A new study published on the preprint server medRxiv* uses the popular social media platform Twitter to examine public attitudes towards the vaccines developed to fight the coronavirus disease 2019 (COVID-19) in the United States of America. The researchers found that the nature of the discussion of the COVID-19 vaccines differed with the stage of development of the vaccine and the phase of the pandemic, which, in turn, varied with the geographic location and demographic characteristics of the Twitter user.
Study: Public Perception of COVID-19 Vaccines on Twitter in the United States. Image Credit: rafapress / Shutterstock.com
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
Background
COVID-19, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has caused the deaths of over 4.9 million around the world, over 700,000 of which resided in the United States. The worst affected by COVID-19 mortality included the frail, elderly, Blacks, and Hispanics. In order to arrest viral transmission, vaccine development was accelerated, with high priority being accorded to this project.
The vaccine development process culminated in the emergency use authorization of COVID-19 vaccines at the end of 2020, within a year when the COVID-19 pandemic began. Scientists have estimated that at least 70% of the population needs to be vaccinated to reach population immunity; however, the degree of acceptance of the vaccines must be high for this goal to be reached.
Unfortunately, the chances that a given person would take the vaccine decreased from April to December 2020, from about 75% to about 55%. The advantages and limitations of the vaccine development process, potential dangers of the vaccine, as well as alleged hoaxes behind the propagation of vaccine information led to extensive and detailed discussions on the vaccine.
The current study discusses the public debate about the COVID-19 vaccine in an attempt to capture popular opinion on this matter. Social media provides a wealth of information that accurately conveys public opinions on a variety of topics.
The researchers used DeepFace, a computer vision technique, to look at the demographic factors of Twitter users who tweeted about COVID-19 vaccines. First, the total number of tweets from March 5, 2020, to January 31, 2021, was calculated.
The levels of these tweets remained stable, except for two peaks on November 9, 2020, when Pfizer announced that its vaccine was expected to be up to 90% effective, and on December 8, 2020, when the U.S. counted more than 15 million total COVID-19 cases.
The researchers found that the Twitter users who actively discussed the COVID-19 vaccines were from the states at the country's east, west, and south areas. This includes California, with over 390 users who mentioned the vaccines per 100,000 people, followed by Washington, at 370 and Texas, at 362 users per 100,000. However, the average number of tweets related to the vaccine per Twitter user did not vary by state.
Of approximately 4.5 million tweets that mentioned COVID-19 vaccines in the U.S., over a third were positive about the vaccines, a third were negative, and the rest neutral. When examined over time, the researchers found that the sentiment score, positive or negative, showed significant differences from week to week at the beginning of the pandemic, but then became stable.
The period from March 8, 2020, to March 14, 2020, showed the highest mean score for sentiments about the vaccine, with over 54,000 tweets and a score of 0.155. Conversely, from May 31, 2020, to June 6, 2020, the average score dropped to -0.351, with almost 59,000 tweets.
Further analysis of these trends showed that the topics during the first (positive) peak included COVID-19 vaccine development, which was mentioned in over half the tweets, and vaccine availability for all at 31%, as well as vaccine fight COVID-19 in almost a fifth. Half the tweets dealt with Black Lives Matter and the vaccine during the second period, with approximately a quarter each mentioning ‘Bad COVID-19 pandemic’ and ‘COVID-19 vaccine availability.’
The average sentiment scores of tweets mentioning the vaccines varied from state to state. In general, these scores were higher in states on the east coast, such as Maine and New York, at 0.025 and 0.026, respectively, but low in Oregon or Colorado on the west coast, at -0.014 and -0.017, respectively.
Average sentiment score of tweets related to COVID-19 vaccines from different US states. Avg. Sent score stands for average sentiment score.
Of the over 1.1 million Twitter users who tweeted about COVID-19 vaccines in the U.S., the authors obtained approximately 59,000 valid images from the user profiles and classified them using the DeepFace algorithm. The male users outnumbered females by 2:1.
Over 66% of these users were between the ages of 30-49 years, with just under 35% being in the 25–29-year age group. Almost two-thirds of these users were White, with just about one in seven being either Black or Asian, and another 17% being of Other races.
Demographic characteristics of Twitter users who discussed COVID-19 vaccines in the U.S.
Overall, about 44% of users had a favorable perception of the vaccine; however, a third saw it negatively and over one in five were neutral towards the vaccine. Racial or age-based characteristics did not affect the type of perception, nor did the sex of the person.
Implications
The researchers found that COVID-19 vaccines became a highly debated topic on Twitter within the U.S., with the trend increasing longitudinally. The largest stimuli to the discussion of this topic were the course of the pandemic in this country and the development of the new vaccines. The latter factor was of greater interest to Twitter users in coastal states, with those from east coast states tending to accept the vaccine most favorably.
COVID-19 vaccines were more often the subject of tweets among males aged 30-49 years. While the proportion of Whites among the Twitter users who discussed the vaccine reflects the actual proportion in the general Twitter user population, Blacks and Asians were less strongly represented in this area as compared to their general proportions at approximately 6% and 9% versus 14% and 18%, respectively.
The timing of the two highest sentiment scores, both positive and negative, coincided with the development of the vaccine and the number of COVID-19 cases surpassing 15 million.
While these figures corroborate those from most earlier studies, one study showed that the weekly trend of tweets related to the vaccine remained positive but increased over time. This difference may be due to the use of different keywords and different geographical boundaries.
The positive peak in March 2020 could be due to the positive news being released about the vaccine, while the negative peak in early June 2020 may have been due to low vaccine availability and the rise of the Black Lives Matter movement.
The finding that Whites and males in one age group talk about this topic more often provides a focus for targeted information campaigns to boost vaccine confidence. The use of Twitter also provides a broad picture of discussions about the vaccine in the U.S.
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
New vaccine shows potential in preventing recurrence of triple-negative breast cancer