Knowledge, Attitudes and Uptake Related to Inuenza Vaccine Among Healthcare Workers in Tunisia

Background: The inuenza vaccine (IV) is considered the most effective strategy to prevent seasonal inuenza infection and annual vaccination of healthcare workers (HCWs) is recommended by the World Health Organization given their high mixing with patients. We assessed IV uptake among HCWs in the 2018-2019 season and explored their knowledge and attitudes regarding inuenza immunization. Methods: A cross-sectional study was conducted in 150 representative Tunisian health facilities from March to May 2019. We recruited 1,231 HCWs with direct patient contact using self-weighted multistage sampling. Univariate and multivariate logistic regression analyses permitted to assess the factors associated with IV uptake in the 2018 –2019 inuenza season. Results:

Given their close and regular contact with ill patients, healthcare workers (HCWs) are at a high risk of developing in uenza and may transmit the disease to their patients. As such, HCW infections may result in nosocomial outbreaks, with an increased risk of mortality among immunocompromised hospitalized patients [6]. In uenza infection among health professionals was also associated with a high economic burden mainly related to absenteeism [7].
The in uenza vaccine (IV) has been available since 1945 and remains the most effective tool to prevent in uenza infection and its complications [1,8]. HCWs immunization is a cost-effective method proven to reduce in uenza-related deaths among high-risk patients [9,10]. Furthermore, vaccination of HCWs can protect patients who cannot receive the vaccine or those who respond poorly to vaccination [11].
Given the aforementioned reasons, the World Health Organization (WHO) and the U.S. Advisory Committee on Immunization Practices (ACIP) have recommended annual vaccination of HCWs [1,11].
Despite these recommendations and the e cacy of HCW in uenza vaccination in improving self and patients' safety, vaccine coverage among healthcare professionals remains low mainly in developing countries [12,13]. Although the IV is provided free of charge to health professionals in Tunisia, the estimated vaccination uptake proportion among HCWs remains low [14]. However, this indicator has never been accurately assessed at the national level and its determining factors are not fully understood.
We performed this nationwide study to measure the vaccine uptake and to understand the knowledge and attitudes related to in uenza vaccination among HCWs, in order to propose evidence-based strategies to address these gaps in Tunisia.

Study design and population:
A cross-sectional study was conducted in Tunisian primary health care centers, regional and district hospitals between March and May 2019.
HCWs were recruited from 150 health facilities (66 in northern Tunisia, 62 in the center and 22 in southern Tunisia). The study included all HCWs with direct patient contact at participating health facilities. Direct patient contact was de ned as in-person, face-to-face contact between a healthcare provider and a patient, including patient registration, education, counseling, treatment, or any other aspect of patient health care.

Sampling process:
HCWs were recruited according to self-weighted multistage sampling. Stage 1 used strati ed sampling according to Tunisian regions (north, center, and south). Eight of the 24 Tunisian governorates were selected randomly: four in the north (Ariana, Ben Arous, Bizerte, Siliana), three in the center (Kairouan, Mahdia, Sousse), and one in the south (Gafsa) (Fig. 1). Stages 2 and 3 utilized strati ed sampling according to governorates and cluster sampling of healthcare facilities, respectively.
The calculated sample size was distributed according to the distribution of Tunisian HCWs between the three Tunisian regions, to the weight of each governorate in the corresponding region and to the distribution of physicians and other HCWs in each of the selected governorates.

Sample size:
The sample size was estimated for a 2.5 design effect and a 20% non-response rate, using Slovin's formula: n = N / (1 + N(exp) 2 ), N: Total population of the target group; exp: Desired precision.
Assuming a precision of 0.05 and expected target population sizes in primary health care centers, regional and district hospitals of between 10,000 and 100,000, the estimated sample size was ≃1200 HCWs.

Data collection:
A 27-question survey composed of two sections was administered to participants of our study. The rst section focused on HCW uptake of IV and their attitudes and knowledge regarding in uenza and IV. Openended questions were used to assess reasons for vaccine acceptance or refusal and knowledge of priority target groups for vaccination. The responses to general statements related to knowledge and attitudes were assessed using a 5-point Likert scale (1: strongly disagree, 2: disagree, 3: neither agree nor disagree /I don't know, 4: agree, 5: strongly agree). Respondents were also asked to rate their con dence regarding the ability of the IV to prevent in uenza among HCWs on a scale of 1 to 5 (from 1: not at all to 5: very much). The second section contained questions about HCW sociodemographic characteristics.
We conducted a pilot study to train investigators and assess the clarity and comprehensibility of the survey.
Trained staff approached HCWs to solicit their interest and consent. Those who agreed to participate were asked to respond anonymously to the survey questionnaire.
Agreements and approvals from the health authorities were sought prior to data collection to meet regulatory requirements and ensure maximal proportion's response.

