Health Belief Model in studies of influenza vaccination among health care workers

Vaccines prevent numerous infectious diseases, including influenza. Despite their significant contribution to controlling influenza, vaccine coverage against this disease varies among health care workers. The Health Belief Model could thus help to understand the reasons why these workers accept (or not) the immunobiological. The aim of this study was to describe the main results of research performed on influenza vaccination among health care workers using the Health Belief Model. This is an integrative literature review. Data search took place in October 2020 in the PubMed database, with the following descriptors: “influenza vaccine“; “health professionals”; “Health Belief Model,” and their synonyms. Eleven studies were included in this review. The main dimensions of the model (susceptibility, severity, benefits, and barriers) were more explored by the studies, and self-efficacy was the least studied dimension. Moreover, we observed a relationship between the theory’s dimensions (susceptibility, severity, benefits, barriers, cues to action, and self-efficacy) and influenza vaccination in health care workers. In conclusion, this review identified profiles of beliefs for each dimension of the Health Belief Model, which has traditionally been an ally for determining refusal or acceptance of the influenza vaccine among health care workers.


INTRODUCTION
To this day, vaccination is the most effective and low-cost method for preventing numerous infectious diseases.Since their discovery, they have been saving countless lives and contributing to improvements in health and well-being worldwide. 1The World Health Organization 2 recommends immunization against influenza primarily for the following target populations: pregnant women, children aged 6 to 59 months, older adults, individuals with chronic diseases, and health care workers.
The influenza virus is capable of producing a flu-like syndrome with an epidemic character and high morbidity, presenting high hospitalization rates among older adults and people with chronic diseases.Its transmission occurs mainly through droplets released by infected individuals when talking, coughing, or sneezing, and also through contact with one's secretions; immunization is the best preventive strategy. 3,4herefore, influenza vaccination in health care workers presents the following benefits: decreased infection rates among workers; decreased possibility of contaminating the users of health systems due to reduced transmission at health care facilities; reduced probability of infecting the workers' family members; and reduced absenteeism, which would benefit employers. 4,5espite its significant contribution to controlling influenza, vaccine coverage against this disease among health care workers varies.Notably, average coverage rates are seen in Europe (29.5%) 6; in Israel, vaccine coverage is 42% 7 ; in the United States, 78.4% 8 ; in Brazil, 91.25%. 9ome studies indicate aspects that contribute to greater adherence to influenza vaccination, such as: offering vaccines free of charge; with convenient access, including at health facilities; knowledge on the disease and production process of immunobiologicals; belief in vaccine effectiveness; and risk perception. 5,10n the 1950s, the Health Belief Model (HBM) was formulated by social psychologists based on two theoretical sources: the theory of cues to action (CTA) and the cognitive theory, for exploring different health-related behaviors.According to the HBM, for an individual to act preemptively, he or she needs to believe that: a) he or she is susceptible to an undesirable disease or condition; b) its occurrence presents severity to some aspect of his or her life; c) performing a prophylactic intervention is effective for reducing disease susceptibility or severity (benefits); and d) the preventive action does not imply in too many barriers. 11,12he HBM was further reformulated by Rosenstock et al., 13 when two other categories were included: e) cues to action; and f) self-efficacy or health motivation. 11,12Nowadays, HBM is useful for understanding influenza vaccine acceptance or not among different categories of workers in the health care sector. 14his model was used both in the study of the reasons why health care workers accept influenza vaccines (or not) 15 and in the development of interventions for improving adherence. 16onsidering this reality, the aim of this study was to describe the main results of research performed on influenza vaccination among health care workers using the HBM.

METHODS
This integrative review was based on the methodological assumptions of Whittemore & Knafl, 17 being divided into the following stages: problem identification, literature search, data evaluation, data analysis, and presentation of the integrative review.
For identifying the problem, this study had the following research question: "What are the results presented by studies on influenza vaccination in health care workers through the use of the HBM?" For answering this question, our search was performed in October 2020 in the National Library of Medicine (PubMed) database with the following Medical Subject Headings/Descritores em Ciências da Saúde (MeSH/DeCS): "influenza vaccine"; "health professionals"; "Health Belief Model" and their synonyms (in English).We used the Boolean operator AND, aiming to reach the content of all established descriptors.
The following inclusion criteria were employed: original articles with full text available, regarding the HBM applied to influenza vaccination in health care workers.We highlight that no restrictions of for time since publication were established, in order to broaden the number of studies being reviewed.
In the first search, we selected 319 studies.Therefore, we used the Rayyan tool for evaluating the obtained data. 18Considering the inclusion criteria and after reading the titles and abstracts, we excluded 213 studies and a total of 106 articles remained.There were no duplicates; however, 6 studies could not be fully retrieved.The full text of these 100 studies was then analyzed and, after applying the aforementioned criteria, 11 studies were selected for this review, as shown in Figure 1.
The articles were classified according to the level of evidence proposed by the Agency for Healthcare Research and Quality (AHRQ). 19This categorization qualifies articles into six levels, as follows: level 1meta-analysis of controlled studies; level 2 -study with experimental design; level 3 -study with a quasiexperimental design, such as a non-randomized or casecontrol study; level 4 -study with a non-experimental design, such as descriptive correlational and qualitative research or case studies; level 5 -case report or data obtained systematically; and level 6 -expert opinions.
For data collection and organization, we used an instrument constructed by the authors with two sets of questions: the first one comprised items related

