A high rate of COVID-19 vaccine hesitancy in a large-scale survey on Arabs
Vaccine hesitancy can limit the benefits of available vaccines in halting the spread of COVID-19 pandemic. Previously published studies paid little attention to Arab countries, which has a population of over 440 million. In this study, we present the results of the first large-scale multinational study that measures vaccine hesitancy among Arab-speaking subjects.
An online survey in Arabic was conducted from 14 January 2021 to 29 January 2021. It consisted of 17 questions capturing demographic data, acceptance of COVID-19 vaccine, attitudes toward the need for COVID-19 vaccination and associated health policies, and reasons for vaccination hesitancy. R software v.4.0.2 was used for data analysis and visualization.
The survey recruited 36,220 eligible participants (61.1% males, 38.9% females, mean age 32.6 ± 10.8 years) from all the 23 Arab countries and territories (83.4%) and 122 other countries (16.6%). Our analysis shows a significant rate of vaccine hesitancy among Arabs in and outside the Arab region (83% and 81%, respectively). The most cited reasons for hesitancy are concerns about side effects and distrust in health care policies, vaccine expedited production, published studies and vaccine producing companies. We also found that female participants, those who are 30–59 years old, those with no chronic diseases, those with lower level of academic education, and those who do not know the type of vaccine authorized in their countries are more hesitant to receive COVID-19 vaccination. On the other hand, participants who regularly receive the influenza vaccine, health care workers, and those from countries with higher rates of COVID-19 infections showed more vaccination willingness. Interactive representation of our results is posted on our project website at https://mainapp.shinyapps.io/CVHAA.
Our results show higher vaccine hesitancy and refusal among Arab subjects, related mainly to distrust and concerns about side effects. Health authorities and Arab scientific community have to transparently address these concerns to improve vaccine acceptance.
This study received no funding.
It has been recognized early that the race to produce COVID-19 vaccines will not halt the pandemic unless there is a general acceptance by the public to take the vaccine (Neumann-Böhme et al., 2020; Burgess et al., 2021). Therefore, COVID-19 vaccination hesitancy has been studied heavily before and since the early stage of vaccine availability, with high variation in the willingness to be vaccinated among different communities (Sallam, 2021; Feleszko et al., 2021; Lin et al., 2020). Several large multinational studies on vaccine acceptance have been conducted in 19 countries (n = 13,426) (Lazarus et al., 2021), 15 countries (n = 18,526) (Ipsos, 2020), 15 countries (n = 13,500) (Mega, 2021), 14 countries (n = 12,777) (Ipsos, 2021), and seven countries (n = 7662) (Neumann-Böhme et al., 2020). Interestingly, none of which covered Arabic-speaking nations. A smaller multinational study that surveyed 3414 participants was conducted in Jordan and Kuwait, with minor participation from some other Arab countries (Sallam et al., 2021). This study showed that vaccine acceptance was low (29.4%) and was lower in females, individuals with lower academic education, and individuals with no chronic diseases.
Other studies have been conducted in Saudi Arabia (n = 1000) and (n = 3101) (Al-Mohaithef and Padhi, 2020; Magadmi and Kamel, 2020), Egypt (n = 559) (Abdelhafiz et al., 2020), Jordan (n = 3100) (El-Elimat et al., 2020), and the UAE (n = 1109) (Muqattash et al., 2020). With the Arab nations having significant variations socioeconomically, politically and in the measures taken to control the pandemic, the study of reactions to and acceptance of the vaccine becomes necessary. Also, the authorization of the use of Sinopharm vaccine by some Arab countries, despite the lack of sufficient safety and effectiveness evidence (Cyranoski, 2020), may have an impact on the public’s trust in the vaccine and the health policies in these countries. Furthermore, attitudes toward the vaccines are affected by complex and dynamic interplaying factors, and considerable changes over time have been observed in acceptance and hesitancy rates (Lazarus et al., 2021; Ipsos, 2020; Mega, 2021; Ipsos, 2021; Wang et al., 2021; Imperial College London, 2021; Lewis, 2020). For all of these reasons, the earlier local studies cannot be generalized to the Arab world, and further larger studies will present a clearer picture of the region.
Arab countries and territories (23 in total) span a large geographical area in North Africa and West Asia with a population of over 440 million (Worldometer, 2021). The total reported number of COVID-19 cases in the region until the mid-February 2021 was more than 4.1 million with 70.7 thousand deaths (Dong et al., 2020; Appendix 1—figure 1). Yet, the Arab region is understudied, despite the geographical spread, the number of residents, and the number of cases and deaths. So, a large-scale multinational study for this area is necessary.
Our study aims to fill the gaps by investigating vaccine acceptance using a large-scale survey targeting the relatively understudied Arab populations living in different countries around the world following vaccine availability and administration. Secondly, to unveil the barriers leading to vaccine hesitancy and their prevalence among the participants using an extensive updated list of barriers against vaccine acceptance. Thirdly, the study compares the answers of the respondents residing in and outside the Arab world to evaluate the effect of socioeconomic, cultural, health policies and political differences on their reported attitudes and barriers to acceptance.
