Welcome to the website of Influenza Vaccine Effectiveness Estimation Research Group (FluVEE) at the Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, Atlanta, GA, U.S.A!

The research described in this website was funded by the NIH-NIAID grant R01AI110474: ‘Study Designs for Estimating the Effectiveness of Vaccination against Influenza’. The Principle Investigator was Michael Haber, PhD.

Background

Seasonal and pandemic influenza outbreaks cause significant morbidity and mortality in the U.S.A. and worldwide, and vaccination is the most effective method to reduce the number of cases of the disease. Accurate and precise assessment of the effectiveness of influenza vaccination is important for the following reasons: (a) Understanding the relationship between antigenic match and mismatch and vaccine effectiveness, (b) Assessing the ongoing impact of vaccination efforts in the setting of antigenic drift and periodic vaccine reformulation, (c) Evaluation of vaccination programs and strategies in terms of individual and population-wide benefits, and (d) Identifying risk factors for vaccine failure to assist in determining strategies to improve effectiveness in such groups (e.g., higher doses). However, accurate estimation of influenza vaccination effectiveness (VE) is challenging for the following reasons: (a) Predominant influenza virus types and strains change from one season to the next, necessitating a new vaccine with different strains each season. As a result, VE must be re-estimated in every season. (b) Influenza vaccination is now recommended for almost every person above 6 months of age in the U.S.A, therefore randomized controlled clinical trials are unethical and one must resort to observational studies. (c) Confounding and other sources of bias are a major problem in observational VE studies, and some of the confounders are difficult to measure. (d) It is not easy to find all or most influenza patients in a given community, as symptoms are frequently mild and many patients do not seek medical care to alleviate them. (e) Symptoms of influenza are non-specific, hence many patients who develop an acute respiratory illness (ARI) are not infected with the influenza virus. (f) Special expensive laboratory tests are required to confirm influenza infection, and these tests are not 100% sensitive and specific. For all these reasons, observational studies to estimate influenza VE must be designed very carefully to minimize the various sources of bias. Therefore, it is essential to evaluate and compare characteristics (bias and precision) of existing observational study designs and develop new study designs producing more robust VE estimates.

Objective

The main objective of this research project focused on evaluation and comparison of the bias of influenza VE estimates derived from different observational study designs. We considered different types of cohort and case-control studies. In particular, we carefully examined the test-negative (TN) design, were ARI patients who seek medical care and test negative for the presence of the influenza virus are used as controls. This relatively new design (Orenstein et al, 2007) has rapidly become the most commonly used study design for estimating influenza VE. In addition, we developed a new maximum likelihood method for estimation of influenza VE from household studies. This method provides separate estimates of VE against transmission of influenza from an infected household member and from the community at large.