1. INTRODUCTION

According to the World Health Organization (WHO), influenza epidemics result in 250,000 to 500,000 deaths annually worldwide [1]. In 2009, the H1N1 influenza pandemic posed a worldwide public health threat [2]. Although the WHO International Health Regulations Emergency Committee declared an end to the global pandemic in 2010, the virus itself continues to be transmitted during seasonal outbreaks [3].

Person-to-person transmission of influenza via droplets is common inside and outside health care facilities. Healthcare Workers (HCWs) are at higher risk of occupational acquisition and spreading to healthy staff and vulnerable patients [4,5]. Elderly persons, children, immunocompromised individuals, and pregnant women are at increased risk of infection, leading to several complications and potential deaths [4].

Influenza vaccination decreases the risk of infection in the population, prevents nosocomial infections, decreases morbidity and mortality among patients, and minimizes absenteeism among HCWs [5–7]. It has been estimated that implementing a vaccination program with 100% adherence for all individuals at risk across the European Union would save 15 million euros spent on primary care visits and approximately 1.6 billion euros due to reduced hospitalizations [8]. Annual Influenza vaccination has been recommended by the US Centers for Disease Control and Prevention (CDC) since the early 1980s ( https://www.cdc.gov/vaccines/adults/index.html). Nevertheless, coverage remains largely inadequate reaching 65% of the elderly US population [9] and only 35.4% of the at risk population in the European Union [3].

Seasonal Influenza vaccination policies targeting healthcare workers, pregnant women, children above the age of 6 months and elderly and others are implemented in different Eastern Mediterranean countries including Lebanon, Saudi Arabia, Qatar and others. Despite the above policies, vaccination coverage remains low. For instance, vaccination rates in the United Arab Emirates (UAE), Kuwait, and Oman have been reported at 24.7%, 67.2% and 46.4%, respectively [10] while in Turkey, various studies have cited rates ranging from 5.9% to 27.3% [11]. Among HCWs in the middle east region, vaccination rates have varied between 12.7% and 23.1% in Turkey [12], and a similar study in Saudi Arabia reported a vaccination rate of 41% [13].

In this study, we aim at identifying factors associated with vaccine use or refusal, in addition to knowledge, perceptions, attitudes and behavior toward influenza vaccination among three different populations: healthy college students, HCWs (physicians and registered nurses), and clinic patients. The rationale behind targeting these three populations was to capture subjects with different and distinct backgrounds, namely young healthy adults with college students, older adults with various health statuses with clinic patients, and professionals with background knowledge on influenza with HCWs. To the best of our knowledge, this study remains among the few studies of its kind to be conducted in the middle east region in recent times and is the first in Lebanon.

2. MATERIALS AND METHODS

3. RESULTS

4. DISCUSSION

In this cross-sectional study, self-reported vaccination was low among patients, students and HCWs in Lebanon, reaching no more that 36% among HCWs, despite agreement about the seriousness of influenza and a high perceived likelihood of contracting the infection. This finding is validated by the University Health Services data reporting a vaccination uptake of 38.5% and 33.4% in 2016 and 2017, respectively. Compared to neighboring countries, our HCWs vaccination uptake falls within regional data and ranks lower than Kuwait (67.2%), Oman (46.4%) and Saudi Arabia (41%) [10,13]. In contrast, they supersede those reported in Turkey (12.7–23.1%) and the UAE (24.7%) [10–12]. HCWs overestimate their knowledge about influenza while scoring lowest on the objective knowledge component. Low levels of knowledge about the vaccine were previously shown to cause lower vaccination rates in nurses compared to physicians [5,14,15]. The low rates of adherence with vaccination among physicians may be traced back to a tradition of neglecting their own health in favor of their many professional and personal obligations [16]. The relatively small number of HCWs in this study does not allow for subgroup analysis to delineate differences between nurses and physicians. A recent study in Pakistan showed low vaccination rates among physicians against influenza (8.84%), with lack of awareness about the vaccine and the recommendations being the major obstacles for adherence [17].

Vaccination uptake was lowest among students (10.2%), with a significantly lower likelihood of regular vaccination (p = 0.001), despite responses of high likelihood of contracting influenza (75.0%). Similar low rates of influenza vaccination among students were reported in studies from European countries [18,19]: students could perceive themselves as healthy individuals susceptible to influenza but at lesser risk for its complications. They also visit healthcare providers less frequently. In fact, people who view themselves as healthy are usually less likely to take the vaccine, while those with poorer self-rated health and higher co-morbidities have a higher vaccination rate [20].

Previous studies have shown that among HCWs, adherence to vaccination manifested in their recommendation to patients [5,15,21,22]. In our cohort, HCWs demonstrated the lowest objective knowledge scores despite reporting a significantly higher level of self-rated knowledge. These results are supported by another study showing that while 85% of nurses believed they received the necessary information to make an adequate decision about influenza vaccination, only 9.6% correctly answered knowledge questions [23]. During the last H1N1 pandemic (2009–2010), adequate basic knowledge of HCWs about the vaccine was strongly associated with recommending it to patients [24] and correlated with higher rates of vaccination in older community-dwelling adults [25].

