Financing TB control
Financial data were received from 156 out of 212 (74%) countries (Table 19), continuing the year-on-year increase in reporting since the start of data collection in 2002 (the total in Global tuberculosis control 2006 was 140 countries).1 Complete budget data for 2006 were provided by 98 countries (up from 87 in last year’s report), 87 countries provided complete budget data for 2007, and 83 provided complete expenditure data for 2005 (compared with 73 that provided complete expenditure data for 2004). The countries that provided financial reports accounted for 96–100% of the regional burden of TB in four WHO regions, with lower figures of 85% and 81% for the Region of the Americas and the European Region, respectively. Overall, countries that reported financial data accounted for 98% of the global burden of TB.
Data were received from all 22 HBCs, including South Africa for the first time (Table 20). Complete budget data for 2006 were provided by 21 countries (the exception was Thailand), and complete budget data for 2007 were provided by 19 countries (the exceptions were Thailand, UR Tanzania2 and Zimbabwe). Complete expenditure data for 2005 were provided for 19 countries, with data missing for Thailand, Uganda and Zimbabwe. A total of 21 countries provided data on the utilization of health services and made projections of the number of cases they would treat in 2006 and 2007. While considerable clarification and verification of data by WHO are still required, the quality of the data when first submitted is improving: Bangladesh, Brazil, China, India, Indonesia, South Africa and UR Tanzania provided timely and exemplary data that required almost no follow-up.
NTP budgets and funding
High-burden countries, 2002–2007
NTP budgets in 21 of the 22 HBCs have increased during the period 2002–2007, sometimes by substantial amounts (Figures 34–35; Table 21). There are insufficient data to make an assessment for Thailand. The total combined budget for the 22 HBCs in 2007 is US$ 1.25 billion, 2.5 times the US$ 509 million budgeted in 2002. The Russian Federation has by far the largest budget (US$ 513 million), followed by China (US$ 200 million), South Africa (US$ 95 million), India (US$ 75 million) and Indonesia (US$ 59 million), making a combined total that is 75% of the NTP budgets reported by HBCs. There are three countries with budgets in the range US$ 30–50 million and four with budgets in the range US$ 20–30 million; the rest (10 countries, half of which are in Africa) have budgets of under US$ 20 million.
In absolute terms, the budgetary increase in the Russian Federation dwarfs that in any other HBC, at US$ 351 million since 2002; the second largest increase (in China) was US$ 103 million. In relative terms, the increases in nine countries (Afghanistan, Brazil, DR Congo, Kenya, Myanmar, Nigeria, Pakistan, the Russian Federation and Zimbabwe) stand out, with three- to eight-fold increases over six years (Table 21). Countries with relatively small increases are Ethiopia, the Philippines, UR Tanzania and Viet Nam. Across all 22 HBCs, DOTS has consistently accounted for the largest share of NTP budgets,3 but since 2004 an increasing share of these budgets has been accounted for by MDR-TB treatment and new approaches such as PPM, community TB care, ACSM and PAL (Figure 34). NTP budgets for collaborative TB/HIV activities remain small, although Kenya is an exception (see Annex 1).
These large budget increases have been accompanied by big improvements in available funding (Figures 34–35; Table 21). For all HBCs, funding for NTP budgets has increased by US$ 592 million since 2002, reaching US$ 1 billion of the US$ 1.25 billion needed in 2007. Kenya and Viet Nam are the only countries where projected funding for 2007 is less than in 2002, although in the case of Kenya this is because the NTP is unsure about whether funding theoretically available in GFATM grants will be approved for disbursement and because multi-year grants with bilateral donors need to be renegotiated during 2007.4 While most of the extra US$ 592 million has come from HBC governments (US$ 404 million including loans), this overall statistic conceals the fact that most of the additional domestic funding comes from three countries only: China, the Russian Federation and South Africa (an extra US$ 340 million including loans since 2002). Although most other HBC governments have also increased their domestic funding (the six exceptions are Afghanistan, Cambodia, Ethiopia, Kenya, the Philippines and Viet Nam), the remaining increase in funding is largely due to the GFATM. Funding from the GFATM in 2007 amounts to US$ 168 million compared with zero in 2002, and all HBCs have now secured GFATM grants (although Myanmar’s grant has been terminated and funding ended in 2006). The largest grants are held by Bangladesh, China, India, Indonesia, Nigeria and the Russian Federation (worth US$ 10–30 million in 2007); in other HBCs, grants are worth in the range US$ 1–8 million in 2007. In relative terms, the most impressive improvements in funding overall (from all sources) have occurred in Indonesia, Myanmar and Zimbabwe (Figure 36), mainly due to GFATM funding in Indonesia and Zimbabwe and GDF funding in Myanmar.
