A statistical model to estimate the impact of a hepatitis A vaccination programme

Abstract A program of routine hepatitis A + B vaccination in preadolescents was introduced in 1998 in Catalonia, a region situated in the northeast of Spain. The objective of this study was to quantify the reduction in the incidence of hepatitis A in order to differentiate the natural reduction of the incidence of hepatitis A from that produced due to the vaccination programme and to predict the evolution of the disease in forthcoming years. A generalized linear model (GLM) using negative binomial regression was used to estimate the incidence rates of hepatitis A in Catalonia by year, age group and vaccination. Introduction of the vaccine reduced cases by 5.5 by year ( p -value < 0.001), but there was a significant interaction between the year of report and vaccination that smoothed this reduction ( p -value < 0.001). The reduction was not equal in all age groups, being greater in the 12–18 years age group, which fell from a mean rate of 8.15 per 100,000 person/years in the pre-vaccination period (1992–1998) to 1.4 in the vaccination period (1999–2005). The model predicts the evolution accurately for the group of vaccinated subjects. Negative binomial regression is more appropriate than Poisson regression when observed variance exceeds the observed mean (overdispersed count data), can cause a variable apparently contribute more on the model of what really makes it.