Financing tuberculosis control: the role of a global financial monitoring system
Katherine Floyda, Andrea Pantojaa, Christopher Dyea
Control of tuberculosis (TB), like health care in general, costs money. Although some components of diagnosis and treatment can be provided free of charge, such as patient support provided by community volunteers, most inputs – such as national TB programme staff, drugs, laboratory supplies and primary health-care staff and infrastructure – need to be paid for.
Since sustaining TB control at current levels and making further progress to achieve global targets require money, data about funding needs, sources of funding, funding gaps and expenditures are important at global, regional, national and sub-national levels. Such data can be used to answer questions such as: How much money is needed for TB control globally, both in total and for different components of control? How much of the required funding is available, what are the major sources of this funding, and how is the total amount of funding changing over time? Which regions and countries have the biggest needs and funding gaps? To what extent are countries able to spend the funding mobilized for TB control? Have changes in funding for TB programmes had an impact on their performance and the burden of TB?
In common with efforts to assess funding needs, and track funding and expenditures for other global health priorities,1,2 WHO has established a system for global financial monitoring of TB control.3–7 In this paper, we explain why this system was set up in 2002 and how it works in practice. We then illustrate its value by presenting major findings concerning the period 2002–2007, including results that are relevant to the achievement of the global TB control targets set for 2005 and 2015. Finally, we analyse the system’s strengths and limitations, and its relevance to other health-care programmes.
Why was the global financial monitoring system established?
Establishment of a system for global financial monitoring of TB control reflected a major step-up in national and international commitment to TB control dating from around the year 2000. The Stop TB Partnership, which now consists of more than 400 partners, was formed in 2000. In March 2000, 20 of the 22 high-burden countries, which collectively account for approximately 80% of the world’s TB cases, committed to achieving WHO’s TB control targets through implementation of the internationally recommended DOTS strategy.8 The targets were to detect 70% of new smear-positive cases and to cure 85% of detected smear-positive cases by 2005.9 In July 2000, the G-8 countries pledged additional resources and set new targets for the control of HIV infection, TB and malaria.10 The United Nations Millennium Development Goals (MDGs), which provide a framework for development efforts as a whole for the period 2000–2015, were also launched in 2000. They include targets related to TB control, the principal one being to reverse the incidence of TB by 2015.11 The Global Fund to Fight AIDS, Tuberculosis and Malaria (referred to subsequently as the Global Fund) was established in 2002, providing a major new opportunity to improve the financing of TB control.
These developments were associated with an increased demand for information about various aspects of funding for TB control, particularly in high-burden countries. During 2001, WHO made it a priority to assess funding needs and funding gaps in high-burden countries and other low- and middle-income countries for the period 2001–2005, using strategic plans, a standardized questionnaire and costing studies; these findings were published in March 2002.12 Building on these efforts, but recognizing that systematic assessment of financing for TB control would be preferable to more ad hoc studies, WHO established a system for routine monitoring of financing for TB control in late 2002.
The global financial monitoring system in practice
The WHO global financial monitoring system has been operating for five years (five rounds of data collection, 2002–2006; and five rounds of annual reporting, 2003–2007). The system is explained below.
The type of data collected, and the methods and sources used, have evolved over time (Table 1, available at: http://www.who.int/bulletin). Data on national TB programme (NTP) budgets, utilization of general health services by TB patients during treatment (to capture “health system” costs not reflected in NTP budgets), and funding sources and funding gaps have consistently been collected for the 22 high-burden countries (ranked by their number of cases, these are: India, China, Indonesia, Nigeria, South Africa, Bangladesh, Pakistan, Ethiopia, the Philippines, Kenya, the Democratic Republic of the Congo, the Russian Federation, Viet Nam, the United Republic of Tanzania, Uganda, Brazil, Afghanistan, Thailand, Mozambique, Zimbabwe, Myanmar and Cambodia).