Statistical Analysis:
HCWs were categorized as physicians, paramedics (nurses, assistant nurses, midwives and healthcare technicians), or other HCWs (healthcare assistants, administrative staff, psychologists, and pharmacists).
To facilitate the interpretation of study results, Likert scale responses to general statements related to knowledge and attitudes were dichotomized by grouping "I don't know" responses with "neutral," "disagree," and "strongly disagree," as "other responses" and "strongly agree" responses with "agree." Likewise, we dichotomized responses to the question related to con dence toward IV e cacy as follows: answers rated from 1 to 3 were assigned to low con dence while those from 4 to 5 were assigned to high con dence. "I don't know" answers were not included in the univariate analysis except for knowledge and attitude questions that were assessed using a 5-point Likert scale. χ 2 tests for univariate analysis and logistic regression for multivariate analysis were used to identify factors signi cantly associated with IV uptake in the 2018-2019 in uenza season among Tunisian HCWs.
Data were entered and analyzed using Epi Info version 7.2.2.6 (Developed by Centers for Disease Control and Prevention, U.S (US CDC)).

Iii. Results:
Overall, 1,264 questionnaires were collected. Among them, 33 were removed owing to high percentages of missing responses and not meeting the inclusion criteria. The remaining 1,231 questionnaires (97%) were eligible for analysis. Most of the participants were women (80.0%). Their mean age was 44.5 ± 9.3 years, ranging from 22 to 64 years. More details on the participants' profession and sociodemographic characteristics are presented in Table 1. Univariate analysis showed that HCWs belonging to primary healthcare centers were 2.4 times more likely to be vaccinated against in uenza in 2018-2019 than those working in regional and district hospitals (p < 10 − 3 ). We also observed a signi cant association between the participants' educational level and vaccination status (p = 0.005), with the highest percentage of vaccine uptake among those with the lowest educational level. Besides that, the vaccination rate increased with age (p < 10 − 3 ), professional experience (p < 10 − 3 ) and with the average number of patients seen per day (p = 0.038). Gender and type of occupation were not signi cantly associated with 2018-2019 vaccination uptake (Appendix A.1).
HCWs who highly trust the vaccine e cacy in preventing in uenza, and those who believed that HCW vaccination against in uenza could reduce severe illness and deaths in patients, as well as those who support a mandatory IV among HCWs were more likely to be vaccinated than their counterparts (OR = 3.5, p < 10 − 3 ; OR = 1.7, p = 0.001 and OR = 2.8, p < 10 − 3 respectively).
Respondents who were aware that annual in uenza vaccination is recommended for HCW had signi cantly higher vaccination rates than other respondents (p < 10 − 3 ) and vaccine uptake in 2018-2019 was higher among those who had received the vaccine at least once in the four preceding seasons (42.2% vs 5.0%) ( Table 2). The independent factors associated with IV uptake in 2018-2019 among Tunisian HCWs are summarized in Table 3. Iv. Discussion: We assessed IV coverage among Tunisian healthcare professionals working in primary healthcare centers, district and regional hospitals in the 2018- educational level, belief that the IV should be mandatory for HCWs, and con dence regarding IV e cacy in preventing in uenza.
The reported vaccine uptake among our respondents for the 2018-2019 season was much lower than those reported in studies conducted in some countries of the Middle East and North Africa (MENA) region, where the IV is provided freely to health professionals. In Saudi Arabia, two studies conducted in similar healthcare settings reported an increase in HCW vaccination rates between the 2012-2013 and 2016-2017 seasons -from 38-67.6% [22,23]. Similarly, a study conducted in a community hospital in Qatar found that more than half of health professionals were vaccinated against in uenza in 2011-2012 and exceeded 70% during the 2012-2013 in uenza season [25]. In addition, the administrative vaccination coverage of health professionals against in uenza in Morocco was estimated 54% in 2016 [26].
Despite international recommendations for annual vaccination of health professionals and the provision of IV free of charge to HCWs in Tunisia, the vaccination rate is low. This might be explained by the lack of appropriate u awareness campaigns in Tunisian health care facilities. Thus, health authorities should pay more attention to raise awareness of health professionals toward the necessity of in uenza immunization through annual educational programs that could be delivered online. Facilitating access to in uenza immunization by making time of vaccine delivery more exible should also be considered [27]. Reminder messages through social networks and media before and during the in uenza season may also help to increase vaccine coverage [27].
The most important predictor of IV uptake was the previous vaccination during the previous 4 years.
However, we did not observe a signi cant association between vaccination status and professional category. Surprisingly, HCWs with the lowest educational level were more likely to be vaccinated against in uenza in 2018-2019 compared to their counterparts. Our results are in contrast to those reported by Hammour et al [28], in which HCW vaccine uptake increased with educational level. In our study, the highest vaccine uptake among those with the lowest educational level might be explained by a lower concern about IV side effects than those with higher education level. Vaccine uptake in 2018-2019 was higher among healthcare professionals working in primary care centers than among those working in regional and district hospitals despite their more frequent contact with patients at high risk of complications. This could be explained by higher exposure of general practitioners to the recommendations of the national in uenza control program. Therefore, national educational programs should focus mainly on health professionals in contact with vulnerable patients.
Vaccine uptake among HCWs was associated with their willingness to recommend IV to their patients (p = 0.001). Our results are consistent with those of Joseph et al. [16] who reported a positive association between IV uptake among French general practitioners and vaccine coverage among patients aged 65 years and above. The low in uenza immunization coverage in Tunisia may be partly attributed to low level of con dence regarding vaccine e cacy which is translated in low vaccine acceptance among prescribers. In agreement with Petek et al [18], we observed an independent positive association between high con dence in vaccine e cacy and IV uptake in the 2018-2019 season (OR: 2.07).
Main barriers to vaccine acceptance for HCWs were fear of IV side effects, low perceived risk of severe in uenza disease and doubt about vaccine e cacy. As expected, these reasons corroborate those reported globally [17_19, 21,23,29] and support the presence of misconceptions regarding in uenza and in uenza vaccines among HCWs. Indeed, almost half of our study sample believed that IV could cause a person to develop in uenza. These results are in agreement with those from Saudi Arabia, where we found that 48.2% of health professionals believed mistakenly that the IV included live virus and could cause in uenza [23]. Although IV containing live viruses does exist, it cannot cause in uenza. Indeed, viruses contained in these vaccines are attenuated [30].
Despite the ACIP and WHO recommendations, less than one-third of Tunisian respondents identi ed HCWs as a target group for in uenza immunization. These results are in contrast with previous studies in India and Saudi Arabia, in which the majority of surveyed HCWs were aware that the IV was recommended to HCWs. [12,22,23].
Our results underscore the urgent need to educate HCWs about IV target groups and the vaccine composition. Tunisian vaccination awareness programs should also include information on the rates of IV side effects in addition to those of severe illnesses and deaths averted due to vaccination to raise HCW con dence regarding in uenza immunization. In addition, COVID-19 pandemic might be a great opportunity to promote IV. Indeed, COVID-19 and in uenza are both infectious respiratory diseases that have some symptoms and complications in common causing respiratory distress, hospitalization in intensive care units and even deaths mainly among vulnerable persons. HCWs encouraging the use of a vaccine against SARS-COV-2 just released after emergency use authorization, may nd it hard to justify their negative attitude regarding u shot that is already available for decades.
As educational programs alone may not be su cient to increase vaccine uptake among health professionals, the Healthcare Infection Control Practices Advisory Committee (HICPAC) and the ACIP recommend the requirement of a signed declination from HCWs if they refuse to receive the IV without having any medical contraindications [31].
Mandatory vaccination of Tunisian health workers may be a solution to increase the uptake of IV among HCWs; however, only 42.8% of the present study respondents agreed with this action. Other studies observed higher percentages of mandatory vaccine acceptance among healthcare providers [13,32].
Although more than half of respondents believed that vaccinating HCWs can reduce severe illness and deaths among patients, fewer than one-third mentioned patient protection as a main reason for vaccine acceptance. Indeed, self and family protection appeared to be more motivating than patient safety. These results are consistent with those of other studies, in which self-protection was identi ed as the main reason for vaccine acceptance [21,33].