Reasons for exclusion n 213 ( = )
Reasons for exclusion n 95 ( = ) Figure 1.Flowchart of the article selection process for the integrative review, according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist. 18o article identification (title, authors, and year of publication), place (country), research participants, and type of influenza; and the second contained the identification of HBM dimensions and the main conclusions regarding them.Finally, we presented the results, with a synthesis of findings about HBM on influenza vaccination among health care workers, which were systematized into two categories: the main HBM dimensions and the two new HBM dimensions.

RESULTS AND DISCUSSION
Out of the 11 selected articles (Table 1), 10 studies used a quantitative approach [20][21][22][23][24][25][26][27][28][29] and only 1 employed a qualitative method. 30Considering the levels of evidence, all studies were classified as level 4. The articles were published between 2005 and 2020, but mainly on 2019, which presented three publications. 21,22,29The sample sizes of the studies varied between 30 and 3,971 workers, with a mean value of 1,328 and a median value of 601.
As to the country where the studies were conducted, one of them was carried out in Jordan, 20 one in Brazil, 21 one in Singapore, 22 two in Israel, 23,24 two in Canada, 25,26 two in the United States, 27,28 and one in Greece. 30In addition, one study had a multicenter design (Hong Kong, Singapore, and Brunei). 291][22]26 Regarding the type of influenza, most articles considered the seasonal 20,22,23,[25][26][27]29 and pandemic types.20,[25][26][27]30 Some studies also mentioned avian flu, 27,30 and other three articles did not specify the flu type; however, considering the date of data collection, we may infer that it was seasonal influenza.21,24,28

MAIN HBM DIMENSIONS: SUSCEPTIBILITY, SEVERITY, BENEFITS, AND BARRIERS
Most articles (82%) presented in this review approached the four main dimensions of the HBM.However, one of them did not analyze the perceived barriers dimension, 29 and another one did not present perceived severity 28 (Table 2).
The susceptibility category may be defined as the perceived probability of contracting a disease or an undesired condition. 11,12All quantitative studies associated this dimension with higher influenza immunization rates, whereas the conclusion reached by the qualitative study 30 was that most participants in the focal group were not vaccinated and a large proportion did not consider themselves susceptible to the flu.
The literature review performed by Hofmann et al. 31 on influenza immunization in health care workers indicated that the lack of perceived susceptibility increased vaccine hesitancy.Moreover, the systematic review by Corace et al. 14 on interventions based on structures such as HBM for increasing vaccination rates among health care workers also concluded that vaccinated workers were more likely to report that they were susceptible to influenza.
It is important to highlight that, during their work activities, health care workers are exposed to multiple and varied risks related to chemical, physical, biological, psychosocial, and ergonomic hazards, of which biological hazards are the main generators of threats to workers' health.These workers interact with several pathologies such as the flu, which exposes them to varied health risks and hazards, making them a vulnerable group. 32he severity dimension can be defined as the perceived disease severity, that is, the medical and clinical consequences -such as death, disability, and pain -and the social consequences -such as effects on work and social relationships. 11,12egarding the perceived severity, six studies associated it to vaccination [22][23][24][25][26][27] ; in two articles, the severity of the flu was acknowledged, but no advances were made in the severity/immunization relation. 20,21wok et al. 29 did not find an association between this dimension and vaccination, and participants of  The workers who intended to get vaccinated had a significantly higher probability of recognizing the increase in susceptibility of their ill patients, in addition to their own risk.
The participants recognized the severity of the flu and its complications.
The group of workers who intended to receive the vaccine endorsed its benefits.
The participants who did not intend to receive the vaccine were more likely to perceive barriers as detrimental.
A2 21 The knowledge that the influenza vaccine does not provide protection for many years was associated with vaccination, possibly because these individuals feel more vulnerable to influenza.
Most participants recognized the severity of the flu.
A statistically significant association with vaccination was observed for recognizing the vaccine as safe and a borderline association for acknowledging vaccine efficacy.
Not fearing adverse events was associated with vaccination.
A3 22 Having family members aged less than 16 years was negatively associated with vaccination, possibly because these young families had other priorities and/ or believed they were not at risk of contracting the flu, since they were usually healthy.
The belief in a potential severity of the flu was a predictor of vaccine acceptance.
The belief in vaccine safety was a predictor of vaccine acceptance.
Adverse effects were the main reason for not receiving the vaccine.
A4 29 An association between susceptibility and vaccine uptake was observed.
No relationship was observed between severity and vaccination.
An association between benefits and vaccine uptake was observed.
-A5 23 Significant correlations were observed between susceptibility and willingness to receive the vaccine.