Among all HBCs, national governments will provide US$ 758 million (61%) of the funding required by NTPs in 2007 and US$ 241 million (19%) will be funded by donor agencies (Table 21). This leaves a reported funding gap of US$ 251 million (20%). In absolute terms, the largest funding gaps (as in 2006) are those reported by China, Kenya, Pakistan and the Russian Federation (US$ 186 million, or 74% of the total gap). Proportionally, the largest gaps are in Afghanistan, Cambodia, DR Congo, Kenya, Mozambique, Myanmar and Pakistan (with gaps representing 40–93% of the required budget).
Further details, including charts showing trends in NTP budgets by funding source and line item for each HBC during the period 2002–2007, are provided in Annex 1.
All countries by region, 2007
The Global Financial Monitoring Project started to collect data from all countries (rather than focusing only on the 22 HBCs) in 2003 and to report on these data in 2004. Since there is variation in the set of countries that report complete data each year, presentation of needs for all countries over time is difficult. For this reason, Figure 37 presents NTP budgets by source of funding for 2007 only. In 2007, 90 countries (22 HBCs and 68 other countries) that collectively account for 90% of the global burden of TB submitted complete data.5 These countries accounted for almost all of the regional burden of TB in the Eastern Mediterranean, South-East Asia and Western Pacific regions, for 87% of the regional burden in the African Region, 57% of the burden in the Region of the Americas, and 65% of the regional burden in the European Region. These figures mean that the reporting of complete financial data to WHO has been maintained (compared with 2006) in the South-East Asia and Western Pacific regions, and improved in all regions except the Region of the Americas.6
NTP budgets in 2007 in these 90 countries total US$ 1.6 billion, with a funding gap of US$ 307 million (both figures higher than for 2006). Budgetary funding gaps as a proportion of the total budget are higher in HBCs compared with other countries, except in the African Region and the Region of the Americas. Overall, NTP budgets per TB case (estimated annual incidence) were lower for HBCs compared with non-HBCs in four regions; in the African Region, budgets were very similar (US$ 138 per case and US$ 135 per case for HBCs and non-HBCs respectively), and in the European Region the budget for the Russian Federation was higher than the average for the other 16 countries that reported data.
Total costs of TB control
High-burden countries, 2002–2007
NTP budgets include only part of the resources needed for TB control. In particular, they do not include the costs associated with general health-service staff and infrastructure, which are used when TB patients are hospitalized or make outpatient clinic visits for DOT and monitoring. For the 22 HBCs combined, the total cost of TB control is projected to be almost US$ 1.7 billion in 2007, compared with US$ 644 million in 2002 (Figures 38–40; Table 22). The figures for total costs 2002–2006 are lower than those reported in Global tuberculosis control 2006, due to downward revisions of the costs estimated for South Africa following the reporting of financial data and related estimates of health services utilization (hospitalization and clinic visits) to WHO for the first time in 2006. Notably, the financial report for South Africa included lower estimates of the frequency and duration of hospitalization compared with the costing studies conducted in the mid-late 1990s that were used to produce cost estimates for previous reports in this series.