Study strengths and limitations
To our knowledge, this is the rst study of its kind in Tunisia as well as in North Africa; thus, it provides benchmark data for use by health authorities to tailor and improve local IV strategies. Moreover, assessing IV acceptance among HCWs will help to forecast vaccine supply needs in order to avoid vaccine wastage and its related costs.
Given the COVID-19 pandemic, understanding barriers toward IV uptake among Tunisian HCWs may also help to implement effective awareness campaigns to promote SARS-CoV-2 vaccine.
Despite the fact that we conducted a national study among a randomized sample of Tunisian HCWs, we could not verify the representativeness of our sample since we don't have an updated information of the sociodemographic characteristics of health professionals working in primary healthcare centers, regional and district hospitals in Tunisia overall. Besides that, for feasibility reasons, we did not include health professionals from the private sector and those working in university hospitals which indicate that our ndings are not generalizable to all Tunisian HCWs.
Future studies focusing on the tertiary care level health professionals and those in the private sector are planned to get a more comprehensive picture.

V. Conclusions:
Despite recommendations, the vaccination rates were low among Tunisian health professionals. The low vaccination uptake may be related to a lack of con dence regarding IV e cacy and misconceptions about in uenza immunization. These ndings highlight the need for educational programs to raise HCWs' awareness of vaccine e cacy and safety. Mandatory vaccination policies in healthcare facilities may also be considered. Findings from the present study can be useful to overcome potential barriers against the uptake of COVID-19 vaccine among HCWs which are identi ed as a priority group. Distribution of selected health facilities by governorates. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.