What causes COVID-19 vaccine hesitancy? Ignorance and the lack of bliss in the United Kingdom

The study examines the drivers of COVID-19 vaccine hesitancy in the United Kingdom. Set against the rapid development and deployment of COVID-19 vaccines and widespread misinformation online, the authors investigate whether knowledge about vaccines and social media use influence willingness to vaccinate. The research question centers on whether anxiety (e.g., concerns about safety/efficacy and pandemic-related worry) rather than familiarity/knowledge better explains vaccine hesitancy, and whether geopolitical trust cues (e.g., vaccines from Russia/China vs Western sources) and social media platforms shape perceived knowledge and acceptance.

Prior work identifies demographic and psychosocial correlates of vaccine hesitancy, including lower acceptance among Black/African groups, unemployed individuals, and those with lower income, education, and younger age. Common reasons for refusal include general anti-vaccine attitudes, safety concerns about rapidly produced vaccines, doubts about efficacy, low perceived severity of COVID-19, lack of trust, belief in prior immunity, and skepticism about provenance. Studies across China, middle-income countries, Portugal, Italy, and the US consistently find safety/side-effect and efficacy concerns as prominent predictors of hesitancy, alongside needs for more information and political ideology effects. In Ireland and the UK, women, younger cohorts, and lower-income individuals showed higher hesitancy/resistance. The review highlights the role of misinformation and social media in amplifying doubt, conspiracy beliefs, and distrust in scientific expertise—factors directly relevant to COVID-19 vaccination contexts.

Design and sample: A cross-sectional online survey of UK adults (N = 537) conducted in August 2021. A soft-launch via social media yielded 37 respondents (included in analyses); the main nationally representative sample was recruited via Prolific. Ethical approval obtained from Kingston University; informed consent collected. Participants were compensated £1.29. Measures: Demographics (education, age, gender, race/ethnicity, UK region), religious/spiritual identity, COVID-19 risk status, alumni status, and frequency of social media use (Twitter, Reddit, Facebook, Instagram). Political ideology assessed with two sliders (0 = very liberal to 100 = very conservative) for social and economic dimensions. Vaccination-related items: Vaccination status; if vaccinated, self-reported anxiety about taking the vaccine; if not vaccinated, willingness to accept when offered (immediately, delay, refuse). Additional items: affect toward receiving a vaccine (very positive to very negative), actions if vaccine available at local pharmacy, encouraging others and being encouraged by others to vaccinate, perceived importance of vaccination. Knowledge rating (from “never heard of this vaccine” to “know a lot about this vaccine”) for Novavax, Janssen-J&J, Moderna, Pfizer-BioNTech, Oxford-AstraZeneca, Sinovac, Sputnik V, and two fictitious vaccines (“Theranos” and “Medicare”). Willingness to take vaccines from the listed producers assessed on Likert scales. Willingness to vaccinate by location (hospital, church, mobile ambulance, pharmacy, university, mosque, Jewish/Hindu/other religious sites, or self-administered). Trust in and responsiveness to religious/spiritual leaders on health matters. Perceived COVID-19 risk severity (0–100). Identity fusion measured via 7-item verbal fusion scales for nation and religion (Gómez et al., 2011). Analysis: Conducted in R using OLS regression models with two-tailed tests (95% confidence). Models assessed: (1) demographic, political, and social media predictors of self-reported knowledge about vaccines (approved and unapproved/fictive), and (2) predictors of willingness to take specific vaccines (approved, unapproved foreign, and fictitious). Dropout was minimal (3.2% at consent; 1.6% at a lengthy mid-survey page).

- High willingness for UK/Western approved vaccines: 89.22% would take Oxford-AstraZeneca; 88.24% would take Pfizer-BioNTech. - Paradoxical acceptance of fictitious vaccines: Respondents showed more willingness to take non-existent US-associated vaccines (Medicare, Theranos) than to take Russian (Sputnik) and Chinese (Sinovac) vaccines. - Anxiety vs familiarity: Data indicate anxiety is a more critical factor in hesitancy than familiarity/knowledge per se. - Social media effects on perceived knowledge: Greater Twitter use was associated with higher self-reported knowledge of Janssen-J&J and Moderna; economic conservatism positively related to feeling knowledgeable about Oxford-AstraZeneca; social conservatism negatively related to feeling knowledgeable about Moderna and Oxford-AstraZeneca. For unavailable vaccines, Twitter use increased perceived knowledge of Sputnik; economic conservatism also increased perceived knowledge of Sputnik. - Social media effects on willingness (approved vaccines): Higher Twitter use was positively associated with willingness to take Janssen-J&J, Moderna, and Novavax; higher Instagram use showed a negative association with willingness to take approved vaccines (except no significant effect for Oxford-AstraZeneca). Social conservatism negatively related to willingness to take Pfizer-BioNTech. - Social media effects on willingness (unavailable/fake vaccines): Higher Twitter use positively predicted willingness to take Sputnik, Sinovac, and the fictitious Medicare and Theranos vaccines (Table 4).

Findings show that social media—especially Twitter—shapes both perceived knowledge and willingness to vaccinate, including willingness to accept fictitious vaccines. Twitter users reported feeling more knowledgeable yet also displayed greater willingness to accept vaccines that were unapproved or non-existent, suggesting susceptibility to misinformation and/or anxiety-driven decision-making. Instagram use was associated with a generalized anti-vaccination stance (lower willingness for approved vaccines), whereas Facebook use showed no discernible relationship and Reddit use showed no consistent association with knowledge or hesitancy. The pattern that UK respondents were more willing to accept fictitious US-associated vaccines than real vaccines from Russia or China suggests geopolitical trust cues and perceived rigor of US/European regulatory systems influence acceptance. Overall, anxiety surrounding the pandemic appears to have outweighed actual familiarity with vaccines in driving hesitancy and acceptance judgments. These results highlight how platform-specific information environments and broader trust dynamics can differentially impact vaccine attitudes beyond traditional demographic predictors.

This study contributes evidence that COVID-19 vaccine hesitancy in the UK is driven more by anxiety and trust cues than by familiarity alone, and that social media—particularly Twitter—can increase both perceived knowledge and willingness to accept even fictitious vaccines. Acceptance patterns indicate a preference for vaccines perceived as Western/US-aligned and skepticism toward Russian/Chinese vaccines, pointing to geopolitical trust as a key factor. Policymakers and health communicators should target anxiety reduction, correct misinformation, and consider platform-specific strategies. Future research should implement randomized experiments that vary vaccine names/sources (including neutral, non-institutional labels), integrate behavioral trace data to validate self-reported social media use, and employ richer multi-dimensional measures of political ideology to disentangle social vs economic components in predicting hesitancy.

- Self-reported social media use could not be externally validated; future work should include behavioral data. - Non-experimental design limits causal inference; randomized designs varying vaccine names/approval status are recommended. - Use of pre-existing institutional names for fictitious vaccines (Medicare, Theranos) may have introduced recognition biases; future studies should use neutral labels. - Political ideology measured with two sliders may not capture full multidimensionality; richer ideological measures are warranted.