Materials and methods
The Survey of Arab COVID-19 Vaccine Acceptance (SACVA) is an open online survey that was conducted using the online platform https://www.surveyplanet.com/ from 14 January 2021 to 29 January 2021. The sample population was a convenience sample targeted through a digital campaign using social media platforms. Institutional Review Board (IRB) approval was obtained from the last author’s institution. Unique IP addresses are allowed to participate once on the survey platform to prevent multiple entries. Consent to participate was obtained at the first entry of the survey portal for each participant. The platform allows participants to move through screens only when answers were obtained, which prevents missing entries. The survey consisted of 17 questions, including the consent to participate. All questions were written and validated in the Arabic language – an English translation of the questions can be found on Supplementary file 1. Questions two to nine captured demographics and current health status; question 10 was about the annual influenza vaccine; question 11 was about available vaccine(s) in each country (if known), and answers to subsequent questions were directed based on the type of available vaccine(s). Questions 12 and 13 queried whether the participant received the COVID-19 vaccination and if they had any side effects. Those who had already taken the vaccine were not allowed to answer question 14 that queried participant’s acceptance/hesitance toward COVID-19 vaccine; these participants were not included in the analysis reported in this paper. Questions 15 and16 surveyed participants' attitudes toward the need for COVID-19 vaccination and associated health policies. Question 17 was a detailed question that evaluated 29 barriers, which potentially influenced the decision to receive the vaccine in addition to ‘I do not have any reservations about taking the vaccine’ option. We also allowed participants who answered ‘Yes’, meaning they are willing to be vaccinated, to choose from the 29 barriers.
Questions were discussed thoroughly among authors and other colleagues. Face validity was tested by the third author, who has expertise in the domain. A pilot survey was then posted online and 100 individuals participated following direct contacts by authors. Analyzing responses and comments of this pilot survey helped in refining the final survey and confirming its validity and reliability.
The survey data was analyzed using R software v.4.0.2. Descriptive statistics and analytical graphs were used as needed. Participants were also subcategorized based on country of residence. Arab countries with less than 100 participants (Somalia, Djibouti, and Comoros) were grouped together in one category and labeled ‘Other Arab countries’. The non-Arab countries where the Arabic-speaking respondents were residing were classified into groups: European countries (n = 30), North American countries (n = 3), Turkey, and the rest of non-Arab countries as others (n = 88). The answers to the 14th question, ‘Do you intend to take the vaccine?’, were used as a dependent variable and were analyzed using binary logistic regression. Two of the answer choices (‘Yes’, ‘Depends on the type of vaccine’) were used to define vaccine acceptance, while the other three (‘No’, ‘Not sure’, ‘I will wait and see its effects on others’) were labeled as ‘Vaccine Hesitancy’. Responses to the question of the barriers to acceptance (Question 17) were compared for gender, academic education, and country of residence using chi-square. Our acquisition and analysis of the results followed the guidelines of the CHERRIES checklist (Eysenbach, 2004).
As for COVID-19 cases and death statistics, we used the COVID19 package v2.3.2 that collects data from different sources to provide up-to-date COVID-19 statistics (Guidotti and Ardia, 2020). The total number of confirmed cases and deaths were correlated with vaccine acceptance in different Arab countries using the Spearman correlation. The results of the survey can be found on the project’s website at https://mainapp.shinyapps.io/CVHAA, while the data and the R code written for the analysis can be found on the project’s GitHub repository (Qunaibi, 2021).
Our online survey raw data were downloaded on 29 January 2021; there were 38,485 participants who started filling the survey of whom 36,958 consented and proceeded with the survey. A total of 738 participants reported receiving COVID-19 vaccination before filling the survey and were excluded from further analysis in this report bringing the total respondents who qualify for analysis to 36,220. The participants cover all the 23 Arab countries and territories (n = 30,200, 83.4%) and Arabs who live in 122 other countries (n = 6020, 16.6%). Participants from countries out of the Arab region were clustered into four groups: Europe (N = 3130, 52%), North America (n = 748, 12.4%), Turkey (n = 1630, 27.1%), and others (n = 512, 8.5%).
The mean age was 32.6 years (±10.8). There were more males (n = 22,040, 61.1%) than females (n = 14,180, 38.9%) – Appendix 1—figure 2. Chronic diseases were reported by 5839 participants (16.1%). Previous COVID-19 infection – suspected or confirmed – was reported by 6637 (18.3%) participants; 11,458 (31.6%) other participants were not sure if they had contracted the virus. Among the 4494 participants who reported testing for COVID-19, there were 2792 participants with positive test results (62.1% positivity). Only 908 (2.5%) participants reported annual influenza vaccine, while 28,040 (77.4%) reported never receiving it. More than half of the participants had a bachelor’s degree or higher (22,236, 61.4%). Being a health care worker (HCW) was reported by 5708 participants (15.8%). When asked about the type of vaccine available in their countries, 15,057 (41.6%) did not know the type, while vaccines made in China and the United States were reported by 12,374 and 12,254 participants, respectively. Detailed participant characteristics are shown in Table 1.