Increases in projected costs during the period 2002–2007 arise because of the large increases in NTP budgets (described above) and because of the higher costs of clinic visits and hospitalization that are associated with treating more patients. As in previous years, the largest costs in 2007 are for the Russian Federation and South Africa, which together account for US$ 829 million, or almost exactly half of the total cost of US$ 1.7 billion (Figure 40; Table 22). South Africa is a middle-income country, and the high costs are mainly explained by the higher prices for items such as hospitalization and outpatient visits, compared with those typical in low-income countries, as well as a relatively large budget for treatment of MDR-TB (US$ 43 million for about 6000 patients). The high costs in the Russian Federation reflect continued staffing and maintenance of an extensive network of TB hospitals and sanatoria, a large budget for second-line anti-TB drugs to treat many MDR-TB patients (US$ 91 million, with an estimated total of about 34 000 cases) and continued use of fluorography for mass population screening. China (US$ 200 million), India (US$ 119 million), Brazil (US$ 74 million) and Indonesia (US$ 64 million) rank third to sixth. These six countries account for 78% of the total cost of TB control in the 22 HBCs. An additional nine countries have total costs in the range US$ 23–52 million in 2007, and the remaining seven have costs of US$ 19 million or less.
The countries with by far the largest projected absolute increases in annual costs between 2002 and 2007 are the Russian Federation and China (US$ 423 million and US$ 139 million respectively). They are followed by increases in the range US$ 36–81 million in Brazil, India, Indonesia, Nigeria and South Africa. The smallest absolute changes are projected for Cambodia, Ethiopia, Uganda, UR Tanzania, and Viet Nam. The biggest proportional increases are for Afghanistan, Kenya, Myanmar, Nigeria and Pakistan.
Funding for the general health-service staff and infrastructure used by TB patients during clinic visits and hospitalization is assumed to be provided by governments. This assumption, together with the implicit assumption that health systems have sufficient capacity to support the treatment of growing numbers of patients in 2007, means that the resources available for TB control are estimated to have increased from almost US$ 644 million in 2002 to US$ 1.4 billion in 2007 (Figure 39; Table 22). The contribution by HBC governments to the total cost of TB control in 2007 is 70% on average, which is larger than their contribution to NTP budgets. However, this high average figure conceals important variations among countries. There are 10 HBCs that are dependent on grants to cover more than one-third of the total costs of TB control (Bangladesh, Cambodia, DR Congo, Ethiopia, Indonesia, Myanmar, Nigeria, Uganda, UR Tanzania and Zimbabwe), and a further four (Afghanistan, Kenya, Mozambique and Pakistan) that are likely to rely on grant funding to a similar or greater extent to fill reported funding gaps (Figure 41). The share of the total costs provided by HBC governments is closely related to average income levels (Figure 42), although the government contribution relative to income levels is comparatively high in Ethiopia, India, South Africa, UR Tanzania and Viet Nam, and comparatively low in Indonesia, Kenya, and Pakistan. For all HBCs, the estimated gap between the funding already available and the total cost of TB control is US$ 251 million in 2007, i.e. the NTP budget gap reported above.
Further details, including charts for each country that show trends in total TB control costs by line item for each year 2002–2007, are shown in Annex 1.
High-burden countries: country reports compared with the Global Plan
The Global Plan has set out what needs to be done between 2006 and 2015 to achieve the MDG and related Stop TB Partnership targets for TB control. For the Global Plan to be successfully implemented, country-level planning and budgeting for TB control needs to be in line with the seven regional plans and budgets that are described in the Global Plan; plans need to be fully funded; and planned interventions and activities need to be fully implemented. For the 22 HBCs as a whole, planned costs and available funding for 2006 and 2007 according to country reports are compared with those derived from the Global Plan,7 as well as with planned costs, available funding and actual expenditures in 2005, in Figure 43. This shows that while planned costs and available funding reported by countries are higher in 2006 and 2007 compared with 2005, they are much less than the funding requirements included in the Global Plan. For example, in 2007 the Global Plan indicates that US$ 2.2 billion is required in the 22 HBCs, while country reports indicate planned costs of US$ 1.7 billion, and available funding of US$ 1.4 billion. The discrepancy is mostly due to lower planned costs for collaborative TB/HIV activities (especially in the African region – see Annex 1) and ACSM. Exceptions where planned costs in country reports are either in line with or more ambitious than the Global Plan include Brazil, China, Kenya, the Philippines and Viet Nam (see Annex 1).
