Vaccine coverage

Vaccines are a powerful tool for saving lives and improving health, but their power can be diluted by gaps in coverage. We study where routine childhood immunizations are being used – and where they aren’t – to understand and address these gaps. 

Photo by Heather Hazzan for SELF magazine, Flickr.

Over 99% of measles cases and deaths occur in low- and middle-income countries; all of these could be prevented by vaccination.
Only 11 out of 204 countries and territories reached at least 90% coverage for 9 major vaccines in 2019.
17 million children missed routine vaccinations in 2020 due to the COVID-19 pandemic.
236,000 deaths in 2019 were caused by meningitis, a vaccine-preventable disease.

How can I download vaccine coverage data?

Our vaccine coverage data are the latest estimates available and may differ from previous estimates published in journals. After entering your email, you will be redirected to a Nextcloud folder where the data are free to download.

Which data sources did you use?

Our data sources include both survey data and administrative data, recorded in the Nextcloud folder and our data catalog, the Global Health Data Exchange (GHDx).

  • Survey data includes the Demographics and Health Survey and others.
  • Administrative data refers to official data reported by countries to WHO.

How do you address lower-quality data or gaps in data? 

We check for data quality issues before running our models both by analyzing the data and by looking at external sources of information about the data. Where possible, we use statistical techniques to quantify and adjust for biases in the data – for instance, we adjust for the difference between survey data and official country-reported data. 

In some cases, when there are significant concerns about data quality, those sources may be excluded from the models. Our models use trends in time and predictive covariates to produce estimates of vaccine coverage for locations and years in which no data are available. 

We also produce estimates of the uncertainty for our predictions – when data are sparse or conflicting, our estimates are more uncertain.

While these models provide valuable insights into vaccination coverage in settings where data are missing or of lower quality, they aren’t a replacement for high-quality data. We support the broader efforts of the immunization community to strengthen data quality. 

How can I contribute data or share feedback on estimates?

To learn how you can contribute to this project, email us at [email protected]