COVID-19 vaccination hesitancy and related factors
When asked about their willingness to receive COVID-19 vaccine if the option is available to them, 4548 (12.6%) of the respondents answered ‘Yes’; 1615 (4.5%) answered ‘Depends on the type of vaccine’; 7552 (20.9%) answered ‘I will wait and see its effects on others’; 7856 (21.7%) answered ‘I am not sure’; and 14,649 (40.4%) chose ‘No’. The first two choices were considered acceptance to receive a vaccine, while the last three were labeled as vaccine hesitancy (Figure 1). Variations in responses were analyzed using different factors as covariates (Figure 2).
Respondents from the Arab Gulf countries (Qatar, Oman, Kuwait, Bahrain, Saudi Arabia, and UAE) plus Libya and Sudan showed the highest willingness for vaccination, while those who showed the least willingness are participants from the west region (Algeria, Tunisia, Mauritania, and Morocco) (Figure 1). Arabic-speaking participants living in North America were more willing to receive vaccination than those in the other three clusters (Figure 1, Table 2).
Several factors (shown in Figure 3, Table 3) were tested in a binomial logistic regression model to examine their correlation with vaccine hesitancy. Univariate and multivariate analyses showed that almost all tested factors were significant predictors for vaccine hesitancy, reflecting the large sample size tested. Odds ratio (OR) showed the stronger effect of the following factors on participants’ hesitance: Never (OR, 4.04) or rarely (OR, 2.69) receiving the influenza vaccine, not knowing the vaccine type available (OR, 1.93), female gender (OR, 1.91), and outside of the health care system (OR, 1.84). Vaccine acceptance in each Arab country was correlated with the number of confirmed COVID-19 cases and deaths using Spearman correlation. It was found out that the number of cases (p=0.0047) but not deaths (p=0.3) correlated significantly with vaccine acceptance (Appendix 1—figure 3).
Barriers to acceptance
There were 3905 participants who chose acceptance but yet had one or more barrier(s) selected. Of the 29 barriers, the most common responses were ‘I am afraid side effects of the vaccine will develop, other than what has been disclosed’ – 22,235 (61.4%), ‘Not enough time has passed to verify the vaccine’s safety’ – 20,172 (55.7%), ‘The vaccine production has been rushed, making me doubt the credibility of the producing company’ – 16,698 (46.1%), ‘I do not trust the health care policies applied in my country’ – 14,151 (39.1%), and ‘I do not trust the published studies, nor the company producing the vaccine’ – 11,968 (33%) (Figure 4).
Comparison of participants inside and outside the Arab World
Participants in the Arab World were slightly more likely to have vaccine hesitancy when compared to those living outside (83.3% vs. 81.2%) (Figure 1). Those living in North America were the least hesitant (76.3%), while those living in Turkey had the highest hesitancy (83.6%). Additionally, participants living in Arab countries were more likely to report ‘I do not trust the health care policies applied in my country’, ‘There are no published studies on the vaccine’, ‘I do not trust the published studies, nor the company producing the vaccine’, ‘No need for the vaccine as rates of viral infection are decreasing’, and ‘No need for the vaccine as most people in my country have already been infected’ (chi-square test, p<0.0001, with difference >5% for all) (supplementary file 2, Figure 4—figure supplement 1).
Attitudes toward vaccination policies and need
When asked about their opinions regarding suggested vaccination policies, participants’ responses were to let people choose if they want to take it or not (59.5%); to mandate it on populations in which the vaccine was proven to be effective and safe as per clinical studies (13.6%), not sure (10.9%), should not be given to anybody (6.1%); and to give work and transportation privileges to whomever takes the vaccine (3.9%). When asked who needs the vaccine, responses were as follows: whomever – the vaccine was proven to be effective and safe as per clinical studies (35.4%), specific categories of people need it, but they are not the majority (30.5%), I don’t know (24.9%), and no one needs it (9.2%) (Figure 5).
This study presents the largest online survey on vaccine hesitancy that covered a heterogeneous population of Arabic people living all over the globe. In addition, it bridged the gap in knowledge on vaccine hesitancy in the Arab region. It shows low rates of vaccine acceptance in the face of the ongoing pandemic. Only one in eight respondents (12.5%) reported their willingness to take the vaccine. One in 22 (4.4%) based their decision on whether to take the vaccine or not on the type of the vaccine, acknowledging that the vaccine type they prefer may not be available in the near future and even when available, they may not be given the choice of selecting that vaccine type. These results are of unique significance because the study has been conducted after the vaccine has become available and administered to millions of people worldwide, and while about 70 different vaccine candidates are currently under development.
The study also showed a clear correlation between acceptance and gender, academic background, attitudes toward the flu shot, having been previously suspected of – or confirmed with – COVID-19 infection and knowledge of the vaccine type. Females were more hesitant to take the vaccine, while the previous studies were inconclusive on the correlation of hesitance with gender, where women were found to have higher (Fisher et al., 2020; Grech et al., 2020; Kreps et al., 2020), equal (Abdelhafiz et al., 2020), or lower (Lazarus et al., 2021; Ward et al., 2020) hesitancy compared with men. Our results show lower acceptance in participants with current or previous suspected or confirmed COVID-19 infection (data) when compared with Lazarus et al. (n = 13,426) who found no significant correlation (Lazarus et al., 2021). On the other hand, our results are consistent with the literature in terms of lower acceptance in people who do not get influenza vaccination (Lin et al., 2020) and who have lower academic education (Lin et al., 2020; Lazarus et al., 2021).