In 2004, the system was expanded to cover all countries as well as expenditure data. This expansion was facilitated by integrating a financial questionnaire into a WHO TB data collection form that was already being sent to all countries every year for the purposes of collecting epidemiological data (such as case notifications and treatment outcomes). The Global Fund was included as a specific category for reporting funding sources in 2004, to reflect its increasing prominence as a source of external financing. In 2006, budget categories were re-evaluated and revised to allow for monitoring according to the Stop TB Strategy and the Global Plan to Stop TB, 2006–2015, which were launched in 2006 by WHO and the Stop TB Partnership, respectively.13,14
Table 1. Data collected by global financial monitoring system for tuberculosis (TB) control, 2002–2007
Data management, follow-up and validation
Data management, follow-up and validation are typically done between August and December each year. Data are entered into a database and follow-up queries sent to NTP managers. Particular attention is given to the 22 high-burden countries, and global and regional meetings as well as country missions are used to discuss data.
Various methods are used to check and validate data. These include cross-checking with Global Fund proposals and databases and a recent independent survey of donor financing for TB control;15 comparisons with recent costing studies and with data reported for previous years; and calculation of indicators that can be predicted (e.g. the price of first-line drugs per patient treated). In addition, WHO country staff who have first-hand knowledge of the NTP are often involved in producing the financial reports.
If National Health Accounts (NHAs) included a subcomponent for TB, they would also be a useful way to validate data. NHAs are designed to provide a comprehensive estimate of national health expenditures, including public, private and donor funds (see: http://www.who.int/nha/what/en/). However, while several sub-accounts have been produced for HIV,16 they do not yet exist for TB.
Standard analyses and related graphs or tables are prepared for individual countries, generally from October to December each year. Summary analyses are also prepared for the 22 high-burden countries combined, and for the six WHO regions. These analyses have evolved and improved over time (Table 2, available at: http://www.who.int/bulletin). By 2007, major analyses included trends in NTP budgets by line item and funding source for 2002–2007; trends in total TB control costs by line item and funding source for 2002–2007; total TB control costs (NTP budgets plus health system costs not included in NTP budgets) as a percentage of total government health expenditures; trends in total costs, budgets, available funding and expenditures on a per-patient basis for 2002–2007; comparisons of planned costs, available funding and actual expenditures; assessment of the relationship between increased funding and programme performance; and comparisons of total costs based on country reports with costs according to the Global Plan to Stop TB, 2006–2015.3–7 Examples are provided in the next section.
Some analyses rely on databases managed by WHO’s Health Financing and Stewardship department, notably estimates of the total cost of TB control and the percentage of total government health expenditures used for TB control. Total TB control costs are estimated by adding the costs of hospital admissions and outpatient visits to health facilities by TB patients to the NTP budgets reported on the WHO TB data collection form. Two essential inputs to these calculations are the unit cost of a hospital bed per day and of an outpatient clinic visit, which are taken from a database that includes estimates for every country (see: http://www.who.int/choice/country/en/index.html). Comparisons of total TB control costs with total government health expenditures are made using NHA data, which include estimates of total government health expenditure up to 2005 (see: http://www.who.int/nha/en/).
Table 2. Data analysis and presentation based on the global financial monitoring system for TB control, 2002–2007
How results are presented and disseminated
Results are primarily disseminated via the WHO report Global tuberculosis control: surveillance, planning, financing, which is published in March each year.3–7 Results are presented both within country profiles for each high-burden country that cover surveillance, planning and financing, and within the main body of the report (Table 2; available at http://www.who.int/bulletin). Results are also presented at meetings and conferences, and country profiles are distributed to all countries that report data but for which a profile does not appear in the global report.
Completeness of reporting
The number of countries submitting data has increased over time (Table 3; available at http://www.who.int/bulletin). In the last two years, countries reporting complete data have accounted for over 90% of the global burden of TB.
