Progress in global health will need high-quality evidence

In 2015, the international community made a historic pledge to end the epidemics of AIDS, tuberculosis, malaria and other infectious diseases by 2030. Countries committed to achieving universal health coverage, promising to ensure that everyone, everywhere has access to safe and affordable medicines and vaccines. These pledges formed the third goal in the United Nations Sustainable Development Goals, which focuses on healthy living and well-being for all. Achieving the goals by the UN deadline of 2030 will always be difficult. But the commitment of world leaders to these and other goals was no small achievement.

Some progress was recorded in the first five years after that historic moment. There were fewer deaths among newborns and children under five. New HIV infections have declined, and the proportion of the world’s population with access to universal health care has continued to rise, albeit at a slower pace than before 2015. But as a result of the COVID-19 pandemic, wars and other factors, global life expectancy is now falling for the first time in decades. Polio has reappeared, after it was on the verge of being eradicated. Malaria cases have risen since 2016.

It is unacceptable that more efforts are not made, but more can be done. We are in nature – and journals via Nature Portfolio – remain determined to play whatever small role we can in achieving all the Sustainable Development Goals. Objectives are an important focus for our publisher, Springer Nature, as well.

Synthetic data can benefit medical research, but the risks must be acknowledged

Today we witness the publication of a new magazine Nature healthwith the mission of “bridging the implementation gap from health research to policy and practice.” The first editorial of the magazine¹ It states: “We will prioritize research with real-world impact, especially when conducted in resource-limited settings, whether in low- and middle-income countries or in disadvantaged communities elsewhere.”

The launch of the journal – and the challenges the publication seeks to help solve – comes at a time when scientific knowledge in medicine and healthcare is advancing at the fastest pace in human history. From gene editing to 3D bioprinting, from high-resolution imaging to robotic surgery, doctors have tools that could revolutionize the diagnosis, prevention and treatment of disease. But there are stark disparities in health outcomes and life expectancy between people with high and low incomes. In low- and middle-income countries in particular, routine public health services such as clean water and sanitation are beyond the reach of billions². There is an “unprecedented crisis” in global health financing, Tedros Adhanom Ghebreyesus, director-general of the World Health Organization, wrote in an article. Nature healthFirst issue³. In 2025, international funding for health in Africa alone is expected to fall below US$40 billion, from US$80 billion in 2021, as richer countries cut health aid budgets, according to a news article.⁴.

Tedros adds that the current crisis represents an opportunity for countries to reduce, if not eliminate, their dependence on foreign aid. In the process of building health systems, all countries must harness the talents of their researchers. The first group of Callistos Juma Science Leadership Fellows, a group of scientists based in Africa and supported by the Gates Foundation in Seattle, Washington, proposed focused actions in six thematic areas. One of their most important recommendations, detailed in the commentary article, is for the private sector to lead research and development⁵.

This is a welcome call to action from the business community, not only to fill the funding gaps left by departing international donors, but also to provide more longer-term support to their countries’ public sectors. It provides an opportunity for business leaders to recognize the necessity of a strong and effective public sector – to support research, innovation and regulation – and that they bear some responsibility for helping to achieve this.

Companies are also rapidly developing artificial intelligence for the healthcare sector. In a perspective article also published in Nature healthIn the new edition, the authors highlight some case studies in which the application of large language models (LLMs) improves diagnosis and supports decision-making in low- and middle-income countries.⁶.

As we reported in A nature In an editorial last year, AI models are increasingly trained on “synthetic” data generated by algorithms⁷. The promise of these data for low- and middle-income countries is that they should be able to mimic the statistical properties of real-world data, which are rare, difficult or expensive – and do not need as much ethical review.

However, as the authors of the Perspective article point out, low- and middle-income countries face significant barriers to wider adoption of LLM programs in this field, including limited digital infrastructure and the absence of relevant laws and regulations. The researchers wrote that there is an “urgent need” for evaluation studies, such as those that use randomized controlled trials to independently verify the claims made for MBAs, as well as economic analyzes that can evaluate any cost-effectiveness claims.

in naturein Nature health Across Nature Portfolio magazines, we remain committed to playing our part in creating a world where everyone can experience health and well-being.