How the Gates Foundation cut childhood deaths in half in two decades

Executive overview

Ten million children died annually from preventable diseases in 2000. Markets had no incentive to fund cures for diseases only poor people get, and the data infrastructure to even measure the problem barely existed.

The Gates Foundation built a playbook around four moves: fund independent measurement institutions, solve the delivery problem before scaling new vaccines, partner with willing governments rather than bypass them, and use grand engineering challenges to attract talent to neglected problems.

The core insight: you can't just invent a better vaccine — you have to fix the data, the delivery system, and the political will at the same time.

The data problem nobody was solving

• Causes of childhood death were routinely unrecorded or listed as vague symptoms like "diarrhea" — not root causes.
• Rotavirus caused over half of diarrheal deaths, pneumococcus caused ~40% of pneumonia deaths, but the breakdown was unknown.
• Autopsies — the standard tool for determining cause of death — are expensive and culturally unacceptable in many developing countries.
• The foundation funded minimally invasive autopsies (two lung samples, two gut samples, no visible disfigurement) and secured consent from 80% of bereaved mothers.
• This led to the creation of CHAMPS (Child Health and Mortality Prevention Surveillance) as an independent data organisation.
• WHO couldn't publish honest country-by-country comparisons — its member states were also its political managers.
• The foundation funded the Institute for Health Metrics and Evaluation at the University of Washington in 2007, led by Chris Murray, to provide independent global health data.

The delivery problem Bill didn't see coming

• Bill's initial assumption: find missing vaccines and drugs, fund their development, done.
• The real gap: rich-world vaccines already existed for diseases like rotavirus — but weren't reaching the children most at risk.
• Primary healthcare systems in the poorest countries were severely degraded: salaries unpaid, funds misallocated or stolen.
• The foundation spent years convincing itself that someone else would fix the financing and delivery problem — eventually concluded no one would.
• Pivoting to delivery was essential; inventing vaccines that couldn't be distributed would have been "almost Pyrrhic."

Working with governments

• Governments in poor countries must ultimately own and fund health systems — no philanthropic workaround is sustainable long-term.
• Ethiopia became the model partner: the government allocated funds correctly, installed measurement systems, and accepted bad data rather than rejecting it.
• Nigeria was the harder case: half the world's polio cases concentrated there, with vaccine access especially poor in the north.
• A parallel campaign used 200,000 volunteers to immunise 45 million Nigerian children, circumventing the government where necessary.
• Nigeria has recorded no new polio cases for three years, but sustaining coverage in northern states requires Bill personally running multiple governor conferences per year.
• Key lesson: find willing governments, highlight exemplars (Tanzania, Rwanda, Senegal achieved child mortality rates a third lower than peers at the same GDP), and use public accountability to drive change.

Scaling lessons from Microsoft vs. the foundation

• Geographic scaling differs from engineering scaling: geography assumes 90% of the approach is consistent, with 10% local variation; engineering scaling means adding parallel product groups.
• The foundation replicated Microsoft's engineering model — separate programme groups for malaria, tuberculosis, HIV — but underestimated the need for horizontal functions.
• Drug creation requires shared expertise in manufacturing and regulatory approval across all science groups; the foundation now dedicates over 30% of headcount and budget to these horizontal capabilities (vs. ~5% at Microsoft).
• Co-founding the foundation with Melinda required different dynamics than co-founding Microsoft with Paul Allen: equal voice rather than a managed hierarchy.

The reinvent-the-toilet challenge

• Sewer-based sanitation requires upfront infrastructure costs that price out low-income cities — trickle-down doesn't work for systems with this cost structure.
• A billion people live without adequate sanitation; poor sanitation kills one million children a year.
• In 2011, the foundation launched the Reinvent the Toilet Challenge, funding university prototypes and corporate partnerships.
• Concepts include solar-powered toilets, worm-based waste processing, and nano-membrane toilets that burn waste to ash.
• The target: a self-contained unit costing five cents per day that matches the performance of a flush toilet.
• The challenge was modelled on Hilbert's unsolved problems in mathematics — create visibility and funding for a neglected field to attract top talent.

The 50% reduction in childhood deaths

• From 2000 to the mid-2020s, annual childhood deaths fell from 10 million to just over 5 million.
• This year, 4.5 million children who would have died before age five are alive.
• The foundation was the largest philanthropic contributor to the UN Millennium Development Goals, contributing $17.9 billion.
• Governments contributed hundreds of billions more — the foundation's role was to move them, not replace them.
• Future inflection points already in view: HIV and TB vaccines expected within 10–15 years, making current delivery-system investment doubly important.