The Use of Quasi-experimental Designs for Vaccine Evaluation

• Published in: Clinical infectious diseases Vol. 68; no. 10; pp. 1769 - 1776
• Main Authors: Lopez Bernal, James A, Andrews, Nick, Amirthalingam, Gayatri
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 02.05.2019
• Subjects: Biomedical Research; Humans; Immunization Programs - standards; Immunization Programs - statistics & numerical data; Interrupted Time Series Analysis; Non-Randomized Controlled Trials as Topic - methods; Research Design; Vaccination; Vaccines - standards; evaluation; interrupted time series; regression discontinuity; quasi-experimental designs; immunizations
• DOI: 10.1093/cid/ciy906
• ISSN: 1058-4838

Abstract Randomized, controlled trials are not always possible to evaluate interventions targeting infectious disease. This is frequently the case when evaluating the population-level impact of vaccines or when evaluating interventions aiming to increase vaccine uptake. Under such circumstances, an array of quasi-experimental designs is increasingly being used to evaluate the effects of vaccines on a wide range of morbidity and health service outcomes. These studies can provide valuable information on the impact of vaccination programs and other related interventions in real-world settings. Nevertheless, not all quasi-experimental designs are equal, and it is important that authors and readers are aware of their relative strengths and potential sources of bias. In this paper, we discuss what a quasi-experimental design is, when they might be used for vaccine evaluation, their strengths and limitations, and examples of their application. This article reviews and critically appraises the use of quasi-experimental designs for the evaluation of vaccines. We highlight where quasi-experimental designs can be applied, provide examples, and discuss the relative strengths and limitations of each design.