Theoretical Basis of the Test-Negative Study Design for Assessment of Influenza Vaccine Effectiveness.

摘要: Influenza viruses undergo frequent antigenic changes. As a result, the circulating change within and between seasons, composition of influenza vaccine is updated annually. Thus, estimation vaccine's effectiveness not constant across seasons. In order to provide annual estimates effectiveness, health departments have increasingly adopted "test-negative design," using enhanced data from routine surveillance systems. this design, patients presenting participating general practitioners with influenza-like illness are swabbed for laboratory testing; those testing positive virus defined as cases, negative form comparison group. Data on patients' vaccination histories confounder profiles also collected. Vaccine estimated odds ratio comparing among vaccinated unvaccinated patients, adjusting confounders. The test-negative design purported reduce bias associated confounding by health-care-seeking behavior misclassification cases. paper, we use directed acyclic graphs characterize potential biases in studies design. We show how can avoid or minimize where may be introduced particular study variations.