In early and mid-March 2021, a number of European countries paused vaccination against COVID-19 with the ChAdOx1 nCoV-19 (AZD1222) vaccine from Oxford-AstraZeneca [1, 2] following reports of thromboembolic events, some with fatal outcomes, among people who had received this vaccine [3, 4]. On 18 March2021, after extensive scrutiny of these events, the European Medicines Agency concluded regarding the Oxford-AstraZeneca Covid-vaccine that “benefits still outweigh the risks despite possible link to rare blood clots with low blood platelets” . This led some European countries to immediately reinitiate vaccination with the Oxford-AstraZeneca COVID-19 vaccine, whereas other countries – including Denmark – decided to await further examination of the potential relationship between the vaccine and the cases of thromboembolisms . This train of events was covered extensively by the news media [6-8].
Prior studies have suggested that vaccine willingness is highly dependent on public trust in the safety of vaccines [9, 10]. Therefore, it seems likely that vaccine willingness may have dropped in the wake of the reported cases of thromboembolic events in relation to the Oxford-AstraZeneca COVID-19 vaccine. This would potentially threaten the success of the global COVID-19 vaccine programme, which is crucial in the management of the COVID-19 pandemic. In the present study, we examined the vaccine willingness and perceived safety of the two most widely used COVID-19 vaccines in Denmark ; the Oxford-AstraZeneca [1, 2] and the Pfizer-BioNTech  vaccine. Specifically, we addressed the following five research questions: i) Has the COVID-19 vaccine willingness dropped following the thromboembolic events reported in relation to the Oxford-AstraZeneca COVID-19 vaccine? ii) What is the perceived safety of the COVID-19 vaccines from Oxford-AstraZeneca and Pfizer-BioNTech, and does it differ between the two? iii) Do people who are vaccine-hesitant perceive the safety of the COVID-19 vaccines from Oxford-AstraZeneca and Pfizer-BioNTech to be lower than those who are vaccine-willing? iv) Does trust in the Health Care System system differ between those who are vaccine-willing and vaccine-hesitant? and v) Is the perceived safety of the COVID-19 vaccines related to trust in the Danish healthcare system?
To investigate these research questions, we used data from the two most recent waves of a five-wave panel survey conducted by the Danish survey company Epinion (invitees sampled from Epinion’s respondent pool), the first three of which are described in detail elsewhere [13-15]. In the fourth wave, which was fielded from 4 to 21 February 2021 (i.e. prior to the reports of thromboembolic events in relation to the Oxford-AstraZeneca COVID-19 vaccine), we targeted the 2,458 respondents from the first wave . In addition to questions regarding psychological well-being and social and political attitudes/perceptions for other research purposes, the participants were asked: “Have you been offered vaccination against coronavirus?”. The response options were: “Yes”, “No” and “Do not wish to answer”. Participants who answered “Yes” were then asked “Have you accepted this offer?”. Participants who responded “No”, were asked: “Will you accept vaccination against coronavirus, once it is offered to you?”. For both of these questions, the response options were: “Yes”, “No” and “Do not wish to answer”. In the fifth wave, which was fielded from 15 to 25 March 2021 (i.e. after the reports of thromboembolic events in relation to the Oxford-AstraZeneca COVID-19 vaccine), we included the same questions on vaccine willingness as in wave four and added a question on COVID-19 vaccine safety, randomising the vaccine in question. Specifically, the respondents were randomly assigned one of the following two questions: “How do you consider the safety of the Pfizer vaccine against coronavirus?” and “How do you consider the safety of the AstraZeneca vaccine against coronavirus?”. The response to these questions was provided on a Likert-like scale ranging from 0 – “Very low safety” to 10 – “Very high safety”. The respondents could also choose a “Don’t know” response. The fifth wave of the survey also included questions regarding institutional and inter-individual trust (posed in the latter part of the questionnaire). For the present study, we included the response to the question on trust in the Danish healthcare system (Danish: Sundhedsvæsenet): “On a scale of 0-10, how much do you personally trust the healthcare system?”. The response was provided on a Likert-like scale ranging from 0 – “No trust at all” to 10 – “Complete trust”. The respondents could also choose a “Don’t know” response.
Respondents and weighting
A total of 1,654 invitees responded to the survey wave collected prior to the reports of thromboembolic events in relation to the Oxford-AstraZeneca COVID-19 vaccine, and 1,657 responded to the survey wave collected after these reports (1,491 responded to both waves). The analyses corresponding to the five research questions were based on respondents answering both survey waves. To minimise the effect of potential sample selection bias, we weighted the responses to all questions in the survey using inverse propensity weighting  by sex, age, level of education, region and political party choice at last general election (5 June 2019). This weighting renders the respondents representative of the Danish population on these variables.
