Financing DOTS expansion
The total cost of TB control in the HBCs was about US$ 850 million in 2002, with a large increase in planned expenditure to US$ 1 billion in 2003. In both years, funds came primarily from governments (through domestic revenues and loans), and to a lesser extent from grants. The funding shortfall reported by HBCs in 2003 was only US$ 41million, about 4% of the total, and lower than in 2002.
But summary statistics of this kind conceal a diversity of financial needs among the countries that carry the largest burdens of TB. Our analysis of budgets and expenditures puts the 22 HBCs into broadly three groups. The first, most progressive group contains 10 countries that have planned to significantly increase spending from 2003 onwards, in order to meet the global targets for case detection and treatment success by 2005. Encouragingly, this group includes four of the countries with the most TB cases: India, China, Indonesia, and Bangladesh. India's projected budgetary growth should allow the rapid increase in patient recruitment to continue, while maintaining the same per patient expenditure that has yielded high cure rates under DOTS. China, Indonesia and Bangladesh aim to improve case detection while spending more on the management of each patient. In Ethiopia, Kenya, Cambodia, Uganda, and Myanmar, the total increases in planned costs are smaller but, as for the larger countries, they are linked to plans for scaling up and improving the quality of DOTS. The Russian Federation plans a major increase in activities and costs in 2004. All of these forward-looking countries, with the exception of India, will need some extra money to put their plans into action. Kenya, Cambodia, Uganda, and Myanmar report the largest budgetary shortfalls relative to their needs. However, once approved funding from the GFATM is disbursed in full, the deficits in Myanmar and Uganda will be eliminated. Some of the country budgets are well-reasoned and consistent with recommended policy; others are less so. The Russian Federation errs towards the latter, where a large part of the need is generated by the purchase of X-ray equipment and by the costs of refurbishing hospitals.
In the second group of countries are Brazil, the Philippines, Thailand and Viet Nam, where a large proportion of patients are already treated in the public sector, either by DOTS or non-DOTS programmes. They probably do not require large budget increases to meet targets, and funding gaps are low or non-existent.
The remaining eight countries are in a third group, where NTPs are not yet close to reaching targets, and apparently have neither plans nor budgets that will get them to the targets by 2005. Some of these countries provided no data either for 2002 or 2003; for others the planned increase in costs was small. Some members of this group did plan budgetary increases, but without explanation. If the 13 constraints that emerged from our review of planning are genuinely obstacles to TB control, we would expect to see large and well-justified budgets to overcome them. In the absence of new sources of money, we would also expect to see larger funding gaps.
In general, the governments of richer countries pay a larger fraction of the costs of TB control. For the poorer countries that have identified greater needs, progress in TB control will be closely linked to the flow of funds from grants, especially those recently awarded by the GFATM. The GFATM has rapidly become a major donor for TB control, but our analysis raises difficulties of two kinds. First, payments from the Fund have so far been small compared with the size of grants awarded. During 2003, only 16% of the total approved for TB and TB/HIV activities in the first 2 years was paid to countries. Second, it is questionable whether large influxes of new money can be immediately and effectively used in countries that have little experience of rapidly scaling up health interventions, and weak capacity for developing effective plans. The HBCs have together planned a sizeable 18% increase in expenditure for 2003. The GFATM grants to Bangladesh, Ethiopia and Myanmar would (at least) double the annual funding available for TB control in these countries in 2004. As external donors contribute more to TB control, filling the current holes in budgets, attention will turn to the absorption capacity of the poorest countries.
A strength of comparative, cross-country analysis is that it suggests various ways in which TB control in the HBCs could be improved. For example, the government contribution to funding is lower in China than in Viet Nam, even though China has a higher GNI. The comparatively high costs per patient treated in South Africa and the Russian Federation can be explained by their over-reliance on hospital care and expensive diagnostics. In other HBCs, a higher proportion of patients are successfully treated at lower cost outside hospitals and clinics. Although the Russian Federation has a relatively high GNI, the government foresees a large funding gap for 2004 and 2005. Some of these need could perhaps be met from domestic resources.
There remains much variation among HBCs in the way they report data on budgets and expenditures. Several countries, including India, Brazil, China, Viet Nam, and Indonesia, provided complete data and little or no follow-up was required from WHO. For others, much discussion with NTP managers and WHO country staff was needed to satisfactorily complete the questionnaire. During 2003, a large number of low-burden countries submitted data, but the poor quality of some of these data made them unusable. The reporting problems in high- and low-burden countries included the following: aggregate budget and expenditure totals were given with no breakdown by line item and funding source; information about GFATM proposals and awards was excluded, and data contained in GFATM proposals was inconsistent with data submitted to WHO; loans providing support to the health sector as a whole (e.g. from the World Bank in Brazil, Indonesia, and Tanzania) were not mentioned; the costs of dedicated NTP staff were not accurately calculated, or not calculated at all; and drug budgets were apparently inconsistent with the number of patients to be treated (often due to the existence or purchase of a drug buffer stock). The budgeting exercise has been made difficult in some countries with decentralized TB control, because funds for TB control are allocated at sub-national level and there is limited transparency or reporting of line items to national level.
While some of these complications are understandable, they raise questions about the capacity of NTPs to plan strategically, and to adequately fund and implement a DOTS programme. During 2004, WHO will address the difficulties that respondents faced in completing the financial questionnaire. The questionnaire itself will need revision: it is not yet clear, for example, what countries are budgeting for TB/HIV activities and for the treatment of MDR-TB cases, because they are not line items on the questionnaire. For the same reason, it is generally unclear what countries would wish to budget for external technical assistance. Technical assistance is needed to support a variety of activities, including the effective use of grants from the GFATM. Based on the observation that many proposals to the GFATM appear to be rich in financial data, there is no doubt that it will be possible to gather more budgetary data of higher quality from more countries, and with greater efficiency.
As the WHO database grows, the investigative techniques applied to these data will need to be refined and developed. On refinement, the projections of costs for 2004 and 2005 in the 22 HBCs assume, among other things, that the cost per patient treated will remain constant as the number of cases detected increases. This would underestimate resource requirements if the cost per patient increases as additional cases become harder to find, or more difficult to treat. On development, there is no general procedure, as yet, for calculating the expected percentage of a country's total health spending that should be used for TB control. These are two examples of the analytical challenges facing the financial monitoring project.
In summary, the estimated cost of TB control in the HBCs was about US$ 1 billion in 2003, and rising. Ten of the 22 HBCs project budgetary increases that are in line with plans for a major expansion of DOTS coverage. But some of these countries need to find significantly more money, and to find ways of efficiently disbursing this money, if they are to turn these plans into patients diagnosed, treated and cured. Four of the HBCs probably do not need much more money to reach the targets because most TB patients are already treated in the public sector, if not always under DOTS. The stated funding needs and funding gaps for the remaining Eight countries are almost certainly too low. These countries need sharply-focused strategic plans to overcome the constraints laid out in this report. For some of these countries, the planning and implementation of DOTS will come too late to reach the targets by 2005.