JMIR Publications

ABSTRACT

Background:

Despite pediatric populations representing a smaller proportion and less severe prognosis of COVID-19 case counts, similar demographics such as racial and ethnic minority status place certain groups at an increased risk of developing more severe COVID-19-related outcomes, including hospitalization. Universal vaccine coverage is crucial to the pandemic preventative and mitigation efforts, yet since COVID-19, vaccine hesitancy has increased and routine immunizations among pediatrics have decreased. Limited research exists on how current vaccine hesitancy may contribute to low pediatric COVID-19 vaccine uptake among racial and ethnic minority populations.

Objective:

To characterize COVID-19 vaccine-related discussion and sentiment among Twitter users, particularly among racial and ethnic populations.

Methods:

We used the Twitter API to collect tweets and replies. Tweets were selected by filtering for keywords associated with COVID-19, vaccines, and pediatric-related terms. From this corpus of tweets, we used Biterm Topic Model to output topics and examined the top 200 retweeted tweets which were coded for pediatric COVID-19 vaccine relevance. Relevant tweets were analyzed using an inductive coding approach to characterize pediatric COVID-19 vaccine-related themes. Replies to relevant tweets were collected and coded. User metadata was assessed for self-reporting of race or ethnic group affiliation and verified account status.

Results:

863,007 tweets were collected from October 2020 - October 2021. After outputting BTM topics and reviewing the 200 most retweeted tweets, 208,666 tweets and 3,905 replies were identified as pediatric COVID-19 vaccine-related. Among signal tweets and replies, we identified 418 users who self-identified as a racial minority and 40 who self-identified as an ethnic minority. The majority (72.0%) of tweets expressed vaccine-related concerns. Among tweets discussing vaccine confidence, user replies expressing agreement were significantly outweighed by those expressing disagreement (32.7% vs 67.3%; P<.001). The main themes identified in Twitter interactions included conversations regarding vaccine-related concerns (i.e., adverse side effects and the need for more testing) and conversations associated with vaccine-related confidence (i.e., the vaccine is protective).

Conclusions:

Vaccine-related concerns can have negative consequences on diverse clinical trial participant recruitment and retention, as well as ensuring the development of safe and effective vaccines, especially among minority populations.