Methods for Evaluating Vaccine Efficacy

有效估计疫苗接种的直接、间接、总和总体影响对于监管和预防接种至关重要

Valid estimates of the direct, indirect, total, and overall effects of vaccination are crucial for regulatory and policy decisions. New vaccines pose new challenges for novel statistical methods and study designs. We propose an ambitious research agenda in a collection of five main aims to develop cutting-edge methods for currently available and future vaccines. Our methods are motivated by particular vaccines, but in most cases are applicable to vaccines with similar characteristics. In Aim 1, we will develop novel cluster- randomized designs for evaluating the effects of new dengue vaccines and for evaluating the indirect effects of malaria transmission blocking candidates. In Aim 2, we will develop novel statistical methods for estimating efficacy of multivalent vaccines under heterogeneity and analyze trials of current dengue vaccine candidates. In Aim 3, we will develop statistical methods for evaluating and comparing general surrogates of protection from a collection of trials. We will use these methods to explore candidate general surrogates of rotavirus vaccines, We will develop methods for validating the immune correlate of protection of the new meningococcal A vaccine. In Aim 4, we will develop methods to estimate vaccine efficacy for infectiousness from a study with two levels of clustering and estimate this quantity for influenza vaccine. We will develop a new measure of risk called potential exposure and delineate its characteristics. In Aim 5, we will develop two novel designs for evaluating new candidate tuberculosis vaccines. The first is a randomized case-referent design that will be a cost-effective method to evaluate new tuberculosis vaccines in large populations. The second is a design that takes into account prior sensitization and will allow differentiation of the effects of vaccine on disease due to primary infection or reactivation. Our novel methods of design and analysis will have direct consequences for implementation of actual studies. The results will contribute directly to regulatory and public health policy decisions, as well as be used "as inputs in computer simulations to explore optimal vaccination strategies. Our research will take place in partnership with individuals and organizations directly involved with vaccine studies. RELEVANCE (See instructions): Accurate assessment of direct, indirect, total, and overall effects of vaccines is crucial for regulatory and policy decisions. The research proposed here would have direct consequences for implementing actual studies to evaluate effects of currently available and future vaccines, and thus contribute directly to public health. The results will also be useful for computer simulations of optimal vaccination strategies.