Respondents who did not know the vaccine type available to them showed increased hesitancy. This may be attributed to the fact that some Arab countries were first to approve the Sinopharm vaccine despite lack of affirmative data (Cyranoski, 2020). The impact of vaccine efficacy on attitudes toward vaccination has been echoed in the study of Harapan et al. (n = 1359) (Harapan et al., 2020), conducted before vaccine availability, where 93.3% of respondents chose to be vaccinated with a 95% effective vaccine, but this acceptance rate decreased to 67.0% in the case of a vaccine with 50% effectiveness. The results showed a level of mistrust in health care policies in Arab countries (44%) which can also be attributed to the selection of certain vaccines, as well as the inability to choose which vaccine to take. All of these factors may contribute to high hesitance when the vaccine type is unknown to the participant.
Consistent with previous studies (Harapan et al., 2020; Detoc et al., 2020), HCW were more accepting the vaccine, although with still low proportions of about one in four (18% yes and 7.1% depending on the type). One study on Congolese HCW (n = 613) conducted in March–April, 2020 reported a similar notably low rate of acceptance (only 27.7%) (Kabamba Nzaji et al., 2020). In addition, consistent with the results of the multinational study by Lazarus et al. (n = 13,426) (Lazarus et al., 2021), participants from countries with higher per million COVID-19 cases were more likely to welcome the vaccine. Participants between the ages of 30 and 59 were less willing to receive vaccination compared with older participants, an expected result given the fact that COVID-19 severity is associated with older age. However, younger participants (<29 years old) showed more willingness to be vaccinated.
Among the 29 different reasons for vaccine rejection/hesitancy, the top two reasons selected by the respondents reflected concerns about safety, while the next three most prevalent reasons were issues of distrust. This is consistent with the literature, which showed high levels of distrust and concern about safety (Lin et al., 2020).
The three forms of distrust (in health care policies, in vaccine expedited production and in published studies) were notably higher among respondents residing in the Arab countries than those living outside the Arab world. The same applies to the belief that the vaccine has not been tested on a large enough number of people, just tens or hundreds, which reflects less awareness of the vaccine development process in the Arab countries and highlights the need to educate the general public on the subject. Similarly, more residents of the Arab world believe that the vaccine is not necessary anymore because most people in the participant’s country ‘have already been infected’ or because the infection rate is decreasing. The infection rate is in fact decreasing (Appendix 1—figure 1A), but the public may need to be made aware that future outbreaks are still a possibility.
With the high rate of distrust, any form of coercion to take the vaccine may have negative impacts. Lazarus et al.’s large-scale study indicated that promoting voluntary acceptance is a better route and that coercion should be avoided (Lazarus et al., 2021). Similarly, a systematic review indicated that ‘mandates could increase resistance' (Lin et al., 2020). In our study, the majority of participants (59.5%) believed that vaccination should be left to individual choices and only a minority believed it should be mandated for certain categories of people (6.1%) or on populations in which the vaccine was proven to be effective and safe as per clinical studies (13.6%).
Approximately one-fifth of our respondents chose ‘The vaccine might lose its efficacy against the new viral strains’ as a reason for hesitation. The survey was published shortly after reports of the new viral strains in the UK and South Africa have been made public, and with the recent reports of decreased efficacy of some vaccines (Kabamba Nzaji et al., 2020; The Guardian, 2021; Knoll and Wonodi, 2021), this concern of efficacy is expected to increase among the public.
Several factors appear to contribute to the low level of vaccine acceptance in the current study compared with the previous works. First of all, the response to the question of willingness is broken down from Yes/No or Likert scale (in previous works) to a spectrum of choices which could more accurately detect the hesitant respondents who could have otherwise chosen (Yes) or (Agree/Strongly agree). In a large-scale survey conducted in October 2020 that included 18,526 adults across 15 countries (Ipsos, 2020), 73% strongly agreed or agreed that ‘if a vaccine for COVID-19 were available, I would get it’. However, of those, only 22% agreed that they would become vaccinated ‘immediately after the vaccine is available’, while some others chose that they would wait for a year and even longer. The same study found out that there is less certainty about getting vaccinated among those who will wait. Thus, the affirmative nature of the Yes/No and 5-point Likert scales do not seem to reflect the true nature of hesitancy and whether or not it changes its nature over time.
In 2015, the World Health Organization (WHO) Strategic Advisory Group of Experts on Immunization defined vaccine hesitancy as a ‘delay in acceptance or refusal of vaccination despite the availability of vaccination services’ (MacDonald, 2015). We, therefore, question considering those who intend to take the vaccine after a prolonged time of availability as ‘Accepting’ since this may interfere with the targeted achievement of collective immunity.