Significant correlations were observed between severity and the willingness to receive the vaccine.
A strong significant correlation was observed between perceived benefits and the intention of getting vaccinated.
A significant correlation was observed between perceived barriers and a low intention of receiving the vaccine.
A6 25 A significant association was observed between personal susceptibility and receiving the vaccine.
A significant association was observed between perceived severity and vaccination.
The vaccinated interviewees were more likely to agree that the vaccine's risks outweigh its benefits.
An association was seen between perceived barriers and a reduction in the probability of vaccination.
A7 26 Vaccinated individuals were more likely to perceive a high personal risk for influenza.
Those who accepted the vaccine were more likely to perceive influenza as personally dangerous.
The confidence in vaccine safety promoted its acceptance.
Unvaccinated workers were more likely to cite barriers to vaccination.
A8 27 Increased perceived susceptibility was a predictor of vaccine acceptance.
The severity of the flu was one of the reasons for vaccine acceptance.
Favorable beliefs regarding vaccine efficacy were predictors of vaccine acceptance.
One of the reasons for vaccine hesitancy was its unavailability at convenient hours.
A9 24 Vaccinated workers perceived themselves as more susceptible to the disease.
Vaccinated workers perceived the flu as a more severe disease.
The perceived benefits were predictors of vaccine acceptance.
Vaccinated workers perceived less barriers.A10 30 Most participants were not vaccinated, and a large portion did not consider themselves susceptible to the flu.
The participants acknowledged that influenza could cause a global outbreak of severe disease.
The participants acknowledged that vaccinating health care workers could protect patients.
The lack of vaccine availability and distribution were the main reasons for vaccine hesitancy.
A11 28 One of the reasons for receiving the vaccine was the concern about being at risk of exposure.
-The perceived benefits were the main contributors for vaccine acceptance.
One of the reasons for vaccine hesitancy were concerns about side effects.
the qualitative study conducted by Raftopoulos 30 indicated that influenza could cause a global outbreak of severe disease, even though most participants were not vaccinated.This way, we note that the association between perceived severity and vaccination was not unanimous.
We highlight that, sometimes, the acknowledgment of one of the dimensions is not translated into action, as it was shown in the qualitative study of a review. 30his was also demonstrated in the publication by Hidiroglu et al., 33 where a focal group with 33 health care workers in Turkey concluded that, although the participants considered themselves at risk of contracting H1N1, most of them had not received the available immunization.
In addition, other authors also confirmed that perceived severity increased vaccine acceptance: the rapid evidence assessment by Jenkin et al., 4 comprising 60 publications and aiming to examine data on the flu in health care environments and impacts of the vaccination of health care workers against influenza; and the mixed study performed in Kenya, 10 which aimed to evaluate the knowledge, attitudes, and practices on infection and vaccination among health care workers against H1N1 and noticed that health care workers were more likely to accept vaccination when they believed H1N1 could lead to death.
Another dimension analyzed in different publications are the perceived benefits, which may be defined as the belief in the efficacy of a prophylactic intervention in reducing disease susceptibility or severity. 11,12The results of this review indicate that authors who used a quantitative methodology associated perceived benefits to influenza vaccination.However, the article by Raftopoulos, 30 with a qualitative methodology, revealed that the participants of the focal group acknowledged that vaccinating health care workers could protect patients, even though most of them were not vaccinated.The perceived benefits dimension also presented an association in other similar studies. 4,10,14t is important to note that health care workers are at greater risk of infections by influenza when compared to the general population due to their close contact with infected patients at health care facilities. 34Nevertheless, the vaccination of health care workers brings, as benefits, reductions in morbidity and absenteeism related to influenza in these workers, contributing to a safer work activity. 5erceived barriers can be defined as potentially negative aspects of the preventive action, such as the cost and pain. 11,12All studies that adopted the quantitative method associated perceived barriers with lower immunization rates, and the qualitative 30 study noticed that the lack of vaccine availability and distribution were the main reasons for vaccine hesitancy.
The study by Hidiroglu et al., 33 with 689 health care workers and which aimed to determine the knowledge and opinions on the influenza vaccine and its acceptance in the Medical University of South Carolina, also confirmed that the perceived barriers decreased immunization rates, in addition to evidence from studies by Oria et al. 10 and Hofmann et al. 31 A relevant aspect involves some of the reasons presented for not accepting vaccination, such as the fear of adverse effects -pointed out by four of the analyzed studies. 21,22,26,30This barrier was corroborated by a literature review comprising various categories of workers 31 and by the qualitative study by Jaiyeoba et al., 35 which investigated the nursing team, attending and resident physicians, and medical students.
We highlight that adverse events can occur after influenza vaccination, but most of them are mild and self-limited.This vaccine has an excellent safety profile, being very effective in these workers, as well as in other categories and age groups, in addition to presenting significant tolerance. 3,4

