Information and vaccine hesitancy: Evidence from the early stage of the vaccine roll-out in 28 European countries

Abstract

The success of mass vaccination programs against SARS-CoV-2 hinges on the public's acceptance of the vaccines. During a vaccine roll-out, individuals have limited information about the potential side-effects and benefits. Given the public health concern of the COVID pandemic, providing appropriate information fast matters for the success of the campaign. In this paper, time-trends in vaccine hesitancy were examined using a sample of 35,390 respondents from the Eurofound's Living, Working and COVID-19 (LWC) data collected between 12 February and 28 March 2021 across 28 European countries. The data cover the initial stage of the vaccine roll-out. We exploit the fact that during this period, news about rare cases of blood clots with low blood platelets were potentially linked to the Oxford/AstraZeneca vaccine (or Vaxzeveria). Multivariate regression models were used to analyze i) vaccine hesitancy trends, and whether any trend-change was associated with the link between the AstraZeneca vaccine ii) and blood clots (AstraZeneca controversy), and iii) the suspension among several European countries. Our estimates show that vaccine hesitancy increased over the early stage of the vaccine roll-out (0·002, 95% CI: [0·002 to 0·003]), a positive shift took place in the likelihood of hesitancy following the controversy (0·230, 95% CI: [0·157 to 0·302]), with the trend subsequently turning negative (-0·007, 95% CI: [-0·010 to -0·005]). Countries deciding to suspend the AstraZeneca vaccine experienced an increase in vaccine hesitancy after the suspensions (0·068, 95% CI: [0·04 to 0·095]). Trust in institutions is negatively associated with vaccine hesitancy. The results suggest that SARS-CoV-2 vaccine hesitancy increased steadily since the beginning of the vaccine roll-out and the AstraZeneca controversy and its suspension, made modest (though significant) contributions to increased hesitancy.

Fig 1. Sample selection.

Notes: The initial sample of 57,621 respondents is obtained by eliminating those individuals who declare themselves to be older than 100 or who have missing age information.

Fig 2. Vaccine hesitancy time-trend.

Notes: the graph shows the time-trend of the vaccine hesitancy during the period from 12 February 2021 to 28 Marrh 2021 The blue dashed line represents the moment of the AstraZeneca controversy, coinciding with 11 March 2021, while the green dashed line represents the date when the AstraZeneca vaccine was suspended across a number of countries (16 March 2021).

Fig 3. Google trends in the search of “AstraZeneca vaccine” and “thrombosis”.

Notes: the graph presents the mean interest for two keywords, "thrombosis" and "AstraZeneca vaccine", during the period from 1 February 2021 to 21 March 2021. Search volumes for these keywords have been retrieved from Google Trends in seven languages: English, French, German, Italian, Spanish, Polisch and Danish.

Fig 4. Time-trend in Trust in a) Government b) The EU c) Pharmaceutical Companies and d) Healthcare Systems.

Notes: the graphs show the time trend of four trust measures (trust in government (a), trust in EU (b), trust in pharmaceutical firms (c), trust in the healthcare system (d)) during the period from 12 February 2021 to 28 March 2021. The blue dashed lines represent the moment of the AstraZeneca controversy, coinciding with 11 March 2021, while the green dashed lines represent the date when the AstraZeneca vaccine was suspended across a number of countries (16 March 2021).

Fig 5. Linear prediction vaccine hesitancy.

Notes: the graph presents the linear prediction of vaccine hesitancy. The red dashed line coincides with the date of the AstraZeneca controversy (11 March 2021).