We first compared willingness to accept a future COVID-19 vaccination offer across the two waves. Specifically, we compared the proportions answering in the affirmative to the question "Will you accept vaccination against coronavirus once it is offered to you?" using a logistic regression with standard errors adjusted for repeated observations and two-sided p-values. This analysis only included respondents who reported that they had not yet been offered vaccination in both waves (83.6% of the respondents). Second, we compared the mean levels of perceived safety of the two vaccines, as defined by the randomly assigned version of the question ("Pfizer" versus "AstraZeneca"), using a two-sample t-test. Third, we compared the perceived safety of the two vaccines between respondents who were vaccine-hesitant and vaccine-willing (the latter group comprised respondents who had answered in the affirmative to one of the two questions about vaccine willingness in the fifth wave of the survey, whereas the former group comprised those who answered “No” or “Do not wish to answer”), using a two-sample t-test. Fourth, the level of trust in the Danish healthcare system was compared between the vaccine-willing and vaccine-hesitant respondents using a two-sample t-test. Fifth, the relationship between the perceived safety of the COVID-19 vaccines and the level of trust in the Danish healthcare system was examined by linear regression.
Participation in the survey was based on consent. Under Danish Law, ethical review board approval is not required for survey studies. Data were stored and handled in accordance with the General Data Protection Regulation.
Trial registration: not relevant.
Table 1 shows the characteristics (unweighted and weighted) of the 1,491 individuals having responded to both survey waves.
The results of the analyses corresponding to the three first research questions are shown in Figure 1. Figure 1A shows that vaccine willingness in Denmark was very high both before (89.3%) and after (89.2%) the thromboembolic events reported in relation to the Oxford-AstraZeneca COVID-19 vaccine (p = 0.961). Figure 1B shows that the perceived safety of the Oxford-AstraZeneca COVID-19 vaccine was substantially lower than that of Pfizer-BioNTech (mean safety perception on the scale of 0-10 of 5.35 versus 8.26, p < 0.001). Figure 1C shows that it was particularly those who were vaccine-hesitant who perceived the safety of the vaccines to be low – especially the Oxford-AstraZeneca COVID-19 vaccine (mean safety perception of 2.55 compared with 5.86 for Pfizer-BioNTech, p< 0.001).
As regards the fourth and the fifth research question, the mean level of trust in the Danish healthcare system on the scale of 0-10 was 6.30 among the vaccine-hesitant and 8.09 among the vaccine-willing (p < 0.001). Furthermore, a positive correlation was observed between the level of trust in Danish healthcare system and the perceived safety of the Oxford-AstraZeneca COVID-19 vaccine (standardised linear regression coefficient: 0.36, p < 0.001) and the Pfizer-BioNTech COVID-19 vaccine (standardised linear regression coefficient: 0.38, p < 0.001).
The results of this study are reassuring with regard to COVID-19 vaccine uptake in Denmark following the thromboembolic events reported in relation to the Oxford-AstraZeneca COVID-19 vaccine. However, the perceived safety of the Oxford-AstraZeneca COVID-19 vaccine is substantially lower than that of Pfizer-BioNTech, especially among those who are vaccine-hesitant. These results are compatible with – although not proof of – perceived safety being associated with vaccine willingness, which is in line with the literature on this topic [9, 10]. Furthermore, the present study replicated the finding of a positive association between trust in healthcare systems and vaccine willingness [17, 18], and reported that trust in the Danish healthcare system is also positively associated with perceived safety of the COVID-19 vaccines. Taken together, these findings suggest that maintenance of vaccine uptake will likely benefit from health authorities investing in increasing trust in the safety of the Oxford-AstraZeneca COVID-19 vaccine.
Some limitations need to be considered when interpreting the results of this study. First and foremost, the assessment of vaccine willingness is based on a survey questionnaire. Whereas this method is widely used [9, 10, 19], data on actual vaccine uptake (and rejection) would be superior as they will not be affected by social desirability bias (reporting vaccine willingness without eventually accepting the vaccine). Such studies should be pursued as data on vaccine uptake become available. Second and relatedly, although the results of this study are weighted on key sociodemographic variables, we cannot rule out that non-response bias may have resulted in overestimation of vaccine willingness. Third, we did not collect data on the perceived safety of the COVID-19 vaccines from Oxford-AstraZeneca and Pfizer-BioNTech in the survey wave fielded prior to the reports of thromboembolic events in relation to the Oxford-AstraZeneca COVID-19 vaccine. Therefore, we cannot know whether the low level of perceived safety of the Oxford-AstraZeneca COVID-19 vaccine observed after these events would also have been found prior to them, had these questions been asked then. Given the severity of the thromboembolic events [3, 4], the paused use of the vaccine [3, 4, 6], and the associated media coverage [6, 7, 8], this, however, seems unlikely. Fourth and finally, vaccine willingness in Denmark is among the highest worldwide , and the results regarding sustained vaccine willingness may therefore not necessarily generalise to countries with a lower “baseline” vaccine willingness. Specifically, in countries with lower institutional trust than Denmark – where institutional trust is among the highest in the world  – it may prove challenging for the health authorities to uphold support for COVID-19 vaccination, given the recent events.
The (high) vaccine willingness of Danes does not seem to have been affected by the reports of thromboembolic events in relation to the Oxford-AstraZeneca COVID-19 vaccine. However, the perceived safety of the Oxford-AstraZeneca COVID-19 vaccine is low – especially among vaccine-hesitant individuals. Consequently, maintenance of vaccine uptake will likely benefit from health authorities investing in increasing trust in the safety of the Oxford-AstraZeneca COVID-19 vaccine.
