Rotavirus Gastroenteritis Infection Among Children Vaccinated and Unvaccinated With Rotavirus Vaccine in Southern China

Key Points Question What is the effect of the Lanzhou lamb rotavirus vaccination? Findings In this cross-sectional, ecological study of 33 407 patients with rotavirus gastroenteritis from 2007 to 2015 seasons in China, vaccination was associated with a 4-month increase in median age at onset and with delays in onset, peak, and cessation of incidence. The incidence rate ratio among children younger than 4 years and among children ineligible for vaccination decreased as citywide vaccination coverage increased, and the adjusted odds ratio for rotavirus gastroenteritis among unvaccinated infants decreased in areas with higher vaccination coverage. Meaning The Lanzhou lamb rotavirus vaccination can provide population health benefits in preventing rotavirus gastroenteritis, including herd effects.


eAppendix 1. How to Choose the Most Significant Cumulative Coverage
Using district-wide electronic vaccination data for April 2004 through March 2016, we obtained the number of administered LLR vaccine doses during the prior 3, 6, 9, 12, 24 and 36 months and calculated monthly cumulative coverage from May 2007 to April 2016 among those < 4 years of age. We did not include vaccination during the current month as the rotavirus vaccine typically begins to provide protection 2 weeks after immunization.
To determine which of the prior vaccination periods, 3, 6, 9, 12, 24 and 36 months, was most significantly associated with changes in RV-GE incidence, we fit separate models for each of these periods. For use in subsequent models, we chose the prior period with the most significant Incidence Rate Ratio (IRR).

eAppendix 3. Vaccination Benefits in Different Populations
Vaccination benefits were seen across age groups and sex, but were most pronounced in urban districts and in the 2013-15 seasons. There are two possible explanations for the difference between the rural and urban districts. First, inoculation techniques, including cold chain maintenance and selection of those eligible for vaccination, may vary between the urban and rural districts. Health providers in urban districts are thought to provide better inoculation services and thus rotavirus vaccination may be more reliable than in rural areas. As a consequence, the direct impact of vaccination may be more significant in urban districts 1 . Second, as urbanization increases in China 2 , population density increases in urban districts faster than in rural districts. In addition to higher urban vaccination coverage (data not shown), it is possible that social contacts at day care or community centers in urban districts provide more indirect protection to young children.
However, such an association has not been evaluated.

eAppendix 4. Studies Which Have Evaluated the Indirect Effects of RV Vaccination
As for natural infection, which generates serotype-specific immunity but also crossprotection against moderate-to-severe disease due to other serotypes, vaccination with a live attenuated vaccine potentially can reduce rotavirus transmission and induce herd protection 3,4 . For example, a dynamic transmission model projected the population-level impact of RV vaccination in children within national immunization programs in Europe. Using an assumption of 79% VE against RV-GE, the model predicted that, 5 years after implementation of a vaccination program with vaccination coverage rates of 70%, 90%, and 95%, herd protection would induce an additional reduction in RV-related GE incidence of 25%, 22%, and 20% respectively 5 . In the US, marked declines have been noted in GE in all age groups during peak rotavirus season; for example, data from the National Inpatient Sample during 2008 showed that hospitalizations due to caseunspecified GE deceased for populations up to 24 years of age 6 , and a 48% decline in the prevalence of rotavirus was observed in adults who had stool sent for bacterial stool culture from 2006-2007 to 2008-2010 7 . In the United States, Lopman et al. found a remarkable 71% reduction in hospitalizations coded as rotavirus and a 30% reduction in diarrhea of unspecified causes among older children who were not vaccinated and should have been immune due to early childhood infections 6,8 .     NOTE. Low mortality countries were defined as those in the lowest quartile of under-5 child mortality rates, with medium mortality countries as those in the second quartile, and high mortality as those countries in the highest two quartiles 9 .