Understanding health spending for SDG 3

In The Lancet, the Global Burden of Disease Health Financing Collaborator Network¹ present new estimates of historical global health spending and future estimates for 2030 and 2050 using ensemble modelling techniques. Additionally, their contribution breaks down health spending for Sustainable Development Goal (SDG) 3, healthy lives and wellbeing, and for pandemic preparedness. They note that since the development and implementation of the SDG agenda in 2015, health spending has increased. Spending on SDG3 has also increased, but not in all countries, and progress towards meeting targets has been mixed.

Tracking spending for SDG3 is relevant to policy makers and complements previous efforts by Stenberg and colleagues² who estimated the additional funds needed to reach SDG3 in low-income and lower-middle-income countries. Furthermore, the long-term projections of health spending to 2030 and 2050 provided here¹ contribute a useful overview of global and regional developments in health expenditure. We believe that more can be done to increase policy relevance and contextual understanding of data such as those presented—eg, by going into more depth regarding trends in current and future global health spending. An improved understanding of these trends is valuable for policy making and useful for stakeholder groups engaged in following up governments' commitment to health.

Another takeaway from the present Article is that increased funding is needed to achieve SDG3, in particular in low-income countries, and the authors argue that interest in domestic resource mobilisation has been renewed as a key strategy for generating resources for SDG3. We would go even further and argue that current trends show that improved domestic resource mobilisation will be the only way for these countries to mobilise the resources needed for SDG3. The present Article and previous publications from the Global Burden of Disease Health Financing Collaborator Network clearly indicate the diminishing importance of development assistance for health (DAH) for financing services at the country level, even for disease areas commonly perceived as highly dependent on DAH.1, 3, 4 The projections to 2030 indicate that DAH will continue to be important in some low-income countries at the end of the SDG period, but this finding is more a consequence of the lack of growth in domestic spending than an effect of increasing DAH.

The Article also raises questions about the future role of DAH, suggesting that a need exists to shift attention to using DAH for funding “so-called global public goods” for health and collective ability to respond to global health threats.¹ This suggestion has been emphasised before and we agree that a shift in funding is needed,⁵ and this has become increasingly clear in light of the current coronavirus disease 2019 pandemic. Furthermore, global commitment to achieving universal health coverage⁶ requires a gradual move from siloed funding for specific diseases, earmarking of funds, and limitations on what DAH can be spent on (eg, many development partners are still reluctant to finance salaries). Instead DAH, like governments' own health spending, should be allocated to a broad set of interventions addressing the most common health needs of the population. Resource allocation would typically be defined in a country's essential health-care package, and we argue that such country-owned packages should be the centrepiece to determine how resources, including DAH, should be allocated. While countries' efforts towards universal health coverage are increasingly formalised, future analyses from the Global Burden of Disease Health Financing Collaborator Network could hopefully also track progress towards this goal.

© 2020 Vlad Sokhin/Panos Pictures

In the present Article, the authors also present funding estimates for specifically relevant areas for SDG3—ie, HIV/AIDS, tuberculosis, malaria, and universal health coverage.¹ Although these estimates might be useful for a global dialogue around how SDGs are prioritised in DAH, the approach is less relevant for increased understanding of, and improved decision making in, the health sector of countries included in the study. The authors also acknowledge this aspect as a limitation of their study. Consequently, the Article does not take a health systems approach to the data. Instead, aggregated data on health spending are reported followed by spending in disease categories or, in the case of universal health coverage, covering several diseases. For decision makers in a ministry of health, this way of dividing the funds is not particularly helpful because budgeting systems are commonly constructed on the basis of entities that cut across the health system by expenditure categories (eg, staff costs, medical equipment, rent, pharmaceuticals) and not on disease categories. Further work should seek to develop formats for resource requirements that are aligned with standard government budgeting and we look forward to following the authors work in developing such methods.