In the present study, 20.8% of the participants chose (I will wait and see its effects on others) – many of whom could have possibly chosen (Yes) or (Agree) had the waiting choice been eliminated. A systematic review noted that ‘When answer options included different timings for vaccination, more people chose to wait than get it as soon as possible’ and that the two answer choices (Yes/No) received relatively high affirmatives (Lin et al., 2020). Only 3 of about 70 studies and polls in this review included the choice of ‘wait a while until others have taken it’ for the question of vaccine acceptance. In these three polls, conducted before vaccine availability, the percentage of those who chose that they would take the vaccine as soon as they can (or as soon as possible) was low (21–28%). This indicates that vaccine acceptance may be overestimated in many studies and highlights the need to redefine vaccine acceptance in a uniform way among different studies.
Another factor that may explain the lower rate of acceptance observed in the study is the nature of our survey population. Social and political differences were found to have a prominent effect on COVID-19 vaccine acceptance, especially that many people assumed political interference in the vaccine and in the pandemic itself (Lin et al., 2020). Vaccine acceptance was lower in Arab countries in previous studies: Jordan (37.4%) (El-Elimat et al., 2020), Saudi (64.72%) (Al-Mohaithef and Padhi, 2020) and (44.7%) (Magadmi and Kamel, 2020), and in a small multinational survey that included several countries, mainly Jordan and Kuwait (29.4%) (Sallam et al., 2021).
The chronological analysis of vaccine acceptance and time – in local and multinational studies – does not show a linear relationship – if anything, the public’s acceptance can be best described as fluctuating. Several surveys conducted in the last third of 2020 have shown a decrease in vaccination acceptance compared with previous surveys (Lin et al., 2020; Ipsos, 2020; Wang et al., 2021; Kreps et al., 2020). For example, the intent to vaccinate has declined in 10 of the 15 countries from August to October 2020 (Ipsos, 2020). A systematic review of publications until 20 October 2020 showed declining vaccine acceptance (from >70% in March to <50% in October) with demographic, socioeconomic, and partisan divides observed (Lin et al., 2020). However, a more recent multinational survey conducted from November 2020 to mid-January 2021 in 15 countries (Imperial College London, 2021) and a study conducted from 28 to 31 January 2020 in 14 countries Ipsos, 2021 have both shown an increase in vaccine acceptance.
As for Arab countries, the more recent studies (Sallam et al., 2021; El-Elimat et al., 2020; Hussein, 2021) show lower acceptance rates than the earlier ones. For example, A study conducted in Egypt (n = 559) during March 2020 found out that about 73.0% were looking forward to getting the vaccine when available (Al-Mohaithef and Padhi, 2020; Magadmi and Kamel, 2020; Abdelhafiz et al., 2020). However, the more recent study on HCW in Egypt (n = 496) during December 2020 concluded that only 13.5% totally agree to receiving the vaccine and 32.4% somewhat agree (Hussein, 2021). In our survey that is more recent than the mentioned studies, 17.0% and 24.0% of participants in Egypt (general [n = 5339] and HCW [n = 1250], respectively) were willing to take the vaccine.
A Saudi Arabia study published in May 2020 observed that 64.7% of participants (n = 1000) showed interest to accept the COVID-19 vaccine if it is available (Al-Mohaithef and Padhi, 2020), and another study conducted during May (n = 3101) showed a 44.7% acceptance rate (Magadmi and Kamel, 2020). However, in a more recent study in Saudi Arabia too, published in December 2020, 31.8% of participants (n = 154) showed acceptance (Sallam et al., 2021). In our survey that is more recent than the mentioned studies, 19.8% of participants from Saudi Arabia (n = 3588) showed acceptance.
Health care workers (HCWs) are at increased risk of acquiring and transmitting COVID-19 infection. Moreover, they present role models for communities with regards to attitudes towards COVID-19 vaccination. Therefore, vaccine hesitancy in this group is of a special concern, and we discussed hesitancy in Arab healthcare workers elsewhere (Qunaibi et al., 2021).
This study comes with few limitations. Similar to several previous surveys (Lin et al., 2020), participants were recruited through social media. Being an online survey, our study may have under-represented certain groups of individuals, including members of older age groups and those who are not active on social media. We cannot rule out selection bias that might have affected our results. Other high-risk groups such as people with chronic diseases are well represented (n = 5839) or even over-represented (HCW, n = 5708). Our sample size was not pre-planned but was rather arbitrary reflecting a convenience sample. We believe that the large number of participants and the consistency of results in different countries that were geographically close and similar socioeconomically confirm the reliability of our survey.
Our results show high COVID-19 vaccine hesitancy among Arab respondents residing inside and outside the Arab world after millions of people around the world have received the vaccine. The main reasons for hesitancy are concerns about safety and distrust in health care policies, vaccine expedited production, and published studies, with the distrust being notably higher among respondents residing in the Arab countries. Given that the vaccine is being purchased from state expenditure, the high vaccine hesitancy could further compromise the economies of Arab countries in addition to the pandemic health hazard. At the same time, mandating the vaccine is not a desirable choice and could further increase the distrust. With the highly dynamic nature of the pandemic and vaccine production process and the interplay of ever-changing factors that affect vaccine acceptance, our study needs to be replicated at a later time to measure the change in public acceptance. The high proportion of people willing to wait until others have received the vaccine and the unavailability of the preferred vaccine for others show a need to create a uniform definition for vaccine acceptance in the surveys to avoid misestimation.