THE TWO NEW DIMENSIONS OF HBM: CUES TO ACTION AND SELF-EFFICACY
The new cues to action and self-efficacy dimensions were added to the HBM in 1994 by Rosenstock et al. 13 The cues to action dimension can be defined as devices capable of promoting and initiating attitudes for the acceptance of preventive actions (the use of media campaigns, for example). 11This dimension was presented in nine articles included in this review; in two studies, no information was found 21,30 (Table 3).
Therefore, the authors who evaluated this dimension associated it with influenza immunization, that is, when cues to action were present, the immunization rates were higher.
Other studies also confirmed this association through a systematic review by Corace et al., 14 who included various categories of workers, indicating that many cues to action were able to predict influenza vaccine acceptance; the qualitative study by Hidiroglu et al., 33 with physicians, nurses, midwives, and sanitarians, presented a relationship between cues to action, such as a medical recommendation, and higher willingness to accept the H1N1 vaccine.
Self-efficacy can be defined as the belief to be sufficiently capable of overcoming difficulties inherent to the preventive attitude. 11This dimension is present in only one of the studies reviewed in this study, 24 indicating that participants who were vaccinated against influenza had higher self-efficacy levels than the unvaccinated ones.
In this regard, two studies also concluded that the self-efficacy dimension of HBM justified the adoption of other preventive measures.The first study, performed by Shewasinad Yehualashet et al. 36 with a total sample of 683 inhabitants of Ethiopia, aimed to identify predictors of adherence to preventive measures against COVID-19 and showed a significant association between self-efficacy and the level of adherence to safety measures.In the second study, Pribadi & Devy 37 investigated 58 young adult women smokers in Indonesia, aiming to analyze the correlation between the intention of quitting smoking a HBM factors, and finding as a result a correlation between willingness to quit smoking and self-efficacy.
Regarding which of the dimensions would be the most important for accepting influenza vaccination, no agreement was found between authors.Two of them agree that the perceived benefits dimension is the most relevant, 20,28 one author believes it is perceived barriers, 22 and another reports the perceived benefits for experienced workers and cues to action for unexperienced workers. 24The other articles in this review did not draw conclusions on which were the most important dimensions.
Considering the application of HBM in influenza vaccination among health care workers, the authors who expressed general conclusions on this theory stated that the results validated the applicability of the model for understanding vaccination behavior 25 ; the theory was able to predict the willingness to receive the vaccine 23 ; the levels of the dimensions among vaccinated individuals were significantly higher, except for the barriers category 24 ; almost all of the model's main constructs were significantly associated with the intention (or not) of getting vaccinated 20 ; and their results are strongly correlated with HBM principles. 28ble 3. Identification of the cues to action and self-efficacy dimensions of the Health Belief Model among the 11 studies

Study
Cues to action Self-efficacy A1 20 The professionals who intended to get vaccinated were positively motivated by the cues to action.-A2 21 --A3 22 One of the reasons mentioned for vaccination by those who received the vaccine was peer pressure.
-A4 29 An association was observed between cues to action and immunization.-A5 23 A strong significant correlation was observed between cues to action and willingness to receive the vaccine.
-A6 25 Cues to action were associated with higher chances of getting vaccinated.-A7 26 Encouragement by the physician and supervisor were factors that promoted vaccination.-A8 27 The recommendation of vaccination by the employer was a predictor of vaccine acceptance.-A9 24 Cues to action were predictors of vaccine acceptance.Vaccinated individuals had higher self-efficacy levels.

Table 1 .
Characterization of the studies included in the integrative review * Although not specified by the author, data were collected in 2015, referring to the 2014 campaign, which employed vaccines against seasonal influenza only.† Although not specified by the author, data were collected between November 2005 and January 2006, when only vaccines against seasonal influenza were employed.‡ Although not specified by the author, data were collected in 2004, when only vaccines against seasonal influenza were employed.

Table 2 .
Identification of the four main dimensions of the Health Belief Model among the 11 studies