Understanding spending trends for tuberculosis.

As the tuberculosis community strives to work towards tuberculosis elimination goals, financing and spending continue to be crucial issues. Tuberculosis usually affects the most poor and vulnerable populations and resources have always been few and strained. Year after year, reports from WHO, STOP TB, and other advocacy groups show that tuberculosis spending is inadequate to diagnose and treat existing cases, and recent meetings such as the 2018 UN high-level meeting on tuberculosis have elicited pledges to improve resources and finances available in the fight against tuberculosis.3, 4 Such pledges might become even more crucial as resources, funding, and manpower initially dedicated towards tuberculosis control efforts are redirected to support efforts to fight the coronavirus disease 2019 (COVID-19) pandemic, which is now affecting many, if not all, high tuberculosis burden countries.

Modelling analyses, such as the one published in The Lancet infectious Diseases by Yangfang Su and colleagues, provide useful information to assess and monitor total tuberculosis spending across low-income and middle-income countries. In their study, Su and colleagues used modelling techniques from the Global Burden of Diseases study to generate comprehensive estimates of total tuberculosis spending from all sources across 135 low-income and middle-income countries between 2000 and 2017, allowing for comparisons across countries and over time. The authors estimate total spending for both notified and non-notified cases. These data can be helpful in understanding financial contributions from different sources including government, pre-paid private spending, out-of-pocket medical expenses, and development assistance for health funding, and in capturing the burden experienced by households and communities not typically captured by more traditional reports focused only on notified cases and government and donor spending. They also disaggregated spending estimates by function (eg, outpatient visits, pre-diagnosis visits, private drug spending).

Su and colleagues found that total tuberculosis spending increased for 2000–17, driven primarily by government and national tuberculosis programme spending on notified cases, and that spending on non-notified cases also increased. Total out-of-pocket spending decreased over the same period; however, although the authors captured direct out-of-pocket spending on medical expenses, they did not include non-medical costs including loss of income, transport, and indirect economic costs due to tuberculosis (many of which are now being collected through WHO patient cost surveys) in their analysis. The authors' findings show that three countries with strong private sectors—Democratic Republic of the Congo, Nigeria, and Pakistan—have out-of-pocket medical expenses as the primary source of tuberculosis spending.

Prepaid private and out-of-pocket spending contributions were found to be relatively small and that many governments in low-income and middle-income countries finance most national spending on tuberculosis. Several, but not all, high tuberculosis burden countries were middle-income countries and cost data is often skewed, driven by increased costs in these countries—eg, average outpatient visits were estimated to cost US$35·92 per visit, while the median cost was only $4·24 per visit, meaning that in half of the countries the cost per visit was lower than $4·24.

Conversely many high tuberculosis burden low-income countries are still heavily reliant on development assistance for health spending. Notably, the authors present mean values weighted by population size or number of incident cases for the region, which is in line with the larger global burden of disease approach; however, such presentation requires careful interpretation. For instance, when looking at data for a specific country, the average total spend per incident case might be driven up by a few key countries in that region with large populations or numbers of incident cases, or both.

Although the results might be an overestimation or underestimation of specific estimates, extensive sensitivity analyses done by the authors show that the qualitative results are robust. Therefore, trends over time and across countries can be used to monitor fluctuations in total tuberculosis spending and assess needs across regions. This work also raises many important questions around reasons for changes in tuberculosis spending trends. Future work in this area should investigate reasons why tuberculosis spending levels have changed over time, with a particular focus on understanding decreasing costs in specific countries.