All data are available at our study website: https://mainapp.shinyapps.io/CVHAA. All the project's data and code has been deposited on Github. E. Qunaibi (2021) COVID-Vaccine-Arab-Survey. GitHub. https://github.com/MoHelmy/COVID-Vaccine-Arab-Survey. 5d1a881 (copy archived at https://archive.softwareheritage.org/swh:1:rev:5d1a8814d9dc1e3a87fb873ef994f9e0af2fce98).
Knowledge, Perceptions, and Attitude of Egyptians Towards the Novel Coronavirus Disease (COVID-19)Journal of Community Health 45:881–890.https://doi.org/10.1007/s10900-020-00827-7
Determinants of COVID-19 vaccine acceptance in Saudi Arabia: a Web-Based national surveyJournal of Multidisciplinary Healthcare 13:1657–1663.https://doi.org/10.2147/JMDH.S276771
An interactive web-based dashboard to track COVID-19 in real timeThe Lancet Infectious Diseases 20:533–534.https://doi.org/10.1016/S1473-3099(20)30120-1
Improving the quality of web surveys: the checklist for reporting results of internet E-Surveys (CHERRIES)Journal of Medical Internet Research 6:e34.https://doi.org/10.2196/jmir.6.3.e34
Attitudes toward a potential SARS-CoV-2 vaccine : a survey of U.S. adultsAnnals of Internal Medicine 173:964–973.https://doi.org/10.7326/M20-3569
COVID-19 data hubJournal of Open Source Software 5:2376.https://doi.org/10.21105/joss.02376
Acceptance of a COVID-19 Vaccine in Southeast Asia: A Cross-Sectional Study in IndonesiaFrontiers in Public Health 8:381.https://doi.org/10.3389/fpubh.2020.00381
WebsiteCOVID-19 vaccination intent is decreasing globallyAccessed February 26, 2021.
WebsiteGlobal attitudes : covid-19 vaccinesAccessed February 26, 2021.
A global survey of potential acceptance of a COVID-19 vaccineNature Medicine 27:225–228.https://doi.org/10.1038/s41591-020-1124-9
Once we have it, will we use it? A European survey on willingness to be vaccinated against COVID-19The European Journal of Health Economics 21:977–982.https://doi.org/10.1007/s10198-020-01208-6
WebsiteReal time world statisticsAccessed February 26, 2021.
Margaret StanleyReviewing Editor; University of Cambridge, United Kingdom
Eduardo FrancoSenior Editor; McGill University, Canada
In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.
The current SARS Cov-2 pandemic can only be effectively controlled by the delivery of effective vaccines to non-immune populations but a significant hurdle to this is vaccine hesitancy and the consequent failure to take up the offer of vaccination. To counter this public health authorities need to have robust data on the extent of vaccine rejection and the underlying reasons for this. The paper by Qunaibi and colleagues addresses this question in a large scale online study of over 36,000 participants from all the Arab speaking countries plus > 100 other countries with Arabs living in them. The data represent the largest study to date and the detailed questionnaires provide context behind the acceptance or rejection of a highly effective vaccine against a life threatening infectious disease. As one of the reviewers comments the study "provides important dataset for the health authorities in Arab and non-Arab countries to plan mitigation strategies to improve vaccine acceptance".
Decision letter after peer review:
Thank you for your submission entitled "A High Rate of COVID-19 Vaccine Hesitancy in a Large-scale Survey on Arabs". Your paper has been evaluated by two reviewers and by a Reviewing Editor and a Senior Editor. As is customary in eLife, the reviewers have discussed their critiques with one another and with the editors. We are pleased to inform you that your article has been accepted for publication in eLife.
Please take note of the points below, in case you may wish to revise the paper prior to production. We hope you will continue to support eLife.
The authors conducted a large-scale online survey that covers the region of the Arab countries and the Arabic-speaking people residing outside the Arab region. The main focus of the survey is to assess the vaccination acceptance in the Arab region accurately and in details. The work has multiple strength points:
1. The large sample size. The authors successfully recruited over 36K participants which makes this survey the largest worldwide, to date.
2. The detailed differentiation between the levels of acceptance/hesitation came in 5 different detailed levels, unlike other surveys that make them from 1 to 10 or yes/no.
3. The very detailed reasoning of the vaccine hesitancy. The authors asked the participants to detail their reasons for hesitation or rejection by giving them 29 different reasons that cover almost all possible hesitancy reasons.
4. The study is the first multinational study in the region. The Arab region was not studied at all, except a few local studies with limited participation. So, this study comes to fill a gap in the data about the region.
The results of the survey show a high rate of vaccination hesitancy, which comes in agreement with the escalated rates worldwide post-vaccination campaigns and news about unexpected side effects.
The manuscript "A High Rate of COVID-19 Vaccine Hesitancy 1 in a Large-scale Survey on Arabs" submitted by Eyad A. Qunaibi presents detailed description on hesitancy against COVID-19 vaccine with thorough analyses of factors influencing such hesitancy across the entire Arab world. The analysis conveniently included a small, yet reasonably-sized, sample of Arabs living in non-Arab countries, which provided a contrasting substance to compare with the main study. The manuscript is helpful in understanding the spectrum of barriers limiting people's acceptance of vaccination against COVID-19, and further highlights major barriers consistent with previously published literature, yet at a much larger scale.