Staffing and operational funding
The development phase of the system in 2002–2003 was funded by a grant of around US$ 200 000 from the Rockefeller Foundation. Subsequently, funding for staff and operational activities has been provided from a variety of sources, the major one being the United States Agency for International Development (USAID). Operational funding used to date amounts to less than US$ 50 000 per year. Two staff members are principally responsible for the financial monitoring system (one for about 50% of their time, the other for about 25% of their time).
Major findings and implications for reaching TB control targets
Examples of some of the major results produced by the global financial monitoring system are shown in Figs. 1–5 and Table 4 (available at: http://www.who.int/bulletin). These are adapted and simplified versions of figures and tables that are presented in the annual WHO report for 2007.7
Fig. 1 shows how the total cost of TB control has changed in the 22 high-burden countries since 2002, by major cost categories. Actual (2002–2005, based on expenditures) and planned costs (2006–2007, based on budgets) amount to US$ 6.5 billion over six years, increasing from US$ 0.6 billion in 2002 to US$ 1.7 billion in 2007. DOTS treatment for new cases of TB accounted for most of these costs. Costs for clinic visits and hospitalization illustrate the contribution made by general health system resources to TB control. These overall figures conceal considerable variation in total costs and funding gaps among countries, as well as in the cost per patient treated (Table 4; available at http://www.who.int/bulletin). They also show where there may be scope for improving efficiency – the Russian Federation being a notable example. Here, costs are much higher than in other countries, including those with a similar income level, due to extensive reliance on hospitalization and on screening and diagnostic methods that do not conform to international guidelines.
Fig. 2 demonstrates how NTP budgets have grown in the 22 high-burden countries over the last six years, from around US$ 0.5 billion in 2002 to US$ 1.25 billion in 2007. Most of the funding has come from the governments of these countries, but the contribution from the Global Fund has steadily grown since 2002, reaching almost US$ 200 million in 2007. Grant funding from other sources has remained comparatively stable. Funding gaps have existed every year, including the target year of 2005. These gaps help to explain why the targets set for 2005 were not achieved in most countries. For example, in 2005 the funding gap reported by high-burden countries was US$ 172 million.
From 2006 onwards, comparisons between country plans and related budgets, and the Global Plan, assume greater importance. This is because the Global Plan sets out what needs to be done in each year during the period 2006–2015 to achieve the MDG and related Stop TB Partnership targets set for 2015. For 2006 and 2007, Fig. 3 shows that funding requirements based on country plans are about US$ 500 million less than those according to the Global Plan. The discrepancy is mostly due to the Global Plan’s much higher estimates of the funding required for advocacy, communication and social mobilization, and collaborative TB/HIV activities. Besides suggesting a need for plans and budgets that are more in line with the Global Plan, Fig. 3 also highlights a second challenge related to achieving targets set for 2015: filling existing funding gaps. A further challenge – spending the available money – is illustrated in Fig. 4. In 2005, expenditures were less than available funding in three WHO regions, particularly Africa and the Eastern Mediterranean.
Finally, Fig. 5 shows the relationship between increased spending and changes in programme performance. This is measured by comparing changes between 2003 and 2005 in NTP expenditures and the number of new smear-positive patients treated for a selection of 12 high-burden countries. Greater expenditure was strongly associated with improved case-finding in Bangladesh, China, India and Indonesia. However, there was no systematic relationship between increased expenditure and improved case detection across all high-burden countries. Further investigation of this relationship is needed on a country-by-country basis.
Table 4. Total funding required, funding gaps and cost per patient treated for 21 high-burden countriesa in 2007, based on country reports
After five years of operation, the global financial monitoring system has improved such that by 2007, data could be analysed and reported for countries representing 90% of the global burden of TB cases. Factors contributing to this level of reporting include an established system for annual collection of epidemiological, planning and financial data; a network of WHO staff in regions and countries who help to complete and follow-up on reports at country level; working relationships with NTP managers that have been built up since 2002; and timely opportunities for follow-up of data in person.
