Understanding hesitancy towards vaccination against SARS-COV2 among Health professionals in Tunisia

Background: Vaccines that were authorized for use in a wide brought a However, sucient coverage is conditioned by the people’s acceptance of these vaccines especially by health Indeed, they represent the of the against COVID-19. Several studies focused on this issue in developed countries. However, few were reported from developing ones including Tunisia. Objective: The current study aimed to estimate the prevalence and the predictors of hesitancy towards the vaccination against the SARS-COV2 among the Tunisian health professionals. Methods: A cross-sectional study was led online between the 7 th and the 21 th of January 2021 among Tunisian health professionals. A number of at least 460 participants was required. Snowball sampling method served to recruit participants. Data were collected using a pre-established and pre-tested questionnaire recorded in a free Google form. The link of the questionnaire was disseminated online to be self-administered anonymously to the participants. The generated online Google Sheet was uploaded and exported to SPSS software. Responses of non eligible participants were deleted before analysis.


Page 4/21
The new corona virus disease  has drastically altered people's lives (1). One year after declaring COVID-19 as Public Health emergency, the number of deaths caused by this new disease exceeded two millions worldwide (2). Measures such as lockdowns, social distancing, traveling restrictions and wearing protection tools for long time; limited people's freedom, trigged psychological issues, reduced the income of the disadvantaged groups and worsened the existing social and health inequalities (1,3). Until now, there is no speci c treatment to beat this new disease. Nonetheless, several vaccines were developed in a record time and were authorized for emergency use in a wide range of countries (4). Indeed, speed vaccination of people is required not only to cut the spread of the severe acute respiratory syndrome corona virus 2 (SARS-COV2) but also to tackle the emergence of new variants threatening the e cacy of these vaccines (5). However, hesitancy towards these vaccines represents a major barrier to obtain su cient vaccination coverage and to control the current COVID-19 pandemic (6).
Indeed, in 2019, the World Health Organization listed vaccine hesitancy as one of the top 10 threats to world Health (7).
Health professionals represent the leaders of their communities in term of adherence to COVID-19 vaccination (8). Their recommendations can bolster support for the vaccinations in the community (8). Furthermore, they represent a high risk group that must be prioritized in terms of vaccination as they are at the frontline of the war against the SARS-COV2 and may increase the spread of COVID-19 among the users of healthcare facilities (9).
Hesitancy towards COVID-19 vaccination among them would be therefore challenging for achieving coverage for population immunity. Reasons behind this reluctance may be related to the quick production of vaccines, concerns about long-term safety of the new vaccines, the politicization of vaccination, altruism towards higher-risk populations...etc.) (10,11). A recent scoping review of 35 studies, led mainly in developed countries, reported an average worldwide prevalence of hesitancy towards COVID-19 vaccination of 22.51% in healthcare workers with variable rates ranging from 4.3-72% (12). It is therefore relevant to lead more studies about vaccine hesitancy among health professionals especially in developing countries which are also suffering from scarcity of vaccines (13).
In Tunisia, a developing country that has undergone a rapid epidemiological transition thanks to its successful national vaccination program (14), is now facing a new challenge: To succeed in its vaccination strategy against SARS-COV2. In fact, there was a delay in obtaining vaccines doses (15).
Estimation of the prevalence and predictors of hesitancy towards SARS-COV2 vaccines among Tunisian health professionals would guide the national vaccination campaign against SARS-COV2 and orient international organizations for a more equal access to the vaccines.

Objective
To determine the prevalence and the predictors of hesitancy towards COVID-19 vaccine among Tunisian health professionals.

Study design:
A cross-sectional study was led online between the 7th and the 21th of January 2021 among Tunisian health professionals in order to evaluate their willingness to get vaccination against COVID-19.