Main barriers cluster around lack of knowledge that likely results in mistrust and worry. This is highlighted in the high proportion of participants not receiving the annual influenza vaccine and the concerns with the safety of COVID-19 particularly due to the speedy emergency use authorization of such vaccines.
One major value is the thoroughness of barriers included in the manuscript, which provides important dataset for the health authorities in Arab and non-Arab countries to plan mitigation strategies to improve vaccine acceptance.
The only optional recommendation that may improve on the quality of the manuscript, if data is available, is to understand similarities and differences in response to barriers between countries that show higher vs. lower hesitancy to COVID-19 vaccine. For example, was there a response pattern among Qatar, Oman, and Kuwait that clusters uniquely compared with response pattern among Tunisia, Algeria, and Morocco? Similarly, were there similar pattern of responses to barriers in countries with high acceptance such as in Qatar, Oman, and Kuwait compared with non-Arab countries of match acceptance such as North America? Such data may provide additional guiding details that will allow health authorities to devise more accurate policies to minimize such country-specific barriers and improve the rate of vaccine acceptance.https://doi.org/10.7554/eLife.68038.sa1
[…] The only optional recommendation that may improve on the quality of the manuscript, if data is available, is to understand similarities and differences in response to barriers between countries that show higher vs. lower hesitancy to COVID-19 vaccine. For example, was there a response pattern among Qatar, Oman, and Kuwait that clusters uniquely compared with response pattern among Tunisia, Algeria, and Morocco? Similarly, were there similar pattern of responses to barriers in countries with high acceptance such as in Qatar, Oman, and Kuwait compared with non-Arab countries of match acceptance such as North America? Such data may provide additional guiding details that will allow health authorities to devise more accurate policies to minimize such country-specific barriers and improve the rate of vaccine acceptance.
Thank you for the nice recommendation. Per the reviewer’s comment, we compared the response to barriers between countries that show higher hesitancy (Algeria, Mauritania, Tunisia) vs. lower hesitancy (Kuwait, Oman, Qatar) to COVID-19 vaccine. We found that the most prominent difference was in choosing the following barrier:
“I do not trust the healthcare policies applied in my country”.
The response rate to this barrier was as follows:
Countries with high hesitancy: Algeria 51.7%, Mauritania 46.5%, Tunisia 38.8%.
Countries with low hesitancy: Kuwait 18.5%, Oman 16.6%, Qatar 4.7%.https://doi.org/10.7554/eLife.68038.sa2
Article and author information
No external funding was received for this work.
Human subjects: This study was approved by the institutional review board (IRB) at King Hussein Cancer center (Approval No. 21 KHCC 0.34).
- Eduardo Franco, McGill University, Canada
- Margaret Stanley, University of Cambridge, United Kingdom
- Received: March 3, 2021
- Accepted: May 18, 2021
- Accepted Manuscript published: May 27, 2021 (version 1)
- Version of Record published: June 15, 2021 (version 2)
- Version of Record updated: December 7, 2021 (version 3)
© 2021, Qunaibi et al.
This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.
- Page views
Article citation count generated by polling the highest count across the following sources: Scopus, Crossref, PubMed Central.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
- Epidemiology and Global Health
Short-term forecasts of infectious disease burden can contribute to situational awareness and aid capacity planning. Based on best practice in other fields and recent insights in infectious disease epidemiology, one can maximise the predictive performance of such forecasts if multiple models are combined into an ensemble. Here, we report on the performance of ensembles in predicting COVID-19 cases and deaths across Europe between 08 March 2021 and 07 March 2022.
We used open-source tools to develop a public European COVID-19 Forecast Hub. We invited groups globally to contribute weekly forecasts for COVID-19 cases and deaths reported by a standardised source for 32 countries over the next 1–4 weeks. Teams submitted forecasts from March 2021 using standardised quantiles of the predictive distribution. Each week we created an ensemble forecast, where each predictive quantile was calculated as the equally-weighted average (initially the mean and then from 26th July the median) of all individual models’ predictive quantiles. We measured the performance of each model using the relative Weighted Interval Score (WIS), comparing models’ forecast accuracy relative to all other models. We retrospectively explored alternative methods for ensemble forecasts, including weighted averages based on models’ past predictive performance.
Over 52 weeks, we collected forecasts from 48 unique models. We evaluated 29 models’ forecast scores in comparison to the ensemble model. We found a weekly ensemble had a consistently strong performance across countries over time. Across all horizons and locations, the ensemble performed better on relative WIS than 83% of participating models’ forecasts of incident cases (with a total N=886 predictions from 23 unique models), and 91% of participating models’ forecasts of deaths (N=763 predictions from 20 models). Across a 1–4 week time horizon, ensemble performance declined with longer forecast periods when forecasting cases, but remained stable over 4 weeks for incident death forecasts. In every forecast across 32 countries, the ensemble outperformed most contributing models when forecasting either cases or deaths, frequently outperforming all of its individual component models. Among several choices of ensemble methods we found that the most influential and best choice was to use a median average of models instead of using the mean, regardless of methods of weighting component forecast models.