All countries: country reports compared to the Global Plan
The financial data submitted to WHO allow total TB control costs for 2007 to be estimated for 84 of the 172 countries that were included in the Global Plan (22 HBCs and 62 other countries).9 These 84 countries account for 90% of all new cases arising each year, while the 172 countries included in the Global Plan account for 98% of such cases. A regional comparison of costs and available funding based on (a) country reports and (b) the Global Plan is shown for these 84 countries in Figure 44. Overall, country reports indicate planned costs of US$ 2.3 billion, compared with US$ 3.1 billion in the Global Plan. As for the 22 HBCs, the main discrepancy is the higher costs for collaborative TB/HIV activities and ACSM that are included in the Global Plan. However, Figure 44 also illustrates that this overall discrepancy is mostly accounted for by the African and (to a lesser extent) South-East Asia regions. In the Western Pacific Region, costs based on country reports are similar to those set out in the Global Plan. In the Region of the Americas and the Eastern Mediterranean Region, higher costs in the Global Plan reflect higher projections of the number of patients that need to be treated in DOTS programmes (both regions) and, in the Eastern Mediterranean region, an NTP budget that is not increasing in line with country projections of patients to be treated (notably in Pakistan). In the European Region, planned costs based on country reports are higher than those in the Global Plan. These differences mean that while the funding gap reported by countries amounts to US$ 307 million in 2007, the funding gap would be US$ 1.1 billion if the available funding of US$ 2.0 billion is compared with the funding requirements of US$ 3.1 billion set out in the Global Plan.
Budgets and costs per patient
Budgets and costs per patient in HBCs are shown in Table 23. The budget for first-line anti-TB drugs is lowest in Bangladesh (US$ 13) and highest in South Africa (US$ 61). In most countries, the budget is in the range US$ 16–35. The relatively high figure of US$ 51 for Kenya is due to the purchase of a one-year buffer stock; it is possible that the comparatively high figures for Mozambique and UR Tanzania have a similar explanation.
The budget per patient, including all line items, also varies. Three countries have budgets below US$ 100 per patient (Ethiopia, India and Pakistan). A total of eight countries have budgets in the range US$ 100–200 per patient, five are in the range US$ 200–300 and four are in the range US$ 300–550.10 The Russian Federation is the only country with a budget above US$ 1000 per patient. The total cost per patient treated in 2007 is below US$ 100 in Ethiopia and India, in the range US$ 100–300 in 12 countries, and US$ 300–500 in three countries. There are four countries with much higher costs: Afghanistan, Brazil, the Russian Federation, and South Africa. Afghanistan’s relatively high costs reflect the need to rebuild the basic infrastructure required for TB control,11 as well as a plan for 2006–2010 that incorporates all elements of the new Stop TB Strategy and follows the planning and costing framework used for the Global Plan. The other three countries are middle-income countries with generally higher prices for the inputs needed for TB control and in the Russian Federation, as noted above, a further explanation is the continued reliance on lengthy hospitalization of patients as well as mass population screening using fluorography. Among the low-income countries, there is no clear-cut relationship between the cost per patient treated and GNI per capita: for example, in India and Pakistan the cost per patient treated is low relative to income levels, while in DR Congo and Mozambique the cost per patient treated is relatively high compared with GNI per capita (data not shown). Overall, budgets and costs per patient are generally increasing, with a median increase of 90% per patient for budgets and of 80% for total costs (though the median for first-line drugs shows no change since 2002).
