Methods: We used school-based records of 505,015 kindergarteners to describe the percent of undervaccinated students with personal belief exemptions or conditional enrollment during the 2010-2011 school year, and conducted an ecologic analysis of the effect of county-level pertussis vaccine coverage on county-level pertussis incidence rates from Jan – Dec 2010. County-level vaccine coverage was estimated as the average of the percentage of vaccinated kindergarteners at each school in the county. County-level incidence rates were defined as the number of pertussis cases per 100,000 population as reported by the California Department of Public Health. Quartile distributions were used to differentiate between high and low levels of vaccine coverage and incidence rates. Data on covariates were collected from the 2010 US Census. Logistic regression was used to estimate crude and adjusted odds ratios of the effect of low vaccine coverage compared to high vaccine coverage on the incidence of pertussis.
Results: Among 46,670 (9%) undervaccinated students, 76% had conditional enrollment and 24% had personal belief exemptions. Using a cut-point of 86.5% to define low coverage, 14 (24%) counties were classified as having low vaccine coverage and 44 (76%) counties were classified as having high vaccine coverage. The 15 (26%) counties with incidence rates ³34.6 per 100,000 were classified as having high incidence rates. After adjusting for the county-level percentage of the population <5 years-old and living in poverty, counties with low vaccine coverage had 8.4 times the odds of having high county-level pertussis incidence compared with counties with high vaccine coverage (aOR 8.4; 95% CI 1.1, 61.3).
Conclusion: Low county-level pertussis vaccine coverage in California was associated with increased risk of high county-level pertussis incidence rates during the 2010 epidemic.