Has Gavi lived up to its promise? Quasi-experimental evidence on country immunisation rates and child mortality

Abstract

INTRODUCTION: Gavi, the Vaccine Alliance, was set up in 2000 to improve access to vaccines for children living in the poorest countries. Funding has increased significantly over time, with Gavi disbursements reaching US $1.58 billion in 2015. We assess whether Gavi's funding programmes have indeed increased immunisation coverage in 51 recipient countries for two key vaccines for 12-23 month olds: combined diphtheria, pertussis and tetanus (DPT) and measles. Additionally, we look at effects on infant and child mortality. METHODS: Taking a difference-in-differences quasi-experimental approach to observational data, we estimate the impact of Gavi eligibility on immunisation coverage and mortality rates over time, using WHO/UNICEF figures covering 1995-2016. We control for economy size and population of each country as well as running a suite of robustness checks and sensitivity tests. RESULTS: We find large and significant positive effects from Gavi's funding programmes: on average a 12.02 percentage point increase in DPT immunisation coverage (95% CI 6.56 to 17.49) and an 8.81 percentage point increase in measles immunisation coverage (95% CI 3.58 to 14.04) over the period to 2016. Our estimates show Gavi support also induced 6.22 fewer infant deaths (95% CI -10.47 to -1.97) and 12.23 fewer under-five deaths (95% CI -19.66 to -4.79) per 1000 live births. CONCLUSION: Our findings provide evidence that Gavi has had a substantial impact on the fight against communicable diseases for improved population and child health in lower-income countries. In this case, the health policy to verticalise aid-specifically development assistance for health-via a specialised global fund has had positive outcomes.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 24 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.