Further details, including charts that show five per patient indicators (costs, budgets, available funding, expenditures and budget for first-line anti-TB drugs) for each year 2002–2007 for each HBC, are provided in Annex 1. Data have also been compiled and analysed for all other countries that reported data, and are available upon request.
Expenditures compared with available funding and case detection
For countries that have received large increases in funding, there are two important challenges: to spend the extra money, and to translate extra spending into improved case detection and treatment success rates. To date, we have been able to conduct analyses for the HBCs only.
The ability to translate additional funding into spending can be assessed by comparing expenditures with available funding (Table 24; Figure 45). Complete sets of data on budgets, funds and expenditures for 2005 were available for 18 HBCs (the exceptions being South Africa, Thailand, Uganda and Zimbabwe). When budget and funding data were prospectively reported for 2005, five of these 18 HBCs had fully-funded budgets (Afghanistan, Brazil, India, Indonesia and Viet Nam). Among these five countries, Brazil, India, and Viet Nam spent all the available funds; in Brazil and India, expenditures included the spending of funds that were mobilized in excess of the original budget. China was also successful in mobilizing additional funding during 2005, and spent funds that were in excess of the original budget. Apart from these six countries, budgets were not fully funded and, except for Myanmar, Nigeria and the Russian Federation, expenditures were almost always less than available funding. Expenditures were particularly low in relation to available funding in Afghanistan and Pakistan. For three African countries highlighted as spending less than 50% of the funding available to them in 2004, there was an improvement in 2005. In 2005, Kenya, Mozambique and UR Tanzania spent 98%, 66% and 78% of available funding, respectively.
China was also successful in mobilizing additional funding during 2005, and spent funds that were in excess of the original budget. Apart from these six countries, budgets were not fully funded and, except for Myanmar, Nigeria and the Russian Federation, expenditures were almost always less than available funding. Expenditures were particularly low in relation to available funding in Afghanistan and Pakistan. For three African countries highlighted as spending less than 50% of the funding available to them in 2004, there was an improvement in 2005. In 2005, Kenya, Mozambique and UR Tanzania spent 98%, 66% and 78% of available funding, respectively.
The ability to translate spending into improved case detection can be assessed by comparing changes in expenditures 2003–2005 with changes in the number of patients treated 2003–2005 (Figure 46; 2005 is the most recent year for which both case notification and expenditure data are available). Of the 18 countries for which data were available, all but one that increased spending between 2003 and 2005 also increased the number of cases (both new smear-positive and new cases as a whole) that were detected and treated in DOTS programmes (the exception was UR Tanzania). However, the relationship was variable. In Brazil and the Russian Federation, the increase in the number of patients treated under DOTS was far in excess of the increase in expenditures, probably because increasing the number of cases treated under DOTS requires a substitution of DOTS for non-DOTS treatment rather than an increase in total notifications. There was a close to one-to-one relationship between increased expenditures and increased notifications of new smear-positive cases under DOTS in China, while the percentage increase in notifications of new smear-positive cases under DOTS was 56–87% of the percentage increase in expenditures in Bangladesh, DR Congo, India, Indonesia, Kenya, Myanmar and Nigeria (with a range of 46–81% when all forms of new case are considered). There were four countries where the percentage increase in the number of cases treated in DOTS programmes was small compared with the increase in expenditures (Cambodia, Mozambique, UR Tanzania and Viet Nam). In three countries, reported expenditures fell while the number of cases treated increased (Afghanistan, Ethiopia and Pakistan). This fall in expenditures combined with an increase in the number of cases treated is plausible in Ethiopia, since large capital expenditures occurred in 2003, but the data for Afghanistan and Pakistan suggest that expenditures are being underreported. Finally, in the Philippines there were relatively small absolute changes in both expenditures and cases (all forms) treated (2% and 3% respectively).
