Background: The 2025 US measles resurgence was concentrated in communities with low vaccination coverage. Aggregate national and state measles-mumps-rubella (MMR) coverage can mask local susceptibility: pockets of unvaccinated children persist where average coverage appears adequate. School-entry nonmedical exemptions (NMEs) are collected routinely and may capture vaccination behavior and clustering of susceptible children. We examined whether county-level kindergarten NME rates were associated with reporting at least 3 measles cases in 2025, independently of average county MMR coverage.
Methods: Cross-sectional ecological study of US counties. Confirmed cases reported January 1 to December 31, 2025, came from the Johns Hopkins University US Measles Tracker. Kindergarten MMR coverage and NME rates (school year 2023-2024 or, if unavailable, 2022- 2023) came from public state reports and published county-level compilations. The prespecified outbreak proxy was 3 or more versus 0-2 reported cases per county, a count threshold without verified epidemiologic linkage. A secondary, exploratory outcome was total case burden among counties with 3 or more cases. Firth penalized logistic regression estimated odds ratios (ORs) for occurrence; negative binomial regression with a log county-population offset and state-clustered robust standard errors estimated incidence rate ratios (IRRs) for burden. Models adjusted for percentage foreign-born, metropolitan status, population density, percentage uninsured, college education, 2024 presidential vote share, and primary care physician supply; estimates are per 1-percentage-point increase. Sensitivity analyses varied the threshold and adjustment set and examined spatial dependence, international airport proximity, state exclusions, and modified Poisson estimation. Only aggregate, publicly available data were analyzed; the study was exempt from institutional review.
Results: The sample comprised 1914 counties in 30 states (237 million residents, 71.6% of the US population), capturing 1821 of 2214 reported 2025 cases (82.2%); 66 counties (3.4%) met the outbreak proxy. Mean (SD) MMR coverage was 91.1% (7.2%) and mean NME rate 4.58% (4.37%). In the joint model, each 1-percentage-point higher NME rate was associated with higher odds of meeting the proxy (OR 1.16; 95% CI 1.09-1.24; P<.001), whereas average MMR coverage was not (OR 1.01; 95% CI 0.96 1.07; P=.78). The exemption association persisted across thresholds of 1 to 10 or more cases (ORs 1.12-1.17), with reduced adjustment (OR 1.15), airport-proximity adjustment (OR 1.16), and modified Poisson estimation (risk ratio 1.14; 95% CI 1.06-1.23). Residual spatial autocorrelation was present (Moran I 0.0758; P=.001), and adding neighboring counties' exemption rates attenuated the OR to 1.12 (95% CI 1.03-1.20). In the exploratory burden model, the IRR was 1.08 (95% CI 1.00-1.16; P=.046), with confidence intervals crossing the null in several specifications.
Conclusions: Across US counties in 2025, kindergarten nonmedical exemption rates were associated with clustered measles case reports independently of average MMR coverage, which carried little additional information once exemptions were considered. Estimates are ecological, not individual-level or causal; the count proxy did not establish epidemiologic linkage, and exposure definitions varied across states. Routinely collected school-entry exemption data may nonetheless complement local knowledge when pediatricians, school health programs, and health departments prioritize school-level review and outbreak preparedness.

