Geospatial analysis of nonmedical vaccine exemptions and pertussis outbreaks in the United States

Abstract

Because of increased numbers of recorded pertussis cases in the United States, this study sought to understand the role of nonmedical vaccine exemptions and waning immunity may have had on the resurgence of pertussis in the United States at the community level. We used geospatial scan statistics, SaTScan, version 9.4, to analyze nonmedical vaccine exemptions of children entering kindergarten in 2011 and 2012 and reported pertussis cases in 2012 for children in age groups 5 years and younger and 10 to 14 years. Eight statistically significant clusters of nonmedical vaccine exemptions in kindergarteners and 11 statistically significant clusters of pertussis cases in children and adolescents were identified and geospatially linked. Forty-five percent of the counties in the study had high rates of nonmedical vaccine exemptions. The proportion of kindergarteners with nonmedical vaccine exemptions was 2.8 times larger in the identified exemption clusters. In addition, 31 counties had geographic clusters of high rates of pertussis in children ages 10 to 14 years old, consistent with waning immunity. Our findings are consistent with the view that geographic clusters of nonmedical vaccine exemptions and waning immunity may have been factors contributing to community-level pertussis outbreaks.

What if protective immunity is antigen—driven and not due to so‐called “memory” B and T cells?

“… There is ample epidemiological evidence for the relatively rapid decline of protective (neutralizing) high affinity antibodies in the controlled absence of antigen (eg, currently observed decline of protective immunity by Rubella or acellular pertussis vaccines) There is a clear correlation between persistence of antigen in extra lymphatic sites and neutralizing antibody or T cell‐dependent protection: Herpes viruses, TB (vs shorter‐lived BCG), leprosy or LCMV and many chronically persisting parasitic infections (eg, leishmaniosis, malaria etc.…”

“…This has been clearly shown by the superior herd immunity achieved with the Sabin live vaccine vs the Salk inactivated vaccine against polioviruses, reviewed in . This also becomes now suddenly apparent with new forms of pertussis/whooping cough or rubella infections in very young newborns at one or to 3 months of life or even before birth. These are first wake‐up calls to evaluate the usually popular and easy to understand explanation of immunological memory to reflect an altered‐system behavior.…”

“…This problem of course is a consequence of the efficiency of both hygiene and of vaccination. Because they have continually reduced, endemically/epidemiologically recirculating levels of infection and therefore formally occurring re‐exposure by these circulating below clinical‐level infections cannot maintain individual and herd immunity any longer . This has been clearly shown by the superior herd immunity achieved with the Sabin live vaccine vs the Salk inactivated vaccine against polioviruses, reviewed in .…”

Getting personal: how vaccination exemptions shape herd immunity

“…To analyze the spatial heterogeneity at the county level, and how the heterogeneity varied following policy changes, we computed Moran’s I [25] for each state and year. We performed a spatial clustering analysis for each state before and after the change in policy using SatScan [26] with the Bernoulli model [27, 28], as this model is adapted to our situation where individuals have or do not have an exemption. This method detects clusters of counties with high exemption rates relative to the rest of counties in a state; the mean rate of “high” clusters thus varies between states and between clusters.…”

The landscape of childhood vaccine exemptions in the United States

“…However, this spatial scale is largely inadequate as both exemptions and vaccine preventable childhood disease cases tend to cluster at smaller spatial scales 11 . Many prior studies have been performed at smaller scales, from the county to the school levels 3,[12][13][14][15] . However, the datasets used in these studies tend to be limited both in time and in terms of the spatial area they cover.…”