GFATM contribution to TB control
In HBCs, the GFATM is the single most important source of external financing, with nine countries (Bangladesh, Cambodia, DR Congo, Ethiopia, Indonesia, Nigeria, the Philippines, Uganda and Zimbabwe) relying on the GFATM to fund more than 25% of their NTP budgets. After six rounds of proposals, the total value of approved proposals in the HBCs is US$ 1.3 billion (Figure 25). The amounts in the Phase 1 grant agreements (i.e. the grants that cover the first two years of the proposal) total US$ 519 million.
By the end of 2006, US$ 324 million had been disbursed. For each country, we can compare the actual and expected rates of disbursal, where the expected rate assumes that disbursements should be spread evenly over the two or five year period of the grant agreement following the programme start date (Table 25).3 Across all grants and countries, the actual disbursement rate is similar to the expected rate. However, for half (19 out of 38) of the grants the actual disbursement rate is below the expected rate, and for half it is above the expected rate. Disbursements are particularly low in relation to the expected disbursement of funds in India (round 2 but not rounds 1, 3 and 4), Kenya (round 2 but not round 5), South Africa (rounds 1 and 2) and Viet Nam (round 1). The main delay in the initial flow of funds to countries is the time taken to sign the grant agreement after proposal approval; the median time is 11 months (range 2–34 months), which is in line with GFATM expectations that it takes about one year to prepare and finalize the Phase 1 grant agreement and related documentation. Once grant agreements are signed, disbursements are usually made within 2 months.
In six funding rounds between 2002 and 2006, the GFATM approved proposals worth a total of US$ 2.1 billion for control of TB and TB/HIV in 92 countries, including all 22 HBCs. The total for TB proposals was US$ 1.9 billion. The African Region has the single largest share, at 39% (Figure 47), which is higher than its share of the global burden of TB (28%). The South-East Asia and Western Pacific regions have the second and third highest funding in absolute terms, but less than might be expected given their share of the global burden of TB. The funding approved for the Eastern Mediterranean Region is in line with its share of the global burden of TB (7%), while the share of funding for the European Region and the Region of the Americas is higher than these regions’ share of the global burden of TB.
The value of approved proposals for TB control was relatively high in rounds 5 and 6 compared with rounds 1–4, as was the proposal approval rate (Figure 48).14 The approval rate for TB proposals submitted to the GFATM was 50% in round 5 and 62% in round 6, up from 37–40% in rounds 1–4.
1 Global tuberculosis control: surveillance, planning and financing. Geneva, World Health Organization, 2006 (WHO/HTM/TB/2006.362).
2 As in previous years, the planning cycle in UR Tanzania means that we did not expect budget data for 2007 to be reported.
3 See Methods for definition of the budgetary line items included in the category DOTS.
4 If multi-year grants are successfully renegotiated and GFATM grants are disbursed on schedule, then the funding available in 2007 will be higher than in 2002.
5 Data in 2007 assumed to be as for 2006 in Thailand, UR Tanzania and Zimbabwe.
6 This is because Peru reported data in the 2005 round of data collection, but not the 2006 round of data collection used for this report.
7 Nonetheless, the capacity of health systems to manage an increasing number of TB patients warrants further analysis, particularly in countries where the number of patients will need to increase substantially to achieve the MDG and related Stop TB Partnership targets for TB control.
8 See Methods for explanation of how costs for individual countries were derived from the Global Plan.
8 Six of the 90 countries that reported complete data were not considered in the Global Plan cost estimates.
10 Figures were not calculated for Thailand because the budget and health services utilization data reported to WHO were incomplete.
11 While we have reported these costs as part of the NTP budget, they will help to strengthen the health system as a whole.
12 This explains why the value of expenditures in 2005 as a percentage of the available funding prospectively reported in 2005 (final column of Table 24) is above 100.
13 For other countries, a summary table with the same indicators as those shown for the HBCs is available upon request.
14 Calculated as the number of proposals approved divided by the number of proposals reviewed by the GFATM’s Technical Review Panel.