Why high vaccination coverage doesn't always mean low measles risk
Clusters of unvaccinated children can leave communities vulnerable to measles even when vaccination coverage across the wider area appears high, new research suggests.
- 5 October 2026
- 6 min read
- by Linda Geddes
At a glance
- An analysis of vaccination data from more than 50,000 US schools suggests that broader county-level figures for measles immunisation coverage could mask pockets of susceptibility in individual schools and local communities.
- Researchers found that average transmission potential at school level had risen above the level needed to sustain outbreaks by 2022–2023, a shift that remained hidden when vaccination data was analysed at county level.
- More localised analysis could help public health authorities identify vulnerable communities and target vaccination efforts, with potential implications for how measles vaccination coverage is monitored in other countries.
As immunity against measles weakens in populations around the world, a new study suggests that monitoring vaccination coverage at a more local level could help identify growing pockets of vulnerability where outbreaks are most likely to take hold and spread.
Researchers analysed vaccination data from tens of thousands of US schools and found that broader county-level figures for measles immunisation coverage could mask substantial differences between individual schools and communities.
This could potentially explain why measles outbreaks have occurred in areas where overall vaccination coverage appeared relatively high.
The findings could have implications for how vaccination coverage is monitored, as declining vaccine uptake creates increasingly uneven pockets of susceptibility to measles and other vaccine-preventable diseases.
Dr Alexis Robert, an infectious disease modeller at the London School of Hygiene & Tropical Medicine, UK, who was not involved in the research, said the analysis demonstrated the importance of having access to accurate and detailed vaccination data:
“This is valid for a variety of settings, and even pathogens, but especially important for measles in places where transmission risk is concentrated in pockets of susceptibility, because coarse geographical units will give us a false impression of safety,” Robert said.
Can vaccination data reveal where measles outbreaks are most likely?
Measles is extremely contagious, which is why the threshold for herd immunity is so high. Around 95% of a community needs to have been vaccinated with two doses of measles vaccine to prevent the disease from continuing to spread if a case is introduced.
Yet vaccination coverage has fallen or stagnated in many countries in recent years, partly because of disruption to routine immunisation during the COVID-19 pandemic, compounded by growing vaccine hesitancy and misinformation. As immunity gaps have widened, measles has resurged in some countries that had previously achieved elimination status.
As in many countries, vaccination coverage in the US is typically reported for relatively broad administrative areas (usually states or counties), reflecting the way vaccination programmes are organised. This makes sense for many purposes, including allocating funding and planning vaccination campaigns, Bento said. However, it may not always be the best scale for assessing the risk of measles transmission.
Even before the recent surge in cases, Dr Ana Bento, a disease ecologist at Cornell University in Ithaca, US, and her colleagues had noticed a pattern of highly localised US measles outbreaks.
This made them wonder whether monitoring vaccination coverage at state or county level might conceal more localised clusters of susceptible individuals. “We wanted to test this hypothesis that perhaps there was an averaging problem,” Bento said.
What did school-level vaccination data reveal about measles risk?
To investigate, Bento and her colleagues compiled school-entry MMR vaccination records from more than 50,000 US schools, covering 13,000 school districts and 3,000 counties, between 2013 and 2025.
They used this data to estimate how the proportion of children susceptible to measles – and the potential for the virus to spread if a case was introduced – changed over time at school, school-district and county level.
Their analysis, published in Nature Medicine, revealed that by 2022–2023 average measles transmission potential at school level had risen above the epidemic threshold, meaning there was an increased risk of outbreaks if the virus was introduced.
Yet this shift remained hidden when the data was analysed at county level. The model estimated transmission potential, rather than predicting when outbreaks would occur.
Bento stressed that existing surveillance systems weren’t to blame. But while the recent large outbreaks had come as a surprise to some, the warning signs of increased risk had been there all along, she said: “We were just looking in the wrong place, at the wrong scale.”
The researchers also compared their estimates with seven locations in the US that experienced measles outbreaks between 2017 and 2026. All seven ranked among the top 10% for estimated transmission risk when analysed at school level, but only three were identified as high-risk at county level.
For instance, in South Carolina, where a large outbreak began in Spartanburg County in October 2025, the team identified several schools with extremely low vaccination coverage, including one with only 21.3% MMR vaccination coverage. County-level estimates flagged Spartanburg as at risk, but the school-level data showed which schools faced the greatest risk.
How could pockets of low measles immunity impact neighbouring populations?
In a separate analysis, the researchers found that such pockets of susceptibility could also increase the risk in neighbouring, better-vaccinated populations.
Some school districts cross county boundaries, so children from different counties mix at school. The modelling suggested that these connections could push transmission potential above the epidemic threshold even in areas with relatively high vaccination coverage.
Have you read?
“Even if you have a county or a school district that has very high coverage, if your next-door neighbour, county or school is at a much lower level, there is a risk of spillover,” Bento said.
What could these findings mean for measles surveillance worldwide?
The findings suggest that analysing vaccination coverage at a finer scale could help public health authorities identify vulnerable communities before measles arrives, allowing vaccination efforts to be targeted towards places where an imported case would be most likely to trigger an outbreak.
In low- and middle-income countries, the approach could help vaccination teams decide where to focus outreach and catch-up campaigns. This would need local vaccination and population data that also cover children who do not attend school.
Similar analyses could be used for other vaccine-preventable diseases, with adjustments for how each disease spreads and how well its vaccine protects, said Bento.
Whether the same approach would be useful elsewhere will depend partly on what data is available, Robert said. In England, for example, historical vaccination coverage is publicly available at local-authority level.
Although these areas are geographically smaller than US counties, they can contain hundreds of thousands of people and still conceal substantial differences in vaccine uptake. Analysing school-level data could therefore be useful, provided sufficiently detailed data was available.
However, there may be limitations. Catch-up vaccinations received years after their scheduled delivery may not appear in school entrance surveys, while adults accounted for around 30% of measles cases in England and 40% in the US in 2026. “Identifying pockets of susceptibility in adults is a major challenge due to population movement and infection history,” Robert said.
Nor is vaccine hesitancy the only reason why pockets of measles susceptibility may emerge. “Access [to vaccination] has also repeatedly shown to be a driver and is very likely to be socially and geographically clustered in vulnerable groups, which will lead to pockets of susceptibility,” Robert said.
Despite such challenges, Bento believes the approach could help to reveal similar pockets of susceptibility in other countries - and potentially for other diseases. "The principle is transferrable," she said, although the most useful scale for monitoring risk is likely to differ between countries.
Her team is now analysing measles vaccination data from Mexico, Canada, Brazil and parts of southern Europe to investigate whether looking at coverage at a finer scale reveals pockets of hidden vulnerability.
A separate study of Mexico’s 2025–2026 measles outbreak by researchers at the University of Guadalajara, Mexico, suggested that cases clustered in under-vaccinated communities, highlighting the importance of identifying and reaching populations missed by routine immunisation, Bento said.
