmatic interventions.

A lower share of the Global Fund HIV investment, as well as lower per capita funding, was targeted to countries experiencing concentrated and low-level epidemics where the recorded infection was largely confined to individuals with risk behaviours, for example, sex workers, people who inject drugs and men who have sex with men. Our analysis showed variability in the Global Fund funding for prevention interventions by type of epidemics. All Global Fund countries prioritized behaviour change communication interventions in their prevention activities, reaching about half of all prevention funds in countries with low-level epidemics. However, cumulatively, only 11% of all of such interventions targeted most-at-risk populations, which are more effective in settings where HIV burden is high among risk groups [20‐24].

The next priority for Global Fund recipients was social marketing of condoms and HIV counselling and testing. While there is some evidence of success in turning around generalized HIV epidemics by changing sexual behaviour, this turns out to be most effective in risk groups in concentrated epidemics [25‐29]. Several studies show only modest evidence for the effectiveness of counselling and testing activities in generalized epidemics settings compared with concentrated epidemics, but concluded that it should not negate the need to expand them [30‐35]. Its great potential should be weighed against other interventions in allocating prevention funding.

In 2002-2010, about 10% of the Global Fund's cumulative approved funding for HIV prevention was allocated to interventions targeting sex workers, people who inject drugs and men who have sex with men. In countries with concentrated and low-level epidemics, funding for interventions targeting prevention in most-at-risk populations account for 18% and 13% of all prevention activities, respectively. The rest of the preventive funds were invested in interventions for the general population that did not address the epidemiological context of the concentrated epidemics. New evidence suggests that targeted approach in funding allocated to the major risks of transmission and acquisition of HIV infection in the concentrated epidemics provides the greatest effect and substantial changes might be possible with a few appropriately targeted efficacious interventions [36].

Although there was low funding for the most-at-risk populations, a review of the UNAIDS country reports on HIV financing in 2005-2009 showed that the Global Fund was the only or the major funding source targeting risk groups for HIV prevention activities for most-at-risk populations in many countries of the Eastern Europe and Central Asia region (such as Albania, Armenia, Bulgaria, Croatia, Georgia, Kazakhstan, Kyrgyzstan, the former Yugoslav Republic of Macedonia, Romania, Tajikistan and Ukraine), as well as in countries of other regions (such as Algeria, China, Ecuador, Madagascar, Mongolia, Swaziland and Thailand) [1, 7, 10, 37].

The Global Fund resource allocation model seeks to ensure that funding is going to where it is most needed. For the purposes of this analysis, the need is interpreted in terms of HIV burden and national income [38]. The observed relationships between the HIV funding per capita, national HIV prevalence and prevalence in MARPs indicate that the Global Fund resource allocations to HIV programmes best correspond to the HIV prevalence in the applicant countries.

Our analysis shows that the Global Fund eligibility criteria resulted in allocating more funds to countries with lower national income. In 2002-2010, the Global Fund provided more support to low- and low-middle income countries (52% and 34% of cumulative funding and US$2.3 and $1.3 per capita, respectively), which is in line with the equity principles of the Global Fund [15]. Country GNI per capita, although positive, was not statistically significant with regards to the Global Fund allocations per capita, except for the Eastern Europe and Central Asia region and within the group of countries with generalized epidemics. For some upper-middle-income countries, mostly representing the Eastern Europe and Central Asia and the Latin America and Caribbean regions, the funding per capita was comparable to those in low-income countries, disregarding the higher cost of living and higher unit cost of HIV interventions in the concentrated HIV transmission settings of these regions. This demonstrates that the Global Fund invests in HIV programmes in countries with the least financial ability to address the problem.

However, within this group, the HIV funding does not linearly correspond to the country's national income. The national HIV prevalence and prevalence in MARPs predict the magnitude of the Global Fund investment, acknowledging the focus of the Global Fund programmes not only on the income level of the countries, but also in prioritizing the most-in-need countries and population groups; the latter was addressed in Round 10 (2010). A targeted response to concentrated epidemics is being achieved through revised prioritization criteria adopted by the Global Fund Board for Round 10 that allowed upper-middle-income countries to access funding solely for most-at-risk populations.

This expansion of the Global Fund eligibility criteria for upper-middle income countries allowed the organization to overcome one of the drawbacks of the use of the GNI per capita Atlas method indicator as one of the eligibility criteria as it is affected by annual fluctuations in the value of the respective domestic currencies in relation to the US dollar [39, 40] and excludes some countries in need from being eligible to receive support from the Global Fund. The use of the GNI per capita indicator as a criteria for eligibility for Global Fund support does not account for the sub-national distribution of income, which is part of the social policy in many upper-middle-income applicant countries, where sub-national averages of income significantly deviate from national averages and affect subsequently equity in resource allocation by income [41‐44]. The regression analysis we conducted using purchasing power parity did not bring significant difference in the results; thus, we are not presenting them in this paper. We have not adjusted our analysis to control for the variations in the unit cost of service delivery in the countries with different income level that might evidence a stronger correlation between GNI and the Global Fund funding.

This study assessed only some of the considerations that predict the Global Fund's funding decisions. These include HIV prevalence, prevalence of risk factors and national income. However, there are other factors that influence Global Fund resource allocation, as well as the country's demand for HIV funding, such as the potential for a rapid increase in burden of disease due to the current trends, size of population at risk, and extent of cross-border and internal migration.

The Global Fund resource allocation decisions are also based on the levels of national contributions to the financing of the proposal and contributions of other key funders, such as PEPFAR, the World Bank and the Bill & Melinda Gates Foundation, in order to ensure that Global Fund support for HIV is as additional to other sources as possible. The country capacity to implement the grant and existence of supportive national policies play a vital role in the distribution of the Global Fund's resources. These are the areas that should be further explored to ensure an evidence- and performance-based resource allocation for HIV control in the Global Fund recipient countries.

The Global Fund resource allocation model allows for the scale up of investment in HIV prevention, treatment, care and support programmes, and its funding is aligned with HIV burden and national income. However, prevention in most-at-risk populations still does not have an urgent enough priority in most of the country programmes supported by the Global Fund. The intensified and targeted response to HIV control in these populations was further addressed through revised prioritization criteria adopted by the Global Fund Board for Round 10. More guidance is being provided for Round 11 to strategically focus demand for Global Fund financing, which is crucial in the present resource-constrained environment.