| Dengue fever is a vector-borne disease currently demonstrating patterns of re-emergence. With approximately 2.5 billion people at risk in over 100 countries and 50-100 million annual infections, it has become the most rapidly spreading arbovirus.;Good understanding of the transmission dynamics of dengue will be essential to plan and target successful control strategies. Our results, from age-stratified serological surveys conducted in Rayong, Thailand; Recife, Brazil; and Chennai, India, provide some insight about the mechanisms that might be driving shifts observed in different settings. While our findings are consistent with the demographic transition as a major driver of the changes in the age distribution of dengue cases in Thailand, they suggest that the epidemiologic shift observed in Brazil might reflect the expected transition from re-emergence to hyperendemicity. Our results from the seroprevalence study conducted in Chennai argue against current endemic transmission of dengue in this setting, but suggest that all of the risk-factors are in place for transmission to become endemic.;We present current estimates of the basic reproductive number ( R¯0) and critical vaccination fraction ( V) for Recife, Brazil and Rayong, Thailand. According to our results, R¯0 ranges from 1.5 to 4.6 in Rayong and from 2.1 to 3.3 in Recife, translating to V ranging from 33% to 78% and from 52% to 70% respectively. We also present current serotype-specific estimates of R0 (R0 i) for Rayong, Thailand. While R02 seems to be significantly higher than that of other serotypes, implying that it would require a higher level of vaccination to control transmission, other sources of heterogeneity seem to be more important than differences between serotypes.;We compare estimates of the force of infection (lambda( t)) and R¯0 obtained using serological data vs. case data. We find that case data can provide valid mean estimates of lambda(t) and R¯ 0 even in complex multiserotype diseases like dengue. This allows utilization of the large amount of case data that is available nationally and regionally.;Overall, results of this dissertation hope to contribute to the understanding of recent shifts in the epidemiology of dengue and to refine the toolbox available to estimate lambda(t) and R0 when characterizing transmission dynamics in different settings. |