Requests for data and customized analyses that go beyond those done for the annual WHO global TB control report have grown considerably, demonstrating that the system is producing useful data. Clients include the Global Fund, nongovernmental organizations, bilateral donor agencies and the media. The uses to which the data can be applied have also grown. A recent example is the production of cost estimates for the Global Plan to Stop TB, which made extensive use of the data compiled through the financial monitoring system (see: http://whqlibdoc.who.int/publications/2006/924159487X_eng.pdf). Analyses tailored for use at country level are also being done; for example, in support of resource mobilization efforts. Although separate statistics for the financial data are not available, there were over 300 000 “hits” for the 2006 TB control report as a whole over a one-year period, and it ranked second in terms of downloads of TB-related material from the WHO web site.
The system has several limitations related to data collection, analysis and dissemination.
A noteworthy limitation related to data collection is the system’s focus on provider costs within the public health-care system. No data are compiled on the costs of private-sector provision (unless financed through public health budgets), nor on expenditures by patient/households, i.e. user costs. The reason is that collection of these data is much more difficult and time-consuming, and therefore not feasible on a routine basis.
Within the system itself, collection of data requires considerable follow-up with some countries to ensure that a complete financial report is provided. Expenditure data have consistently been more difficult to collect than budget data. Where development and management of TB budgets and expenditures are decentralized, it is difficult to obtain aggregated national data. Reporting by European countries is relatively poor. Budgets and funding for some of the recommended collaborative TB/HIV activities (e.g. antiretroviral treatment for HIV-positive TB patients) are often not reflected in data reported by NTPs. Budgets and funding for collaborative TB/HIV activities reported by NTPs therefore fall short of the total budgets and funding available for these activities, and are being underestimated. This could be addressed by working more closely with resource tracking work undertaken by the United Nations Joint Programme on HIV/AIDS (UNAIDS). Finally, more opportunities to validate data would be useful, for example via country missions during which data could be reviewed and discussed in detail.
Data analysis is undertaken in a short period from October to December, to allow for publication of results in WHO’s annual report on TB control in March, on World TB Day. With only two staff working part-time on financial monitoring (see above), this means that some potentially useful analyses are not yet being done. An example is analysis of how funding gaps could be filled, based on trends in domestic government health expenditures, a country’s income level and economic growth forecasts, and comparisons of existing funding among countries with similar resources.
The system is also not able to provide information on the funding needed to strengthen health systems as a whole; for example, to improve primary health-care infrastructure and to increase the existing stock of health-care workers. Such costs cannot be budgeted or estimated for TB control alone, nor are they captured in the unit cost estimates used to produce estimates of the total cost of TB control.
Data dissemination could be extended to customize the financial data for particular audiences and to support resource mobilization efforts.
These limitations illustrate the main areas where the existing system for global financial monitoring could be improved, as well as work that is needed beyond this system. Addressing them requires additional staff and operational funding.
Relevance to other health-care programmes
The financial monitoring system that we have discussed is specific to TB control. However, it has broader relevance, including providing useful experience on which other programmes could draw. The need for financial data and the reasons for establishing a system for financial monitoring of TB control are not unique to TB control. For example, some other major health-care priorities are identified by the MDGs: HIV, malaria, maternal and neonatal health, and child health. Systematic resource tracking work is carried out for HIV, and to a lesser extent for immunization programmes. However, all five priorities are likely to require additional funding to achieve the MDG targets.
The TB system also demonstrates that it is feasible to collect financial data, despite initial doubts about countries’ willingness and capacity to report such data. At the same time, important general lessons have been learned. These include the need to keep data collection forms short and simple, with clear accompanying explanatory notes. Other lessons are that careful follow-up is essential, with discussions in person often necessary at the beginning; that expenditure data are more challenging to collect than budget data, with both being harder to obtain in decentralized systems; that the quality of financial data reported is often a good indicator of overall programme quality, especially planning and management capacity; that the quantity and quality of data improve over time as people being increasingly familiar with what is being requested; that the presence of WHO staff at country level makes data collection much easier; and that combined with other sources of data, health system as well as programme-specific costs can be assessed.