Study population:
All Tunisian health professionals represented the target population. The following formula: n= [(Zα/2) 2 x p x (1-p)]/i 2 was used to calculate the required sample size. A proportion (p) of vaccine hesitancy towards COVID-19 vaccination of 50%, a precision (i) of 5%, a rst species risk (α) of 5% and a loss of 20% due to non-eligible participants were considered which gave a required sample of at least 460 participants.
Given that no updated national list of Tunisian health professionals with contact details was available in Tunisia, random sampling was not possible. Accordingly, the study was led using a snowball sampling. Initially, the investigators: an Associate Professor in Occupational Medicine, an Assistant Professor in Public Health, a Residency Trainee in Family Medicine and a Doctor of Dental Medicine working in different hospital wards, disseminated the survey online. They used their own mailing lists to send emails and their Facebook pro les to send messages and to post publications in the Facebook groups of Tunisian health professionals (62 Facebook groups were integrated by the investigators). In fact Facebook is the most popular social media in Tunisia (16). They targeted Medical Doctors, Pharmacists, Dentists, Health Technicians and Nurses who are tenured or in tenure-track. They also recommended to their colleagues in and out of their hospital wards as well as to the participants to disseminate the online survey. In total almost 2500 e-mails were sent with a daily sharing in 62 Facebook groups.

Data collection:
The investigators, based on their experience and a literature review, designed a questionnaire written in French, as it is the public administration and education language in Tunisia. The questionnaire included parts exploring socio-demographic (age, gender), professional characteristics ( eld of activity, position, sector, geographic location of health activities and direct contact with hospitalized COVID-19 patients), medical history (chronic disease, allergy, vaccination against in uenza for the current season), sources of information about SARS-COV2 vaccine, perceptions and attitudes related to the vaccination against SARS-COV2. The questionnaire was given to two other experts (a Public Health Professor and an Occupational Professor) who were familiar with the assessment methods of content validity. They evaluated the items with respect to appropriate wording, grammar, clarity, understandability and relatedness to Tunisian culture. They were also required to review the items with respect to their relevance.
The questionnaire was then pre-tested on a convenience sample of 30 health professionals to assess the acceptability and the understandability of the items. Unclear items and those that were di cult to understand by two or more health professionals were reformulated taking into account their comments and the experts 'opinion. The nal version of the questionnaire was recorded in a free Google form with two sections: one for the consent and one for the entire questionnaire. A question was added at the end of the form to determine whether the participant has responded to the same questionnaire previously in order to could identify duplicated responses. In order to limit missing data, all questions were mandatory to reply before sending the lled form. The link of the questionnaire was disseminated online to be selfadministered anonymously to the participants.
De nition of the hesitancy towards vaccination against SARS-COV2: To measure the hesitancy towards the vaccination against SARS-COV2, the following question was used: "When the vaccine against SARS-COV 2 (the virus responsible for the COVID 19 disease) would be available in Tunisia, will you accept to be vaccinated?" The possible responses were: "Yes, certainly", "Yes, probably", "I do not know yet", "Probably no", "Certainly no", "No I have already contracted the COVID-19", "No it is contra-indicated for me" and "Other response".
The responses: "Yes probably", "I do not know yet", "Probably no" and "Other response" were re-coded to "yes" to indicate hesitancy towards the SARS-COV2 vaccine. The other responses ("Yes, certainly", "Certainly no", "No I have already contracted the COVID-19", "No it is contra-indicated for me") were recoded to "no" to indicate no hesitancy.