Our results support the use of combining forecasts from individual models into an ensemble in order to improve predictive performance across epidemiological targets and populations during infectious disease epidemics. Our findings further suggest that median ensemble methods yield better predictive performance more than ones based on means. Our findings also highlight that forecast consumers should place more weight on incident death forecasts than incident case forecasts at forecast horizons greater than 2 weeks.
AA, BH, BL, LWa, MMa, PP, SV funded by National Institutes of Health (NIH) Grant 1R01GM109718, NSF BIG DATA Grant IIS-1633028, NSF Grant No.: OAC-1916805, NSF Expeditions in Computing Grant CCF-1918656, CCF-1917819, NSF RAPID CNS-2028004, NSF RAPID OAC-2027541, US Centers for Disease Control and Prevention 75D30119C05935, a grant from Google, University of Virginia Strategic Investment Fund award number SIF160, Defense Threat Reduction Agency (DTRA) under Contract No. HDTRA1-19-D-0007, and respectively Virginia Dept of Health Grant VDH-21-501-0141, VDH-21-501-0143, VDH-21-501-0147, VDH-21-501-0145, VDH-21-501-0146, VDH-21-501-0142, VDH-21-501-0148. AF, AMa, GL funded by SMIGE - Modelli statistici inferenziali per governare l'epidemia, FISR 2020-Covid-19 I Fase, FISR2020IP-00156, Codice Progetto: PRJ-0695. AM, BK, FD, FR, JK, JN, JZ, KN, MG, MR, MS, RB funded by Ministry of Science and Higher Education of Poland with grant 28/WFSN/2021 to the University of Warsaw. BRe, CPe, JLAz funded by Ministerio de Sanidad/ISCIII. BT, PG funded by PERISCOPE European H2020 project, contract number 101016233. CP, DL, EA, MC, SA funded by European Commission - Directorate-General for Communications Networks, Content and Technology through the contract LC-01485746, and Ministerio de Ciencia, Innovacion y Universidades and FEDER, with the project PGC2018-095456-B-I00. DE., MGu funded by Spanish Ministry of Health / REACT-UE (FEDER). DO, GF, IMi, LC funded by Laboratory Directed Research and Development program of Los Alamos National Laboratory (LANL) under project number 20200700ER. DS, ELR, GG, NGR, NW, YW funded by National Institutes of General Medical Sciences (R35GM119582; the content is solely the responsibility of the authors and does not necessarily represent the official views of NIGMS or the National Institutes of Health). FB, FP funded by InPresa, Lombardy Region, Italy. HG, KS funded by European Centre for Disease Prevention and Control. IV funded by Agencia de Qualitat i Avaluacio Sanitaries de Catalunya (AQuAS) through contract 2021-021OE. JDe, SMo, VP funded by Netzwerk Universitatsmedizin (NUM) project egePan (01KX2021). JPB, SH, TH funded by Federal Ministry of Education and Research (BMBF; grant 05M18SIA). KH, MSc, YKh funded by Project SaxoCOV, funded by the German Free State of Saxony. Presentation of data, model results and simulations also funded by the NFDI4Health Task Force COVID-19 (https://www.nfdi4health.de/task-force-covid-19-2) within the framework of a DFG-project (LO-342/17-1). LP, VE funded by Mathematical and Statistical modelling project (MUNI/A/1615/2020), Online platform for real-time monitoring, analysis and management of epidemic situations (MUNI/11/02202001/2020); VE also supported by RECETOX research infrastructure (Ministry of Education, Youth and Sports of the Czech Republic: LM2018121), the CETOCOEN EXCELLENCE (CZ.02.1.01/0.0/0.0/17-043/0009632), RECETOX RI project (CZ.02.1.01/0.0/0.0/16-013/0001761). NIB funded by Health Protection Research Unit (grant code NIHR200908). SAb, SF funded by Wellcome Trust (210758/Z/18/Z).
- Epidemiology and Global Health
Affectionate touch, which is vital for mental and physical health, was restricted during the Covid-19 pandemic. This study investigated the association between momentary affectionate touch and subjective well-being, as well as salivary oxytocin and cortisol in everyday life during the pandemic.
In the first step, we measured anxiety and depression symptoms, loneliness and attitudes toward social touch in a large cross-sectional online survey (N = 1050). From this sample, N = 247 participants completed ecological momentary assessments over 2 days with six daily assessments by answering smartphone-based questions on affectionate touch and momentary mental state, and providing concomitant saliva samples for cortisol and oxytocin assessment.
Multilevel models showed that on a within-person level, affectionate touch was associated with decreased self-reported anxiety, general burden, stress, and increased oxytocin levels. On a between-person level, affectionate touch was associated with decreased cortisol levels and higher happiness. Moreover, individuals with a positive attitude toward social touch experiencing loneliness reported more mental health problems.
Our results suggest that affectionate touch is linked to higher endogenous oxytocin in times of pandemic and lockdown and might buffer stress on a subjective and hormonal level. These findings might have implications for preventing mental burden during social contact restrictions.
The study was funded by the German Research Foundation, the German Psychological Society, and German Academic Exchange Service.