After five years of operation, the WHO system for global financial monitoring of TB control is able to report systematically on the financial resources required for TB control, as well as on sources of funding and funding gaps. It is also able to assess the extent to which financial needs estimated by countries, and related resource mobilization and actual spending, are sufficient to achieve global targets for TB control. It is able to do this for countries that represent about 90% of the global burden of TB, and that between 2002 and 2007 accounted for more than US$ 7 billion of actual or planned spending on TB control. While the system has limitations and related areas for improvement, it provides an example of how financing can be routinely monitored that has relevance to other health-care programmes, particularly those concerned with disease control. ■
- B Schwartlander, J Stover, N Walker, L Bollinger, JP Gutierrez, W McGreevey, et al. AIDS. Resource needs for HIV/AIDS. Science 2001; 292: 2434-6.
- Global Polio Eradication Initiative. resource requirements 2006-2008. Geneva: WHO; 2006 (WHO/Polio/2006.1).
- Global tuberculosis control: surveillance, planning, financing. Geneva: WHO; 2003 (WHO/CDS/TB/2003.287).
- Global tuberculosis control: surveillance, planning, financing. Geneva: WHO; 2004. Available from: http://whqlibdoc.who.int/publications/2004/9241562641.pdf
- Global tuberculosis control: surveillance, planning, financing. Geneva: WHO; 2005 (WHO/HTM/TB/2005.349).
- Global tuberculosis control: surveillance, planning, financing. Geneva: WHO; 2006 (WHO/HTM/STB/2006.362). Available from: http://whqlibdoc.who.int/publications/2006/9241563141_Rev_eng.pdf
- Global tuberculosis control: surveillance, planning, financing. Geneva: WHO; 2007 (WHO/HTM/STB/2007).
- Amsterdam Declaration to Stop TB. Available at: http://www.stoptb.org/stop_tb_initiative/assets/documents/decla.pdf
- Fifty-third World Health Assembly. Resolution WHA 53.1. Geneva: WHO; 2000.
- G8 countries, chair’s summary, Okinawa International Conference on Infectious Diseases. Available from: http://www.stoptb.org/resource_center/assets/documents/G8OkinawaReport.htm.
- The Millennium Development Goals Report 2006. New York: United Nations; 2006. Available from: http://mdgs.un.org/unsd/mdg/Resources/Static/Products/Progress2006/MDGReport2006.pdf
- K Floyd, L Blanc, MC Raviglione, JW Lee. Resources required for global tuberculosis control. Science 2002; 295: 2040-1.
- MC Raviglione, MW Uplekar. WHO’s new Stop TB Strategy. Lancet 2006; 367: 952-5.
- Stop TB. Partnership. The Global Plan to Stop TB, 2006-2015 (WHO/HTM/STB/2006.35) Geneva: WHO; 2006. Available from: http://whqlibdoc.who.int/publications/2006/9241593997_eng.pdf
- Trends in international funding for TB control. Geneva: Stop TB Partnership and HLSP Institute; 2005. Available from: http://www.stoptb.org/resource_center/assets/documents/FINAL%20REPORT%20TB%20FUNDING%20(final).pdf
- De S, Dmytraczenko T, Chanfreau C, Tien M, Kombe G. Methodological guidelines for conducting national health accounts: subanalysis for HIV/AIDS. Bethesda: Partners for Health Reformplus; 2004. Available from: http://www.phrplus.org/Pubs/Tech044_fin.pdf
- Stop TB Department, World Health Organization, 20 avenue Appia, 1211 Geneva 27, Switzerland.