“Better safe than sorry”: Behaviorally informed digital campaigns and their association with COVID-19 vaccine interest and take-up in Belize

Abstract

Abstract Background: Increasing vaccine take-up has been key to curb the COVID-19 pandemic globally. However, vaccine acceptance and take-up has been a challenge in some regions such as the Caribbean countries. This paper analyzes the performance of different behaviorally informed social media campaigns on people’s engagement with information about the vaccination process and vaccine take-up in Belize, a country where despite of enough vaccine supply take-up was very low. Methods: Using Belize’s Facebook platform (>230,000 users) and national administrative vaccination data at the district level, we conduct three types of analyses. First, we run linear regression models to analyze which of five behaviorally informed and sequentially displayed Facebook campaigns were better at predicting i) social media engagement with COVID-19 vaccination; and ii) COVID-19 vaccination uptake. Then, within one of the campaigns (e.g. the side-effects one) we conduct a randomized experiment to measure the effect on social media engagement of exposure to either positively (“Majority did not report discomfort”) or negatively framed messages, using words (“Few persons reported discomfort”) or numerical proportions (“3 out of 100 reported discomfort”). Social media engagement is measured by the number of people that clicked on the respective Facebook ad and visited the official vaccination site (“Clicks”) and the number of people that interacted with the ad through likes and different emojis (“Engagements”). Results: Campaigns highlighting COVID-19 vaccine safety best predicted vaccination uptake (second and booster doses) and social media engagement. Emphasizing the vaccine´s effectiveness was associated with a higher uptake of first doses. These results match previous research showing that safety was one of the main concerns among Belizean citizens and that efficacy beliefs were important to increase the likelihood of uptake among the unvaccinated. Finally, informational messages related to side effects that were framed positively (“Majority did not report discomfort”) and using words (“Few persons reported discomfort”) were better at generating "Clicks” compared to those negatively framed and using numbers (“3 out of 100 reported discomforts”). Conclusions: Highlighting COVID-19 vaccines´ safety and framing side-effect information positively (i.e., focusing on those who will not get them) with words might increase vaccination uptake and interest in low-uptake regions like the Caribbean.