Profiling vaccine attitudes and subsequent uptake in 1·1 million people in England: a nationwide cohort study

Despite highly effective vaccines against SARS-CoV-2, COVID-19 vaccine hesitancy persisted in some populations in England during the pandemic, with rates and motivations for hesitancy varying by demographic group. Addressing the drivers of vaccine hesitancy through targeted interventions in hesitant groups is a public health priority for better and more rapid control of disease spread. We aimed to characterise the determinants and subtypes of vaccine hesitancy and identify more persistent forms of hesitancy via analysis of vaccine uptake in a large cross-sectional cohort with linked National Health Service (NHS) data.

We conducted an initial cross-sectional analysis of vaccine hesitancy at baseline, followed by a longitudinal analysis of vaccine uptake in the hesitant cohort. We analysed survey data from the Real-time Assessment of Community Transmission (REACT) studies, which monitored the prevalence of SARS-CoV-2 in England during the COVID-19 pandemic at regular intervals from May 1, 2020, to March 31, 2022, in random samples of the population. Participants self-reported detailed sociodemographic information, vaccination status, and attitudes towards vaccination. Participants were classified as hesitant if they reported that they had refused, planned to refuse, or had not yet decided whether to receive the COVID-19 vaccine. Participants who said they were unvaccinated when NHS records showed that they had been vaccinated were excluded from further analysis. The primary outcome of the cross-sectional analysis was vaccine hesitancy. Longitudinal analysis of vaccine uptake was done in participants in the hesitant cohort who consented to the use of linked NHS vaccination records to track their vaccination history after the survey, with post-survey vaccination as the outcome. Consensus clustering was used to categorise reasons for vaccine hesitancy, and cross-sectional and longitudinal analyses used logistic regression models to identify demographic predictors of vaccine hesitancy and subsequent vaccination.

Our analyses included 1 137 927 adults (aged 18 years and older) surveyed between Jan 6, 2021, and March 31, 2022. Across the whole study period, 37 982 (3·3%) participants indicated some form of vaccine hesitancy. Hesitancy rates peaked at 8·0% in early 2021, subsequently decreasing to a low of 1·1% at the start of 2022, and increasing to 2·2% in early 2022. Of the 24 229 participants who indicated hesitancy and consented to NHS data linkage, 15 744 (65·0%) went on to receive one or more vaccinations. Cluster analysis identified eight stable categories of vaccine hesitancy, including concerns about effectiveness and side-effects, perception of low risk from COVID-19 and mistrust of vaccine developers, and fear of vaccines and reactions. The most prevalent categories of hesitancy, related to effectiveness and health concerns, declined substantially over the roll-out period and were not strongly associated with the likelihood of later vaccination. Some forms of hesitancy, primarily related to low trust, low risk perception, and general anti-vaccine sentiment, were more resistant, rebounded in 2022, and were strongly associated with a lower likelihood of subsequent vaccination.

Our findings suggest that most COVID-19 vaccine hesitancy was rooted in concrete concerns that can be addressed and successfully overcome with time and increasing availability of information. These findings should help future vaccination roll-outs to encourage vaccine acceptance.

UK National Institute for Health and Care Research, UK Research and Innovation, and the UK Department of Health and Social Care.