Data Analysis:
The generated online Google Sheet was uploaded and exported to the Statistical Package for the Social Sciences (SPSS) 10.0 software (IBM Inc, Chicago, IL) for analysis. Responses of non eligible participants were deleted. Descriptive statistics were reported as frequencies for categorical variables and as means and standard deviations for quantitative ones. Differences between groups were examined using the Chisquared (χ 2 ) test to compare proportions.
To determine predictors of hesitancy towards vaccination against SARS-COV2, binary logistic regressions were performed. The dependent variable was "hesitancy towards vaccination against SARS-COV2"; all factors that were revealed associated with the dependent variable with a signi cance level less than 25% were included in a multivariable model. Then, a stepwise backward approach was used to identify predictors of hesitancy towards vaccination against SARS-COV2. Observations with missing data about some variables that were used in the different regression models were deleted. Results of the regression models were expressed as odds ratios (ORs) with con dence interval (CI) of 95%. All statistical tests were two-tailed, and p-values < 0.05 were considered statistically signi cant.
Ethical Considerations: The current study was carried out in accordance with the ethical principles of the Declaration of Helsinki. An introducing paragraph explaining the purpose and the conduct of the study preceded the two sections of the Google form. Anonymity of responses was highlighted. Participants had to give consent to access to the rest of the questionnaire by clicking on the response "yes" to the following question: "Do you agree to participate in the study?" In case of responding by "No", the rest of questions were not administered to the user of the link.
The response option "I do not want to answer" was added to the questions about the gender, the age and the medical history. Furthermore, to ensure anonymity, rst and last names were not collected and e-mail addresses were not collected.

