Communication about vaccine efficacy and COVID-19 vaccine choice: Evidence from a survey experiment in the United States.

Abstract

While mass vaccination campaigns against COVID-19 have inoculated almost 200 million Americans and billions more worldwide, significant pockets of vaccine hesitancy remain. Research has firmly established that vaccine efficacy is an important driver of public vaccine acceptance and choice. However, current vaccines offer widely varying levels of protection against different adverse health outcomes of COVID-19. This study employs an experiment embedded on a survey of 1,194 US adults in June 2021 to examine how communications about vaccine efficacy affect vaccine choice. The experiment manipulated how vaccine efficacy was defined across four treatments: (1) protection against symptomatic infection; (2) protection against severe illness; (3) protection against hospitalization/death; (4) efficacy data on all three metrics. The control group received no efficacy information. Subjects were asked to choose between a pair of vaccines-a one-dose viral vector vaccine or two-dose mRNA vaccine-whose efficacy data varied across the four experimental treatment groups. Efficacy data for each vaccine on each dimension were adapted from clinical trial data on the Johnson & Johnson/Janssen and Pfizer/BioNTech vaccines. Among all respondents, only modest preference gaps between the two vaccines emerged in the control group and when the two vaccines' roughly equivalent efficacy data against hospitalization and death were reported. Strong preferences for a two-dose mRNA vaccine emerged in treatments where its higher efficacy against symptomatic or severe illness was reported, as well as in the treatment where data on all three efficacy criteria were reported. Unvaccinated respondents preferred a one-dose viral vector vaccine when only efficacy data against hospitalization or death was presented. Black and Latino respondents were significantly more likely to choose the one-shot viral vector vaccine in the combined efficacy treatment than were whites. Results speak to the importance of understanding how communications about vaccine efficacy affect public preferences in an era of increasing uncertainty about efficacy against variants.