County-level measles risk is no longer staying put
Measles is reasserting itself in places where elimination once seemed durable, and the new pattern is different from the old one. Across U.S. county-level case and vaccine coverage data from 2011 through 2025, epidemiologic behavior shifted in recent years in a way that points to re-emergence rather than isolated flare-ups. Early in the study window, outbreaks tended to remain locally contained and only weakly tracked with county vaccination levels. In the later years, especially 2021 through 2025, lower vaccine coverage lined up much more tightly with larger outbreaks, while spread into neighboring counties became more common.
A single county with underimmunized pockets no longer explains the whole story.
That matters because measles is among the most transmissible human viruses, and once transmission gains enough foothold, local clustering can spill outward into adjacent communities, schools, and social networks. The study’s central message is not merely that vaccination gaps are dangerous, but that the geography of danger is changing. Transmission appears to be moving beyond small, socially clustered communities and becoming more sustained across the broader population. In practical terms, that means county-by-county risk is starting to behave like a connected network rather than a set of isolated nodes.
Why vaccine coverage began to matter more after 2021
The first decade of the analysis, spanning 2011 to 2020, showed outbreaks that were only weakly associated with county-level vaccine coverage. Many of those episodes stayed confined to single counties, suggesting that local introductions could ignite brief chains of transmission without generating large regional spread. The later period told a different story. From 2021 to 2025, county vaccine coverage became strongly associated with outbreak size, and spillover into neighboring counties increased. In epidemiologic terms, the signal moved from sparse, localized ignition to a more interconnected transmission pattern.
That shift is consistent with waning population immunity in the setting of declining immunization rates. Measles does not need much help once herd protection erodes; a few vulnerable communities can create the conditions for a larger outbreak, especially when movement between counties provides repeated opportunities for seeding. The paper’s framing suggests that the virus is increasingly finding enough susceptibility across wider swaths of the population to sustain itself beyond a single enclave. The biology is unforgiving: when enough susceptible hosts accumulate, the epidemic curve steepens quickly.
The analysis does not just describe where outbreaks occurred. It also links those changes to how well future outbreaks can be anticipated from county-level data. That predictive component is where the work becomes especially relevant for public health planning.
Model performance favored spatial targeting over simple coverage thresholds
Investigators built a spatial statistical model to predict outbreak risk and size in all U.S. counties and validated it over 2021 through 2025. The model identified counties with at least 10 cases with a sensitivity of 95%, with a year-to-year range of 75% to 100%, and a specificity of 83%, ranging from 64% to 98%. Those are strong performance characteristics for a geographically complex infectious disease problem, especially one driven by movement, clustering, and heterogeneous vaccine uptake.
Mechanistic models that relied on county vaccine coverage alone performed less well, particularly when the task was to forecast larger outbreaks. In other words, vaccine coverage remains essential, but it is not enough by itself to capture where risk is concentrating. Spatial structure adds information that raw coverage totals miss. Counties do not exist in isolation, and measles transmission respects that reality.
The study then asked a policy-oriented question: where would interventions have the greatest effect? Prioritizing the highest-risk counties identified by the model, such as the top decile of U.S. counties, outperformed strategies focused on entire states or on low coverage alone. That result is clinically and operationally important because public health resources are finite. A county-level strategy can focus outreach, vaccination campaigns, and surveillance where they are most likely to interrupt chains of transmission before they widen.
Modest coverage gains produced outsized national benefit
One of the most striking findings came from the simulation work. Raising vaccine coverage by only 5 percentage points in counties within the top decile of model-predicted risk reduced long-term measles risk nationwide by more than 90%. That is a large effect from a relatively modest increase in local coverage, and it reflects how transmission accelerates once the highest-risk counties are strengthened as barriers against spread.
The result also explains why targeted prevention can outperform broader but less precise strategies. Measles transmission is highly sensitive to small changes in susceptibility, especially in places that are already positioned to seed spread into surrounding counties. Boosting immunity in the counties most likely to function as launch points disrupts both local outbreaks and the downstream cascade into neighboring regions. The intervention is not simply about protecting one county’s residents; it is about cutting off transmission pathways that radiate outward through a connected network.
This is the kind of finding that changes how surveillance should be organized. Instead of treating vaccination coverage as a static threshold, the study argues for dynamic risk mapping that incorporates geography, adjacency, and outbreak history. County-level risk prediction becomes a tool for deciding where to intensify immunization, where to watch more closely for early cases, and where to prepare for secondary spread.
The preprint’s limits are straightforward. It remains unreviewed, and its projections depend on the quality of county-level case and vaccine data, along with assumptions built into the spatial model. Even so, the trajectory it describes is difficult to ignore: measles is reappearing in a more connected and more geographically diffuse form, and the counties most likely to drive that spread can be identified with useful accuracy. The next question is whether public health systems can turn that predictive signal into faster, more precise vaccination campaigns before the next cluster widens.