Results
A total of 546 responses to the online questionnaire were obtained with 23 refusal and 523 acceptances. Among those who accepted to participate, 28 were not health professionals and two were not Tunisian. Accordingly, the retained participants accounted for 493.  were sure to accept it when it will be available in Tunisia.
Proportion of health professionals under the age of 40 years was signi cantly superior (72.3%) among those hesitating to get the vaccine than those not hesitating (59.5%) (p = 0.003). Similarly, females represented 74.6% of those who hesitate against 65.4% in those who do not with a p value of 0.047. Concerning the professional activity, having it in the North of Tunisia or in the public sector was signi cantly associated with more hesitancy towards the SARS-COV2 vaccine (Table 2). However, having been infected by the SARS-COV2 was negatively associated with hesitancy towards the vaccine (0.4% among hesitating participants versus 12.2% among those not hesitating (p = < 0.001)).   The current survey, led between the 7th and the 21th of January 2021, revealed that 51.9% (95% CI: 47.5-56.3) of the Tunisian health professionals hesitate to uptake vaccine against SARS-COV2. Being affected in the south or in the central of Tunisia, the female gender and the fear of components in the upcoming vaccines predicted more hesitancy among them. While a previous episode of SARS-COV2 infection and the use of the national site for information about COVID19 predicted less hesitancy among them.
Results of the current study should be interpreted with taking into account some limits. Firstly, the cross sectional nature of the study did not allow to report causal relationships but only statistical associations.
Besides, random sampling was not possible as no lists of national or regional health professionals were available. However, the required sample size was reached. In addition, the main categories of the health professionals were represented. Finally, attitudes and perceptions were self-reported by participants, this might lead to a social desirability bias. However, data were collected anonymously and participation was voluntary.
The hesitancy rate (51.9%) revealed by the current study was higher than that reported after an online opinion survey conducted almost at the same period (between the 10th and the 20th of January 2021) and which showed a lower hesitancy rate of 33.6% but higher refusal rate of 23.5% (17). This opinion survey was conducted by the "BEDER Association for Citizenship and Fair Development" with ve questions about the age, the eld of activity, the intention and the motivation to get the vaccine and its recommendation for others. Results of the survey were published on the website of the association (SAUVE.tn) while precisions about the methodology of this online opinion survey were not reported (17). Among our participants, 59.2% were physicians, 15.8% were dentists, 14.2% were pharmacists and 10.8% were from the paramedical stuff which was not far from the opinion survey (17). The French language of the questionnaire may explain the reluctance among some categories of the paramedical staff to respond to the survey. An available national updated contact list of the Tunisian health professionals is necessary to allow wider periodic evaluation of the willingness to get the vaccine against the SARS-COV2. In this way, policy makers would be able to adapt their information strategy.  (24). Analyzing hesitancy among participants according to the gender showed that female gender is a predictor of hesitancy among health professionals towards the SARS-COV2 vaccine. This result is harmonious with the majority of previous similar studies (12). The higher male acceptance of vaccine may be due to an innate male propensity for risk taking towards the novel vaccine (25).
Older respondents were signi cantly less hesitant to uptake the SARS COV2 vaccine. While having a chronic condition or allergy did not seem to contribute to this hesitancy among them. A recent scoping review reported that individuals of older age are more likely to accept COVID-19 vaccines (12). This was explained by a perception of greater vulnerability to SARS-COV2 infection but also by higher education and greater experience in healthcare (12).
Health professionals from different Tunisian regions responded to our questionnaire. Thirty-nine point eight were working in the north, 38.1% in the center and 21.3% in the south. Having its professional activity far from the north of the country (where is located the capital) predicted more hesitancy among participants. In line with this result, lower vaccination rates among deprived groups were observed in several surveys (18,20,26). More efforts should be provided in the Tunisian interior regions to overcome regional disparities in terms of vaccination against SARS-COV2.
Professionals from private sector were signi cantly less hesitant to get the SARS-COV2 vaccine. This joins the results of a study led in Hong Kong (27). This may be explained by economic reasons as in private sector sick leave in case of COVID19 episode is not regularly paid.
Having been previously infected by SARS-COV2 predicted less hesitancy among participants. A study conducted in Saudi Arabia among healthcare workers did not show signi cant association between previous personal SARS-COV2 infection and willingness to receive a COVID-19 vaccine (28). Otherwise, among Italian patients recovered from COVID-19, the majority were hesitant or undecided towardsSARS-CoV-2 vaccine (11,29). Similar result was reported in Chicago (30). Lack of knowledge concerning the duration of protection against the SARS-COV2 after infection may explain this uctuation between countries (31).
Among participants, 81.7% reported lack of information about SARS-COV2 vaccines. Social media was the most source of information reported by participants which joins the results of the Egyptian study (21).
Use of social media and lack of information about the SARS-COV2 vaccines were signi cantly associated with more hesitancy among participants. These results corroborate those in the healthcare workers of Egypt and Italia (11,21). Fear from harmful components was signi cantly associated with vaccine hesitancy among participants. In fact, doubts concerning the vaccines safety seems to be a global phenomenon that in uence vaccine uptake as it was mentioned in Italy (11) . Furthermore, proportion of health professionals that were vaccinated until now is not available. Although, there is scarcity of COVID-19 vaccines in Tunisia in addition to poor resources explaining the slowness of vaccination, an effective information strategy should be implemented as soon as possible. Facebook may represent a good channel for disseminating valid messages and tackling misinformation especially that Facebook is the most social media used in Tunisia (16). This would ensure rapid coverage of the population when the vaccines will be more affordable. Engaging health care professionals in social media to counter the vaccine related misinformation would boost the national information strategy. In addition, reporting the results of the pharmaco-vigilance surveillance would improve the vaccine acceptability among health professionals and the general population as well. More attention should be paid to female health professionals, the youngest ones and those in the regions far from the capital. A special network designed for the Tunisian health workers would facilitate access to them and getting feed back from them in return. The SAUVE.tn website may represent a suitable basis for this network.
Otherwise, regarding the emergence of new strains of SARS-COV2 and the slowness of the vaccination process, reinforcement of the non-pharmaceutical interventions is necessary until su cient coverage by vaccination will be reached in Tunisia. This underlines once again the necessity of an effective information strategy with multi-sectorial actions and a lobbying for a global COVID-19 vaccine equity.

Conclusion
The hesitancy rate towards SARS-COV-2 vaccine is high among Tunisian health professionals. An effective national information campaign represents the only solution to overcome this crisis and regain a normal life. International solidarity is strongly recommended to increase vaccine affordability in developing countries such Tunisia.