A distrust regarding the vaccine is prevalent in our study with doubts regarding its effectiveness and ambivalence as to its safety. A third of participants stated fear of potential adverse effects as a deterrent from vaccination. Many studies have recently investigated adverse effects of the seasonal influenza vaccine, with concerns being raised regarding potential effects on pregnancy [26–28], neurological diseases [29–31], and potential solid organ transplantation rejection [32]. These studies have not been conclusive enough, and the CDC has not modified its recommendation of yearly vaccination. Considering these doubts and misconceptions, HCWs could act as a source of accurate information regarding the vaccine efficacy and safety. In fact, 77% of participants stated their willingness to take the vaccine if recommended by a physician. Studies in the United Kingdom, France, Germany, and Mexico showed that HCWs and the general population alike consulted their practitioner before obtaining the free-of-charge influenza vaccine [19,24,33]. The lack of recommendation by a physician was further shown to significantly hinder immunization [24,33,34].

Unlike the objective knowledge score, a self-reported knowledge of “very good” was the strongest independent predictor of vaccination (OR = 16.3; 95% CI 1.4–194.2). Therefore, perceived rather than objective knowledge seems to drive participants to vaccination. Efforts should however be centered on bridging the gap between the self-reported knowledge level and actual knowledge among HCWs.

The positive association between having children and vaccine uptake is a new finding that has not been addressed in previous studies. It can be attributed to the sense of responsibility individuals feel toward their children. This aspect could be further exploited in awareness campaigns using “protect your children” as a message.

Overall, having a positive attitude toward vaccination was associated with increased uptake of regular or yearly vaccination. Patients who reported not being afraid of the flu were 80% less likely to receive the vaccine, and this was supported by other studies in healthy adults and HCWs [21,24]. Thus, clarifying the potential severity of influenza should be a key element in future vaccination campaigns. In a prior assessment of influenza vaccine uptake at our institution, having a positive attitude consistently predicted regular vaccination in individual sub-groups and across all participants (OR = 38.7; 95% CI 38.7–38.8). These determinants have been echoed in regional studies performed in Turkey and several Arab countries mentioned above [10–13], thus highlighting a possible cultural context to be considered while building future campaigns.

In a recent study by Yeung et al. [35], a systematic review of 23 articles assessed factors associated with yearly influenza vaccination uptake. Perceptions about vaccine efficacy (OR 2.7–10.5), safety and adverse events (OR 10.5) were more influential than the level of knowledge itself. Advice from doctors was also a key predictive factor of vaccination. Moreover, in the US, individuals with higher education, better income, and insurance coverage have higher vaccination rates [17,36].

Based on our findings, we propose a model for a locally effective influenza vaccination awareness campaign using the framework of the Health Belief Model (HBM) [37]. Such an approach was undertaken in a Chinese study to provide a description of the high-risk sexual behavior in men who have sex with men [38]. The model presents a framework of constructs that, when adapted to the influenza disease and vaccination, consists of the following elements: (i) perceived susceptibility (increasing awareness of risk); (ii) perceived severity (stressing potential complications); (iii) perceived benefits (protection of self and others, especially children); (iv) perceived barriers (accurate depiction of vaccine adverse events and effectiveness); and (v) cues to action (using frequent reminders). The model is based on the assumption that a person’s belief in a personal threat of an illness or disease together with his/her belief in the effectiveness of the recommended health behavior will predict the likelihood the person will adapt the behavior. We believe the HBM can help explore or identify approaches that may be used to increase adherence to vaccination. Even though multiple factors in our surveys were shown to correlate with vaccination, we believe that these factors require a further exploration through descriptive qualitative study designs using focus groups and in-depth interviews.

The current vaccination uptake is higher compared to our 2010 findings but remains significantly lower than US data (68.5% for HCWs and 40% for the general population) and data from neighboring countries, namely Kuwait, Oman and Saudi Arabia [10]. These rates are higher than earlier findings whereby adherence rates were 25.5% and 7.1% for HCWs and students, respectively (data from the University Health Services at AUB).

Whether vaccination should be mandatory to all HCWs remains debatable: among 97 facilities in the US, there was a 25.5% increase in vaccination rate between pre-mandates and post-mandates in hospitals with consequences for non-adherence, versus 15% in hospitals without consequences [39]. Consequences ranged from wearing a mask during the influenza season to termination. Currently, AUBMC adopts a strategy of strongly encouraging influenza vaccination rather than making it compulsory.

Our study is limited by the small sample size and low response rates from HCWs. This raises two important potential biases: the under coverage of the population, and the non-response bias, especially given the low response rate from HCWs. Participants self-reported their vaccination status, and one cannot rule out the possibility of over reporting the vaccination uptake as a manifestation of social desirability. Our population consists of a convenience sample that may not be representative of the larger population of students, patients and HCWs at our institution. Therefore, large multicenter and population-based studies are needed to produce generalizable data. Lastly, and despite the measures taken to ensure confidentiality, bias based on social desirability and conformity especially amongst students cannot be ruled out given the study design.

5. CONCLUSION

Low uptake of influenza vaccination remains a global problem. In this study, we identified three major independent predictors of regular or yearly vaccination uptake: having children, having a “very good” self-reported knowledge of influenza, and being afraid of contracting the flu virus. These, along with socio-demographic background should be considered to develop an efficient national influenza vaccination campaign. Based on our findings we have presented a model for a locally effective campaign based on the HBM framework.

CONFLICTS OF INTEREST

The authors declare they have no conflicts of interest.

AUTHORS’ CONTRIBUTION

Project inception: NE, HH and ZK. Data collection: KC, JE, NE, HH, NS, MY and AR. Statistical analysis: KC, SA, NE and ZK. Write-up: KC, NE, HH and ZK. Manuscript revision: SK and ZK.

FUNDING

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

ACKNOWLEDGMENT

The authors would like to acknowledge Dr. Jamil Kazma for his help in formatting and finalizing the manuscript.