Correspondence Søren Dinesen Østergaard. E-mail: email@example.comAccepted 29 March 2021Conflicts of interest Potential conflicts of interest have been declared. Disclosure forms provided by the authors are available with the article at ugeskriftet.dk/dmjReferences can be found with the article at ugeskriftet.dk/dmjCite this as Dan Med J 2021;68(5):A03210292
Ramasamy MN, Minassian AM, Ewer KJ et al. Safety and immunogenicity of ChAdOx1 nCoV-19 vaccine administered in a prime-boost regimen in young and old adults (COV002): a single-blind, randomised, controlled, phase 2/3 trial. Lancet 2021;396:1979-93.
Voysey M, Costa Clemens SA, Madhi SA et al. Single-dose administration and the influence of the timing of the booster dose on immunogenicity and efficacy of ChAdOx1 nCoV-19 (AZD1222) vaccine: a pooled analysis of four randomised trials. Lancet 2021;397:881-91.
Wise J. Covid-19: European countries suspend use of Oxford-AstraZeneca vaccine after reports of blood clots. BMJ 2021;372:n699.
Vogel G, Kupferschmidt K. ScienceMag.Org. ‘It’s a very special picture.’ Why vaccine safety experts put the brakes on AstraZeneca’s COVID-19 vaccine. 2021. https://www.sciencemag.org/news/2021/03/it-s-very-special-picture-why-vaccine-safety-experts-put-brakes-astrazeneca-s-covid-19 (27 Mar 2021).
European Medicines Agency. COVID-19 vaccine AstraZeneca: benefits still outweigh the risks despite possible link to rare blood clots with low blood platelets. 2021. https://www.ema.europa.eu/en/news/covid-19-vaccine-astrazeneca-benefits-still-outweigh-risks-despite-possible-link-rare-blood-clots (27 Mar 2021).
Gronholt-Pedersen J. Denmark reports two cases of serious illness, including one death, after AstraZeneca shot 2021. https://www.reuters.com/article/us-health-coronavirus-denmark-idUSKBN2BC0A4 (27 Mar 2021).
Grady D, Robbins R. Should you be concerned about blood clots, bleeding and the AZ-Vaccine? 2021. https://www.nytimes.com/2021/03/15/health/astra-zeneca-vaccine-blood-clots-bleeding.html (27 Mar 2021).
Dreiager IS. Employee in the Capital Region dead after recent Astrazeneca-vaccination (translated from Danish) 2021. https://jyllands-posten.dk/indland/ECE12847662/ansat-i-region-hovedstaden-er-doed-efter-nylig-astrazenecavaccination/ (27 Mar 2021).
Karlsson LC, Soveri A, Lewandowsky S et al. Fearing the disease or the vaccine: The case of COVID-19. Pers Individ Dif 2021;172:110590.
Karlsson LC, Lewandowsky S, Antfolk J et al. The association between vaccination confidence, vaccination behavior, and willingness to recommend vaccines among Finnish healthcare workers. PLoS One 2019;14:e0224330.
Statens Serum Institut. Daily monitoring of vaccine uptake in Denmark 2021. https://experience.arcgis.com/experience/1c7ff08f6cef4e2784df7532d16312f1 (27 Mar 2021).
Polack FP, Thomas SJ, Kitchin N et al. Safety and efficacy of the BNT162b2 mRNA COVID-19 Vaccine. N Engl J Med 2020;383:2603-15.
Sønderskov KM, Dinesen PT, Santini ZI et al. The depressive state of Denmark during the COVID-19 pandemic. Acta Neuropsychiatr 2020;32:226-8.
Sønderskov KM, Dinesen PT, Santini ZI et al. Increased psychological well-being after the apex of the COVID-19 pandemic. Acta Neuropsychiatr 2020;32:277-9.
Sønderskov KM, Dinesen PT, Vistisen HT et al. Variation in psychological well-being and symptoms of anxiety and depression during the COVID-19 pandemic: results from a three-wave panel survey. Acta Neuropsychiatr (online 11 Jan 2021).
Cole SR, Hernán MA. Constructing inverse probability weights for marginal structural models. Am J Epidemiol 2008;168:656-64.
Murphy J, Vallières F, Bentall RP et al. Psychological characteristics associated with COVID-19 vaccine hesitancy and resistance in Ireland and the United Kingdom. Nat Commun 2021;12:29.
Larson HJ, Clarke RM, Jarrett C et al. Measuring trust in vaccination: a systematic review. Human Vaccines Immunother 2018;14:1599-609.
Lindholt MF, Jørgensen FJ, Bor A et al. Willingness to use an approved COVID-19 vaccine: cross-national evidence on levels and individual-level predictors. PsyArXiv 2020. doi: 10.31234/osf.io/8kn5f.
Sønderskov KM, Dinesen PT. Trusting the state, trusting each other? The effect of institutional trust on social trust. Political Behavior 2016;38:179-202.